diff --git "a/notes/Bailey and loves short practice of surgery 27 edi_7.txt" "b/notes/Bailey and loves short practice of surgery 27 edi_7.txt" new file mode 100644--- /dev/null +++ "b/notes/Bailey and loves short practice of surgery 27 edi_7.txt" @@ -0,0 +1,10319 @@ + + + +Figure 56.39 Operative appearance of a large, non-ruptured infra-renal abdominal aortic aneurysm. + + + +(a) + +narrowing or tortuosity. The usual technique is to expose both femoral arteries (under general or local anaesthetic), which allows access to the aorta. Then, under radiological control, guidewires and catheters are used to cross the aneurysm and an angiogram performed to mark the level of the renal arteries. The endovascular prosthesis (often termed a ‘stent graft’) +is usually made up of three separate parts – a main body (Fig-ure 56.41a) and two limbs, which are enclosed in separate delivery catheters (Figure 56.41b). Some types have only two pieces, a main body with ipsilateral limb attached and a sepa-rate contralateral limb. The prosthesis is made from Dacron® or PTFE, with integral metallic stents for support. The deliv-ery catheter is inserted in the aneurysm sac and the stent-graft deployed by withdrawal of the delivery system. Most systems now have hooks or barbs to anchor the prosthesis in the aortic wall and some surgeons inflate a moulding balloon catheter in the stent-graft to ensure the hooks and barbs are engaged and a good seal is obtained (Figure 56.42). Although the top edge of the fabric of the stent-graft has to be deployed below the renal arteries (infrarenal fixation), some systems have addi-tional bare metal stents at the proximal end of the main body that lie across the renal arteries to give better support and fixation (suprarenal fixation). Blood flows between the metal struts of the stent into the renal arteries. Success is dependent on a good seal between the stent-graft and the proximal and distal ‘landing zones’ in the aorta and iliac arteries. Failure to achieve a good seal results in an endoleak, that means that the aneurysm is not excluded from the circulation and may +still expand and rupture. Patients who undergo EVAR require + + + +(b) + + + + + + + + + + + + + + + + + + + + + + + + +Figure 56.40 (a) Aneurysm sac opened. Note that the posterior wall of the aorta immediately above and below the sac is not divided. A Dacron tube graft is laid in place within the sac ready for suture. (b) Graft sutured in place and vascular clamps removed. +PART 10 | VASCULAR +964 CHAPTER 56 Arterial disorders + + +(a) + + + + + + + + + + + + +(b) + + +Figure 56.43 Duplex ultrasound scan post endovascular aneurysm repair (EVAR), showing the aortic sac in cross-section and two limbs of EVAR (red ovals). There is a type II endoleak from the inferior mes-enteric artery, with blood flowing retrogradely into the aneurysm sac (arrow) + + + + +Figure 56.41 (a) Endovascular prosthesis main body, with separate limbs (b). + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 56.42 Spiral computed tomogram showing an endoluminal aortoiliac ‘stent-graft’. The metallic stent structure is clearly observed. + + +life-long follow-up and surveillance with duplex or CT scans to detect endoleak, disconnection of the components and migration of the stent-graft, all of which predispose to late +rupture (Figure 56.43). + +Ruptured abdominal aortic +aneurysm +Abdominal aortic aneurysms can rupture anteriorly into the peritoneal cavity (20%) or posterolaterally into the retroper-itoneal space (80%). Less than 50% of patients with rupture survive to reach hospital. Anterior rupture results in free bleeding into the peritoneal cavity; very few patients reach hospital alive. Posterior rupture, on the other hand, produces a retroperitoneal haematoma (Figure 56.44). Often a brief period ensues when a combination of moderate hypotension and the resistance of the retroperitoneal tissues arrests fur-ther haemorrhage and may allow transport to hospital. The patient may remain conscious but in severe pain. If no oper-ation is performed, death is virtually inevitable. Operative mortality is around 50% and the overall combined mortality (community and hospital) is around 80–90%. +Ruptured abdominal aortic aneurysm is a surgical emer-gency; it should be suspected in a patient with the triad of severe abdominal and/or back pain, hypotension and a pul-satile abdominal mass. If there is doubt about the presence of an aneurysm an ultrasound scan may help, but this cannot diagnose rupture. CT scanning should be used to establish the diagnosis and to determine whether an endovascular repair is possible. +Good venous access is needed for infusion of saline or vol-ume expanding fluids, but the systolic blood pressure should not be raised any more than is necessary to maintain con-sciousness and permit cardiac perfusion (<100 mmHg). Many surgeons now adopt a policy of permissive hypotension, where fluids are withheld if the patient is conscious (and cerebral perfusion is therefore adequate) in order to avoid provoking further uncontrolled haemorrhage. After CT scanning, the patient should be transferred immediately to an operating theatre, where a urinary catheter and arterial line are usually +inserted. If the patient appears stable, surgery may be delayed +PART 10 | VASCULAR Aneurysm 965 + + +Symptomatic abdominal aortic +aneurysm +These patients most commonly present with abdominal and/ or back pain but the aneurysm is not ruptured on CT scan. Pain may also occur in the thigh and groin because of nerve compression. Gastrointestinal, urinary and venous symptoms can also be caused by pressure from an abdominal aneurysm. About 3% of all aneurysms cause pain as a result of inflam-mation of the aneurysm itself (Figure 56.45). Finally, a few cause symptoms from distal embolisation of fragments of their intraluminal thrombus. An operation is usually indicated in patients who are otherwise reasonably fit. Pain may be a warn-ing sign of stretching of the aneurysm sac and imminent rup-ture; surgery should be performed as soon as possible (usually on the next available operating list). The operative mortality +of symptomatic aneurysms is usually higher than elective cases. + +Postoperative complications +The most common complications after open repair are car-diac (ischaemia and infarction) and respiratory (atelectasis +and lower lobe consolidation). A degree of colonic ischaemia + + + + + + + + + +Figure 56.44 The retroperitoneal haematoma of a ruptured aortic aneurysm. The aortic pulsation is palpated through the haematoma at its upper limit and fingers are insinuated on each side of the aorta. With finger control, the upper clamp is positioned and closed on the aorta. The procedure is then as for a planned case. In this illustration, the clamp is at the proximal end of the aneurysm; the haematoma has spread from the left paracolic gutter to encircle the aneurysm and the aortic bifurcation. + +until cross-matched blood is available but surgery should com-mence immediately if haemodynamic instability develops. The abdomen is usually prepared and draped with the patient awake. It is important to remember that the treatment of rup- +tured aneurysm is operation, not monitoring and resuscitation. + + + +Summary box 56.5 + +Management of ruptured abdominal aortic aneurysm ●● Early diagnosis (abdominal/back pain, pulsatile mass, shock) +●● Immediate resuscitation (oxygen, intravenous replacement therapy, central line) +●● Maintain systolic pressure, but not >100 mmHg, consider permissive hypotension +●● Urinary catheter +●● Cross-match 6 units of blood +●● Rapid transfer to the operating room + + + + + + + + + + +Figure 56.45 An inflammatory abdominal aortic aneurysm. Note the white ‘icing’ effect. Such lesions can be technically difficult to manage. +PART 10 | VASCULAR +y +966 CHAPTER 56 Arterial disorders + + + +because of lack of a collateral blood supply occurs in about 10% of patients, but fortunately this usually resolves sponta-neousl . Renal failure is an uncommon event after elective procedures but may complicate procedures undertaken for rupture. Renal failure is more likely if there is preoperative renal impairment or considerable intraoperative blood loss. Neurological complications include sexual dysfunction and spinal cord ischaemia. An aortoduodenal fistula is an uncom-mon but treatable complication of abdominal aortic replace-ment surgery. It should be suspected whenever haematemesis or melaena occurs in the months or years after operation. Prosthetic graft infection is also uncommon; it may require explantation of the original graft and replacement with an autologous deep vein (superficial femoral vein) graft limb, or removal of the original graft with oversewing of the aortic stump and limb revascularisation by insertion of an axillobi-femoral bypass. Both techniques are associated with signifi-cant risk of perioperative morbidity and mortality. +Cardiac, respiratory, renal and neurological complications are less common after endovascular repair. However, there are complications that are unique to EVAR such as endoleak, graft migration, metal strut fracture and graft limb occlusion. Life-long surveillance with duplex or CT (together with plain abdominal x-ray for strut fracture) is required to detect endoleak and migration. High-pressure endoleaks may require repeat ballooning or a proximal cuff or distal limb extension to reseal the endograft. Migration may also require extension of the graft. Overall, 10–20% of patients with EVAR will require secondary interventions to treat complications at some future date, although many of the interventions can be performed with a percutaneous approach via the femoral artery in the +angiography suite. + + +Peripheral aneurysm Popliteal aneurysm +Popliteal artery aneurysm accounts for 70% of all peripheral aneurysms classically diagnosed in males in their seventh decade of life; 50% are bilateral. Examination of the abdomi-nal aorta is indicated if a popliteal aneurysm is found because one-third are accompanied by aortic dilatation. +Popliteal aneurysms present as a swelling behind the knee or with symptoms caused by complications, such as severe ischaemia following thrombosis or distal ischaemia as a result of emboli. The diagnosis is usually confirmed with duplex scanning but assessment of the distal vessels (with CT, MR or DSA) is important prior to repair if the foot pulses are diminished or absent. An asymptomatic aneurysm exceeding 20mm in diameter should be considered for elective repair, to prevent future complications. Some surgeons would also offer elective repair for smaller diameters if the sac contains throm-bus, because of a perceived increased risk of distal embolisa-tion. All symptomatic popliteal aneurysms, including those in which single crural vessel embolisation has occurred, should be considered for repair. +Two techniques for surgical repair may be used: exclu-sion bypass and inlay repair. An exclusion bypass involves +a medial approach to the above- and below-knee popliteal + +arteries, ligation of the aneurysm and restoration of flow to the foot with a bypass graft using saphenous vein. Many sur-geons favour this approach because the anatomy is similar to that for a femoropopliteal bypass and therefore familiar. An inlay graft repair is performed through a posterior approach and has the benefits of allowing free ligation of feeding genic-ulate branches as well as aneurysmectomy in cases with neu-rovascular compression. However, the posterior approach limits exposure of the superficial femoral and crural arteries and should only be used when the popliteal aneurysm is con-fined to the popliteal fossa. +In the acute situation, the presentation is usually with a thrombosed aneurysm and an ischaemic foot; popliteal aneurysms very rarely rupture. Surgery is often unsuccess-ful because the distal vessels are thrombosed and difficult to clear. Attempts should be made with a Fogarty catheter and +intra-arterial thrombolysis. The limb loss rate is high (50%). +Femoral aneurysm +True aneurysm of the femoral artery is uncommon. Compli-cations occur in less than 3% so conservative treatment is generally indicated, but it is important to look for aneurysms elsewhere as over half are associated with abdominal or pop-liteal aneurysms. Large aneurysms should be repaired. False aneurysm of the femoral artery occurs in 2% of patients after arterial surgery at this site. Local repair may involve reanas-tomosis of the bypass in the groin under suitable antibiotic cove . However, if infection is the cause, the treatment may involve excision of the infected graft and insertion of a fur-ther bypass routed around the infected area. In the latter case, the failure rate is high, and limb loss may be unavoidable. For false aneurysms caused by femoral artery puncture mea-suring <3cm, thrombin injection under ultrasound guidance may be successful and avoids surgery. False aneurysms measur-ing >3cm are unlikely to be successfully treated by thrombin injection and require open surgical arterial repair with sutur- +r +ing of the puncture site. +Iliac aneurysm +This usually occurs in conjunction with aortic aneurysm and only rarely on its own. When occurring in isolation it is diffi-cult to diagnose clinically, so about half present already rup-tured. Open surgery usually involves an inlay graft but some +iliac aneurysms may be suitable for EVAR. + +Arteriovenous fistula +Communication between an artery and a vein (or veins) may be either a congenital malformation or the result of trauma. Arteriovenous fistulas for haemodialysis access are also cre-ated surgically. All arteriovenous communications have a structural and a physiological effect. The structural effect of arterial blood flow on the veins is characteristic; they become dilated, tortuous and thick walled (arterialised). The phys-iological effect, if the fistula is big enough, is an increase in cardiac output. In extreme circumstances this can cause left ventricular enlargement and even cardiac failure. +A pulsatile swelling may be present if the lesion is superfi- +cial. A thrill is detected on palpation and auscultation reveals +PART 10 | VASCULAR Arteritis and vasospastic conditions 967 + + + +a buzzing continuous bruit (‘machinery murmur’). Dilated veins may be seen, in which there is rapid blood flow. Pressure on the artery proximal to the fistula reduces the swelling and the thrill and bruit cease. +Duplex scan and/or angiography confirms the lesion, which is noteworthy for the speed with which venous filling +occurs. +Management +Treatment is by embolisation. Excision surgery can be advo-cated only rarely, perhaps for severe deformity or recurrent haemorrhage; the assistance of a plastic surgeon is wise. It is important to realise that ligation of a ‘feeding’ artery on its own is of no lasting value and is actually detrimental as it may +preclude treatment by embolisation. + +ARTERITIS AND VASOSPASTIC +CONDITIONS +Thromboangiitis obliterans +(Buerger’s disease) +This is characterised by occlusive disease of small- and medium-sized arteries (plantar, tibial, radial, etc.), throm-bophlebitis of the superficial or deep veins and Raynaud’s syndrome; it occurs in male smokers, usually under the age of 30 years. Often, only one or two of the three manifestations are present. Histologically, there are inflammatory changes in the walls of arteries and veins, leading to thrombosis. Treat-ment is total abstinence from smoking, which arrests, but does not reverse, the disease. Established arterial occlusions are treated as for atheromatous disease, but amputations may +eventually be required. + +Other forms of arteritis +Arteritis occurs in association with many connective tissue disorders, e.g. rheumatoid arthritis, systemic lupus erythema-tosus and polyarteritis nodosa. This is usually the province of the specialist physician, but the surgeon may be called on to carry out minor amputations. Sympathectomy has previously been used but is usually ineffective. +Temporal arteritis is a disease in which localised infil-tration with inflammatory and giant cells leads to arterial occlusion, ischaemic headache and tender, palpable, pulse-less (thrombosed) arteries in the scalp. Irreversible blindness occurs if the ophthalmic artery becomes occluded. The sur-geon may be required to perform a temporal artery biopsy, but this should not delay immediate steroid therapy to arrest and reverse the process before the ophthalmic artery is involved. The length of the biopsy should be at least 2.5 cm. +Takayasu’s disease is an arteritis that obstructs major arter-ies, particularly the large vessels coming off the aortic arch. It +usually pursues a relentless course. + + +Cystic myxomatous degeneration +This is typified by an accumulation of clear jelly (like a syno-vial ganglion) in the outer layers of a main artery, especially the popliteal artery. The lesion may narrow the vessel, caus-ing claudication. Duplex scan is the investigation of choice. Decompression, by removal of the myxomatous material, is often all that is required, but the ‘ganglion’ may recur, neces-sitating excision of part of the artery with interposition vein +graft repair. + +Raynaud’s disease +This idiopathic condition usually occurs in young women and affects the hands more than the feet. There is abnormal sensi-tivity in the arteriolar response to cold. These vessels constrict and the digits (usually the fingers) turn white and become incapable of fine movements. The capillaries then dilate and fill with slowly flowing deoxygenated blood, resulting in the digits becoming swollen and dusky. As the attack passes off, the arterioles relax, oxygenated blood returns into the dilated capillaries and the digits become red. Thus, the condition is recognised by the characteristic sequence of blanching, dusky cyanosis and red engorgement, often accompanied by pain. Superficial necrosis is very uncommon. This condition must be distinguished from Raynaud’s syndrome, which has similar features (see below). Treatment of Raynaud’s disease consists of protection from cold and avoidance of pulp and nail bed infection. Calcium antagonists, such as nifedipine, may also have a role to play and electrically heated gloves can be useful in winter. Sympathectomy has been used in the past but it is +either ineffective or its effects are short-lived. + +Raynaud’s syndrome +Although peripheral vasospasm may be noted in athero-sclerosis, thoracic outlet syndrome, carpal tunnel syndrome, etc., the term Raynaud’s syndrome is most often used for a peripheral arterial manifestation of a collagen disease, such as systemic lupus erythematosus or rheumatoid arthritis. The clinical features are as for Raynaud’ disease but they may be much more aggressive. Raynaud’s syndrome may also follow the use of vibrating tools. In this context, it is a recognised industrial disease and is known as ‘vibration white finger’. +s +Treatment is directed primarily at the underlying con-dition, although the conservative measures outlined above are often helpful. The syndrome when secondary to colla-gen disease leads frequently to necrosis of digits and multi-ple amputations. Sympathectomy yields disappointing results and is not recommended. Nifedipine, steroids and vasospastic antagonists may all have a role in treatment. Patients with +vibration white finger should avoid vibrating tools. + +Acrocyanosis +Acrocyanosis may be confused with Raynaud’s disease but it +is painless and not episodic. It tends to affect young women + + + + +Mikito Takayasu, 1860–1938, Japanese ophthalmologist, described this disease in 1908. +PART 10 | VASCULAR +968 CHAPTER 56 Arterial disorders + + + +and the mottled cyanosis of the fingers and/or toes may be +accompanied by paraesthesia and chilblains. + +Cervical sympathectomy +Open cervical sympathectomy was previously performed for vasospastic conditions affecting the hands and to treat pal-mar (sometimes axillary) hyperhidrosis. The operation is now obsolete, having been replaced by endoscopic transthoracic sympathectomy. Furthermore, it has been increasingly rec-ognised that the vasospastic conditions do not respond to this form of treatment, rendering the endoscopic intervention a +therapy that is suitable solely for hyperhidrosis. + +Lumbar sympathectomy +Lumbar sympathectomy has been used to treat chronic lower limb ischaemia in the past. Lumbar sympathectomy +by open operation has, however, been obsolete for several + +years and even chemical sympathectomy, its minimally inva-sive equivalent, can now be regarded as outdated. Chemical sympathectomy requires the injection of small quantities of a sclerosant into the lumbar sympathetic chain under radio- +graphic control. + +FURTHER READING +Ascher E, Veith FJ, Gloviczki P (eds). Haimovici’ vascular surgery, 6th edn. Oxford: Wiley-Blackwell Publishing, 2012. +s +Bhattacharya V, Stansby G (eds). Postgraduate vascular surgery: the can-didates’ guide to the FRCS. Cambridge: Cambridge University Press, 2011. +Cronenwett JL, Johnstone WK (eds). Rutherford’s vascular surgery, 8th edn. Philadelphia: WB Saunders, 2014. +England T, Nasim A (eds). ABC of arterial and venous disease, 3rd edn. London: BMJ Books, 2014. +Moore WS (ed). Vascular and endovascular surgery: a comprehensive re-view, 8th edn. Philadelphia: WB Saunders, 2013. +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love +57 +Chapter + + + + + +Venous disorders + + + + + + +Learning objectives +To understand: +• Venous anatomy and the physiology of venous return +• The pathophysiology of venous hypertension + + + +• The clinical significance and management of superficial venous reflux +• The management of venous ulceration • Venous thromboembolism + + + + + + +INTRODUCTION +Up to 40% of the adult population in resource-rich coun-tries have diseases of the veins of the leg. This extraordinary prevalence along with the associated impairment in health-related quality of life make it a very important area of surgical practice. Surgical intervention has been revolutionised by the development of endovenous techniques, and level 1 evidence has demonstrated that treatment can be associated with very high clinical- and cost-effectiveness. Despite the consider-able importance placed on lower limb function during the management of orthopaedic and arterial diseases, venous dis-eases are often forgotten or dismissed as cosmetic practice. An understanding of the nature and management of venous disease is critical to address this imbalance and improve the +quality of patients’ lives. + +THE ANATOMY OF THE VENOUS +SYSTEM OF THE LOWER LIMB +The venous system of the lower limb can be divided anatom-ically into the superficial venous system, which is located within the superficial tissues and the deep venous system, beneath the deep fascia of the leg, accompanying the arterial tree. The superficial veins drain into the deep system, either at junctions or via fascial perforating veins, and the deep veins then return blood to the right atrium of the heart. Venous anatomy is characteristically variable. The terminology used below is consistent with international consensus. +The deep veins of the lower limb (Figure 57.1a) include three pairs of venae commitantes, which accompany the three crural arteries (anterior and posterior tibial and pero-neal arteries). These six veins intercommunicate and come together in the popliteal fossa to form the popliteal vein, +which also receives the soleal and gastrocnemius veins. The + + +popliteal vein passes up through the adductor hiatus to enter the subsartorial canal as the femoral vein, which receives the deep (profunda) femoral vein (or veins) in the femoral trian-gle before passing behind the inguinal ligament to become the external iliac vein. The internal iliac vein combines with the external iliac vein in the pelvis to form the common iliac vein. The left common iliac vein passes behind the right com-mon iliac artery to join the right common iliac vein on the right side of the abdominal aorta to form the inferior vena cava, which goes on to the right atrium. +Far more anatomical variations exist within the superfi-cial veins of the lower limb, but there are almost always two trunks or axes, the great and small saphenous veins (Figures 57.1b, c). These lie superficial to the fascia lata (deep fascia) but deep to the saphenous fascia, in the saphenous ‘envelope’. As the sole of the foot is often placed under significant pressure, the majority of the venous drainage of the foot is into the dorsal venous arch, running in the subcutaneous tis-sues over the metatarsal heads. The medial end of this arch drains into the first axis: the great saphenous vein (GSV). This is the longest vein in the body and the most frequently affected by superficial incompetence. The GSV passes anter-ior to the medial malleolus and ascends the leg accompanied by the saphenous nerve in the superficial tissues medial to the tibia, looping posteriorly at the level of the medial condyle of the femur and continuing in the medial thigh. In the groin, it unites with tributaries corresponding to the arterial branches of the common femoral artery, before piercing the cribriform fascia covering the saphenous opening (approximately 2.5cm below and lateral to the pubic tubercle, but often somewhat higher) and terminates by draining into the common femoral vein (CFV) at the saphenofemoral junction (SFJ). Through-out its course the GSV unites variably with other superficial tributaries. The anterior (accessory) saphenous vein (ASV) +is one of the most common. This is often seen originating +PART 10 | VASCULAR +970 CHAPTER 57 Venous disorders + + + + +Common femoral vein + + + +Saphenofemoral junction + +Saphenofemoral junction + + +Great saphenous vein + + + +Supercial femoral vein + + +Profunda femoral vein + + +Anterior (accessory) saphenous vein + + +Posteromedial thigh tributary + + +Popliteal vein + + + + +Anterior tibial veins (usually paired) + + +Posterior tibial veins (usually paired) + +Peroneal veins (usually paired) + + +Anterior tributary of the leg + + +(a) + + +Figure 57.1 (a) Anatomy of the deep veins of the lower limb; (b) anatomy of the superficial veins of the lower limb (great saphenous axis); (c) anatomy of the superficial veins of the lower limb (small saphenous axis). + +around the lateral border of the knee, although it sometimes originates as low as the lateral end of the dorsal venous arch. Occasionally, this vein may also course up the medial aspect of the thigh, anterolateral to the GSV following its course. In this instance, its origin is typically a confluence of small tributaries around the knee. There is usually an in-line GSV axis passing uninterrupted from the foot (in some cases this may be hypoplastic), but this pattern of ASV is commonly mistaken for the GSV itself (some surgeons will call this a duplex GSV; a true duplex GSV is rare). The ASV may drain into the GSV in the thigh, but is typically at or near the junc-tion itself. +The small saphenous vein (SSV) originates from the lateral side of the dorsal venous arch and accompanies the sural nerve as it passes posterior to the lateral malleolus, then upwards in the posterior midline of the leg. In the proximal calf it is usually found sitting in the groove between the two muscular heads of gastrocnemius. Its termination commonly occurs by piercing the fascia of the popliteal fossa to drain into the popliteal vein at the saphenopopliteal junction (SPJ). However, this junction is highly variable and the vein may terminate as low as the mid-calf. The SSV may extend cranially beyond the SPJ, in which case it is known as either a cranial extension of the SSV, which terminates by piercing the fascia in the posterior thigh to drain into the deep system, or the Giacomini vein, which communicates with the GSV system occasionally joining the GSV at or about the SFJ. In some cases, the SSV does not terminate at or below the popli-teal fossa at all, but continues on as described above. +In the calf and thigh there are a number of valved per-forating (communicating) veins that join the superficial to +the deep veins at inconstant sites and which allow blood + + +(b) + + + + + + +Cranial extension of small saphenous (Giacomini) vein + +Saphenopopliteal junction +Small saphenous vein + + + + + +(c) + + +to flow from the superficial to the deep venous system. The most important of these are the direct perforating veins of the medial and lateral calf and the communicating veins around +the knee and in the mid-thigh. + +VENOUS PATHOPHYSIOLOGY +The purpose of the venous system is primarily to return blood back to the heart so that it can be delivered into the pulmo-nary circulation. The venous system contains approximately 60% of the total blood volume, with an average pressure of around 5–10mmHg. Mechanical factors, alongside the +autonomic nervous and endocrine systems, control the rate + + + +Carlo Giacomini, 1840–1898, anatomist, Turin, Ital , on his death left his skeleton to the Anatomical Museum in Turin. +y +PART 10 | VASCULAR Venous pathophysiology 971 + + + +(a) + + + + + + + + +(b) + + +100 +90 +80 Standing +Foot vein pressure (mmHg) +70 60 50 40 30 20 +10 Walking +0 +0 10 + + + + + + + + + + + + + + +Post-thrombotic limb + + + + + +Normal limb + +Standing + +20 30 40 +Time (seconds) + +walking. The foot pump also ejects blood from the plantar veins during walking. As the calf muscles contract, the veins are com-pressed and the valves only allow blood to pass in the direction of the heart. The pressure within the calf compartment rises to 200–300mmHg during muscle contraction. Rapid blood flow in the deep veins at junctions and perforators draws blood from the superficial veins, driving this up the deep veins also. During muscle relaxation, the pressure falls and further blood from the superficial veins enters the deep vein. Each time this occurs the pressure falls in the superficial venous compartment until a threshold is reached, when the venous inflow keeps pace with ejection from the deep veins. This is normally around 30mmHg, a fall of approximately two-thirds of the resting venous pressure. The net reduction in the pressure of the superficial system is dependent on the presence of a pressure gradient between the leg and the thorax and a patent and compliant venous system containing competent valves. An absence of one or more of these results in venous hypertension, which leads to further vein wall damage including loss of compliance, thickening, dilatation and valvular dysfunction. This venous damage goes on to reduce the function of the affected veins, worsening the venous hyper-tension in a vicious cycle. When exposed to high venous and capillary pressures chronically, the soft tissues of the leg will be damaged, causing a spectrum of damage that become irreversible. +The causes of venous hypertension are listed in Table 57.1. + +Figure 57.2 (a, b) Effect of exercise on the superficial venous pres-sure in health and disease. + + + +at which blood is delivered to the right atrium. Through its effects upon myocardial contractility via the Starling mecha-nism, venous return is one of the factors responsible for deter-mining cardiac output. +Blood enters the lower limb through the femoral arteries before passing through arterioles into the capillaries, which have a pressure of about 32mmHg at their arterial ends. This pressure is reduced along the course of the capillaries and is approximately 12mmHg at the venular end of the capillary. The pressure continues to fall in the main veins, and is as low as 5mmHg at the upper end of the vena cava where it enters the right atrium. +The venous pressure in a foot vein on standing is equivalent to the height of a column of blood extending from the heart to the foot, e.g. approximately 100mmHg (Figure 57.2). To enable blood to be returned against gravity in the standing position a pressure gradient must exist between the veins in the leg and the chest. This gradient is created in two ways. Firstly, the increase in thoracic volume during inspiration decreases intrathoracic pres-sure. Secondly, the pressure in the veins of the leg is increased by compression by the surrounding muscles (the ‘calf muscle pump’) and to a lesser extent the tone of the venous wall. The deep veins of the calf are capacious and are joined by blind-ending sacks called the soleal sinusoids, which force blood into the pop- +liteal and crural veins during calf muscle pump contraction, e.g. + +TABLE 57.1 Factors causing venous hypertension. +●● Pressure gradient dysfunction: +Increased abdominal or thoracic pressure: ■● COPD +■● Pregnancy ■● Obesity +■● Large tumour ■● Constipation +Decreased calf muscle pump function: ■● Immobility +■● Ankle joint fusion ■● Paralysis +●● Dysfunction of the venous system: Venous structural deficit: +■● Valvular agenesis +■● Valvular incompetence ■● Venous dilatation +■● Venous tortuosity +■● Loss of vein wall compliance ■● Loss of venous tone +■● Arteriovenous fistula Venous occlusion: +■● Agenesis +■● Thrombosis +■● Iatrogenic/trauma Venous compression: +■● May–Thurner syndrome ■● Pelvic/abdominal tumour +■● Pelvic/abdominal radiotherapy +COPD, chronic obstructive pulmonary disease. + + + +Ernest Henry Starling, 1866–1927, physiologist, University College, London, UK. +Rudolf Virchow, 1821–1902, pathologist Charite Hospital, Berlin, Germany, was the first to be credited with describing iliac vein compression. It was not until 1957 that May and Thurner (Innsbruck, Austria) clearly described compression of the left common iliac vein by the right common iliac artery. +PART 10 | VASCULAR +972 CHAPTER 57 Venous disorders + + +(a) (b) (c) + + + + + + + + + + + + + + + + + + + + + +Figure 57.3 Varicose veins: (a) left leg varicose veins in the distribution of an incompetent great saphenous vein (marked for intervention); (b) right leg varicose veins in the distribution of the small saphenous system with a recent episode of phlebitis; (c) varicose veins in distribution of an isolated incompetent anterior accessory saphenous vein with associated gaiter area skin changes. + + +The majority of patients with venous disease have a prob-lem primarily with the vein wall structure and in most this is confined to the superficial veins. Little is known about the mechanism of initiation of the changes in the vein wall. These changes are complex, but are typified by valvular failure allowing retrograde flow within the vein with gravity (venous incompetence). It is no longer thought that venous incompe-tence is caused by a primary mechanical valvular failure. +The vein wall changes include inflammatory cell infiltra-tion and activation, dysfunctional smooth muscle cell prolif-eration, collagen deposition, decreased elastin content and increased matrix metalloproteinases. These effects typically lead to loss of compliance, dilatation, elongation (causing tor-tuosity) and secondary valvular dysfunction. This process can be initiated anywhere in the venous tree. Secondary varicose veins may develop in patients with post-thrombotic limbs and in patients with congenital abnormalities such as the Klippel– Trenaunay syndrome or multiple arteriovenous fistulae. +The extent and number of incompetent veins governs the extent of the venous hypertension and correlates to the sever-ity of the soft tissue complications seen. Importantly however, neither the reflux burden nor the presence of skin changes, short of ulceration, correlate with the presence or degree of +symptoms. + +CLINICAL FEATURES OF VENOUS HYPERTENSION OF +THE LEG +The following clinical features are commonly seen: +● Varicose vein: subcutaneous dilated vein 3mm in diam-eter or larger. They are frequently elongated and tortu-ous, with intermittent ‘blowouts’, but are defined by the presence of reflux and may be straight and uniform tubes morphologically (Figure 57.3). +● Telangiectasia (thread veins, spider veins, and hyphen webs): represent tiny intradermal venules less than 1mm in diameter (Figure 57.4). +● Reticular vein: small dilated ‘bluish’ subdermal vein 1–2.9mm in diameter, usually tortuous, can be difficult to distinguish this from a normal subdermal vein in someone with white thin transparent skin. +● Saphena varix (Figure 57.5) is a (usually painless) groin swelling apparent on standing. +● Corona phlebectatica (malleolar flare): a fan-shaped pat-tern of telangiectasia on the ankle or foot. It is thought to be an early sign of advanced venous disease. +● Oedema: increased volume of fluid in the skin and soft tissues of the leg. Commonly starts distally and moves more proximally with increasing venous dysfunction. Classically this is ‘pitting oedema’, with firm digital +pressure leaving an indentation in the soft tissues. + + +A gaiter is a leather or cloth covering for the lower leg and ankle. The name is derived from the French ‘guetre’ for the same piece of clothing. Maurice Klippel, 1858–1942, neurologist, La Salpêtrière, Paris, France. +Paul Trenaunay, b.1875, French neurologist. Klippel and Trenaunay described this condition in a joint paper in 1900. +PART 10 | VASCULAR Clinical features of venous hypertension of the leg 973 + + + + + + + + + + + + + + +Figure 57.6 Advanced skin changes – lipodermatosclerosis, eczema and atrophie blanche. + + + +Figure 57.4 Telangiectasia and reticular veins. + + + + + + + + +Figure 57.7 Pigmentation (haemosiderosis) and mild eczema. + + + + + + + + +Figure 57.8 Severe eczema. Figure 57.5 A saphena varix. + + + +● Eczema: an erythematous dermatitis, often appears minor, although it may be associated with significant itching and discomfort. In extreme cases it may progress to blistering and weeping (Figures 57.6–57.8). +● Pigmentation (haemosiderosis): a brownish discolouration of the skin, usually permanent. It is usually seen around the ankle, but is also seen in proximity to varicose veins and incompetent perforators (Figures 57.7 and 57.9) +● Lipodermatosclerosis (LDS): chronic inflammation and fibrosis of the skin and subcutaneous tissues, resulting in a tight, contracted, ‘woody’ leg on examination. It occa-sionally results in significant contractures of the Achilles tendon. This is a sign of severe chronic venous disease (Figures 57.6 and 57.9). +● Atrophie blanche: localised areas of atrophic, white skin, often surrounded by telangiectasia and pigmentation. Some authors distinguish this from the white scarring left by ulceration, others do not. Either way, this is a sign of +severe chronic venous disease (Figure 57.6). + + + + + + + + + + + + +Figure 57.9 Haemosiderosis and mild lipodermatosclerosis of the calf skin. + +● Venous ulcer: full-thickness skin loss, usually around the ankle, which fails to heal spontaneously and is propagated by continuing venous hypertension and the changes asso- +ciated with chronic venous disease (Figure 57.10). +PART 10 | VASCULAR +974 CHAPTER 57 Venous disorders + + +For pathophysiological classification: +● Pr: reflux; +● Po: obstruction; +● Pr,o: reflux and obstruction; +● Pn: no venous pathophysiology identifiable. + + +VARICOSE VEINS +In clinical practice, patients are normally categorised as having ‘varicose veins’ or ‘venous ulcers’. Cases of varicose veins may be uncomplicated or complicated. Complications include superficial thrombophlebitis, bleeding or any of the skin changes listed above. Uncomplicated varicose veins may +be asymptomatic or symptomatic. + + + + + + + + + + +Figure 57.10 Venous ulcer. + + +Classification system +The descriptive CEAP (Clinical-aEtiology-Anatomy-Patho-physiology) classification for chronic venous disorders is widely utilised. +For clinical classification: +● C0: no signs of venous disease; +● C1: telangectasia or reticular veins; ● C2: varicose veins; +● C3: oedema; +● C4a: pigmentation or eczema (some include malleolar flare in this category); +● C4b: lipodermatosclerosis or atrophie blanche; ● C5: healed venous ulcer; +● C6: active venous ulcer. +Each clinical class is further characterised depending upon whether the patient is symptomatic (S) or asymptomatic (A), e.g. C2S. +For aetiological classification: +● Ec: congenital; ● Ep: primary; +● Es: secondary (post-thrombotic); +● En: no venous cause identified. For anatomical classification: +● As: superficial veins; ● Ap: perforator veins; ● Ad: deep veins; +● An: no venous location identified. + +Epidemiology +The adult prevalence of visible varicose veins is between 30% +and 50%. Factors affecting prevalence include: +● Gender: the vast majority of studies report a higher preva-lence in women than men, though community prevalence may differ. +● Age: the prevalence of varicose veins increases with age. In the Edinburgh Vein stud , the prevalence of trunk vari-cosities in the age groups 18–24 years, 25–34 years, 35–44 years, 45–57 years and 55–64 years was 11.5%, 14.6%, 28.8%, 41.9% and 55.7%, respectively. +y +● Ethnicity: does seem to influence the prevalence of vari-cose veins. +● Body mass and height: increasing body mass index and height may be associated with a higher prevalence of vari-cose veins. +● Pregnancy: increases the risk of varicose veins. +● Family history: evidence supports familial susceptibility to varicose veins. +● Occupation and lifestyle factors: there is inconclusive evi-dence regarding increased prevalence of varicose veins in smokers, patients who suffer constipation and occupations +that involve prolonged standing. + + +Symptoms +Varicose veins frequently cause symptoms. Patients describe aching, heaviness, throbbing, burning or bursting over affected areas and sometimes the whole limb. Such symptoms typi-cally increase throughout the day or with prolonged standing, and are relieved by elevation or compression hosiery. Itching is also commonly described, though is more frequent in the presence of complications as is swelling of the ankle. Venous symptoms in the absence of complications can be vague and it may be difficult to ascertain from history alone whether they are truly venous in origin and, therefore, whether treatment will help. A trial of compression hosiery can help as venous symptoms should show some beneficial improvement. +Symptoms can be very severe and interfere with a patient’s +daily activities such as work, recreation and caring for +PART 10 | VASCULAR Varicose veins 975 + + + +children and adults. Such symptoms are independent of the degree of venous incompetence or the presence of compli-cations, including skin changes short of ulceration. Studies have also shown that symptoms are associated with a signif-icant deficit in health-related quality of life, and significant improvements are seen with treatment to remove or ablate the refluxing veins. The maximal benefit is seen in those with uncomplicated symptomatic varicose veins, as skin changes and a proportion of the associated morbidity are frequently irreversible. +Telangiectasia (not associated with malleolar flare) and reticular veins occur very commonly in the absence of sig-nificant reflux or obstruction and in the vast majority do not cause any physical symptoms, though cosmetic treatment is +commonly sought. + + +Signs +The presence of tortuous dilated subcutaneous veins is usually clinically obvious. These are confined to the GSV and SSV systems in approximately 60% and 20% of cases, respectively. The distribution of varicosities may indicate which superfi-cial axis is defective; medial thigh and calf varicosities suggest GSV incompetence (Figure 57.3a), posterolateral calf vari-cosities are suggestive of SSV incompetence (Figure 57.3b), whereas anterolateral thigh and calf varicosities may indi-cate isolated incompetence of the ASV (Figure 57.3c). Any of the clinical features above may be present. Large dilated veins around the SFJ may present as a (usually painless) lump, emergent when standing and disappearing when recumbent. This is a saphena varix (Figure 57.5). Gentle palpation over the varix during coughing may elicit a thrill, though it may be +mistaken for a groin hernia. + + +Investigation +Tourniquet tests and the use of hand-held Doppler have now largely been abandoned. There is good evidence to support the policy of duplex ultrasound scanning for all patients with varicose veins prior to any intervention. The best clinical results come from clinicians who are personally very skilled in the use of duplex ultrasound and use it to design a bespoke treatment for each individual patient, based upon their unique anatomy. +A high-frequency linear array transducer of 7.5–13 MHz is appropriate for the majority of lower limbs in order to obtain good quality images. The B-mode settings (depth, focal zone, overall gain and dynamic gain) should be optimised to ensure the area of interest is in the centre and occupies the majority of the image, and that the lumen of the vein appears as a dark void in the subcutaneous and deep tissues. The pulsed wave spectral or colour Doppler settings should be optimised for the low-flow velocities encountered within veins. It is conven- +tional to use blue to represent antegrade venous flow towards + +the heart and red for the reverse. Visible venous flow can be augmented by a calf squeeze. +The aim of the duplex scan in a patient with varicose +veins is to establish: +● The presence of reflux in the deep and superficial venous system. +● The exact distribution and extent of reflux in the super-ficial venous system including affected junctions and per-forators. +● The presence of obstruction in the deep venous system. ● The suitability of the incompetent superficial veins for +the different treatments available (based upon diameter, extent, tortuosity, saphena varix). +● The presence of thrombus within the superficial veins. +● An indication of a pelvic source of reflux or obstruction. +In order to standardise measurements of venous diameter and reflux, it is recommended that examination of the super-ficial veins is performed with the patient standing. Superfi-cial or crural vein reflux is defined as retrograde flow in the reverse direction to physiological flow lasting for 0.5 seconds or more. The proximal deep veins require a duration of 1 sec-ond or more to be classified as incompetent. Reflux may be elicited by release of a calf or foot squeeze for proximal or calf varicosities, respectively, manual compression over varicosity clusters, pneumatic calf cuff deflation, active foot dorsiflexion and relaxation or the Vasalva manoeuvre. +The patient should stand facing towards the examiner with the leg rotated outwards, heel on the ground and weight +on the opposite limb (Figure 57.11). The use of a platform, + + + + + + + + + + + + + + + + + + + + + + + +Figure 57.11 Patient position for venous duplex examination of the great saphenous system. + + + +Christian Johann Doppler, 1803–1853, Professor of Experimental Physics, Vienna, Austria, enunciated the ‘Doppler Principle’ in 1842. +PART 10 | VASCULAR +976 CHAPTER 57 Venous disorders + + + + + + + + + + + + + + + + + +Figure 57.12 ‘Micky Mouse’ transverse B-mode image of the right common femoral vein (CFV) and artery (CFA), great saphenous vein (GSV) and saphenofemoral junction. + + + + + + + + + + + + + +Figure 57.13 ‘Saphenous eye’ transverse B-mode view of the great saphenous vein in fascial compartments of the thigh. The fascial line above the vein is the saphenous fascia. A true great or small saphe-nous vein will not cross this line, although the fascia may become discontinuous around the knee. The line deep to the vein is the fascia lata, with the muscle beneath. + +ideally with a handle or support bar for the patient and a stool that can drop to a low height, will improve the ergo-nomic comfort of both the sonographer and the patient. The scan should commence in the groin, using a transverse view to identify the GSV and CFV lying medial to the common femoral artery (the ‘Micky Mouse’ sign, Figure 57.12). SFJ competence is assessed in the transverse view and potential destinations for reflux, including the GSV, the ASV and other major thigh tributaries superficial to the saphenous fas-cia, are noted. Any indication of a pelvic source of reflux sug-gests the need for more proximal imaging. The full length of the GSV within its fascial compartment should be examined (Figure 57.13), and its diameter measured if required. The groin is next examined for reflux or obstruction in the CFV, superficial femoral vein, and SFJ using spectral and/or colour Doppler (Figure 57.14). +A loss of phasic flow with respiration in the CFV suggests upstream obstruction and the need for proximal imaging. The presence and competence of thigh and calf perforators should be noted and the crural veins examined for reflux or obstruc-tion. For examination of the SSV and posterior thigh exten- +sion of the SSV (Giacomini vein), the patient is positioned + +Figure 57.14 Spectral Doppler trace of the saphenofemoral junc-tion showing antegrade and retrograde flow. The downward spike on the trace is the antegrade augmented flow and this is followed by approximately 4 seconds of retrograde flow. + + + + + + + + + + + + + + + + + +Figure 57.15 Varicogram + +facing away, knee slightly flexed, heel on the ground and the weight taken on the opposite leg. If the SPJ is incompetent, the level of the SPJ in relation to the knee crease and whether the SSV joins the popliteal vein posteriorly, medially or lat-erally is noted if open surgical ligation is to be entertained. In the transverse view, the SSV vein is followed distally, check-ing its competence and diameter in the proximal, mid and distal calf. Finally, the patency and competence of the popli-teal vein is assessed. +Pelvic and iliac veins may be investigated using trans-abdominal or transvaginal duplex. Very occasionally inves-tigations other than duplex are required, and these may be non-invasive, such as MR venography, or invasive such as contrast venography or intravenous ultrasound (IVUS). The +use of varicography has become historical (Figure 57.15). + +Management +Many patients with asymptomatic varicose veins do not progress to develop complications, although a significant pro-portion do, and little is known about whether treating such +patients prevents the development of future complications. +PART 10 | VASCULAR Varicose veins 977 + + + +There is clear evidence, however, that those with symptoms and/or complications see a significant quality of life benefit from treatment to remove or ablate refluxing superficial veins. When interventional treatment is planned there are con-siderable variations in practice and treatment strategies. A detailed description of the nuances, merits and criticisms of the various options is beyond the scope of this chapter; how-ever, a description of the basic treatment modalities available is presented below. An experienced surgeon will have his/her own preferred methods, but will frequently employ several or all methods in chosen circumstances, not infrequently in the +same patient. + +Compression +Compression hosiery relies on graduated external pressure to improve deep venous return and reduce venous pressures. It may be knee length or thigh length; there is no evidence which length of stocking is more effective and hence below-knee stockings are usually prescribed as they are easier to don and have much better patient acceptance. Compression hosiery are classified according to the pressure they exert: the British classification class 1 stockings exert pressure of 14–17mmHg, class 2 exert 18–24mmHg and class 3 exert 25–35mmHg. +Compression hosiery significantly improves varicose vein symptoms but is not popular with patients, with compliance rates and long-term tolerance being universally poor. There is no evidence to suggest that compression hosiery prevents the occurrence or progression of varicose veins. Further-more, incorrect application of compression hosiery can have serious consequences (pressure necrosis, tourniquet effects); thus assessment, prescription and application of compres-sion hosiery should be limited to those with the appropriate skills and training. There are level 1 trial data to demonstrate that interventional treatment offers superior improvements in quality of life and is cost-effective. Compression is there-fore to be regarded as an adjunct to assessment or treatment, unless by patient choice. + +Endothermal ablation +Endothermal ablation technologies replaced surgical liga-tion and stripping as the gold standard treatment once ran-domised trials demonstrated that they were marginally safer, have extremely high technical efficacy, offer superior quality of life post procedure (with a rapid recovery) and equiva-lent improvements in quality of life in the longer term. The techniques are cost effective as they can be performed as an outpatient under local anaesthetic. The basic concept is that a treatment device is inserted into the incompetent axial vein percutaneously. The vein is surrounded by tumescent local anaesthetic solution. This compresses the vein onto the treatment device, emptying it of blood. It also hydro-dissects tissues such as nerves away from the zone of injury. Finally, it acts as a heat sink, mopping up excess thermal energy to prevent remote damage. The treatment device then produces thermal energy that destroys the structure of the vein, result-ing in permanent occlusion. Two broad technologies exist: laser and radiofrequency ablation. + + +LASER ABLATION +Endovenous laser ablation (EVLA) utilises a small flexible glass fibre that is inserted into the vein. Laser energy (typically at a wavelength of 1470nm) is transmitted down the fibre and is absorbed at the point of treatment at the end of the fibre. Absorption of this radiation results in a vigorous production of thermal energy. The tip of the fibre may be bare, focus-ing the energy in a very small area; divergent forward firing, spreading the energy over a larger area; or divergent side or radial firing. It is postulated that the latter two designs allow a more even distribution of energy, reducing vein wall perfo-rations that are thought to be associated with postprocedural pain and bruising. There is no clear evidence to support one design over another. This procedure is very good for treat-ment of any vein that will allow the passage of a guidewire. No technique has reported a higher technical efficacy rate. +The procedure begins with ultrasound-guided marking of the truncal vein to be treated and the site of proposed cannulation. The varicosities are also marked at this stage if concomitant treatment (phlebectomy or foam sclerother-apy) is to be undertaken. The patient is then positioned on the procedure couch in the reverse Trendelenberg position. For the GSV, the patient is supine with the hip of the leg to be treated externally rotated and slightly flexed. A pillow under the contralateral hip/lower back may improve patient comfort. For the SSV the patient is positioned in the prone position. The vein is then cannulated percutaneously under ultrasound guidance, at the lowest point of reflux. Some devices allow passage of the fibre directly through a short sheath, while others use a wire first, allowing passage of a catheter that then carries the laser fibre. The former is slightly faster with fewer steps, the latter allows greater success with more tortuous veins. Accurate positioning of the fibre tip with ultrasound is crucial (Figure 57.16), but the exact loca-tion is controversial, with some surgeon positioning the tip several centimetres distal to the junction and others aiming for a flush occlusion. Proponents of the former cite that this strategy protects the deep vein from inadvertent damage and/ or thrombosis. Proponents of the latter argue that neoreflux in junctional tributaries is a common pattern of recurrence + + + + + + + + + + + + + + + + +Figure 57.16 Endovenous laser ablation; B-mode image of catheter tip positioning at the saphenofemoral junction. +PART 10 | VASCULAR +978 CHAPTER 57 Venous disorders + + + +and that in expert hands the rate of deep vein injury is no different and the thrombosis rate may be lower (presumably as there is minimal patent stump in which to form throm-bus). Following the administration of perivenous tumescent anaesthesia (Figure 57.17), the ablation can be performed. Practice varies as to the power of the laser and the withdrawal speed, but commonly an energy delivery of around 60 J/cm is used to achieve a durable closure. There is no clear evidence to guide the optimal power and pullback speed. Following treatment compression is applied, but there is no consensus over the method, degree or duration, and this is true of post- +procedural compression with all techniques. + +This is then maintained for a treatment cycle of 20 seconds. The coil is then withdrawn for a set length and another treatment cycle is commenced. Coils of 3cm and 7cm are produced, with the latter increasing the speed of treatment, while still being suitable for most anatomies. +There have been a range of studies comparing EVLA and RFA. The evidence is generally equivocal, with both treat-ments having relative advantages and disadvantages; choice often comes down to personal preference. Both are excel-lent treatment options and can be applied successfully to the majority of patients. +Advantages in favour of EVLA over RFA include: + + + + +RADIOFREQUENCY ABLATION +Radiofrequency ablation (RFA) uses the same treatment principles, but an electromagnetic current is used to create the thermal energy. A range of different devices have been created but the most popular, which has the most supportive evidence, is the ClosureFast™ device (Medtronic) (Figure 57.18). This device has a wire coil on the end of a treatment catheter. The generator passes an electrical current through +the coil until the surrounding temperature reaches 120oC. + +● EVLA can ablate any vein that can take a guidewire suc-cessfully, with a very low rate of recanalisation. While the RFA catheter can be advanced over a 0.014 guidewire, in practice the relatively inflexible tip may not advance through very tortuous veins. Recanalisation rates may be higher for larger-diameter veins as the energy delivery and penetration are limited. +● A standard EVLA fibre may be used to treat perforators, whereas a specific additional device is required for RFA, +increasing costs. + + +(a) + + + + + + + + + + + + + + +(b) + + + + + + + + + + + + + + + + +Figure 57.17 (a) Ultrasound-guided infiltration of perivenous tumescent anaesthetic via a long spinal needle. The anaesthetic solu-tion is infiltrated using an electronic foot operated pump; (b) ultra-sound image of perivenous ‘halo’ of anaesthetic solution around the vein and catheter in transverse section. + + +Figure 57.18 Radiofrequency ablation with ClosureFast™ introduc-ing the treatment catheter through a sheath. The distal 7cm of this device comprises a metal coil. +PART 10 | VASCULAR Varicose veins 979 + + + +● Laser fibres are relatively inexpensive to produce, reduc-ing the procedural costs. +● In a catheter-based system, the catheter can be used to deliver targeted foam sclerotherapy, for instance to areas of neovascularisation, prior to the actual laser fibre being +inserted. +Advantages in favour of RFA over EVLA include: +● RFA has a standardised treatment protocol that is auto-mated, minimising the uncertainty and learning curve for the ablation portion of the procedure and the possibility of a novice making a mistake with the energy delivery. +● RFA does not require a continuous pullback, again reduc-ing the learning curve. This also frees the surgeon’s focus allowing better communication with the patient and indeed, with care, concurrent treatment, e.g. infiltrating local anaesthetic into the tributaries and performing phle-bectomy, reducing procedural times. +● RFA does not require laser safety precautions, reducing the administrative burden associated with setting up a service and location and allowing it to be performed in a wider range of settings with minimal adaptations. +● RFA may be associated with a marginal reduction in pain and bruising, athough this has not been shown to impact +upon periprocedural quality of life or recovery. +As endothermal ablation treats only junctional and trun-cal incompetence, debate exists regarding the management of varicosities. These can be managed concomitantly or sequen-tially by either phlebectomy or sclerotherapy. Concomitant phlebectomy (Figure 57.19) results in a more rapid improve-ment in disease-specific quality of life, and allows the vast +majority of patients to complete treatment in a single visit. +Non-endothermal, non-tumescent ablation +Endothermal ablation was a large step forwards in the man-agement of superficial incompetence; however, all techniques +require the injection of tumescent local anaesthetic solution + + + + + + + + + + + + + + + + + +Figure 57.19 Phlebectomy performed under tumescent anaesthesia following endothermal ablation. + +and this can be uncomfortable for the patient. Other tech- +niques that avoid injection are being developed. + +ULTRASOUND-GUIDED FOAM SCLEROTHERAPY +Sclerotherapy is the original non-endothermal, non-tumes-cent technique and has been performed for over 100 years. It involves the injection of a sclerosing agent directly into the superficial veins. The most commonly used is sodium tet-radecyl sulphate. The direct contact with detergent causes cellular death and initiates an inflammatory response, aiming to result in thrombosis, fibrosis and obliteration (sclerosis). Blood deactivates the action of the sclerosing agent and the doses administered need to be limited to avoid adverse effects, causing a trade-off between poor efficacy and safety. This led to the development of ultrasound-guided foam sclerotherapy (UGFS). The use of foam increases the effective volume of the agent, maximising endothelial contact and displacing any blood that deactivates it. +The procedure commences with the patient standing, and the sites of venous cannulation are selected and marked using ultrasound. With the patient supine, the major venous trunks and superficial varicosities to be treated are then all cannulated using ultrasound guidance prior to any injection (Figure 57.20). Once all injection sites are cannulated the foam can be prepared. The most widely used method is that of Tessari, which utilises two syringes connected using a three-way tap. A 1:3 or 1:4 ratio mixture of sclerosant and air is drawn into one syringe, and is then oscillated vigorously between the two syringes about 10 or 20 times (Figure 57.21). The foam produced in this way is stable for about 2 minutes so it should be injected as soon as it has been made. The leg is then elevated to empty the veins of blood, and injection of foam commences first with superficial varicosities and ends with injection of the GSV or SSV. Only 1 or 2mL of foam should be injected at a time and the distribution of the foam should be monitored and massaged with the ultrasound probe. When the foam is visualised at the site of junctional incompetence no further foam should be injected. The maximum volume of foam that should be injected at a single session should not exceed 10–12 mL as the incidence of complications is directly +related to the volume of foam injected. Compression is then + + + + + + + + + + + + + +Figure 57.20 Foam sclerotherapy; cannulation of veins during ultrasound-guided foam sclerotherapy. + + +Lorenzo Tessari, b.1949, physician Trieste, Italy +. +PART 10 | VASCULAR +980 CHAPTER 57 Venous disorders + + + + + + + + + + + + + + +Figure 57.22 Mechanochemical ablation device (reproduced with permission of Vascular Insights). + + +Figure 57.21 Foam sclerotherapy; Tessari method of foam sclero-sant preparation. + + + +applied as following endothermal ablation. While it is postu-lated that compression may have a larger effect upon efficacy for this treatment, practice is not informed by evidence and a wide variation exists. +Outside of a small number of centres, the efficacy of UGFS is significantly worse than for endothermal ablation, leading to high reintervention rates, and the rates of complications such as phlebitis and pigmentation can be high. UGFS does however carry some significant advantages: +● It avoids tumescent anaesthetic and is therefore a less painful procedure (although postoperative pain is prob-ably similar). +● No axial or tributary veins are too tortuous. +● It also allows the treatment of calf veins with overlying skin damage or ulceration without the need to pierce through damaged skin. +● Consumable treatment costs are very low. +These factors mean that many surgeons using endo-thermal techniques also use foam sclerotherapy as an adjunct in specific circumstances. + +CATHETER-DIRECTED SCLEROTHERAPY AND MECHANICOCHEMICAL ABLATION +The efficacy of sclerotherapy relies on endothelial contact with fresh, undiluted sclerosant. Some have therefore exper-imented with catheter-delivered sclerotherapy rather than trying to milk the sclerosant down the vein lumen. There is no good evidence to date that this increases efficacy and the technique is not in widespread use. +A related technology that has shown more promise is mechanicochemical ablation (Figure 57.22). This involves a treatment device that deploys an angled wire from the end. This attaches to a motorised handle. The catheter is placed within the vein lumen as for endothermal ablation. The trig-ger on the handle is depressed, spinning the wire around and liquid sclerosant is infiltrated via the catheter simultaneously during catheter withdrawal. It is thought that the spinning wire causes physical damage to the endothelium and allows a + + + + + + + + + +Figure 57.23 Endovenous glue device (reproduced with permission of Medtronic Inc.). + + +deeper penetration of the sclerosant into the vein wall. The technique is possible in most cases without tumescent anaes-thesia, although a small number of patients find the procedure uncomfortable and the device can ‘snag’ on the vein, tearing it or rarely stripping it altogether. Early studies suggest simi-lar early efficacy rates as for endothermal ablation, which is encouraging. The axial ablation is usually less painful than endothermal ablation, but this advantage is lost when it is combined with phlebectomy of the tributaries; therefore, it is uncertain whether it can replace endothermal ablation, unless axial ablation is to be performed in isolation. Treat-ing longer veins can also be challenging due to limitations in catheter length and the safe dose of sclerosant. It is a good choice for a patient with needle phobia, who is happy to forgo +treatment of varicose tributaries. + +ENDOVENOUS GLUE +The final non-tumescent technique is the endoluminal appli-cation of cyanoacrylate adhesive (Figure 57.23). Again, this involves a treatment catheter placed within the vein lumen. A handle is used to infiltrate the adhesive in 0.1mL applica-tions via the catheter. The vein is then compressed, sealing the lumen closed. Early efficacy results are similarly promising and patients experience minimal intraprocedural pain. Long-term results and, similar to mechanicochemical ablation, the optimal management of tributaries are unknown. The consumable costs are currently the highest for any venous +ablative technique. +PART 10 | VASCULAR Varicose veins 981 + + + +Open surgery +The principles of traditional ligation and stripping are to fully dissect the point of junctional incompetence and to remove the refluxing axial vein and dilated tributaries. The operation is usually performed under general anaesthesia but +loco-regional anaesthesia is used by some and the infiltration Ligature of tumescent local anaesthesia around the axial vein prior to +stripping may have some advantages, but is not widely used. The role of open surgery as a primary treatment of a +refluxing superficial axis has been considerably reduced with development of the minimally invasive techniques described above, the long-term results of which are at least comparable with open surgery, but with significantly less morbidity and faster recover . Experienced endovenous surgeons do still use open surgery in some circumstances and a venous surgeon needs to be trained and experienced in this area. +y +Surgical adjuncts including phlebectomy and, occasion-ally, perforator ligation are much more commonly used, and the former has been shown to have a significant impact upon +outcome. + + + + + +Femoral vein + + + + +Point of division + +Long saphenous vein + + +SAPHENOFEMORAL LIGATION AND GREAT SAPHENOUS STRIPPING +An oblique groin incision is made at the level of, and lateral to, the pubic tubercle, ideally above the groin crease. The GSV is identified and dissected to the SFJ, which should be clearly established before the vein is divided to avoid disas-trous inadvertent transection of the superficial femoral vein. The anatomy is often variable but six GSV tributaries may be +encountered close to the SFJ: +● Laterally: +● superficial inferior epigastric vein; ● superficial circumflex iliac vein; +● Medially: +● superficial external pudendal vein; ● deep external pudendal vein; +● Distally: +● anterior accessory saphenous vein; +● posteriomedial thigh vein. + +Classically, these are ligated distal to their divisions. A flush SFJ ligation is then performed and the GSV retrogradely stripped to around the knee (Figure 57.24). Phlebectomy is performed as discussed above. +Closure of the cribriform fascia, with sutures or synthetic patches over the ligated SFJ, does not reduce groin recurrence. Stripping to the lowest point of reflux may improve results, but at a cost of increased saphenous nerve complications and is not widely performed. More recently, some surgeons argue that surgical trauma and subsequent inflammation in the groin is associated with neovascularisation, which in turn may lead to recurrence. Furthermore, others hypothesise that it is the loss of the normal groin tributaries that may be responsible for driving the process of neovascularisation. These concepts have led some to believe that ligation of the refluxing vein should be distal to the tributaries and that the junction itself should be left untouched. There is no clear clinical evidence +to support these hypotheses. + + +Figure 57.24 Saphenofemoral junction ligation and great saphe-nous vein stripping. + + +SAPHENOPOPLITEAL JUNCTION LIGATION AND SMALL SAPHENOUS STRIPPING +Preoperative duplex to mark the position of the SPJ is highly recommended (Figure 57.25). The patient is positioned in the prone position, a transverse incision is made over the premarked SPJ, the fascia is divided and the SSV is exposed. The SPJ can then be formally dissected with a flush ligation or the SSV can be gently retracted and ligated as proximally as possible. No good evidence exists to favour one technique over the other; proponents of the flush ligation would argue that it avoids leaving a stump of SSV, a common source of recurrence, while proponents of the simple SSV ligation technique argue it reduces the incidence of the most common +serious complications, nerve injury and popliteal vein injury. +PART 10 | VASCULAR +982 CHAPTER 57 Venous disorders + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 57.25 Preoperative marking of the saphenopopliteal junction and small saphenous vein mapped using duplex scanning. + + +The SSV can then either be stripped or the proximal section of the vein can be resected. Those who strip argue it reduces the incidence of recurrence, while opponents feel it increases the incidence of sural nerve injury. There are no randomised trials comparing these techniques. Once again, phlebectomy +is then performed. +Adjunctive surgical techniques PHLEBECTOMY +This may be performed following treatment of junctional incompetence and axial vein reflux, or as sole treatment under local anaesthetic in patients with isolated tributary incompetence, or possibly in very early axial reflux, which reverts to normal upon occlusion of the refluxing tributary on duplex ultrasound. Phlebectomy is usually performed through small stab incisions using small mosquito forceps and/or phle-bectomy hooks that have been demonstrated to be superior in terms of bruising, pain and generic quality of life than +transilluminated-powered phlebectomy (see Figure 57.19). + +PERFORATOR LIGATION +The majority of studies assessing the role of perforator ligation have been in patients with venous ulcers, analysing the effects on ulcer healing, and even in this situation randomised data are lacking. The role of perforator ligation in patients with uncomplicated varicose veins is even less clear. In uncom-plicated varicose veins, perforators may be ligated through +a small, duplex-guided incision, while in patients with skin + +changes, subfascial endoscopic perforator ligation may be pre-ferred, although the benefits are unproven. Perforators can also be ablated with endovenous techniques. + +Complications of standard varicose vein surgery +Complications (minor and major) are reported in up to 20% of patients who undergo traditional varicose vein sur-gery. Wound infections, the most common complication, are reduced by prophylactic antibiotics. Nerve injury is the most common serious complication. The incidence of saphenous nerve neuralgia is up to 7% following GSV stripping to the knee (the incidence is higher with stripping to the ankle). The incidence of sural nerve neuropraxia and common pero-neal nerve injury may be as high as 20% and 4%, respectively, following SSV surgery. The incidence of venous thromboem-bolic complications is approximately 0.5% following varicose vein surgery; however, patient risk factors must be individually assessed and appropriate prophylaxis administered according to guidelines. + +Recurrent varicose veins +Approximately 10–20% of patients who present to hospital with varicose veins have had previous intervention. Pro-spective data on long-term results following intervention for recurrent varicose veins are sparse and the criteria for defining recurrence are variable. +Significant clinical recurrence 5–10 years following var-icose vein surgery occurs in 10–35% of patients, but minor/ duplex-detected recurrence is much more common, being in the order of 70%. Causes of recurrence include: neovas-cularisation, reflux in residual axial vein, inadequate initial surgery and new junctional reflux. Neovascularisation is the development of new veins within postsurgical tissue. These veins lack valves and over time can span the tissue between a ligated junction and nearby tributary veins. If significant in size and/or number, these may contribute to recurrent venous hypertension. +Recurrence is more common following SSV versus GSV surgery, and in patients with high body mass index, while stripping of the incompetent axial vein reduces recurrence rates. Limited data suggest recurrence rates following endo-venous thermal ablation may be lower than following surgery. Recurrent varicose veins often have an atypical distribution and duplex assessment is mandatory (Figures 57.26 and 57.27). Open surgery for recurrent varicose veins is associ-ated with a high (40%) complication rate, the most common being lymph leak and wound infection, thus endovenous interventions would seem to offer an attractive alternative, where feasible. + +VENOUS LEG ULCER +Venous disease is responsible for around 85% of all chronic lower limb ulcers in resource-rich countries. Community-based prevalence is 0.1–0.3% in adults (2–4% in the elderly). Venous leg ulcer has a disproportionate cost to society, with profound impairment in health-related quality of life for +PART 10 | VASCULAR Venous leg ulcer 983 + + +Summary box 57.1 + +Varicose veins +●● Are one of the most common conditions causing a physical impairment in quality of life +●● Interventional treatment improves quality of life and is highly cost-effective +●● Anatomical and physiological assessment using duplex ultrasound is invaluable in the diagnosis and planning of treatment +●● Ultrasound-guided endovenous ablation has revolutionised treatment, minimising procedural morbidity while being highly effective + + + + + + + + + + + +Figure 57.27 Recurrent varicose veins secondary to an incompetent thigh perforator. + + + + +Figure 57.26 Recurrent anterior abdominal wall varicose veins fol-lowing saphenofemoral junction ligation complicated by iliac deep vein thrombosis. + +both patients and their carers; the dressings alone account for 1–3% of western healthcare expenditure. Furthermore, 15–30% of patients with ‘venous’ leg ulcers have concomi-tant arterial occlusive disease. This is termed a ‘mixed’ ulcer. There are many other causes of leg ulcers and these must be excluded in any patient presenting with ulceration. Causes of leg ulceration include: +● venous disease: superficial incompetence, deep incompe-tence or obstruction; +● arterial ischaemic ulcers; ● vasculitic ulcers; +● traumatic ulcers; ● neuropathic ulcers; ● neoplastic ulcers; +● infections, especially in resource-poor countries. + + +Pathophysiology of ulceration +The exact pathophysiology of ulcer development has not been established. Originally, it was thought that static blood within the superficial veins led to hypoxia, which caused tissue death (stasis ulcers). This was not confirmed by investigation of venous oxygen saturation, which was found to be higher in ulcerated limbs. This led to the concept of arteriovenous fis-tulae, which were thought to develop in response to the high + +venous pressure; however, this could not be confirmed. High venous pressure was found to be associated with a pericapil-lary infiltrate. This includes fibrin and other proteins, which are known to lead to fibrosis. It was hypothesised that these ‘cuffs’ could act as an impediment to diffusion of oxygen and nutrients. +Leukocytes were found to decrease in the venous effluent coming out of dependent limbs. This decrease in leukocyte passage was shown to increase if short-term venous hyperten-sion was induced by application of a tourniquet. This led to the concept of white cell ‘trapping’, which, however, has not been confirmed by further investigation. Polymorphonuclear leukocytes were not found within the tissues, but increased numbers of mast cells, monocytes and lymphocytes have been found in periulcer tissues. +Reactive oxygen species are increased in the ulcer envi-ronment and these may generate free radicals, leading to tis-sue damage. Proteolytic enzymes are also increased in ulcers and the fibroblasts in the ulcer surrounds are also abnormal, being in a ‘senescent’ state. Growth factors may be inhibited, leading to poor repair, and their absence may also lead to ulceration. It is proving difficult to show whether any of these factors is the cause of or the result of an ulcer. +At present, ambulatory venous hypertension is the only accepted underlying cause of venous ulceration. This also explains why venous ulcers are never seen in the upper limb. It is important to try to define the exact mechanism of ulcer development. Venous hypertension may be the result of pri-mary valve incompetence of the saphenous veins, incompe-tence of the perforating veins or incompetence or obstruction +of the deep veins. +PART 10 | VASCULAR +984 CHAPTER 57 Venous disorders + + + +Clinical features +The ulcer must be carefully examined. A venous ulcer usually has a gently sloping edge and the floor contains granulation tissue covered by a variable amount of slough and exudate. Any significant elevation of the ulcer edge should indicate the need for a biopsy to exclude a carcinoma (usually a squa-mous cell). +The venous ulcer of the leg characteristically develops in the skin of the gaiter region, the area between the muscles of the calf and the ankle. This is the region where many of the Cockett perforators join the posterior tibial vein to the surface vein, known as the posterior arch vein. The majority of ulcers develop on the medial side of the calf, but ulcers associated with SSV incompetence may develop on the lat-eral side of the leg. Ulcers can develop on any part of the calf skin in patients with post-thrombotic syndrome (PTS); how-ever, venous ulcers rarely extend on to the foot or into the upper calf and, if there is ulceration at these sites, other diag-noses should be seriously considered. Ulcers often develop in response to minor trauma; many patients notice some itching, perhaps associated with mast cell degranulation, before the ulcers develop. Almost all venous ulcers have surrounding haemosiderosis (seen as pigmentation) and the more chronic ulcers develop lipodermatosclerosis with associated fibrosis of the subcutaneous tissue (see Figure 57.10). This is manifest as thickening, pigmentation, inflammation and induration of the calf skin. The pigmentation comes from haemosiderin and melanin and the haemosiderin itself may be an important factor in ulcer development. +A full examination of the front and back of the limbs with the patient standing should be carried out to assess the presence of varicosities and truncal incompetence of the saphenous systems (note that venous ulcers are not always accompanied by varicose veins). All patients should have their pulses palpated and, if there is any doubt, their arterial Doppler pressures should be measured. Sensation and proprio-ception should be assessed to exclude neuropathy, especially in diabetic patients. A careful examination of the hand and other joints may confirm the presence of rheumatoid arthritis +or osteoarthritis. + +Investigation +Most vascular surgeons will carry out a duplex scan when the patient with an ulcer is first seen to assess the status of the deep and superficial veins. The presence of reflux in these veins does not confirm a venous ulcer, but supports the diagnosis in the absence of another cause and helps direct treatment. +All patients presenting for the first time with a new leg ulcer should have a full blood count, blood glucose, erythro-cyte sedimentation rate (ESR) or C-reactive protein (CRP) and sickle cell test if they have an appropriate racial back- +ground. Anaemia can both cause ulcers (e.g. sickle cell + + +disease and pernicious anaemia) and be a result of ulceration (e.g. iron deficiency anaemia and the anaemia of chronic dis-ease). Polycythaemia is a rare cause of ulceration. An anti-body screen should be obtained if the ulcer appears ‘atypical’ or there is any suggestion of joint disease (e.g. rheumatoid arthritis). All patients presenting with a new ulcer should have their Doppler pressures measured, unless the foot pulses are easily palpable and have been confirmed as such by a vas-cular specialist. +Venous ulcers are characteristically difficult to heal; how-ever, persistence may indicate that there is another or coex-isting cause (e.g. malignancy, rheumatoid arthritis or arterial ischaemia). Biopsies are indicated if malignancy is suspected and it is important to remember that a Marjolin’s type of ulcer (a squamous cell or basal cell carcinoma) can develop in a +chronic long-standing venous ulcer (Figure 57.28). + +Management Compression +The very best results are seen in specialist multidisciplinary ulcer services. The cause of a venous leg ulcer is venous hypertension and the key-stone of management is to decrease this hypertension. The primary way of doing this is with the use of compression. Most patients are suitable for the classic +‘four-layer bandaging system’ (4LB): +● Orthopaedic wool: distributes the pressure and reduces undue pressure on sensitive areas susceptible to pressure damage. Also helps to absorb excess exudate that escapes the primary dressing. +● Cotton crepe: smooths the wool and holds it in place. + + + + + + + + + + + + + + + + + + + + +Figure 57.28 A Marjolin’s ulcer (a squamous cell cancer arising in a chronic venous ulcer). + + + + +Frank Bernard Cockett, 1916–2014, surgeon, St. Thomas’s Hospital, London, UK. +Jean-Nicholas Marjolin, 1780–1850, surgeon, Paris, France, described the development of malignant ulcers in scars in 1828. +PART 10 | VASCULAR +Pelvic congestion syndrome 985 + + + +● Elastic bandage: first compressive layer, contributes about one-third of the interface pressure. +● Cohesive bandage: second compressive layer, increases stiffness and adds approximately two-thirds of the inter-face pressure +The ideal interface pressure in pure venous ulceration is 35–40mmHg. Skilled application of these dressings is essen-tial for both safety and efficacy, and the best results come from specialist nursing teams either based in secondary care or in the community. Newer compression systems are being created with the aim of reducing the learning curve and variability in application, and some are designed to be applied by patients themselves. There remains little evidence that these are set to replace 4LB for the time being. The interval between ban-dage applications is based upon the amount of exudate and the speed of healing and best judged by an experienced nurse. Compression in mixed ulcers is controversial, but emerg- +ing evidence suggests that it is both safe and effective when performed and monitored appropriately. With an ankle bra-chial pressure index (ABPI) of 0.5–0.8 modified compression with an interface pressure of 30mmHg is safe and effective and pressures of up to 40mmHg have been described in studies using inelastic bandages without ill effect. Contrary to conventional thinking, studies have shown an increase in perfusion in patients treated in this way, presumably by a reduction in capillary back pressure. Patients do see respect-able healing rates in this group, but they remain lower than in those patients with an ABPI >0.8. It is not clear whether revascularisation followed by full compression yields better results, but this is common practice. Patients with an ABPI <0.5 or an ankle pressure <60mmHg must undergo revascu-larisation prior to any compression treatment. + +Superficial venous ablation or surgery +The next method of reducing venous hypertension is to treat superficial reflux. There has been only one trial of adequate power and design to answer the question of whether the abo-lition of superficial reflux improves ulcer healing rates. This trial has completed recruitment, but follow-up is ongoing and practice continues to vary. + +Other treatments +The only drug that shows promise to date is pentoxifylline, which increases microvascular perfusion by decreasing plasma cellular viscosity and cytokine inhibition, although its exact mechanism of action is not fully understood. +A number of biological dressings have been developed, including fetal keratinocytes and collagen meshes, which have been shown to improve healing; however, they are not cost-effective for the majority of ulcers. Pinch grafts and ulcer excision with mesh grafting have been shown to provide good early healing with moderate long-term results (50% healed at 5 years). +Antibiotics do not speed ulcer healing in the absence of cellulitis and all other specific ulcer-healing drugs are of dubi-ous validity. A large range of topical therapies and primary dressings have similarly failed to have an impact. + +Prevention of recurrence +Once an ulcer has healed the patient must be re-evaluated in an attempt to prevent recurrence. If not already performed, there is level 1 evidence that patients should undergo treat-ment for their superficial venous incompetence, unless they have deep venous occlusion. Class II below-knee graduated compression stockings should be prescribed for all patients with residual reflux or deep venous occlusion, or those with recurrent ulceration despite not being in this group. These +should be worn for life. + +Prognosis +Nearly all venous ulcers can be healed but, even in those who have successful ablation or wear their stockings reli-giously, there is a 20–30% incidence of reulceration by 5 years. The greatest risk of reulceration is in the post- +thrombotic leg. + +Summary box 57.2 + +Venous leg ulcer +●● Is associated with a profound impairment in quality of life +●● Ulcers are not infrequently difficult to heal and prone to recurrence +●● The treatment of these chronic wounds is associated with high costs to healthcare systems and patients +●● The mainstay of treatment is the reduction in venous hypertension, with compression and increasingly superficial venous ablation + + +PELVIC CONGESTION +SYNDROME +Pelvic congestion syndrome (PCS) is among the differen-tial diagnoses to be considered in female patients presenting with chronic pelvic pain and may be significantly under-diag-nosed. PCS sufferers are typically premenopausal, multiparous women aged 20–45 years, who present with severe dull aching pelvic pain thought to be the direct result of ovarian and pel-vic varicosities. The pain is usually non-cyclical, and may be precipitated by prolonged standing. Other symptoms include dysmenorrhoea, menorrhagia, rectal discomfort or urinary frequency. Signs may include tenderness over the uterus/ ovaries, vulval varicosities and haemorrhoids. There may be vulval and atypically distributed thigh varicosities. The road to a diagnosis of PCS is often a long and laborious one, usu-ally only made following extensive investigations to exclude other more common causes of pelvic pain. Abdominal, pelvic and transvaginal duplex examination allows dynamic visuali-sation of pelvic blood flow and should be the initial investiga-tion of choice, as these are rapid, readily accessible outpatient procedures that are also valuable in excluding other patholo-gies. Alternatives include magnetic resonance (MR) venog-raphy and diagnostic venography. +Medical treatments for PCS include psychotherapy, pro- +gestins, danazol, gonadotrophin receptor agonists (GnRH) +PART 10 | VASCULAR +986 CHAPTER 57 Venous disorders + + +(a) (b) + + + + + + + + + + + + + + + + + + + + + + +Figure 57.29 (a, b) Left ovarian vein incompetence supplying pelvic and pudendal varicosities: (a) diagnostic venogram; (b)therapeutic embolisation. + + + + +with hormone replacement therapy, and non-steroidal anti-inflammatory drugs. Historical open surgical procedures (extraperitoneal resection of ovarian veins) have now largely been superseded by percutaneous pelvic vein embolisation (Figure 57.29), reducing peri- and postprocedural morbidity +while maintaining high success rates. + +VENOUS THROMBOEMBOLISM +Venous thromboembolism (VTE) is an important condition within surgery, and autopsy studies suggest that it is the most common direct cause of death in surgical patients. Venous thrombosis is the formation of a semisolid coagulum within the venous system and may occur in the superficial system (usually described as superficial thrombophlebitis) or the deep system (deep venous thrombosis, DVT). Venous thrombosis of the deep veins of the leg may be complicated by the imme-diate risk of pulmonary embolus and sudden death. Subse-quently, patients are at risk of developing PTS (Figure 57.30) and venous ulceration. While DVT may occur in the upper limb, it is the leg that gives rise to the vast majority of the +morbidity and subsequent complications of this condition. + +● abnormal flow (e.g. stasis); +● abnormal blood (e.g. thrombophilia). +There are many predisposing causes for VTE. These are listed in Table 57.2. The most important factor is a hospital admission for treatment of a medical or surgical condition. Injury, especially fractures of the lower limb and pelvis, preg-nancy and the oral contraceptive pill are other well-recognised predisposing factors. Endothelial damage is now known to be critically important. The interaction of the endothelium with inflammatory cells, or previous deep vein damage, renders the endothelial surface hypercoagulable and less fibrinolytic. Stasis is a predisposing factor seen in many of the condi-tions described in Table 57.2, especially in the postoperative period, in patients with heart failure and in those with arterial +ischaemia. + + +Aetiology +The three factors described by Virchow over a century ago are still relevant in the development of venous thrombosis. +These are: + +● contact of blood with an abnormal surface (e.g. endothe- +lial damage); + + +Figure 57.30 Post-thrombotic leg demonstrating features of eczema, pigmentation and mild lipodermatosclerosis. +PART 10 | VASCULAR Venous thromboembolism 987 + + + +TABLE 57.2 Risk factors for venous thromboembolism. Patient factors +Age Obesity +Varicose veins Immobility Pregnancy Puerperium +High-dose oestrogen therapy +Previous deep vein thrombosis or pulmonary embolism Thrombophilia (see Table 57.3) +Disease or surgical procedure +Trauma or surgery, especially of pelvis, hip and lower limb Malignancy, especially pelvic, and abdominal metastatic Heart failure +Recent myocardial infarction Paralysis of lower limb(s) Infection +Inflammatory bowel disease Nephrotic syndrome Polycythaemia Paraproteinaemia +Paroxysmal nocturnal haemoglobinuria antibody or lupus anticoagulant +Behçet’s disease +Homocystinaemia + +TABLE 57.3 Abnormalities of thrombosis and fibrinolysis (thrombophilia) that lead to an increased risk of venous thrombosis. +Congenital +Deficiency of antithrombin III, protein C or protein S +Antiphospholipid antibody or lupus anticoagulant +Factor V Leiden gene defect or activated protein C resistance +Dysfibrinogenaemias +Acquired +Antiphospholipid antibody or lupus anticoagulant + +(Figure 57.31). This is likely to extend up to the next large venous branch and it is possible for the clot to break off and embolise to the lung as a pulmonary embolus. In this situa-tion the embolus arising from the lower leg veins becomes detached, passes through the large veins of the limb and vena cava, through the right heart and lodges in the pulmonary arteries. This may totally occlude perfusion to all or part of one or both lungs. This results in a clinical spectrum from tachycardia and pain, through respiratory failure (despite adequate ventilation) to cardiovascular collapse and death. Moderate-sized emboli can cause pyramidal-shaped infarcts +on imaging. + +Diagnosis +The most common presentation of a DVT is pain and swell-ing, especially in the calf, usually in one leg; however, bilat- +eral DVTs are common, occurring in up to 30%. When the + + +A number of conditions are associated with increased coagulability of the blood (thrombophilia) (Table 57.3). Defi-ciencies of antithrombin, activated protein C and protein S have all been shown to predispose to venous thrombosis in young patients. Activated protein C deficiency is associated with inheritance of the factor V Leiden gene and may account for the higher incidence of venous thrombosis in Caucasian populations (being present in 6–7%). It results in a small increase in the risk of VTE, although it may act in concert with some of the other predisposing factors. A thrombophilia should be excluded in any patient presenting with an episode of VTE who gives a family history of VTE or in whom there is no other predisposing factor. +Although the development of DVT is probably multifac-torial, immobility (and hence stasis) remains one of the most important factors. DVT is recognised as a complication of +long-haul flights and other forms of travel. + +Pathology +The thrombus commences as a platelet aggregate. Subse-quently, fibrin and red cells form a mesh until the lumen of the vein wall occludes. The coralline thrombus then progresses as +a propagated loose red fibrin clot containing many red cells Figure 57.31 An organised thrombus. + +Hulusi Behçet, 1889–1948, Turkish dermatologist described a disease of inflamed blood vessels in 1937. +PART 10 | VASCULAR +988 CHAPTER 57 Venous disorders + + + + + + + + + + + + + + + +Figure 57.32 Phlegmasia cerulea dolens. + + + + + + + + + + + + + + + + + + + +Figure 57.33 A foot with venous gangrene. The gangrene is sym-metrical involving all of the toes. There is no clear-cut edge and there is marked oedema of the foot. + +swelling is bilateral, DVT must be differentiated from other causes of systemic oedema, such as hypoproteinaemia, renal failure and heart failure. Some patients have no symptoms of thrombosis and may first present with signs of a pulmonary embolus, e.g. pleuritic chest pain, haemoptysis and shortness of breath. Patients may also develop shortness of breath from chronic pulmonary hypertension. Sometimes the leg appears cellulitic and very occasionally it may be white or cyanosed: phlegmasia alba dolens and phlegmasia cerulea dolens (Figure 57.32). This indicates venous pressures that are so high that they are impeding tissue perfusion. Patients who present with venous gangrene (Figure 57.33) often have an underlying neoplasm. +Clinical examination for DVT is unreliable. Physical signs may also be absent. Mild pitting oedema of the ankle, dilated surface veins, a stiff calf and tenderness over the course of the deep veins should be sought. Leg pain occurs in about 50% of patients with DVT but is nonspecific. Homans’ +sign – resistance (not pain) of the calf muscles to forcible + +dorsiflexion – is not specific and should not be elicited. Ten-derness occurs in 75% of patients but is also found in 50% of patients without objectively confirmed DVT. The pain and tenderness associated with DVT does not usually correlate with the size, location or extent of the thrombus. Clinical signs and symptoms of pulmonary embolus occur in about 10% of patients with confirmed DVT. +A low-grade pyrexia may be present, especially in a patient who is having repeated pulmonary embolus. Patients may have signs of cyanosis, dyspnoea, raised neck veins, a fixed split second heart sound and a pleural rub if they have pulmonary emboli causing right heart strain, although these +signs may be subtle or absent. + +Investigation +The diagnosis of DVT and pulmonary embolus should be established by special investigations as the symptoms and signs are non-specific and may be absent. In addition, treat-ment with anticoagulation is not without risk and the diagno-sis must be made with reasonable certainty. +Many centres direct investigation based upon the modi-fied Wells score (Table 57.4), with further imaging dictated by these results. These scores can be unreliable though, especially in hospital inpatients, and should only be considered a guide. +Venous duplex ultrasound is commonly performed to look for evidence of thrombosis throughout the deep or superficial venous system. Ideally, this should be performed by an expe-rienced vascular sonographer, but the volume of cases is such that compression ultrasound is frequently being performed by non-specialists. Compression ultrasound involves applying pressure with the ultrasound probe over the common femoral +and popliteal veins. Under normal circumstances these veins + + +TABLE 57.4 Modified Wells criteria for predicting deep vein thrombosis (DVT). +Variable Score +Lower limb trauma or surgery or immobilisation in a 1 plaster cast +Bedridden for >3 days or surgery in last 4 weeks 1 +Tenderness along the line of femoral or popliteal veins 1 +Entire limb swollen 1 +Calf >3cm larger circumference than the other side +10cm below the tibial tuberosity 1 +Pitting oedema 1 +Dilated collateral superficial veins (not varicose veins) 1 +Previous DVT 1 +Malignancy (including treatment up to 6 months ago) 1 +Intravenous drug abuse 3 +Alternative diagnosis more likely than DVT −2 +Low probability (5%) of DVT (score –2 to 0), moderate probability (17%) of DVT (score 1–2), high probability (17–53%) of DVT (score >2). + + + +John Homans, 1877–1957, Professor of Clinical Surgery, Harvard University Medical School, Boston, MA, USA. Philip Wells, contemporary, physician, University of Ottawa, Canada. +PART 10 | VASCULAR Venous thromboembolism 989 + + + +will compress tightly shut. In the presence of DVT they will not fully compress. It is rapid to both learn and perform, but not ideal and most importantly misses calf vein thrombosis. Calf vein thromboses may propagate to form a more extensive thrombus, which may in turn embolise. The optimal manage-ment of calf vein thrombosis when detected is not clear, and some units use surveillance, with others anticoagulating such patients upon detection. +Ascending venography, which shows a thrombus as a fill-ing defect, is now rarely required unless thrombolysis is being considered (Figure 57.34). MR venography may also be used. Pulmonary embolus is diagnosed definitively by computed tomography (CT) pulmonary angiogram, which will demon-strate the presence of filling defects in the pulmonary arteries (Figure 57.35). Pulmonary angiography is rarely required unless thrombolysis is being considered. +The differential diagnosis of a DVT includes a ruptured Baker’s cyst, a calf muscle haematoma, a ruptured plantaris muscle, a thrombosed popliteal aneurysm and arterial isch-aemia. Duplex scanning will detect many of these conditions but often patients present with non-specific pain in the calf that resolves with no firm diagnosis being made. The differ-ential diagnosis of a pulmonary embolism includes myocardial +infarction, pleurisy, pneumonia and aortic dissection. + +Prophylaxis +Prophylactic methods can be divided into mechanical and pharmacological. A variety of mechanical methods have been tried, but only the use of graduated elastic compression stock- +ings and external pneumatic compression have been shown + + + + + + + + + + + + + + + + + +Figure 57.35 A computed tomography pulmonary angiogram show-ing pulmonary emboli as filling defects (arrow) in the pulmonary artery +. + +to be worthwhile by reducing the incidence of thrombosis. Newer devices, such as electronic nerve stimulators, lack evi-dence of efficacy to date. Compression-based prophylactic measures should be avoided in patients with peripheral vas-cular disease. +Pharmacological methods are more effective than mechanical methods at risk reduction, although they carry an increased risk of bleeding. In the past, low-dose unfraction-ated heparin was used both intravenously and subcutaneously. In the absence of renal impairment, most centres currently use low molecular weight heparin (LMWH) given subcuta-neously. This is given once daily, does not require monitoring and has a lower risk of bleeding complications. +Patients who are being admitted for surgery may be graded as low, moderate or high risk for PE and VTE (Tables 57.5 and 57.6). Patients in the medium- or high-risk groups should be considered for pharmacological prophylaxis with an antico-agulant medication. A combination of mechanical and phar-macological treatment with heparin can be used in patients +considered at high risk. + + + +TABLE 57.5 Modified Wells criteria for predicting pulmonary embolism (PE). +Variable Score +Clinical signs and symptoms of deep vein thrombosis 3 (DVT) (minimum of leg swelling and pain on palpation of +deep veins) +Alternative diagnosis less likely than PE 3 +Heart rate >100 bpm 1.5 +Immobilisation >3 days or surgery within past 4 weeks 1.5 +Previous DVT or PE 1.5 +Haemoptysis 1 + +Figure 57.34 An ascending venogram of a deep vein thrombo-sis seen as filling defects (arrows) with contrast passing around the thrombus. + + +Malignancy (treatment or palliation within past 6 months) 1 +A score of <4 means PE is unlikely (12.4%), >4 is suggestive of PE (37.1%). + + +William Morrant Baker, 1839–1896, surgeon, St. Bartholomew’s Hospital, London, UK, described these cysts in 1877. +PART 10 | VASCULAR +990 CHAPTER 57 Venous disorders + + + +TABLE 57.6 Low-, medium- and high-risk patient groups for venous thromboembolism. +LOW +Minor surgery <30 minutes; any age; no risk factors +Major surgery >30 minutes; age <40; no other risk factors +Minor trauma or medical illness +MODERATE +Major surgery; age 40+ or other risk factors +Major medical illness: heart/lung disease, cancer, inflammatory bowel disease +Major trauma/burns +Minor surgery, trauma, medical illness in patient with previous DVT, PE or thrombophilia +HIGH +Major orthopaedic surgery or fracture of pelvis, hip, lower limb +Major abdominal/pelvic surgery for cancer +Major surgery, trauma, medical illness in patient with DVT, PE or thrombophilia +Lower limb paralysis (e.g. stroke, paraplegia) +Major lower limb amputation + +DVT, deep vein thrombosis; PE, pulmonary embolus. + + +Treatment +Deep vein thrombosis +The management of DVT has in the past been focused upon reducing the risk of pulmonary embolus. Patients who are confirmed to have a DVT on duplex imaging should be rapidly anticoagulated with a ‘treatment dose’ of subcutaneous LMWH. Patients with significant renal impairment should be commenced on intravenous unfractionated heparin. Patients who have a sensitivity towards heparinoids, such as those with heparin-induced thrombocytopenia, should commence on another anticoagulant, such as fondaparinux (an indirect factor Xa inhibitor) or bivalirudin (a direct thrombin inhib-itor). This will achieve rapid anticoagulation and reduce the risk of embolisation. Typically patients will then commence on warfarin for at least 3 months (or longer depending upon the persistence of risk factors or in recurrent cases). Patients who cannot be safely anticoagulated (usually due to bleed-ing risks) should be considered for a temporary inferior vena cava filter, until either they are safe to be anticoagulated or the risk of embolisation has subsided and the filter may be retrieved. Patients with active cancer typically remain on a LMWH. There is a range of newer or ‘novel’ anticoagulants (NOACs). These oral agents either directly inhibit factor Xa (rivaroxaban and apixaban) or thrombin (dabigatran). Work is ongoing to explore their place within patient management. Alongside the risk of pulmonary embolus is the risk to the patient’s leg. Two-thirds of patients will have developed a PTS within 5 years of their DVT. A PTS limb may present +, + +with any of the symptoms, signs and complications of venous +hypertension discussed earlier, but are typically towards the +most severe end of the spectrum and patients face a consid- +erable deficit in their quality of life. A small number of cen- +tres are performing procedures aiming to treat this and these +include venous recanalisation and stenting and sometimes +venous bypass procedures. These procedures can be very chal- +lenging and although good results are possible, many patients +are condemned to lifelong compression with unaddressed +symptoms and complications. As the treatment of PTS is so +challenging, attention is being turned towards prevention +with the use of thrombolysis, endovenous thrombectomy and +stenting. +During thrombolysis, an agent such as tissue plasminogen +activator is administered directly into the thrombus, either +via the popliteal vein or by direct puncture in the groin. +New devices are being marketed that physically disrupt the +thrombus at the same time as local lysis is carried out. Some +thrombi can be compressed by stent grafting, allowing the +venous lumen to be opened, especially in the iliac region. A +meta-analysis of randomised trials has shown that these treat- +ments result in a significant reduction in PTS at 5 years (from +67% to 39%) but at a cost of an increased risk of significant +bleeding complications (from 4% to 10%). Patient selection +is important. Despite such promising data, access to these treatments remains limited. + +Pulmonary embolus +Most pulmonary emboli can be treated by anticoagulation +and observation, but severe right heart strain and shortness +of breath indicates the need for thrombolysis or radiologically +guided catheter embolectomy. + +Superficial thrombophlebitis +This is a superficial venous thrombosis. An abnormal endo- +thelium is a much more common precipitating factor than +in most DVTs. Common causes include external trauma +(especially to varicose veins), venepunctures and infusions of +hyperosmolar solutions and drugs. The presence of an intrave- +nous cannula for longer than 24–48 hours often leads to local +thrombosis. Some systemic diseases such as thromboangiitis +obliterans (Buerger’s disease) and malignancy, especially of +the pancreas, can lead to a flitting thrombophlebitis (throm- +bophlebitis migrans), affecting different veins at different +times. Finally, coagulation disorders such as polycythaemia, +thrombocytosis and sickle cell disease are often associated, as +is a concomitant DVT. +The surface vein feels solid and is tender on palpation. +The overlying skin may be attached to the vein and in the +early stages may be erythematous before gradually turning +brown. A linear segment of vein of variable length can be +easily palpated once the inflammation has died down. +A full blood count, coagulation screen and duplex scan +of the deep veins should usually be obtained. Any suggestion +of an associated malignancy should be investigated using + + +Leo Buerger, 1879–1943, Professor of Urologic Surgery, The New York Polyclinic Medical School, New York, NY, USA, described thromboangiitis obliterans in 1908. +PART 10 | VASCULAR Venous entrapment syndromes 991 + + + +appropriate endoscopy and imaging studies, such as an abdominal CT scan. +Most patients are treated with non-steroidal anti-inflammatory drugs and topical heparinoid preparations and the condition resolves spontaneously. Rarely, infected thrombi require incision or excision. Ligation to prevent propagation into the deep veins is almost never required, although some advocate saphenofemoral ligation when the thrombus is seen on ultrasound to be at the SFJ. Associated DVT or thrombo- +philia is treated by anticoagulation. + + +(a) (b) + + + +Summary box 57.3 + +Venous thromboembolism +●● May be unprovoked, in which case an association with an inherited ‘thrombophilia’ should be considered +●● Is much more commonly seen as a complication of illness or surgery +●● Is associated with both quality of life impairment and a risk of mortality +●● All healthcare professionals should actively assess the risk and consider preventative measures where this risk is increased +●● Management should involve measures to reduce the risk of extension and/or embolisation, typically with systemic anticoagulation +●● Consideration should be given to local thrombolysis to reduce the risk of postphlebitis limb + + +CONGENITAL VENOUS +ANOMALIES +There are four main types of anomaly: +● aplasia; +● hypoplasia; ● duplication; +● persistence of vestigial vessels. +Aplasia is most commonly seen in the inferior vena cava and has a similar presentation to the post-thrombotic limb. Membranous occlusion of the left common iliac vein (May–Thurner syndrome) often develops where the vein passes behind the right common iliac artery (iliac vein com-pression syndrome). This leads to an iliac vein thrombosis, which most commonly affects the left common and external iliac veins. Membranes may also narrow the hepatic veins, which can become totally occluded, leading to a Budd– Chiari syndrome. +Hypoplasia results in a narrow vein, which frequently offers little significant venous function and amounts to a functional venous occlusion, being circumvented by enlarged collateral venous tributaries. Duplications are quite common, with double vena cava, femoral and renal veins; they often +present as an incidental finding. + + + + + + + + + + + +Figure 57.36 Two patients with Klippel–Trenaunay syndrome. (a) This patient has a longer leg and a capillary naevus; (b) this patient has a large lateral anomalous axial vein. + +Klippel–Trenaunay syndrome +This is a combined anomaly of a cutaneous naevus, persistent vestigial veins with varicose veins and soft tissue and bone hypertrophy. The condition is a mesodermal abnormality that is not familial (Figure 57.36). +Segments of the deep veins are hypoplastic or aplastic and there may be an associated obstruction of the lymphatics. The condition must be distinguished from the Parkes-Weber syndrome, in which there are multiple arteriovenous fistulae causing venous hypertension, ulceration and high-output car-diac failure. +Virtually all patients with Klippel–Trenaunay syndrome should be treated conservatively with compression hosiery; however, some will benefit from laser ablation of the naevus, stapling of the bones to avoid leg length discrepancy and occasional removal of large superficial varicose veins, pro-vided the deep veins are patent. LMWH should be given to all patients having surgery as this syndrome is associated with +an increased risk of VTE. + +VENOUS ENTRAPMENT +SYNDROMES +The axillary vein and the popliteal vein are the two veins that are most commonly compressed. The former is com-pressed at the thoracic outlet between the first rib and the clavicle, where it usually presents as an axillary vein thrombo- +sis (see below) (Figure 57.37a). The latter is compressed by + + +George Budd, 1808–1882, Professor of Medicine, King’s College Hospital, London, UK, described this syndrome in 1845. +Hans Chiari, 1851–1916, Professor of Pathological Anatomy, Strasbourg, Germany (Strasbourg was returned to France after the end of World War I, in 1918), gave his account of this condition in 1898. +Frederick Parkes-Weber, 1863–1962, physician, The German Hospital, Dalston, London, UK. +PART 10 | VASCULAR +r +992 CHAPTER 57 Venous disorders + + + +an abnormal insertion of the gastrocnemius muscles. Entrap-ment may cause discomfort and swelling of the limb during exercise before thrombosis develops. T eatment is by surgical decompression, excising the first rib or dividing the abnormal +musculature of the gastrocnemius insertion. + +AXILLARY VEIN THROMBOSIS +Thrombosis of the axillary vein (Paget–Schrötter disease) +may occur following excessive exercise in a patient with an + +(a) + + + + + + + + + + + + + + + + + + + +(b) + + + + + + + + + + + + + + + + + + + +Figure 57.37 Thoracic outlet syndrome: (a) cervical ribs on plain x-ray; (b) elevation of the arm causing occlusion of the axillary vein with collaterals. The patient has had previous surgery to decompress the left side. + +anatomically abnormal thoracic outlet, but is also associated with excessive muscle bulk as found in weight lifters. The vein may be compressed by a cervical rib if this is present (Figure 57.37b). The arm is swollen and painful and, at an early stage, the thrombus can be disrupted by thrombolysis delivered through one of the arm veins. The vein must then be imaged to see if there is any compression on elevation of the arm. If this is confirmed, thoracic outlet decompression +can be carried out by resecting the cervical rib or first rib. + +VENOUS INJURY +Blunt or penetrating trauma almost always damages some small- and medium-sized veins, which can be safely ignored or ligated without causing any problems. Larger axial venous channels have in the past been ligated when injured, but it is now recognised that these axial veins should be repaired whenever possible to reduce subsequent morbidity (pain and swelling in the tissues being drained) and limb loss when associated with a concomitant arterial injury. Many venous injuries remain undiagnosed at the time of injury (e.g. crural vein damage associated with a fractured tibia) and only pres-ent many years later when post-thrombotic changes become apparent. Venous injuries occur from both civilian and mili-tary trauma but the incidence of venous military injuries has been particularly well documented. In total, 40–50% of arterial injuries have concomitant venous injuries, especially in the popliteal fossa. +The mechanism may be laceration, contusion or avulsion (Figure 57.38). Iatrogenic injuries result from damage at the time of surgery and from punctures caused by catheter inser-tion. Thrombosis, haemorrhage and embolisation are all com-mon complications and arteriovenous fistulae may develop when there is a local concomitant arterial injury. +Associated injuries to soft tissue, arteries and bones often overshadow the venous injury. Massive haemorrhage from the pelvic bones or the inferior vena cava can rapidly lead to hypovolaemic shock and death if left untreated. Haematomas are common and engorgement, cyanosis and swelling are also +indicative of a major venous injury. + +Management +As with all traumatic injuries, the management priorities are the assessment and management of issues affecting the air-way, then breathing and then circulation. Venous injuries have the potential to threaten life through massive bleeding and patients require vascular access, circulatory support and blood products. Trauma patients with life-threatening haem-orrhage are at risk of hypothermia, acidosis, functional and consumptive coagulopathy and paradoxical thrombosis, and these issues need to be prevented where possible and managed when present (see Chapter 2). +Venous pressures are low and so where there is access to the site of injury, pressure will control bleeding and in most +cases offer definitive management. Intervention is required + + +Sir James Paget, 1814–1899, English surgeon and pathologist, best known for his description of Paget’s disease of the bone. Leopold von Schrötter, 1837–1908, Austrian physician and laryngologist, Chair of Laryngology, University of V enna, Austria. +i +PART 10 | VASCULAR Venous tumours 993 + + + +(a) + + + +(b) + + + +(c) + + + +(d) + + +Figure 57.38 Types of venous damage: (a) incision; (b) transection; (c) irregular laceration; (d) avulsion of a tributary. + +where pressure cannot be applied, or where the loss of venous function itself threatens life or limb. Intervention can include reduction and stabilisation of a fracture (e.g. pelvis), endove-nous embolisation or stent grafting. +A small proportion of venous injuries will require formal exploration and ligation or repair. Different types of repair +are shown in Figure 57.39; the type of repair carried out + + +(a) + + + + +(b) + + + + +(c) + + + + + +(d) + + + +(e) + + + + +Figure 57.39 Types of venous repair: (a) lateral suture (risk of ste-nosis); (b) patch graft; (c) Carrel triangulation technique of venous anastomosis; (d) panel graft; (e) spiral graft. + +depends on the extent of the venous injury, including how much venous wall has been lost or damaged. Lateral sutures and vein patches are ideal methods of repair and end-to-end anastomosis is satisfactory, provided that it is not carried out under tension. A jump graft may be required. +Vein replacement should be by autogenous tissue when-ever possible, using vein harvested from another site, e.g. the internal jugular vein or the GSV from an undamaged limb. Artificial grafts, such as PTFE grafts, are at risk of infection and have given poor results in recent conflicts. The use of anticoagulants and an arteriovenous fistula to reduce the risk of thrombosis in the vein graft are controversial and depend on the associated injuries that are present. In contaminated wounds, tetanus toxoid and antibiotics should be given. A fasciotomy should always be considered if there is a concomi- +tant arterial and venous injury. + +Prognosis +It is now recognised that repair of a major vein can be carried out with a 70–80% success rate, reducing the morbidity of a combined arterial and venous injury considerably (especially limb loss). Complex repairs should not, however, be carried out if a patient’s life is at risk, when ligation may have to suf- +fice in the short term. + +VENOUS TUMOURS +Venous malformation cavernous +angioma/haemangioma +These malformations are common, representing one end of a spectrum of arteriovenous malformations. They often affect the skin but also extending into the deep tissues, including bones and joints. They usually present with variable swell-ing and dilated veins beneath the skin. Occasionally, there is no visible mass and the complaint is one of pain. Haem-orrhage and thrombophlebitis may exacerbate the pain. A soft compressible mass, which is venous in colour especially if it is under the skin, is usually present (Figure 57.40a). A dark blue tinge is often apparent, even if the malformation is deeply situated. Nodules within the mass usually represent previous episodes of thrombosis. The size and extent of the haemangioma are best visualised by nuclear MR with a short tau inversion recovery (STIR) sequence (Figure 57.40b) or CT scanning with contrast enhancement. Venography rarely shows an abnormality, but direct puncture with contrast injection shows the connections of the malformation. +Treatment is a highly specialised area. Treatment options nowadays rarely initially involve surgical excision as once this is done future embolisation and sclerotherapy are very diffi-cult. No treatment is entirely curative because it is difficult to remove all of the angiomatous tissue or sclerose the angioma completely. Sclerosis can be dangerous when the veins con-nect to the deep system, particularly near the central nervous +system. + + +Alexis Carrel, 1873–1944, a French surgeon who emigrated to work at the University of Chicago, IL, USA. He was awarded the Nobel Prize in Physiology or Medicine in 1912 for pioneering vascular suturing techniques. +PART 10 | VASCULAR +994 CHAPTER 57 Venous disorders + + +(a) + + + + + + + + + + + + + + +Figure 57.41 Inferior vena cava containing a filling defect from a leiomyosarcoma (arrow). + +Cystic degeneration +As in the peripheral arterial system, cystic degeneration of the vein wall is an uncommon cause of venous occlusion. It may be detected by ultrasound. The cyst may be deroofed or the +venous segment excised. + + + +. (b) + + + + + + + + + + + + +Figure 57.40 (a) Venous angioma of the leg; (b) magnetic resonance imaging showing extensive angioma (white) throughout the superficial tissues, anterior and posterior compartments of the left leg. + + + + +Leiomyoma and leiomyosarcoma of +the vein wall +These are extremely rare tumours that are usually slow grow-ing. They present with pain and a mass with signs of venous obstruction, e.g. oedema and distended veins. Duplex scan-ning, CT (Figure 57.41) and MR imaging show a filling defect within the vein wall. Treatment is by resection with replacement by autogenous vein taken from another site. Rarel , a PTFE graft is required. When the tumour affects the vena cava it must be resected and replaced with a prosthetic +y +graft. + +FURTHER READING +Barwell JR, Davies CE, Deacon J et al. Comparison of surgery and compres-sion with compression alone in chronic venous ulceration (ESCHAR study): randomised controlled trial. Lancet 2004; 363: 1857–59. +Brittenden J, Cotton SC, Elders A et al. Clinical effectiveness and cost-effectiveness of foam sclerotherapy, endovenous laser ablation and surgery for varicose veins: results from the Comparison of LAser, Surgery and foam Sclerotherapy (CLASS) randomised controlled trial. Health Technol Assess 2015; 19(27): 1–342. +Carradice D, Mekako AI, Mazari FA et al. Randomized clinical trial of endovenous laser ablation compared with conventional surgery for great saphenous varicose veins. Br J Surg 2011; 98(4): 501–10. +Carradice D, Mekako AI, Mazari FA et al. Clinical and technical out-comes from a randomized clinical trial of endovenous laser ablation compared with conventional surgery for great saphenous varicose veins. Br J Surg 2011; 98(8): 1117–23. +Carradice D, Wallace T, Gohil R, Chetter I. A comparison of the effectiveness of treating those with and without the complications of superficial venous insufficiency. Ann Surg 2014; 260(2): 396–401. +Coleridge-Smith P, Labropoulos N, Partsch K et al. Duplex ultrasound in-vestigation of the veins in chronic venous disease of the lower limbs – UIP consensus statement. Eur J Vasc Endovasc Surg 2006; 31: 83–92. Eklof B, Rutherford RB, Bergan JJ et al. Revision of the CEAP classifica-tion for chronic venous disorders: Consensus statement. J Vasc Surg +2004; 40:1248–52. +Gohel MS, Epstein DM, Davies AH. Cost-effectiveness of tradition-al and endovenous treatments for varicose veins. Br J Surg 2010; 97(12): 1815–23. +Michaels JA, Campbell WB, Brazier JE et al. Randomised clinical tri-al, observational study and assessment of cost-effectiveness of the treatment of varicose veins (REACTIV trial). Health Technol Assess 2006; 10(13): 1–196. +The National Institute for Health and Care Excellence. NICE CG92: Venous thromboembolism: reducing the risk for patients in hospital. 2010. www.nice.org.uk. +The National Institute for Health and Care Excellence. NICE CG168: Varicose veins: diagnosis and management. 2013. www.nice.org.uk. +‘VEINS’ Supplement. Phlebology 2015; 30(2), supplement. +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love +58 +Chapter + + + + + +Lymphatic disorders + + + + + + +Learning objectives +To understand: +• The main functions of the lymphatic system • The development of the lymphatic system + + + +• The various causes of limb swelling +• The aetiology, clinical features, investigations and treatment of lymphoedema + + + + + + +INTRODUCTION +The lymphatic system was first described by Erasistratus in Alexandria more than 2000 years ago. William Hunter in the late eighteenth century, was the first to describe the function of the lymphatic system. Starling’s pioneering work on the hydrostatic and haemodynamic forces controlling the move-ment of fluid across the capillary provided further insights into the function of the lymphatics. However, there is much about the lymphatic system that is not understood and debate continues over the precise aetiology of the most common +, +abnormality of the system, namely lymphoedema. + +ANATOMY AND PHYSIOLOGY OF +THE LYMPHATIC SYSTEM Functions +The principal function of the lymphatics is the return of pro-tein-rich fluid to the circulation through the lymphaticove-nous junctions in the jugular area. Thus, water, electrolytes, low molecular weight moieties (polypeptides, cytokines, growth factors) and macromolecules (fibrinogen, albumin, globulins, coagulation and fibrinolytic factors) from the inter-stitial fluid (ISF) return to the circulation via the lymphatics. Intestinal lymph (chyle) transports cholesterol, long-chain fatty acids, triglycerides and the fat-soluble vitamins (A, D, E and K) directly to the circulation, bypassing the liver. Lym-phocytes and other immune cells also circulate within the +lymphatic system. + + +Development and macroanatomy +In the human embryo lymph sacs develop at 6–7 weeks’ gesta-tion as four cystic spaces, one on either side of the neck and one in each groin. These cisterns enlarge and develop communi-cations that permit lymph from the lower limbs and abdomen to drain via the cisterna chyli into the thoracic duct, which in turn drains into the left internal jugular vein at its confluence with the left subclavian vein. Lymph from the head and right arm drains via a separate lymphatic trunk, the right lymphatic duct, into the right internal jugular vein. Lymphatics accom-pany veins everywhere except in the cortical bony skeleton and central nervous system, although the brain and retina possess cerebrospinal fluid and aqueous humour, respectively. +The lymphatic system comprises lymphatic channels, lym-phoid organs (lymph nodes, spleen, Peyer’s patches, thymus, tonsils) and circulating elements (lymphocytes and other mononuclear immune cells). Lymphatic endothelial cells are derived from embryonic veins in the jugular and perimeso-nephric areas from where they migrate to form the primary lymph sacs and plexus. Both transcription (e.g. Prox1) and growth (e.g. vascular endothelial growth factor-C (VEGF-C)) +factors are essential for these developmental events. + +Microanatomy and physiology Lymphatic capillaries +Lymphatics originate within the ISF space from special-ised endothelialised capillaries (initial lymphatics) or non- +endothelialised channels such as the spaces of Disse in the + + + +Erasistratus of Chios, c.300–250bc, of the Medical School at Alexandria in Egypt is regarded by many as the first physiologist. +William Hunter, 1718–1783, anatomist and obstetrician who became the first Professor of Anatomy at the Royal Academy of Arts, London, UK. He was the elder brother of John Hunter, the anatomist and surgeon. +Ernest Henry Starling, 1866–1927, Professor of Physiology, University College, London, UK. Josef Disse, 1852–1912, a German anatomist. +PART 10 | VASCULAR +996 CHAPTER 58 Lymphatic disorders + + + +liver. Initial lymphatics are unlike arteriovenous capillaries in +that: +● they are blind-ended; +● they are much larger (50 µm); +● they allow the entry of molecules of up to 1000 kDa in size because the basement membrane is fenestrated, tenuous or even absent and the endothelium itself possesses intra-and intercellular pores; +● they are anchored to interstitial matrix by filaments. In the resting state, initial lymphatics are collapsed. When ISF volume and pressure increases, initial lymphatics and their pores are held open by these filaments to facilitate +increased drainage. + +Terminal lymphatics +Initial lymphatics drain into terminal (collecting) lymphat-ics that possess bicuspid valves and endothelial cells rich in the contractile protein actin. Larger collecting lymphatics are surrounded by smooth muscle. Valves partition the lymphat-ics into segments (lymphangions) that contract sequentially +to propel lymph into the lymph trunks. + +Lymph trunks +Terminal lymphatics lead to lymph trunks, which have a struc-ture similar to that of veins, namely a single layer of endothe-lial cells, lying on a basement membrane overlying a media comprising smooth muscle cells that are innervated with sym-pathetic, parasympathetic and sensory nerve endings. About 10% of lymph arising from a limb is transported in deep lym-phatic trunks that accompany the main neurovascular bun-dles. The majority, however, is conducted against venous flow from deep to superficial in epifascial lymph trunks. Superficial trunks form lymph bundles of various sizes, which are located within strips of adipose tissue, and tend to follow the course +of the major superficial veins. + +Starling’s forces +The distribution of fluid and protein between the vascular system and ISF depends on the balance of hydrostatic and oncotic pressures between the two compartments (Star-ling’s forces), together with the relative impermeability of the blood capillary membrane to molecules over 70kDa. In health, there is net capillary filtration, which is removed by +the lymphatic system. + +Transport of particles +Particles enter the initial lymphatics through interendothe-lial openings and vesicular transport through intraendothelial pores. Large particles are actively phagocytosed by macro-phages and transported through the lymphatic system intra- +cellularly. + +Mechanisms of lymph transport +Resting ISF pressure is negative (-2 to -6mmH2O), whereas lymphatic pressures are positive, indicating that lymph flows against a small pressure gradient. It is believed that prograde +lymphatic flow depends upon three mechanisms: + +1 transient increases in interstitial pressure secondary to muscular contraction and external compression; +2 the sequential contraction and relaxation of lymphan-gions; +3 the prevention of reflux because of valves. +Lymphangions are believed to respond to increased lymph flow in much the same way as the heart responds to increased venous return, in that they increase their contractility and stroke volume. Contractility is also enhanced by noradren-aline, serotonin, certain prostaglandins and thromboxanes, and endothelin-1. Pressures of up to 30–50mmHg have been recorded in normal lymph trunks and up to 200mmHg in severe lymphoedema. Lymphatics may also modulate their own contractility through the production of nitric oxide and other local mediators. Transport in the thoracic and right lymph ducts also depends upon intrathoracic (respiration) and central venous (cardiac cycle) pressures. Therefore, car-diorespiratory disease may have an adverse effect on lym-phatic function. +In summary, in the healthy limb, lymph flow is largely due to intrinsic lymphatic contractility, although this is augmented by exercise, limb movement and external com-pression. However, in lymphoedema, when the lymphatics are constantly distended with lymph, these external forces +assume a much more important functional role. + +ACUTE INFLAMMATION OF THE +LYMPHATICS +Acute lymphangitis is an infection, often caused by Strepto-coccus pyogenes or Staphylococcus aureus, which spreads to the draining lymphatics and lymph nodes (lymphadenitis) where an abscess may form. Eventually this may progress to bacte-raemia or septicaemia. The normal signs of infection (rubor calor, dolor) are present and a red streak is seen in the skin along the line of the inflamed lymphatic (Figure 58.1). The part should be rested to reduce lymphatic drainage and ele-vated to reduce swelling, and the patient should be treated with intravenous antibiotics based upon actual or suspected +, +sensitivities. Failure to improve within 48 hours suggests + + + + + + + + + + + + + + +Figure 58.1 Acute lymphangitis of the arm. Erythematous streaks extend from the site of primary infection on the volar aspect of the forearm to epicondylar nodes at the elbow, and from there to enlarged and tender axillary lymph nodes. +PART 10 | VASCULAR Lymphoedema 997 + + + +inappropriate antibiotic therapy, the presence of undrained pus or the presence of an underlying systemic disorder (malig-nancy, immunodeficiency). The lymphatic damage caused by acute lymphangitis may lead to recurrent attacks of infection and lymphoedema; patients with lymphoedema are prone to +so-called acute inflammatory episodes (see below). + +LYMPHOEDEMA Definition +Lymphoedema may be defined as abnormal limb swelling caused by the accumulation of increased amounts of high protein ISF secondary to defective lymphatic drainage in the +presence of (near) normal net capillary filtration. + +The scope of the clinical problem +At birth, 1 in 6000 people will develop lymphoedema with an overall prevalence of 0.13–2%. The condition is not only associated with significant physical symptoms and complica-tions but is also a frequent cause of emotional and psycholog-ical distress, which can lead to difficulties with relationships, +education and work. + + +Summary box 58.1 + +Symptoms frequently experienced by patients with lymphoedema +●● Swelling, clothing or jewellery becoming tighter ●● Constant dull ache, even severe pain +●● Burning and bursting sensations ●● General tiredness and debility +●● Sensitivity to heat ●● ‘Pins and needles’ ●● Cramp +●● Skin problems, including flakiness, weeping, excoriation and breakdown +●● Immobility, leading to obesity and muscle wasting ●● Backache and joint problems +●● Athlete’s foot +●● Acute infective episodes + + +Despite this significant impact on quality of life, many sufferers choose not to seek medical advice because of embar-rassment and a belief that nothing can be done. Patients who do come forward for help, especially those with non-cancer-related lymphoedema, often find they have limited access to appropriate expertise and treatment. Lymphoedema is often misdiagnosed and mistreated by doctors, who frequently have a poor understanding of the importance of the condition, believ-ing it to be primarily a cosmetic problem in the early stages. However, making an early diagnosis is important because rela-tively simple measures can be highly effective at this stage and can prevent the development of disabling late disease, which is often very difficult to treat. It is also an opportunity for patients +to make contact with patient support groups. + + +Summary box 58.2 + +What every patient with lymphoedema should receive +●● An explanation of why the limb is swollen and the underlying cause +●● Guidance on skin hygiene and care and the avoidance of acute infective episodes +●● Antifungal prophylactic therapy to prevent athlete’s foot +●● Rapid access to antibiotic therapy if necessary, hospital admission for acute infective episodes +●● Appropriate instructions regarding exercise therapy ●● Manual lymphatic drainage (MLD) +●● Multilayer lymphoedema bandaging (MLLB) +●● Compression garments and, if appropriate, specialised footwear +●● Advice on diet +●● Access to support services and networks + + + +The severity of unilateral limb lymphoedema can be clas- +sified as: +● mild: <20% excess limb volume; +● moderate: 20–40% excess limb volume; +● severe: >40% excess limb volume. + +Pathophysiology +The ISF compartment (10–12 litres in a 70-kg man) consti-tutes 50% of the wet weight of the skin and subcutaneous tissues and, in order for oedema to be clinically detectable, its volume has to double. About 8 litres (protein concentration approximately 20–30 g/L, similar to ISF) of lymph is produced each day and travels in afferent lymphatics to lymph nodes. There, the volume is halved and the protein concentration doubled, resulting in 4 litres of lymph re-entering the venous circulation each day via efferent lymphatics. In one sense, all oedema is lymphoedema in that it results from an inability of the lymphatic system to clear the ISF compartment. However, in most types of oedema this is because the capillary filtration rate is pathologically high and overwhelms a normal lymphatic system, resulting in the accumulation of low-protein oedema fluid. In contrast, in true lymphoedema, when the primary problem is in the lymphatics, capillary filtration is normal and the oedema fluid is relatively high in protein. Of course, in a significant number of patients with oedema there is both abnormal capillary filtration and abnormal lymphatic drain-age, as in chronic venous insufficiency (CVI) for example. +Lymphoedema results from lymphatic aplasia, hypoplasia, dysmotility (reduced contractility with or without valvular insufficiency), obliteration by inflammatory, infective or neo-plastic processes, or surgical extirpation. Whatever the primary abnormality, the resultant physical and/or functional obstruc-tion leads to lymphatic hypertension and distension, with further secondary impairment of contractility and valvular com-petence. Lymphostasis and lymphotension lead to the accumu-lation in the ISF of fluid, proteins, growth factors and other active peptide moieties, glycosaminoglycans and particulate +matter, including bacteria. As a consequence, there is increased +PART 10 | VASCULAR +998 CHAPTER 58 Lymphatic disorders + + + +collagen production by fibroblasts, an accumulation of inflam-matory cells (predominantly macrophages and lymphocytes) and activation of keratinocytes. The end result is protein-rich oedema fluid, increased deposition of ground substance, subdermal fibrosis and dermal thickening and proliferation. Lymphoedema, unlike all other types of oedema, is confined to the epifascial space. Although muscle compartments may be hypertrophied because of the increased work involved in limb +movement, they are characteristically free of oedema. + +Classification +Two main types of lymphoedema are recognised: +1 Primary lymphoedema, in which the cause is unknown (or at least uncertain and unproven); it is thought to be caused by ‘congenital lymphatic dysplasia��. +2 Secondary or acquired lymphoedema, in which there is a +clear underlying cause. +Primary lymphoedema is usually further subdivided on the basis of the presence of family history, age of onset and lymph- +angiographic findings (Tables 58.1 and 58.2). + +Risk factors for lymphoedema +Although the true risk factor profile for lymphoedema is not currently known, a number of factors are thought to predis-pose an individual to its development and predict progres- +sion, severity and outcome of the condition (Table 58.3). + +Symptoms and signs +In most cases, the diagnosis of primary or secondary lymph-oedema can be made and the condition can be differentiated from other causes of a swollen limb on the basis of history and examination without recourse to complex investigation (Table 58.4). Unlike other types of oedema, lymphoedema characteristically involves the foot (Figure 58.2). The con-tour of the ankle is lost through infilling of the submalleolar depressions, a ‘buffalo hump’ forms on the dorsum of the foot, the toes appear ‘square’ because of confinement of footwear and the skin on the dorsum of the toes cannot be pinched because of subcutaneous fibrosis (Stemmer’s sign). Lymphoe-dema usually spreads proximally to knee level and less com- +monly affects the whole leg (Figure 58.3). In the early stages, + + + +TABLE 58.1 Aetiological classification of lymphoedema. + +Primary lymphoedema + + + +Secondary lymphoedema + +Congenital (onset <2 years old): sporadic; familial (Nonne–Milroy’s disease) +Praecox (onset 2–35 years old): sporadic; familial (Letessier–Meige’s disease) +Tarda (onset after 35 years old) Parasitic infection (filariasis) Fungal infection (tinea pedis) +Exposure to foreign body material (silica particles) +Primary lymphatic malignancy Metastatic spread to lymph nodes Radiotherapy to lymph nodes Surgical excision of lymph nodes Trauma (particularly degloving injuries) Superficial thrombophlebitis +Deep venous thrombosis + + +TABLE 58.2 Clinical classification of lymphoedema. + +Grade (Brunner) +Subclinical (latent) + +I + + +II + +III + +Clinical features +There is excess interstitial fluid and histological abnormalities in lymphatics and lymph nodes, but no clinically apparent lymphoedema +Oedema pits on pressure and swelling largely or completely disappears on elevation and bed rest +Oedema does not pit and does not significantly reduce upon elevation, positive Stemmer’s sign +Oedema is associated with irreversible skin Figure 58.2 The foot of a patient with typical lymphoedema. +changes, i.e. fibrosis, papillae + + + +Robert Stemmer, 1925–2000, phlebologist, Strasbourg, France. +PART 10 | VASCULAR Lymphoedema 999 + + + +TABLE 58.3 Risk factors for lymphoedema. Upper limb/trunk lymphoedema +Surgery with axillary lymph node dissection, particularly if extensive breast or lymph node surgery +Scar formation, fibrosis and radiodermatitis from postoperative axillary radiotherapy +Radiotherapy to the breast or to the axillary, internal mammary or subclavicular lymph nodes +Drain/wound complications or infection +Cording (axillary web syndrome) +Seroma formation +Advanced cancer +Obesity +Congenital predisposition +Trauma in an ‘at-risk’ arm (venepuncture, blood pressure measurement, injection) +Chronic skin disorders and inflammation +Hypertension +Taxane chemotherapy +Insertion of pacemaker +Arteriovenous shunt for dialysis +Air travel +Living in or visiting an area for endemic lymphatic filariasis + + +Lower limb lymphoedema +Surgery with inguinal lymph node dissection Postoperative pelvic radiotherapy +Recurrent soft tissue infection at the same site Obesity +Varicose vein stripping and vein harvesting +Genetic predisposition/family history of chronic oedema Advanced cancer +Intrapelvic or intra-abdominal tumours that involve or directly compress lymphatic vessels +Orthopaedic surgery +Poor nutritional status +Thrombophlebitis and chronic venous insufficiency, particularly post-thrombotic syndrome +Any unresolved asymmetrical oedema +Chronic skin disorders and inflammation +Concurrent illnesses such as phlebitis, hyperthyroidism, kidney or cardiac disease +Immobilisation and prolonged limb dependency +Air travel +Living in or visiting an area for endemic lymphatic filariasis + +Reproduced with permission from: Lymphoedema Framework. Best practice management of lymphoedema. International Consensus. London: MEP Ltd, 2006. © MEP Ltd 2006. + + +lymphoedema will ‘pit’ and the patient will report that the swelling is down in the morning. This represents a reversible component to the swelling, which can be controlled. Failure to do so allows fibrosis, dermal thickening and hyperkeratosis to occur. In general, primary lymphoedema progresses more slowly than secondary lymphoedema. Chronic eczema, fungal infection of the skin (dermatophytosis) and nails (onychomy-cosis), fissuring, verrucae and papillae (warts) are frequently seen in advanced disease. Ulceration is unusual, except in the presence of chronic venous insufficiency. +Lymphangiomas are dilated dermal lymphatics that ‘blis-ter’ onto the skin surface. The fluid is usually clear but may be blood-stained. In the long term, lymphangiomas throm-bose and fibrose, forming hard nodules that may raise concerns about malignancy. If lymphangiomas are <5cm across, they are termed lymphangioma circumscriptum, and if they are more widespread, they are termed lymphangioma diffusum. If they form a reticulate pattern of ridges then it has been termed lymphoedema ab igne. Lymphangiomas frequently weep (lym-phorrhoea, chylorrhoea), causing skin maceration, and they act as a portal for infection. Protein-losing diarrhoea, chylous asci-tes, chylothorax, chyluria and discharge from lymphangiomas +suggest lymphangectasia (megalymphatics) and chylous reflux. + +Ulceration, non-healing bruises and raised purple-red nodules should lead to suspicion of malignancy. Lymphangio-sarcoma was originally described in postmastectomy oedema (Stewart–Treves’ syndrome) and affects around 0.5% of patients at a mean onset of 10 years. However, lymphangiosarcoma can develop in any long-standing lymphoedema, but usually takes longer to manifest (20 years). It presents as single or multi-ple bluish/red skin and subcutaneous nodules that spread to +form satellite lesions, which may then become confluent. The + + +Summary box 58.3 + +Malignancies associated with lymphoedema ●● Lymphangiosarcoma (Stewart–Treves’ syndrome) ●● Kaposi’s sarcoma (human immunodeficiency virus) ●● Squamous cell carcinoma +●● Liposarcoma +●● Malignant melanoma +●● Malignant fibrous histiocytoma ●● Basal cell carcinoma +●● Lymphoma + + +Ab igne is Latin for ‘from fire’. +Fred Waldorf Stewart, 1894–1991, Chairman of Pathology, Memorial Sloan-Kettering Hospital, New York, NY, USA. An annual award was instituted in his name by the Department of Pathology called the Fred Waldorf Stewart Award. +Norman Treves, 1894–1964, American surgeon. He had a special interest in male breast cancer. Stewart and Treves reported this condition in a joint paper in 1948. +PART 10 | VASCULAR +1000 CHAPTER 58 Lymphatic disorders + + +TABLE 58.4 Differential diagnosis of the swollen limb. + +Non-vascular or lymphatic + + + + + + + + + + + + + + + + + + + + + + + +Venous + + + + + + + + + + + + + + + + + +Arterial + +General disease states + + + + + + +Local disease processes + + + + + + + + +Retroperitoneal fibrosis Gigantism + +Drugs + + +Trauma Obesity + +Deep venous thrombosis + + +Post-thrombotic syndrome + + + + +Varicose veins +Klippel–Trenaunay’s syndrome and other malformations + + + + +External venous compression + + +Ischaemia–reperfusion Arteriovenous malformation +Aneurysm + +Cardiac failure from any cause +Liver failure +Hypoproteinaemia due to nephrotic syndrome, malabsorption, protein-losing enteropathy +Hypothyroidism (myxoedema) +Allergic disorders, including angioedema and idiopathic cyclic oedema +Prolonged immobility and lower limb dependency Ruptured Baker’s cyst +Myositis ossificans +Bony or soft-tissue tumours Arthritis +Haemarthrosis +Calf muscle haematoma Achilles tendon rupture +May lead to arterial, venous and lymphatic abnormalities Rare +All tissues are uniformly enlarged Corticosteroids, oestrogens, progestagens +Monoamine oxidase inhibitors, phenylbutazone, methyldopa, hydralazine, nifedipine Painful swelling due to reflex sympathetic dystrophy +Lipodystrophy Lipoidosis +There may be an obvious predisposing factor, such as recent surgery The classical signs of pain and redness may be absent + +Swelling, usually of the whole leg, due to iliofemoral venous obstruction +Venous skin changes, secondary varicose veins on the leg and collateral veins on the lower abdominal wall +Venous claudication may be present +Simple primary varicose veins are rarely the cause of significant leg swelling Rare +Present at birth or develops in early childhood +Comprises an abnormal lateral venous complex, capillary naevus, bony abnormalities, hypo(a)plasia of deep veins and limb lengthening +Lymphatic abnormalities often coexist +Pelvic or abdominal tumour including the gravid uterus Retroperitoneal fibrosis +Following lower limb revascularisation for chronic and particularly chronic ischaemia May be associated with local or generalised swelling +Popliteal Femoral +False aneurysm following (iatrogenic) trauma + + + + +diagnosis is usually made late and confirmed by skin biopsy. Amputation offers the best chance of survival but, even then, +most patients live for less than 3 years. It has been suggested that + +lymphoedema leads to an impairment of immune surveillance and so predisposes to other malignancies, although the causal +association is not as definite as it is for lymphangiosarcoma. + + + +William Morrant Baker, 1839–1896, surgeon, St Bartholomew’s Hospital, London, UK, described these cysts in 1877. Mauirice Klippel and Paul Trenaunay, two French physicians, described this syndrome in 1900. +PART 10 | VASCULAR Primary lymphoedema 1001 + + +then, there may be considerable delay between the injury and the onset of oedema. Primary lymphoedema is much more common in the legs than the arms. This may be because of gravity and a bipedal posture, the fact that the lymphatic sys-tem of the leg is less well developed, or the increased suscepti-bility of the leg to trauma and/or infection. Furthermore, loss of the venoarteriolar reflex (VAR), which protects lower limb capillaries from excessive hydrostatic forces in the erect pos- +ture, with age and disease (CVI, diabetes), may be important. + +Classification +Primary lymphoedema is usually classified on the basis of apparent genetic susceptibility, age of onset or lymphan-giographic findings. None of these is ideal and the various classification systems in existence can appear confusing and conflicting, as various terms and eponyms are used loosely and interchangeably. This has hampered research and efforts to gain a better understanding of underlying mechanisms, the +effectiveness of therapy and prognosis. + + + + + + + + + + + + +Figure 58.3 The lower leg of a patient with typical lymphoedema. + + +PRIMARY LYMPHOEDEMA Aetiology +It has been proposed that all cases of primary lymphoedema are due to an inherited abnormality of the lymphatic system, sometimes termed ‘congenital lymphatic dysplasia’. However, it is possible that many sporadic cases of primary lymphoe-dema occur in the presence of a (near-)normal lymphatic sys-tem and are actually examples of secondary lymphoedema for which the triggering events have gone unrecognised. These might include seemingly trivial (but repeated) bacterial and/or fungal infections, insect bites, barefoot walking (silica), deep venous thrombosis (DVT) or episodes of superficial thrombo-phlebitis. In animal models, simple excision of lymph nodes and/or trunks leads to acute lymphoedema, which resolves within a few weeks, presumably because of the development of collaterals. The human condition can only be mimicked by +inducing extensive lymphatic obliteration and fibrosis. Even + +Genetic susceptibility +Primary lymphoedema can be familial or sporadic. In familial cases the genetic mutations can be inherited in an autoso-mal dominant, recessive or x-linked pattern. In the last few years the application of Next Generation Sequencing, which allows identification of genetic mutations, even in sporadic cases and smaller families, has increased our understanding of the genetic basis of lymphoedema. +So far, mutations in more than 20 genes have been linked to the development of primary lymphoedema. In individuals with familial mutations the penetrance of lymphoedema can often be incomplete leading to a large variability in clinical presentation. Involvement of other systems (cardiovascular respiratory, nervous, digestive) is common and can indicate a specific mutation. Presence of isolated lower limb lymph-oedema at birth, classically described as Milroy’s disease, suggests mutation in FMS-like tyrosine kinase 4 (FLT4)/ vascular endothelial growth factor receptor-3 (VEGFR-3) or VEGF-C. In these patients, reduced initial uptake and pres-ence of tortuous lymphatic tracts with evidence of rerouting in lymphangiography suggests mutation in the VEGF-C gene. The unique signs and lymphangiographic features of some of the genes currently known to cause primary lymphoedema are summarised in Table 58.5. Lymphoedema can also be present as a minor sign in some well-recognised syndromes. Primary lymphoedema is present in less than 10% of cases of tuberous sclerosis (TSC1 and 2 gene mutations). Lymphoedema has been reported in patients with Noonan syndrome in the pres-ence of PTPN11, SOS1 or KRAS mutations, and in patients with Turner syndrome (monosomy X) and capillary malfor-mation-arteriovenous malformation syndrome (RAS p21 +, +protein activator 1 (RASA1) mutation). + + + +William Forsyth Milroy, 1855–1942, Professor of Clinical Medicine, Columbia University, New York, NY, USA, described the condition in 1892. Jacqueline Noonan, b.1928, paediatric cardiologist, University of Kentucky, USA. +Henry Hubert Turner, 1892–1970, Professor of Medicine, The University of Oklahoma, OK, USA. +PART 10 | VASCULAR +1002 CHAPTER 58 Lymphatic disorders + + +TABLE 58.5 Genes associated with primary lymphoedema. + +Affected gene +FMS- like tyrosine kinase 4 (FLT4) + +Vascular endothelial growth factor C (VEGF-C) +Forkhead box C2 (FOXC2) + +Gap junction protein gamma-2 (GJC2) + +Gap junction protein alpha – 1 (GJA1) + +Kinesin family member 11 (KIF11) + + +Collagen and calcium-binding EGF domain-containing protein 1 (CCBE1) + +Homolog of drosophila FAT tumor suppressor 4 (FAT4) + +Gata-binding protein 2 (GATA2) + + + +Inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase gamma (IKBKG) + +Protein-tyrosine phosphatase non- receptor-type 14 (PTPN14) +SRY-Box 18 (SOX18) + +Hepatocyte growth factor (HGF) +Integrin subunit alpha 9 (ITGA9) + +Unique clinical sign(s) +Isolated lower limb oedema present at birth (Nonne–Milroy disease) +Milroy-like disease + +Distichiasis, cleft lip and/or palate + +Lymphoedema on all four extremities (Meige syndrome) +Oculodentodigital dysplasia + +Microcephaly, chorioretinopathy, lymphoedema and intellectual disability (MCLMR syndrome) +Microcephaly, unusual face and lymphoedema affecting all four limbs (Hennekam syndrome) +Microcephaly, unusual face and lymphoedema affecting all four limbs (Hennekam syndrome) +Myelodysplasia (Emberger syndrome) Warts, recurrent viral and/or bacterial infections +Predisposition to several cancers +Ectodermal dysplasia, anhidrotic, osteopetrosis and immunodeficiency (OLEDAID syndrome) +Choanal atresia + +Hypotrichosis, lymphoedema, telangiectasia and renal defects (HLTR syndrome) +Lymphangectasia +Fetal chylothorax + +Lymphoscintigraphy findings +Failure of initial lymphatic absorption (functional aplasia) +Reduced initial uptake with tortuous lymphatic tracts and evidence of rerouting +Hyperplastic lymphatic vessels with reflux of lymph +Significant reduction in initial uptake, normal lymphatic tracts +Significant reduction in initial uptake, normal lymphatic tracts +Failure of initial lymphatic absorption (functional aplasia) + +Abnormal lymph drainage in all four limbs and the thoracic duct + +Abnormal lymph drainage in all four limbs and the thoracic duct + +Hypoplasia of the lymphatics + + + + + +Most of the proteins encoded by these genes belong to the VEGF-C-VEGFR-3 ligand–receptor signalling complex and RAS/MAPK axis. The VEGF-C–VEGFR-3 signalling pathway is a major regulator of lymphangiogenesis. Down-stream, transcription factors such as FOXC2 and ITGA9 play a major role in the development of valves in the lym-phatic vessels. Mutations in the FLT4 gene that encodes VEGFR-3 protein were the first to be discovered and affect the tyrosine-kinase domain of the receptor, causing congen-ital primary lymphoedema. Mutations in the genes encod-ing the RAS/MAPK pathway cause several rare genetic conditions often called RASopathies. All RASopathies have some common clinical features such as distinct facial features and cardiac abnormalities. In addition to lymphoe-dema, these patients can also present with chylothorax and +chylous ascites. + + +Age of onset +Lymphoedema congenita (onset at or within 2 years of birth) is more common in males and is more likely to be bilateral and involve the whole leg. Lymphoedema praecox (onset from 2 to 35 years) is three times more common in females, has a peak incidence shortly after menarche, is three times more likely to be unilateral than bilateral and usually only extends to the knee. Lymphoedema tarda develops, by definition, after the age of 35 years and is often associated with obesity, with lymph nodes being replaced with fibrofatty tissue. The cause is unknown. Lymphoedema developing for the first time after 50 years should prompt a thorough search for underlying (pel-vic, genitalia) malignancy. It is worth noting that, in such patients, lymphoedema often commences proximally in the +thigh rather than distally (Figure 58.4). + + + +Max Nonne, 1861–1959, a neurologist of Hamburg, Germany, described this disease in 1891. Raoul Hennekam, contemporary, geneticist, Amsterdam Medical Centre, the Nederlands. Jean-Marie Emberger, contemporary, haematologist, Montpellier France. +, +Henri Meige,1866-1940, physician, La Salpêtrière, Paris, France, gave his description of the disease in 1899. +PART 10 | VASCULAR Secondary lymphoedema 1003 + + + + + + + + + + + + + + +Figure 58.4 This patient, in her sixth decade, presented with rapid onset of lymphoedema of the right leg. On further investigation she was found to have locally advanced bladder carcinoma. Note that unlike most cases of lymphoedema, the swelling is greater proximally than distally. + + +Lymphangiographic classification +Browse has classified primary lymphoedema on the basis of lymphangiographic findings (Table 58.6 and Figures 58.5 and 58.6). These findings may be related to the clinical pre-sentations described above. Some patients with lymphatic hyperplasia possess megalymphatics in which lymph or chyle refluxes freely under the effects of gravity against the physi-ological direction of flow. The megalymphatics usually end in thin-walled vesicles on the skin, serous surfaces (chylous ascites, chylothorax), intestine (protein-losing enteropathy), +kidney or bladder (chyluria) (Figure 58.7). + +SECONDARY LYMPHOEDEMA +This is the most common form of lymphoedema. There are several well-recognised causes including infection, inflamma- +tion, neoplasia and trauma (Table 58.7). + + + + + + +Figure 58.5 This patient presented with congenital lymphoedema of the right leg. The lymphangiogram shows lymphatic hypoplasia. + + +Filariasis +This is the most common cause of lymphoedema worldwide, affecting up to 100 million individuals. It is particularly prev-alent in Africa, India and South America where 5–10% of the population may be affected. The viviparous nematode Wuche-ria bancrofti, whose only host is man, is responsible for 90% of cases and is spread by the mosquito. The disease is associated +with poor sanitation. The parasite enters lymphatics from the + + +TABLE 58.6 Lymphangiographic classification of primary lymphoedema. + + +Age of onset Sex distribution Extent Laterality Family history Progression +Response to compression therapy +Comments + +Congenital hyperplasia (10%) Congenital +Male > female Whole leg Unilateral = bilateral Often positive Progressive Variable + +Lymphatics are increased in number; although functionally defective, there is usually an increased number of lymph nodes. May have chylous ascites, chylothorax and protein-losing enteropathy + +Distal obliteration (80%) Puberty (praecox) Female > male +Ankle, calf Often bilateral Often positive Slow +Good + +Absent or reduced distal superficial lymphatics. Also termed aplasia or hypoplasia + +Proximal obliteration (10%) Any age +Male = female Whole leg, thigh only Usually unilateral +No Rapid Poor + +There is obstruction at the level of the aortoiliac or inguinal nodes. If associated with distal dilatation, the patient may benefit from lymphatic bypass operation. Other patients have distal obliteration as well + + + +Sir Norman Leslie Browse, b.1931, formerly Professor of Surgery, the United Medical and Dental Schools of Guy’ and St Thomas’s Hospitals, London, UK; former President of the Royal College of Surgeons of England. +s +PART 10 | VASCULAR +1004 CHAPTER 58 Lymphatic disorders + + +TABLE 58.7 Classification of causes of secondary lymphoedema. + + + + + + + + + + + + + + + + + +Figure 58.6 This patient presented with lymphoedema of the right leg. A bipedal lymphangiogram demonstrated normal lymphatics in the right leg up to the inguinal nodes, but no progression of contrast above the inguinal ligament – a case of proximal obstruction. + +Classification +Trauma and tissue damage + + + + + + +Malignant disease + + + + +Venous disease + + + + +Infection + + + + +Inflammation + + + + + + +Endocrine disease Immobility and dependency + +Factitious + +Example(s) +Lymph node excision Radiotherapy +Burns +Variscose vein surgery/harvesting Large/circumferential wounds Scarring +Lymph node metastases Infiltrative carcinoma Lymphoma +Pressure from large tumours Chronic venous insufficiency Venous ulceration +Post-thrombotic syndrome Intravenous drug use Cellulitis/erysipelas Lymphadenitis Tuberculosis +Filariasis Rheumatoid arthritis Dermatitis +Psoriasis Sarcoidosis +Dermatosis with epidermal involvement +Pretibial myxoedema Dependency oedema Paralysis +Self-harm + +Reproduced with permission from: Lymphoedema Framework. Best practice management of lymphoedema. International Consensus. London: MEP Ltd, 2006. © MEP Ltd 2006. + + + + + + + + + + + + + +Figure 58.7 Lymphangiogram demonstrating reflux from dilated para-aortic vessels into the left kidney in a patient with filariasis who presented with chyluria. + +blood and lodges in lymph nodes, where it causes fibrosis and obstruction, due partly to direct physical damage and partly to the immune response of the host. Proximal lymphatics +become grossly dilated with adult parasites. The degree of + +Summary box 58.4 + +Features of filariasis Acute +●● Fever +●● Headache ●● Malaise +●● Inguinal and axillary lymphadenitis ●● Lymphangitis +●● Cellulitis, abscess formation and ulceration ●● Funiculo epididymo-orchitis + +Chronic +●● Lymphoedema of legs (arm, breast) ●● Hydrocoele +●● Abdominal lymphatic varices Chyluria +Lymphuria +PART 10 | VASCULAR Secondary lymphoedema 1005 + + + +oedema is often massive, in which case it is termed elephan-tiasis (Figure 58.8). Immature parasites (microfilariae) enter the blood at night and can be identified on a blood smear, in a centrifuged specimen of urine or in lymph itself. A comple-ment fixation test is also available and is positive in present or past infection. Eosinophilia is usually present. +Diethylcarbamazine destroys the parasites but does not reverse the lymphatic changes, although there may be some regression over time. Once the infection has been cleared, treat-ment is as for primary lymphoedema. Public health measures to reduce mosquito breeding, protective clothing and mosquito +netting may be usefully employed to combat the condition. + +Bacterial infection +Lymphangitis and lymphadenitis can cause lymphatic destruc-tion that predisposes to lymphoedema complicated by further acute inflammatory episodes. Interestingly, in such patients, lymphangiography has revealed abnormalities in the contra-lateral, unaffected limb, suggesting an underlying, possibly inherited, susceptibility. Lymphatic and lymph node destruc-tion by tuberculosis is also a well-recognised cause of lymph- +oedema, especially in resource-poor countries. + +Malignancy and its treatment +Treatment (surgery, radiotherapy) for breast carcinoma is the most common cause of lymphoedema in resource-rich coun-tries, but is decreasing in incidence as surgery becomes more conservative (see Chapter 53). Lymphoma may present with lymphoedema, as may malignancy of the pelvic organs and external genitalia. Kaposi’s sarcoma, developing in the course of human immunodeficiency virus (HIV)-related illness, may cause lymphatic obstruction and is a growing cause of lymph- +oedema in certain parts of the world. + + +Trauma +It is not unusual for patients to develop chronic localised or generalised swelling following trauma. The aetiology is often multifactorial and includes disuse, venous thrombosis and lym-phatic injury or destruction. Degloving injuries and burns are particularly likely to disrupt dermal lymphatics. Tenosynovitis can also be associated with localised subcutaneous lymphoe-dema, which can be a cause of troublesome persistent swelling +following ankle and wrist ‘sprains’ and repetitive strain injury. + + + + + + + + + + + + +Figure 58.8 Elephantiasis due to filariasis. + +Endemic elephantiasis +(podoconiosis) +This is common in the tropics and affects more than 500 000 people in Africa. The barefoot cultivation of soil composed of alkaline volcanic rocks leads to destruction of the peripheral lymphatics by particles of silica, which can be seen in macro-phages in draining lymph nodes. Plantar oedema develops in childhood and rapidly spreads proximally. The condition is +prevented and its progression is slowed by the wearing of shoes. + +Lymphoedema and chronic venous +insufficiency +It is important to appreciate the relationship between lymph-oedema and CVI as both conditions are relatively common and so often coexist in the same patient, and it can be diffi-cult to unravel which components of the patient’s symptom complex are caused by each. There is no doubt that superficial venous thrombophlebitis (SVT) and DVT can both lead to lymphatic destruction and secondary lymphoedema, espe-cially if recurrent. Lymphoedema is an important contributor to the swelling of the post-thrombotic syndrome. It has also been suggested that lymphoedema can predispose to DVT, and possible SVT, through immobility and acute inflamma-tory episodes. Certainly, tests of venous function (duplex ultrasonography, plethysmography) are frequently abnormal in patients with lymphoedema. +It is not uncommon to see patients (usually women) with lymphoedema in whom a duplex ultrasound scan has revealed superficial reflux (such reflux is present subclinically in up to one-third of the adult population). Although isolated, super-ficial venous reflux rarely, if ever, leads to limb swelling; such +patients are frequently misdiagnosed as having venous disease + + +Moritz Kaposi, 1837–1902, Hungarian dermatologist, described multiple pigmented skin sarcoma in 1872. +PART 10 | VASCULAR +1006 CHAPTER 58 Lymphatic disorders + + + +rather than lymphoedema, and are subjected mistakenly to var-icose vein surgery. Not only will such surgery invariably fail to relieve the swelling, it will usually make it worse as saphenofem-oral and saphenopopliteal ligation, together with saphenous stripping, will compromise still further drainage through the subcutaneous lymph bundles (which follow the major super- +ficial veins) and draining inguinal and popliteal lymph nodes. + +Miscellaneous conditions +Rheumatoid and psoriatic arthritis (chronic inflammation and lymph node fibrosis), contact dermatitis, snake and insect bites and retroperitoneal fibrosis are all rare but well-documented causes of lymphoedema. Pretibial myxoedema +is due to the obliteration of initial lymphatics by mucin. + +Conditions mimicking +lymphoedema Factitious lymphoedema +This is caused by application of a tourniquet (a ‘rut’ and sharp cut-off is seen on examination) or ‘hysterical’ disuse in +patients with psychological and psychiatric problems. +Immobility +Generalised or localised immobility of any cause leads to chronic limb swelling that can be misdiagnosed as lymphoedema; for example, the elderly person who spends all day (and sometimes all night) sitting in a chair (armchair legs), the hemiplegic +stroke patient and the young patient with multiple sclerosis. +Lipoedema +This presents almost exclusively in women and comprises bilateral, usually symmetrical, enlargement of the legs and, sometimes, the lower half of the body because of the abnormal deposition of fat. It may or may not be associated with gen-eralised obesity. There are a number of features that help to differentiate the condition from lymphoedema but, of course, lipoedema may coexist with other causes of limb swelling. It has been proposed that lipoedema results from, or at least is asso- +ciated with, fatty obliteration of lymphatics and lymph nodes. + + + +Summary box 58.5 + +Features of lipoedema that help differentiate it from lymphoedema +●��� Occurs almost exclusively in women +●● Onset nearly always coincides with puberty ●● Nearly always bilateral and symmetrical +●● Involvement of trunk +●● The feet are not involved, leading to an inverse shouldering effect at the malleoli +●● No pitting +●● No response to elevation or compression +●● No skin changes of lymphoedema (negative Stemmer’s sign) +●● MRI shows subcutaneous fat but no fluid accumulation + +INVESTIGATION OF +LYMPHOEDEMA +Are investigations necessary? +It is usually possible to diagnose and manage lymphoedema purely on the basis of history and examination, especially when the swelling is mild and there are no apparent compli-cating features. In patients with severe, atypical and multifac-torial swelling, investigations may help confirm the diagnosis, +inform management and provide prognostic information. + +‘Routine’ tests +These include a full blood count, urea and electrolytes, cre-atinine, liver function tests, thyroid function tests, plasma total protein and albumin, fasting glucose, C-reactive protein, urine dipstick including observation for chyluria, blood smear +for microfilariae, chest radiograph and ultrasound. + +Lymphangiography and isotope +lymphoscintigraphy +Direct lymphangiography remains the ‘gold standard’ for showing structural abnormalities of larger lymphatics and nodes (Figure 58.9). However, it can be technically difficult, it is unpleasant for the patient, it may cause further lymphatic injury and, largely, it has become obsolete as a routine method of investigation. +Isotope lymphoscintigraphy has largely replaced lymph-angiography as the primary diagnostic technique in cases of clinical uncertainty. Radioactive technetium-labelled protein or colloid particles are injected into an interdigital web space and specifically taken up by lymphatics, and serial radiographs are taken with a gamma camera. The technique provides a qualitative measure of lymphatic function rather than quantitative function or anatomical detail. Quantitative lymphoscintigraphy is performed using a dynamic (exercise) component in addition to the static test and provides infor- +mation on lymphatic transport. + +Computed tomography +A single, axial computed tomography (CT) slice through the mid-calf has been proposed as a useful diagnostic test for lymphoedema (coarse, non-enhancing, reticular ‘honeycomb’ pattern in an enlarged subcutaneous compartment), venous oedema (increased volume of the muscular compartment) and lipoedema (increased subcutaneous fat). CT can also be used to exclude pelvic or abdominal mass lesions. It can also be used to monitor response to treatment through serial mea- +surements of cross-sectional area and tissue density. + +Magnetic resonance imaging +Non-contrast magnetic resonance imaging (MRI) is also useful in identifying the classical circumferential reticular +pattern (honeycomb) within the epifascial compartment. +PART 10 | VASCULAR Management of lymphoedema 1007 + + + +Normal Congenital hyperplasia + +Distal obliteration (hypo/aplasia) + +Proximal obliteration (hypo/aplasia) with distal hyperplasia + +Proximal obliteration (hypo/aplasia) with distal obliteration + + +Thoracic duct + + +Nodes + +Para-aortic + +Iliac + +Femoral + + + + + + + +Figure 58.9 Lymphangiographic patterns of primary lymphoedema. + + + +Involvement of the subfascial compartment is more sugges-tive of venous disease. In lipoedema, there is fat deposition in the subcutaneous tissues without the typical reticular pat-tern and lymphatic congestion. Using the principles of spin labelling MRI, the lymphatic flow velocities can be assessed and monitored. MRI lymphography, using gadopentetate dimeglumine contrast, correlates substantially with lympho-scintigraphy and provides detailed anatomical and functional +status of lymphatic vessels and lymph nodes. + +Ultrasound +Ultrasound can provide useful information about venous +function including DVT and venous abnormalities. + +Lympho-fluoroscopies +More recently, fluorescent molecules, such as indocyanine green (ICG), are being used to image the superficial lym-phatic system. When excited using appropriate light energy, these molecules emit photons with a different wavelength that can be detected using dedicated near-infrared imaging devices, providing excellent anatomic and functional details of the lymphatic system. This imaging is increasingly being used to study the lymphatic system and the effects of some +therapies. + +Pathological examination +In cases in which malignancy is suspected, samples of lymph nodes may be obtained by fine-needle aspiration, needle core biopsy or surgical excision. Skin biopsy will confirm the diag- +nosis of lymphangiosarcoma. + +Limb volume measurement +While not helpful in the diagnosis of lymphoedema, limb volume measurement is a useful tool to determine severity of lymphoedema, guide management and assess response to treatment. Limb volume is typically measured at diagnosis, following intensive treatment and at follow-up. In unilateral limb swelling the affected side can be compared to the con-tralateral unaffected limb. In bilateral swelling the volume of both limbs is tracked with time. +Measurements are recorded in millilitres or expressed as a percentage of the normal limb. Water plethysmography (water displacement) is the ‘gold standard’ method, but is limited by practicalities of measurement and hygiene issues. Other options include circumferential limb measurements and perometry (infrared light beams measure the outline of +the limb to calculate volume). + +MANAGEMENT OF +LYMPHOEDEMA Overview +The evaluation of the lymphoedema patient needs to be ‘holistic’ and their care delivered by a multiprofessional team comprising physical therapists, nurses, orthotists, physicians (dermatologists, oncologists, palliative care specialists), sur-geons and social service professionals. Although surgery itself has a very small role, surgeons (especially those with breast and vascular interests) are frequently asked to oversee the management of these patients. Early diagnosis and institution of management are essential because at that stage relatively +simple measures can be highly effective and will prevent +PART 10 | VASCULAR +1008 CHAPTER 58 Lymphatic disorders + + + +the development of disabling late-stage disease, which is extremely difficult to treat. There is often a latent period of several years between the precipitating event and the onset of lymphoedema. The identification, education and treatment of such ‘at-risk’ patients can slow down, even prevent, the onset of disease. In patients with established lymphoedema, the three goals of treatment are to relieve pain, reduce swell- +ing and prevent the development of complications. + + +Summary box 58.6 + +Initial evaluation of the patient with lymphoedema +●● History (age of onset, location, progression, exacerbating and relieving features) +●● Past medical history including cancer history ●● Family history +●● Obesity (diet, height and weight, body mass index) +●● Complications (venous, arterial, skin, joint, neurological, malignant) +●● Assessment of physical, emotional and psychosocial symptoms +●● Social circumstances (mobility, housing, education, work) +●● Special needs (footwear, clothing, compression garments, pneumatic devices, mobility aids) +●● Previous and current treatment ●● Pain control +●● Compliance with therapy and ability to self-care + + + +Relief of pain +On initial presentation, 50% of patients with lymphoedema complain of significant pain. The pain is usually multifacto-rial and its severity and underlying cause(s) will vary depend-ing on the aetiology of the lymphoedema. For example, following treatment for breast cancer, pain may arise from the swelling itself (radiation and surgery induced); nerve (bra-chial plexus and intercostobrachial nerve), bone (secondary deposits, radiation necrosis) and joint (arthritis, bursitis, cap-sulitis) disease; and recurrent disease. Treatment involves the considered use of non-opioid and opioid analgesics, cortico-steroids, tricyclic antidepressants, muscle relaxants, antiepi-leptics, nerve blocks, physiotherapy and adjuvant anticancer therapies (chemo-, radio- and hormonal therapy), as well as measures to reduce swelling, if possible. In patients with non-cancer-related lymphoedema, the best way to reduce pain is to control swelling and prevent the development of complica-tions. Whatever the cause, pain is a somatopsychic experience that is affected by mood and morale. These issues are import-ant in patients with both cancer-related lymphoedema, who are concerned about recurrent disease and non-cancer-related disease, who often have poor self-esteem and problems with +body image and perception. + +Control of swelling +Physical therapy for lymphoedema, comprising bed rest, +elevation, bandaging, compression garments, massage and + +exercises, was first described at the end of the nineteenth century, and through the twentieth century various epony-mous schools developed. Although there is little doubt that physical therapy can be highly effective in reducing swelling, its general acceptance and practice has been hampered by a lack of proper research and confusing terminology. The cur-rent preferred term is decongestive lymphoedema therapy (DLT), which comprises two phases. The first is a short inten-sive period of therapist-led care and the second is a mainte-nance phase in which the patient uses a self-care regime with occasional professional intervention. The intensive phase comprises skin care, manual lymphatic drainage (MLD) and multilayer lymphoedema bandaging (MLLB), and exercises. The length of intensive treatment will depend upon the disease severity, the degree of patient compliance and the willingness and ability of the patient to take more responsi-bility for the maintenance phase. However, weeks rather than months should be the goal. + +Skin care +The patient must be carefully educated in the principles and practice of skin care. The patient should inspect the affected skin daily, with special attention paid to skinfolds, where maceration may occu . The limb should be washed daily; the use of bath oil, e.g. Balneum, is recommended as a moisturiser and the limb must be carefully dried afterwards. A hair drier on low heat is more effective and hygienic, and less traumatic, than a towel. If the skin is in good condition daily application of a bland emollient, e.g. aqueous cream, is recommended to keep the skin hydrated. If the skin is dry and flaky then a bland ointment, e.g. 50:50 white soft paraffin/liquid paraffin (WSP/LP), should be used twice daily and, if there is marked hyperkeratosis, a keratolytic agent, such as 5% salicylic acid, should be added. Many commercially available soaps, creams and lotions contain sensitisers, e.g. lanolin in E45 cream, and are best avoided as patients with lymphoedema are highly susceptible to contact dermatitis (eczema). Apart from caus-ing intense discomfort, eczema acts as an entry point for infection. Management comprises avoidance of the allergen (patch testing may be required) and topical corticosteroids. Fungal infections are common, difficult to eradicate and pre-dispose to acute inflammatory episodes. Chronic application of antifungal creams leads to maceration and it is better to use powders in shoes and socks. Ointment containing 3% benzoic acid helps prevent athlete’s foot and can be used safely over long periods. Painting at-risk areas with an anti-septic agent such as eosin may be helpful. Lymphorrhoea is uncommon but extremely troublesome. Management com-prises emollients, elevation, compression and sometimes cautery under anaesthetic. +r +Apart from lymphangiosarcoma, acute inflammatory epi-sodes are probably the most serious complications of lymph-oedema and frequently lead to emergency hospital admission. About 25% of primary and 5% of secondary lymphoedema patients are affected. Acute inflammatory episodes start rap-idly, often without warning or a precipitating event, with tingling, pain and redness of the limb. Patients feel ‘viral’ +PART 10 | VASCULAR Management of lymphoedema 1009 + + + + +Summary box 58.7 + +Skin care +●● Protect hands when washing up or gardening; wear a thimble when sewing +●● Never walk barefoot and wear protective footwear outside +●● Use an electric razor to depilate +●● Never let the skin become macerated +●● Treat cuts and grazes promptly (wash, dry, apply antiseptic and a plaster) +●● Use insect repellent sprays and treat bites promptly with antiseptics and antihistamines +●● Seek medical attention as soon as the limb becomes hot, painful or more swollen +●● Do not allow blood to be taken from, or injections to be given into, an affected arm (and avoid blood pressure measurement) +●● Protect the affected skin from sun (shade, high-factor sun block) ●● Consider taking antibiotics if going on holiday + + + + +and severe attacks can lead to the rapid onset of fever, rigors, headache, vomiting and delirium. Patients who have suffered previous attacks can usually predict the onset and many learn to carry antibiotics with them and self-medicate at the first hint of trouble.This may stave off a full-blown attack and prevent the further lymphatic injury that each acute inflam-matory episode causes. It is rarely possible to isolate a respon-sible bacterium, but the majority are presumed to be caused by group A β-haemolytic streptococci and/or staphylococci. The diagnosis is usually obvious but dermatitis, thrombophlebitis and DVT are in the differential diagnosis. Oral amoxycillin is the treatment of choice with erythromycin or clarithromy-cin in those with penicillin allergy. Flucloxacillin should be added for those with evidence of S. aureus infection (follic-ulitis, crusted dermatitis). Oral clindamycin is a second-line agent for those with failure to respond to initial therapy. Hos-pital admission is required for patients with: signs of septi-caemia; continuing or deteriorating systemic signs after 48 hours of antibiotic treament; unresolving or deteriorating local signs despite trials of first- and second-line antibiotics. Intravenous amoxycillin or benzyl penicillin with clinda-mycin in penicillin-allergic patients or as second-line therapy is most commonly recommended. Bed rest will reduce lym-phatic drainage and the spread of infection, elevation will reduce the oedema and heparin prophylaxis will reduce the risk of DVT. +Analgesia is often required but non-steroidal anti-inflammatory drugs (NSAIDs) should be avoided as they have been associated with increased complications, includ-ing necrotising fasciitis. Any lymphatic massage should be ceased in the presence of active infection. Amoxycillin can be taken by patients who self-medicate. The use of long-term prophylactic antibiotics is not evidence based, but penicillin V 500 mg daily is probably reasonable in patients who suffer two or more attacks per year. However, the benefits of scru-pulous compliance with physical therapy and skin care cannot +be underestimated. + +Manual lymphatic drainage +Several different techniques of MLD have been described and the details are beyond the scope of this chapter. However, they all aim to evacuate fluid and protein from the interstitial space and stimulate lymphangion contraction, with decon-gestion of impaired lymphatic pathways and development of collateral routes. The therapist should perform MLD daily; they should also train the patient (and/or carer) to perform a simpler, modified form of massage termed simple lymphatic drainage (SLD). In the intensive phase, SLD supplements MLD and, once the maintenance phase is entered, SLD will +carry on as daily massage. + +Multilayer lymphoedema bandaging +and compression garments +Elastic bandages provide compression, produce a sustained high resting pressure and ‘follow in’ as limb swelling reduces. However, the sub-bandage pressure does not alter greatly in response to changes in limb circumference consequent upon muscular activity and posture. By contrast, short-stretch ban-dages exert support through the production of a semi-rigid casing where the resting pressure is low, but changes quite markedly in response to movement and posture. This pressure variation produces a massaging effect within the limb and stimulates lymph flow. Whether the aim is to provide sup-port or compression, the pressure exerted must be graduated (100% ankle/foot, 70% knee, 50% mid-thigh, 40% groin). +Non-invasive assessment of the ankle–brachial pressure index (ABPI) using a hand-held Doppler ultrasound device is usually necessary prior to commencing any form of compres-sion therapy, as it is rarely possible to feel pulses in the lymph-oedematous limb. Standard MLLB and compression is used in patients with ABPI ≥0.8 and modified techniques with lower pressures in those with moderate arterial disease (ABPI 0.5– 0.8). MLLB is contraindicated in severe arterial insufficiency (ABPI <0.5), uncontrolled heart failure and severe peripheral neuropathy. +It is generally believed that non-elastic MLLB is preferable (and arguably safer) in patients with severe swelling during the intensive phase of DLT, whereas compression (hosiery, sleeves) is preferable in milder cases and during the mainte-nance phase. MLLB is highly skilled and to be effective and safe it needs to be applied by a specially trained therapist. It is +also extremely labour intensive, needing to be changed daily. + + + +Summary box 58.8 + +Effects of multilayer lymphoedema bandaging ●● Reduces oedema +●● Restores shape to the affected area +●● Reduces skin changes (hyperkeratosis, papillomatosis) ●● Eliminates lymphorrhoea +●● Supports inelastic skin +●● Softens subcutaneous tissues +PART 10 | VASCULAR +1010 CHAPTER 58 Lymphatic disorders + + + +Compression garments form the mainstay of manage-ment in most clinics. The control of lymphoedema requires higher pressures (30–40mmHg arm, 40–60mmHg leg) than are typically used to treat CVI. These may be reduced to 15–25mmHg in those with moderate arterial insuffi-ciency (Figure 58.10). Confusingly, the British (classes I: 14–17mmHg; II: 18–24mmHg; III: 25–35mmHg) and inter-national (USA) (classes I: 20–30mmHg; II: 30–40mmHg; III: 40–50mmHg; IV: 50–60mmHg) standards are different. The patient should put the stocking on first thing in the morning before rising. It can be difficult to persuade patients +to comply. Putting lymphoedema- grade stockings on and off + + + +Lower limb lymphoedema suitable for compression hosiery + +is difficult and many patients find them intolerably uncom-fortable, especially in warm climates. Furthermore, although intellectually they understand the benefits, emotionally they may find wearing them presents a greater body image problem than the swelling itself. +Enthusiasm for pneumatic compression devices has waxed and waned. Unless the device being used allows the sequen-tial inflation of multiple chambers up to >50mmHg, it will probably be ineffective for lymphoedema. The benefits to the patient are maximised and complications are minimised if these devices are used under the direction of a physical thera- +pist as part of an overall package of care. + + + +Peripheral arterial assessment (ABPI) + + + + + +NO COMPRESSION Refer to vascular specialist + + + +• Prophylaxis +• Early/mild lymphoedema • ISL stages 0–11 +• No or minimal shape distortion +• Maintenance • Palliation +• Elderly/arthritic +• Pressure sensitive • Lipoedema +• Controlled cardiac oedema • Dependency oedema +• Neurological decit + + + +Low: 14–21 mmHg +• Circular or at knit • Ready to wear +• All styles + +Severe arterial disease ABPI <0.5 + + + +• Moderate/severe lymphoe-dema +• ISL late stage II–III +• Some shape distortion* • Phlebolymphoedema +(healed ulcer) • Lipoedema +• Elderly/arthritic • Maintenance + + + + + + +Medium: 23–32 mmHg +• Circular or at knit or combination +• Ready to wear or custom made +• All styles + +Moderate arterial disease ABPI 0.5–0.8 + + + +• Severe lymphoedema • ISL stage III +• Shape distortion* +• Active patients and those at risk of oedema returning +• Phlebolymphoedema (active ulcer) +• Gross forefoot oedema • Retromalleolar swelling + + + + + + +High: 34–46 mmHg +• Flat or circular knit or combination +• Custom made (or ready to wear) +• All styles† + +ABPI >0.8 + + + + +• Severe complex lymphoe-dema +• ISL stage III +• Shape distortion* • ‘Pressure resistant’ +(i.e. medium or high pressure garments do not contain swelling) + + + + + + + +Very high: 49–70 mmHg +• Flat or circular knit or combination +• Custom made (or ready to wear) +• All styles† • MLLB + + + + +Successful outcome +• No increase in swelling +• No deterioration of skin, tissue density or shape +• Improvement in patient/carer involvement and self-management skills + + +*For patients with shape distortion, at knit hosiery is often preferable. +†Including inelastic adjustable compression device. + +ABPI, ankle brachial pressure index; ISL, International Society of Lymphology; +MLLB, multilayer lymphoedema bandaging. + + +Figure 58.10 Compression garments for lower limb oedema and lymphovenous oedema. Reproduced with permission from: Lymphoedema Framework. Best practice management of lymphoedema. International Consensus. London: MEP Ltd, 2006. © MEP Ltd 2006. +PART 10 | VASCULAR Management of lymphoedema 1011 + + + +Exercise +Lymph formation is directly proportional to arterial inflow and 40% of lymph is formed within skeletal muscle. Vigorous exercise, especially if it is anaerobic and isometric, will tend to exacerbate lymphoedema and patients should be advised to avoid prolonged static activities, for example carrying heavy shopping bags or prolonged standing. In contrast, slow, rhyth-mic isotonic movements (e.g. swimming) and massage will increase venous and lymphatic return through the production of movement between skin and underlying tissues (essential to the filling of initial lymphatics) and augmentation of the muscle pumps. Exercise also helps to maintain joint mobil-ity. Patients who are unable to move their limbs benefit from passive exercises. When at rest, the lymphoedematous limb should be positioned with the foot/hand above the level of the heart. A pillow under the mattress or blocks under the bottom of the bed will encourage the swelling to go down +overnight. + +Drugs +There are considerable, and scientifically inexplicable, dif-ferences in the use of specific drugs for venous disease and lymphoedema between different countries. The benzpyrones are a group of several thousand naturally occurring substances, of which the flavonoids have received the most attention. Enthusiasts will argue that a number of clinical trials have shown benefit from these compounds, which are purported to reduce capillary permeability, improve microcirculatory per-fusion, stimulate interstitial macrophage proteolysis, reduce erythrocyte and platelet aggregation, scavenge free radicals and exert an anti-inflammatory effect. Detractors will argue that the trials are small and poorly controlled with short fol-low-up and ‘soft’ end points, and that any benefits observed can be explained by a placebo effect. In the UK, oxerutins (Paroven®) are the only such drugs licensed for venous dis-ease and none has a license for lymphoedema. Diuretics are of no value in pure lymphoedema. Their chronic use is asso-ciated with side effects, including electrolyte disturbance, and should be avoided. +With increasing understanding of lymphangiogenesis pathways there is hope that specific pharmacological targets or gene therapy may become available in the future, but this +remains in the very early stages at present. + +Surgery +Only a small minority of patients with lymphoedema benefit from surgery. Operations fall into three categories: bypass pro- +cedures, liposuction and reduction procedures. + + +Bypass procedures +The rare patient with proximal ilioinguinal lymphatic obstruc-tion and normal distal lymphatic channels might benefit, at least in theory, from lymphatic bypass. A number of methods have been described, including the omental pedicle, the skin bridge (Gillies), anastomosing lymph nodes to veins (Neilubowicz) and the ileal mucosal patch (Kinmonth). More recently, direct lymphaticovenular anastomosis (LVA) has been carried out on vessels of 0.3–0.8mm diameter using super-microsurgical techniques. The procedures are technically demanding and not without morbidity. They are more often attempted in the upper limb following lymph node resection or radiotherapy for breast cancer. The outcomes are best in patients with earlier stages of lymphoedema, for whom the majority can be con-trolled with best medical therapy alone. In those with later stage disease who have failed conservative management, the +outcomes of LVA have generally been disappointing. + +Liposuction +Liposuction has been used in the treament of chronic lymph-oedema. It is usually reserved for patients who have progressed to non-pitting oedema. Case series reported thus far have shown promising results with more than 100% reduction in limb oedema volume, which can be maintained by ongoing use of compression hosiery for at least 1 year. While liposuc-tion appears to be safe, results of long-term efficacy and effects on the incidence of future lymphoedema complications (e.g. +infection) are awaited. + +Limb reduction procedures +These are indicated when a limb is so swollen that it interferes with mobility and livelihood. These operations are not ‘cos-metic’ in the sense that they do not create a normally-shaped leg and are usually associated with significant scarring. Four +operations have been described. + +SISTRUNK +A wedge of skin and subcutaneous tissue is excised and the wound closed primarily. This is most commonly carried out to +reduce the girth of the thigh. + +HOMANS +First, skin flaps are elevated, and then subcutaneous tissue is excised from beneath the flaps, which are then trimmed to size to accommodate the reduced girth of the limb and closed primarily. This is the most satisfactory operation for the calf (Figure 58.11). The main complication is skin flap necrosis. There must be at least 6 months between operations on the medial and lateral sides of the limb and the flaps must not +pass the midline. This procedure has also been used on the + + +Sir Harold Delf Gillies, 1882–1960, plastic surgeon, St Bartholomew’s Hospital, London, UK. Born in New Zealand, he is widely considered the ‘father of Plastic Surgery’, Gillies started his craft to better the lives of the victims of the First World War. Later he became a pioneer in ‘gender reassignment (sex-change) surgery’. He was joined in private practice by his cousin, the other world famous plastic surgeon, Sir Archibald McIndoe. He excelled in most sports – cricket, rowing, golf, and was an accomplished painter. +Jan Nielubowicz, 1915–2000, surgeon, Warsaw, Poland. +John Bernard Kinmonth, 1916–1982, surgeon, St Thomas’s Hospital, London, UK. +Walter Ellis Sistrunk, 1880–1933, Professor of Clinical Surgery, Baylor University College of Medicine, Dallas, TX, USA. John Homans, 1877–1957, Professor of Clinical Surgery, Harvard University Medical School, Boston, MA, USA. +PART 10 | VASCULAR +1012 CHAPTER 58 Lymphatic disorders + + +and deep systems. One skin flap is denuded (shaved of epi-dermis), sutured to the deep fascia and buried beneath the second skin flap (the so-called ‘buried dermal flap’) (Figure 58.12). This procedure has become less popular as pilonidal sinus formation is common. The cosmetic result is no better than that obtained with the Homans’ procedure and there is no evidence that the buried flap establishes any new lym- +phatic connections. + +CHARLES +This operation was initially designed for filariasis and involved excision of all of the skin and subcutaneous tissues down to the deep fascia, with coverage using split-skin grafts (Figure 58.13). This leaves a very unsatisfactory cosmetic result and graft failure is not uncommon. However, it does enable the surgeon to reduce greatly the girth of a massively swollen +limb. + + + + + + + +Figure 58.11 Homans’ procedure involves raising skin flaps to allow the excision of a wedge of skin and a larger volume of subcutaneous tissue down to the deep fascia. Surgery to the medial and lateral aspects of the leg must be separated by at least 6 months to avoid skin flap necrosis. + +Summary box 58.9 + +Surgical options for lymphoedema ●● Bypass procedures +Lymphatic bypass, e.g. omental pedicle, the skin bridge (Gillies) and the ilieal mucosal patch (Kinmonth) +Lymphaticovenular anastomosis (LVA) +Lymph nodes to vein anastomosis (Kinmonth) ●● Liposuction +●● Limb reduction procedures Sistrunk +Homans + + + +1 Subcutaneous 2 fat +Skin + +Deep fascia + +Bone + + + +3 4 + +Thompson +Charles + + + + + + + +Figure 58.12 A cross-sectional representation of Thompson’s reduction operation; red arrows illustrate the buried dermal flap sutured to deep fascia. + + +upper limb, but is contraindicated in the presence of venous +obstruction or active malignancy. + +THOMPSON +This is a modification of the Homans’ procedure aimed to +create new lymphatic connections between the superficial + + +Figure 58.13 The Charles procedure involves circumferential excision of lymphoedematous tissue down to, and including, the deep fascia followed by split-skin grafting. This procedure gives a very poor cosmetic result but does allow the surgeon to remove very large amounts of tissue and is particularly useful in patients with severe skin changes. + + +Frederick Thompson, 1910–1975, plastic surgeon, The Middlesex Hospital, London, UK. +Major-General Sir Richard Havelock Charles, 1887–1934, surgeon in The Indian Medical Service. +PART 10 | VASCULAR +Further reading 1013 + + + +Chylous ascites and chylothorax +These are associated with megalymphatics. The diagnosis may be obvious if accompanied by lymphoedema and lymphangi-oma. However, some patients develop chylous ascites and/or chylothorax in isolation, in which case the diagnosis can be confirmed by aspiration and the identification of chylomicrons in the aspirate. Cytology for malignant cells should also be car-ried out. A CT scan may show enlarged lymph nodes and CT with guided biopsy, laparoscopy or even laparotomy and biopsy may be necessary to exclude lymphoma or other malignancy. Lymphangiography may indicate the site of a lymphatic fis-tula that can be surgically ligated. Even if no localised lesion is identified, it may be possible to control leakage at laparotomy or even remove a segment of affected bowel. If the problem is too diffuse to be corrected surgically, a peritoneal venous shunt may be inserted, although occlusion and infection are import-ant complications. Medical treatment comprising the avoid-ance of fat in the diet and the prescription of medium-chain triglycerides (which are absorbed directly into the blood rather than via the lymphatics) may reduce swelling. Chylothorax is best treated by pleurodesis, but this may lead to death from +lymph-logged lungs as the excess lymph has nowhere to drain. + +Chyluria +Filariasis is the most common cause, with chyluria occurring in 1–2% of cases 10–20 years after initial infestation. It usually presents as painless passage of milky white urine, particularly after a fatty meal. The chyle may clot, leading to renal colic, and hypoproteinaemia may result. Chyluria may also be caused by ascariasis, malaria, tumour and tuberculosis. Intravenous urography and/or lymphangiography will often demonstrate the lymphourinary fistula. Treatment includes a low-fat and high-protein diet, increased oral fluids to prevent clot colic, and lap-arotomy and ligation of the dilated lymphatics. Attempts have also been made to sclerose the lymphatics either directly or via +instrumentation of the bladder, ureter and renal pelvis. + +Lymphocoele +Lymphocoele is a localised collection of lymph without a dis- +tinct epithelial lining. It results from leakage of lymph into the + + +soft tissues or body cavity secondary to surgical disruption of lymphatic vessels. They were first described as a postoperative complication after groin lymph node dissection for gynaeco-logical tumours. Other causes include radical prostatectomy, vascular reconstructive procedures, renal transplant and body contouring surgery after massive weight loss. Clinically, lymphocoele presents as a non-tender, non-pulsatile mobile lump and can be differentiated from a seroma or a localised postoperative oedema by the presence of lymph-rich fluid. Depending upon the site and size they can deform the adja-cent soft tissue structures or cause pressure symptoms. Com-plications include infection, chronic drainage (lymph fistula) and prolonged morbidity In the early stages, repeated needle aspiration and compression can be attempted. However, lymphocoeles are generally refractory to these simple mea-sures and may require insertion of drainage catheters, packing and compression or surgical excision. For a successful surgi-cal result, combining preoperative lymphoscintigraphy with on-table isosulfan blue dye injection technique enhances the chance of accurately identifying the feeding lymphatic channels and ligation. After resecting a large lymphocoele the residual dead space may need muscle flap transposition to reduce recurrence. Other emerging concepts include perform-ing prophylactic muscle flaps in high-risk groups and negative +. +pressure wound therapy. + +FURTHER READING +Cheng M, Chang DW, Patel KM. Principles and practice of lymphedema surgery. Amsterdam: Elsevier, 2016. +Greene AK, Slavin SA, Brorson H (eds). Lymphedema: presentation, diag-nosis and treatment. New York: Springer, 2015. +International Lymphoedema Framework. Best practice for the management of lymphoedema, 2nd edn. Position document. London: ILF, 2012. +Lee B-B, Bergan J, Rockson SG. Lymphedema: a concise compendium of theory and practice. New York: Springer, 2011. +Lymphoedema Framework. Best practice for the management of lymphoede-ma. International consensus. London: MEP Ltd, 2006. +Szuba A, Rockson SG. Lymphedema: anatomy, physiology and patho-genesis. Vasc Med 1997; 2: 321–6. +Twycross R, Jenns K, Todd J. Lymphoedema. Oxford: Radcliffe Medical Press, 2000. +Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love +Bailey & Love + +Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love Bailey & Love +Bailey & Love Bailey & Love +PART +11 + + + + +Abdominal + + + + + +59 History and examination of the abdomen .....................................1016 60 Abdominal wall, hernia and umbilicus............................................1022 +61 The peritoneum, omentum, mesentery and retroperitoneal space............................................................................................................1047 +62 The oesophagus......................................................................................1067 63 Stomach and duodenum.....................................................................1106 64 Bariatric and metabolic surgery ........................................................1144 65 The liver ......................................................................................................1153 66 The spleen.................................................................................................1176 67 The gall bladder and bile ducts.........................................................1188 68 The pancreas............................................................................................1212 69 The small intestine..................................................................................1240 70 The large intestine..................................................................................1258 71 Intestinal obstruction.............................................................................1280 72 The vermiform appendix......................................................................1299 73 The rectum.................................................................................................1318 74 The anus and anal canal.......................................................................1339 +Bailey & Love +Bailey & Love +59 +Chapter + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + + + +History and examination of the abdomen + + + + +Learning objectives +To understand: To identify: + +• The pathway for clinical diagnosis of a patient presenting with an abdominal complaint +• The importance of recognising the organ or system responsible for the clinical features +• The pathophysiological basis of common abdominal symptoms and signs + +• Leading questions based on the organ or system affected and reach the most likely differential diagnosis +• The relevant physical signs and come to an appropriate clinical diagnosis + + + + + +INTRODUCTION +Abdominal symptoms are a frequent cause for surgical con-sultation. The underlying cause may be acute, presenting with the euphemistically termed ‘acute abdomen’, subacute, indicating an evolving disorder, or long-standing, suggesting a functional or degenerative condition. Occasionally symp-toms are due to disorders outside the abdomen, in which case the term ‘referred’ is used, for example epigastric pain experi-enced as a result of a myocardial infarction. +At first presentation, a detailed clinical history and care-ful clinical examination are essential to establish a differential diagnosis which, in turn, leads to appropriate triage into urgent +and non-urgent investigation and subsequent treatment. + +GATHERING INFORMATION +The experienced clinician will recognise the acuity and sever-ity of the patient’s condition even before a history has been taken. Initial observation provides clues to the direction that the history should take: general appearance, gait, position in bed, facial expression and tone of speech all provide useful hints. In an acute presentation, it is important to realise that the patient will feel anxious and vulnerable and may well be in severe pain; therefore, the clinician should introduce him-/ herself, try to comfort the patient and gain his or her con-fidence. This is particularly important in a busy emergency +department where the patient is only one among many. + +Obtaining a history +The patient’s history is obtained during the initial part of +the consultation. It is crucial that the patient be allowed to + + +explain the presenting complaint without interruption, after which carefully directed questions are used to further refine the history; this will then guide subsequent clinical examina- +tion and investigation. + +Presenting complaint +To establish the presenting complaint one should start with an open question inviting the patient to explain the reason for seeking medical advice. In the acute situation, pain is the most common presenting feature. The classic features of site, nature, onset, duration, radiation, and aggravating or reliev-ing features of the pain should be established. In a non-acute presentation anorexia, weight loss, jaundice, altered bowel habit, blood loss and fatigue are all features that should be enquired about. +Once the patient has finished explaining the problem, direct questions should be used to support or exclude possible diagnoses. It should be possible to identify a likely organ or system responsible for the patient’s symptoms and come to a +differential diagnosis. + +Past history +The past history is important because it may have a bearing on the diagnosis and management. A history of previous similar episodes or past abdominal surgery often guides the diagno-sis, for example adhesive small bowel obstruction in a patient with a history of laparotomy or recurrent left iliac fossa pain in a patient with a past history of diverticulitis. Some symp-toms and signs may be due to cardiac, respiratory haemato-logical or musculoskeletal conditions, for example abdominal +, +pain in sickle cell crisis. +PART 11 | ABDOMINAL +Clinical presentation of abdominal problems 1017 + + + +Drug history and allergies +Some drugs will have an effect on the symptoms and signs or may have to be discontinued before surgery. For example, a patient with bleeding who is taking a b-blocking drug will not have tachycardia proportionate to the blood loss, a patient taking long-term corticosteroids will need intravenous steroid supplementation to prevent an adrenal crisis in the periop-erative period and a patient taking anticoagulant drugs may +require reversal of the effects before surgical intervention. +Social history +The use of alcohol and illicit drugs, smoking and occupation is important. Family background will inform potential diffi- +culties with later discharge planning. +Family history +It is important to establish a family history of similar or related conditions, particularly cancer, inflammatory bowel disease, endocrine disease (e.g. hyperparathyroidism causing hyper-calcaemia or renal calculi) and genetic disorders, including +adverse reactions to anaesthetics or medications. +Review of the systems +A system review should highlight any comorbid disease, such +as cardiac, vascular, respiratory or endocrine problems. + +Summary box 59.1 + +Principles of history taking +●● Identify the reason for consultation – the presenting complaint ●● Determine the onset, duration and evolution of the symptoms ●● Deduce the most likely organ or system affected +●● Refine the history with relevant direct questions +●● Establish relevant past, social, family, drug and allergy history ●● Complete with a thorough review of other systems +●● Devise a list of differential diagnoses + +CLINICAL PRESENTATION OF +ABDOMINAL PROBLEMS +Pain, weight loss, anorexia or vomiting, jaundice, alteration of bowel habit, and blood loss or anaemia are the common clinical presentations of abdominal pathology It is import-ant to be vigilant about insidious presentations of malignan-cies. Classic examples are right colon cancer presenting with symptoms of anaemia, metastatic liver cancer with weight loss, gastric cancer with loss of appetite, ovarian cancer with abdominal distension, and malignant obstruction of the +. +extrahepatic biliary tree presenting with jaundice. + +Summary box 59.2 + +Sources of abdominal symptoms +●● Abdominal pain, weight loss, anorexia or vomiting, jaundice, alteration of bowel habit, and blood loss or anaemia are the common clinical presentations of abdominal pathology +●● Occult malignancies may have atypical presentation + +Abdominal pain +Pain is the most common of all abdominal symptoms and may be due to inflammatory, infective, obstructive, neoplas-tic or ischaemic pathology. Sometimes no organic cause can be found, a situation often labelled ‘functional abdominal pain’ for want of a better term. Improved understanding of pain pathways and the relationship with the gastrointestinal microbiome is likely to provide a more precise diagnosis, par-ticularly in common ‘functional’ disorders such as non-ulcer dyspepsia and irritable bowel syndrome (IBS). +It is essential to establish the site, nature and radiation of the pain, the rapidity of onset, and associated or relieving fea-tures. Thus biliary colic will classically result in colicky pain in the right upper quadrant of the abdomen which radiates to the angle of the scapula and is associated with food intake (which results in cholecystokinin release and gallbladder contraction). Acute pancreatitis often has an abrupt onset of severe epigastric pain radiating to the back, which may be similar to pain emanat-ing from peptic ulcer perforation or a leaking aortic aneurysm. Intestinal colic is most frequently associated with periumbilical pain and abdominal distension: the more distal in the intestine the pathology, the greater the degree of distension. Vomiting is an early feature of proximal small bowel obstruction, whereas absolute constipation is an early feature of colonic obstruction. Renal or ureteric colic is intense, located in the flanks and radi-ating towards the lower midline. It is usually associated with either macroscopic or microscopic haematuria. +With regard to altered bowel habit, the onset, nature, dura-tion, type of alteration (constipation or diarrhoea) and its rela-tionship to abdominal pain will help to differentiate organic pathology causing obstruction or inflammation (colon cancer or inflammatory bowel disease [IBD]) from functional conditions such as IBS. When patients complain of diarrhoea, they may imply different meanings – some use the term for loose stools, others may mean frequent but normal stools. A long-standing increase in frequency of stools, with left-sided abdominal pain before defecation, that eases after defecation, is suggestive of IBS. However, if such symptoms are of recent onset or are asso-ciated with blood or mucus in the stools, colonic carcinoma or IBD is more likely. A history of progressive change in bowel habit with an acute presentation with abdominal pain, dis-tension and absolute constipation suggests acute-on-chronic intestinal obstruction, often from a stenotic left colon cancer. Marked distension with tenderness over the caecal area sug- +gests a closed-loop obstruction with impending caecal rupture. + +Summary box 59.3 + +Classic presentations of abdominal pathology +●● Obstructive and inflammatory pathology must be excluded in patients with abdominal pain and altered bowel habit +●● Closed-loop obstruction with tenderness in the right iliac fossa is indicative of imminent caecal rupture +●● Caecal cancer classically presents with anaemia +●● Patients who have had previous abdominal surgery may have adhesions +●● Check carefully for small incarcerated hernias, particularly femoral, in obese patients +PART 11 | ABDOMINAL +1018 CHAPTER 59 History and examination of the abdomen + + + +PATHOPHYSIOLOGICAL BASIS OF COMMON ABDOMINAL +SYMPTOMS AND SIGNS +The abdominal wall and parietal peritoneum are innervated by the somatic nervous system, whereas the abdominal organs and visceral peritoneum are innervated by the autonomic nervous system. Therefore pain may change in its character and distribution as the underlying pathology evolves. Visceral pain from the foregut is generally felt in the epigastrium, in the periumbilical area from the mid-gut and in the suprapubic area from the hind-gut. +The skin and the muscles of the abdominal wall are sup-plied by the lateral and anterior cutaneous branches of the lower six intercostal nerves, the iliohypogastric and ilioin-guinal nerves (Figure 59.1). The dermatome levels of the xiphoid process, umbilicus and pubis are T7, T10 and T12, respectively. The parietal peritoneum is supplied segmentally by the same nerves that innervate the overlying muscles. The central part of the diaphragmatic peritoneum is supplied by the phrenic nerve (C4); therefore, pain arising in this region is referred to the tip of the shoulder as it has the same segmental supply. The peripheral rim of the diaphragmatic peritoneum is supplied by the intercostal nerves. The obturator nerve is +the principal nerve supply of the pelvic parietal peritoneum. +Pain from the viscera is principally due to ischaemia, muscle spasm or stretching of the visceral peritoneum. Unlike somatic pain, autonomic pain is deep and poorly localised. This pain is transmitted via sympathetic fibres and so is referred to the appropriate somatic distribution of that +nerve root from T1 to L2. However, when an inflamed organ + + + + + + + + + + + +T7 + + +Iliohypogastric nerve – L1 T10 Ilioinguinal nerve – L1 T12 +L1 + + + +Figure 59.1 Distribution of the anterior abdominal wall dermatome and nerves. + + +Summary box 59.4 + +Nerves responsible for abdominal pain +●● Abdominal wall and parietal peritoneum are supplied by the somatic nerves +●● Abdominal organs and the visceral peritoneum are supplied by the autonomic nervous system +●● Skin, muscles and parietal peritoneum are supplied by the iliohypogastric and ilioinguinal nerves and the lower six intercostal nerves +●● Afferent pain fibres from the abdominal organs and visceral peritoneum travel with sympathetic nerves + + +touches the parietal peritoneum, the pain becomes sharp and localises to the appropriate segmental dermatome of the abdominal wall. Pain arising from the parietal peritoneum may radiate to the back or the front along the appropriate dermatome. This referral pattern is classically seen in acute cholecystitis when an inflamed gall bladder touches the pari-etal peritoneum. Pain then radiates round to the back along the involved dermatome. The overlying muscle and skin are supplied by the same nerve root, so, when the patient takes a deep breath, the tenderness in the right subcostal region is markedly increased, causing the patient to stop breathing; this is Murphy’s sign. In children with abdominal pain, who hold their right hip in a flexed position to obtain relief from the pain, one should suspect retrocaecal appendicitis causing +irritation of the psoas muscle. + +Summary box 59.5 + +Specific characteristics of abdominal pain +●● Visceral pain arises from ischaemia, muscle spasm or stretching of the visceral peritoneum +●● Autonomic pain, deep and poorly localised, is referred to the equivalent somatic distribution of that nerve root from T1 to L2 +●● When an inflamed organ touches the parietal peritoneum, pain is then localised to the segmental dermatome of the abdominal wall +●● The pain in the parietal peritoneum may radiate to back or front along the dermatome + + +Obstruction +Central colicky abdominal pain is a classic presentation of small bowel obstruction. The central distribution is because of the segmental nerve supply of the mid-gut. When the per-istaltic waves hit an obstruction, the contractions increase to overcome the resistance, producing the colic. The pain reaches a crescendo and then disappears in minutes when the +peristaltic wave passes. This is different from that of biliary + + +John Apley, 1909–1980, consultant paediatrician, Bristol, UK. The further away the chronic abdominal pain in a child is from the umbilicus, the more likely an organic cause. +John Benjamin Murphy, 1857–1916, Professor of Surgery, Northwestern University, Chicago, IL, USA, described his sign in 1903. He was the son of Irish immigrants fleeing the potato famine in Ireland, and was known as the ‘Stormy Petrel’ of American surgery, demonstrating the benefit of appendicectomy over conservative treatment among many things. +PART 11 | ABDOMINAL Examination of the abdomen 1019 + + + +colic. When the gall bladder contracts against a stone, pain is relatively insidious in onset and reaches its peak in about half an hour and then eases off. A basal pain persists between the +bouts of colic. + +Summary box 59.6 + +Colicky abdominal pain +●● Pain of ‘small bowel colic’ comes in waves and disappears completely in minutes when the peristaltic wave ceases +●● Pain of biliary colic is insidious in onset, reaches the peak in half an hour or so and does not ease off completely between spasms +●● Pain of ureteric colic is intense lasting one to two minutes of the ureter + + +Rupture and perforation of organs + + + +1 +Hepatitis Cholecystitis + +4 + +Renal and ureteric pain + + + +7 +Appendicitis + + +2 +Peptic ulcer Pancreatitis + + + +5 (back) Bowel +obstruction + +Aortic aneurysm ( back) +8 + +Pelvic pain + + + +3 +Splenic injury + +6 + +Renal and ureteric pain + + + +9 +Diverticulitis + + + +The urinary bladder, gall bladder and gastrointestinal tract are hollow organs containing fluid. The gastrointestinal system also contains faeces, air and a high concentration of organisms. Trauma, ischaemia or tissue ulceration may cause perforation, with resulting leak of luminal contents, and peri-tonitis, with resulting severe abdominal pain. This may be localised to the area immediately adjacent to the perforation (for example, in a localised perforation of an appendix) or more generalised. The initial site of onset of the pain may give a clue as to the organ involved and so help with the dif-ferential diagnosis. For example, the diagnosis of a perforated peptic ulcer is supported by a past history of ulcer-type pain followed by sudden onset of upper abdominal pain. +The abdomen is divided into nine areas for ease of descrip-tion (Figure 59.2). These regions are demarcated by the mid-clavicular lines in the vertical axis and by the transpyloric and transtubercular lines in the horizontal axis. Figure 59.2 also indicates some of the common organs and pathological processes that commonly cause pain experienced in these +regions. + +EXAMINATION OF THE +ABDOMEN +Abdominal examination must be preceded by a detailed gen-eral examination of the patient as a whole. Physical exam-ination should be systematic using the following sequence: +inspection, palpation, percussion and auscultation. + +General examination +When examining the abdomen it is essential to obtain +the patient’s prior consent and to ensure the presence of a + +(Ovarian cysts, pelvic inammatory disease, etc.) + +Figure 59.2 Nine sites of abdominal pain: 1, right subcostal; 2, epi-gastrium; 3, left subcostal; 4, right flank; 5, periumbilical; 6, left flank; 7, right iliac fossa; 8, suprapubic/hypogastrium; 9, left iliac fossa. (From Bailey and Love, 25th edn., courtesy of Mr Simon Paterson-Brown, Consultant Surgeon, Royal Infirmary of Edinburgh.) + +chaperone if appropriate. The examination should be per-formed in a comfortable environment that ensures privacy. The patient must be lying flat but without causing distress (this may require provision of a pillow) and the abdomen should be adequately exposed. +The examination should be performed sequentially, begin-ning with general inspection looking for evidence of weight loss, dehydration, anaemia, jaundice or abnormal pigmen-tation. Examination of the hands may provide evidence of anaemia or chronic liver disease whereas examination of the head and neck may identify features indicative of liver dis-ease or lymph adenopathy (particularly left supraclavicular) suggestive of intra-abdominal malignancy. The patient’s vital signs (heart rate, blood pressure, respiratory rate and body temperature) should always be noted. It is wise in the elective +setting also to record the patient’s weight at this point. + +Inspection +Scars, abdominal distension, visible peristalsis or abdomi-nal masses, dilated veins, pulsation or abdominal wall swell-ing suggestive of hernia should all be carefully sought. In an abdominal emergency look for Grey Turner’s sign – skin dis-coloration of the flanks due to retroperitoneal haemorrhage in severe acute pancreatitis and leaking abdominal aortic aneu- +rysm. Cullen’s sign – discoloration around the umbilicus – may + + +Vermiculation: meaning wave-like contraction similar to a worm, particularly as referring to peristalsis of the ureter (from the Latin vermiculus – little worm). George Grey Turner, 1877–1951, Professor of Surgery, the University of Durham (1927–1934) and at the Royal Postgraduate Medical School, Hammersmith Hospital, London, UK (1935–1946). He had the surgical club, Grey Turner Surgical Club, named after him. It is said that he dressed shabbily so that when his friends met him they used to ask him to ‘mend the clock’. He had a habit of keeping his cup of tea warm by covering it with his bowler hat! +Thomas Stephen Cullen, 1868–1953, Professor of Gynecology, the Johns Hopkins University Baltimore, MD, USA, described the sign in ruptured ectopic pregnancy in 1916. +, +PART 11 | ABDOMINAL +1020 CHAPTER 59 History and examination of the abdomen + + + + + + + + + + + + + + + + + +Figure 59.3 Cullen’s and Grey Turner’s sign of skin discoloration of flanks and around the umbilicus (courtesy of Mr Pradip Datta, Honor-ary Consultant Surgeon, Wick, Scotland). + + +indicate severe acute pancreatitis, ruptured ectopic pregnancy or trauma to the liver. In these situations, blood tracks to the umbilicus along the ligamentum teres (Figure 59.3). +In a patient with acute abdominal pain, it is important to observe if the abdominal wall moves with respiration. A thin patient with diffuse peritonitis may be unable to lie flat and the abdominal wall will have a ‘scaphoid’ appearance due to protective contraction of the rectus abdominis muscles. It is often appropriate to ask the patient to cough gently – this will evoke sudden discomfort in the area of underlying peritoneal irritation (equivalent to eliciting rebound tender-ness, but not as distressing for the patient). A cough will also help to identify an abdominal wall hernia, if present. Visible abdominal masses, mobility on respiration and peristalsis are all best observed if the clinician kneels by the patient’s bed so that the observer’s eye is at the level of the patient’s anterior abdominal wall. The same position is useful during palpation +for abdominal masses (Figure 59.4). + + + + + + + + + + + + + + + + + +Figure 59.4 Eye at the level of patient’s abdominal wall. + +Palpation +Palpation should be performed in a systematic manner check-ing all nine regions of the abdomen (see Figure 59.2). Pal-pation should start in the region furthest away from the site of pain and the patient instructed to let the examiner know if tenderness is elicited. The examination should be gentle and the hands warm. The patient’s facial expression will imme-diately reveal discomfort. Superficial palpation is followed by deep palpation if tenderness will allow. Finally, palpation during respiration is performed to identify the lower margins +of the liver and spleen as they move with respiration. +Signs of parietal peritoneal irritation (tenderness, guarding, rebound tenderness, rigidity) +In the presence of abdominal pain, the degree of abdominal wall rigidity and involuntary guarding should be assessed. Guarding represents contraction of the abdominal wall mus-cles over the area of pain. This might occur ‘voluntarily’ when the patient wishes to avoid the pain from examination, or ‘involuntarily’ when the muscles go into spasm as the inflamed viscus touches the parietal peritoneum. This produces a reflex spasm of the overlying abdominal wall muscles. The presence of rebound tenderness indicates underlying peritoneal inflam-mation and is examined best using gentle percussion, although pain on coughing is also found when there is rebound tender-ness. When the underlying peritoneal inflammation becomes generalised, the abdomen is ‘board-like rigid’ to palpation, and selective tenderness can no longer be elicited. This sign +represents widespread involuntary guarding. +Abdominal masses +A mass arising from the anterior abdominal wall will usually be mobile when the patient is relaxed. On contracting the abdominal wall muscles (ask the patient to lift his or her legs with the knees extended, or perform Valsalva’s manoeuvre for laterally placed swellings), lumps superficial to the abdominal wall muscles will become more obvious, and those attached to the deep fascia will become less mobile. Those arising within the muscle layer will become fixed and remain unchanged in size. Lumps arising deep to the abdominal wall (i.e. within the peritoneal cavity or behind the peritoneum) will become impalpable or less prominent on tensing the anterior abdom-inal wall muscles. +An intraperitoneal mass in contact with the diaphragm will move on respiration whereas retroperitoneal masses are usually fixed and do not move with respiration; an enlarged kidney is ‘ballotable’ and bimanually palpable. Normal aor-tic pulsations can be both seen and felt in a thin abdomen, but expansile pulsation is characteristic of an abdominal aor-tic aneurysm. This should be differentiated from transmit-ted pulsation of a mass sitting on the aorta (e.g. pseudocyst of the pancreas). When ‘palpating during inspiration’, the examining hand is placed distal to the normal site of the +organ, and is held there until the edge of the organ descends + + +Antonio Maria V lsalva, 1666–1723, Professor of Anatomy in Bologna, of whom Morgagni wrote ‘there is nobody of those times who goes ahead of him, very few who are his equals’. +a +PART 11 | ABDOMINAL +Further reading 1021 + + + +and touches the examiner’s fingers. Liver, spleen, gall bladder and kidneys are best palpated during inspiration. An abdom-inal mass in a female, the lower limit of which cannot be dis-tinguished, is likely to arise from the pelvis. If the mass can be moved in a transverse direction, it is likely to be a uterine +or ovarian mass. +Spleen +In a healthy patient the spleen is not normally palpable. An enlarged spleen descends downwards, forwards and medi-ally. Palpation for an enlarged spleen is best performed in a supine patient. The examining hand should start in the right lower abdomen, with the tips of the fingers pointing upwards and pressed inwards. The patient is then asked to take a deep breath, and if the spleen is enlarged the lower edge with the characteristic notch will touch the fingers. If it is not palpable, then the hand is gradually moved upwards in the direction of the position of the edge of the normal-sized spleen with each breath. If the spleen is still not palpable, the patient is moved to the right lateral position and the +examination repeated. +Liver +In a supine patient, the hand is placed in line with the potential enlarged liver edge lateral to the rectus muscle. The patient is then asked to take a deep breath. If the liver is enlarged sufficiently below the costal margin, then surface +irregularities can also be felt. + + +Percussion +Percussion helps to distinguish distension due to bowel gas from solid masses and free fluid in the abdomen. Percus-sion is most sensitive when the examiner moves from res-onant parts of the abdomen to dull areas. In patients with free fluid in the peritoneal cavity, percussion from the centre to the periphery reveals dullness of flanks. Shifting dullness is elicited if the patient is re-examined lying on his or her side. The margin of dullness is then found to shift when the patient has moved. +Percussion is a very sensitive and refined method of test-ing for rebound tenderness. If the patient winces with pain on +abdominal percussion it denotes underlying peritonitis. + +Auscultation +High-pitched bowel sounds are heard during early stages of mechanical intestinal obstruction. Aortic and iliac bruits are heard when blood flows through a stenosis. A succussion splash is a sound like ‘shaking a half-filled bottle with water’ and is found most often in patients with gastric stasis due to gastric outlet obstruction. In generalised peritonitis, bowel +sounds will not be heard or be very few and far between. + +Inspection of hernia sites, examination of genitalia, inspection of anal region and digital rectal +examination +Abdominal examination is not complete until all external hernia sites and the anal area have been carefully inspected, the genitalia examined and a digital rectal examination per-formed. A vaginal examination may also be needed in females. The thoracic and lumbar spine and renal angles should also +be examined. + +VALUE OF OBSERVATION AND +REVIEW +In the case of acute abdominal pain, there will be a subset of patients in whom, after full clinical assessment, the surgeon remains uncertain about the need for an urgent operation. This is probably the most difficult group to deal with compared with those in whom an urgent operation is either clearly required, or clearly not required, and undoubtedly the one in which the majority of errors occur. Further urgent investigations are obviously important in this group and these are discussed in some detail elsewhere in this book. However, while these are taking place, regular review of the patient is essential. This period of observation has now become an integral part of the +early management of patients with acute abdominal pain. + +FURTHER READING +Lumley JS, D’Cruz AK, Hoballah JJ, Scott-Connor CE. Hamilton Bailey’s demonstrations of physical signs in clinical surgery, 19th edn. London: CRC Press, 2016. +Bailey & Love +Bailey & Love +60 +Chapter + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + + + + + +Abdominal wall, hernia and umbilicus + + + + + + +Learning objectives +To know and understand: +• Basic anatomy of the abdominal wall and its weaknesses • Causes of abdominal hernia +• Types of hernia and classifications +• Clinical history and examination findings in hernia + + + +• Complications of abdominal hernia +• Non-surgical and surgical management of hernia – including mesh +• Complications of hernia surgery • Other abdominal wall conditions + + + + + +THE ABDOMINAL WALL +Basic anatomy and function +related to pathology +The abdominal wall is a complex structure composed primarily of muscle, bone and fascia. Its major function is to protect the enclosed organs of the gastrointestinal and urogenital tracts but a secondary role is mobility, being able to flex, extend, rotate and vary its capacity. Flexibility requires elasticity and stretch, which compromise abdominal wall strength. +The roof of the abdomen is formed by the diaphragm sepa-rating the thoracic cavity above, with negative pressure, from the abdomen below, with positive pressure. Weakness of the diaphragm can lead to much of the bowel being drawn into the chest down this pressure gradient. The bony pelvis forms the floor of the cavity but a muscular central portion, the perineum, may also weaken and allow rectum, bladder and gynaecological organs to bulge downwards, a condition called prolapse. +The overall design of the abdominal muscles is best seen on a transverse computed tomography (CT) scan through the mid-abdomen. Posteriorly the muscles are strong, fur-ther supported by the vertebral column, ribs and pelvis. Two regions called the posterior triangles represent areas of weak-ness which can lead to rare lumbar hernias. Laterally there are three thin muscle layers the fibres of which criss-cross for strength and flexibility. Surgeons can make use of these layers, by making releasing incisions, separating the layers and then sliding one layer over another to increase girth and allow clo-sure of defects in the centre of the abdomen, e.g. the ‘Ramirez slide’ used in large incisional hernia repair (Figure 60.1). +Anteriorly the two powerful rectus abdominis muscles extend vertically from ribs to pelvis. Herniation through +these strong muscles does not occur naturally but their central + + +Rectus abdominis Linea alba Transversalis External fascia +oblique Extraperitoneal Internal fascia +oblique +Parietal peritoneum +Transversus abdominis +Psoas Latissimus major +dorsi +Erector Quadratus Abdominal spinae +lumborum +aorta + +Figure 60.1 A cross-section of the midabdomen showing the muscular layout. + +join, the linea alba, is an area of weakness resulting in epigas-tric and paraumbilical herniation. Divarification of the recti is the condition where the linea alba stretches laterally as the two rectus muscles separate. It occurs in the upper abdomen in middle-aged, overweight men (Figure 60.2) but also as a result of birth trauma in women when it occurs below the +umbilicus. + +Abdominal pressure +The positive pressure within the abdomen is used by a surgeon when drains are placed to allow blood, pus, bile, bowel con-tent and urine to flow outwards down the pressure gradient. However, this constant pressure from within can also lead to the condition of abdominal hernia where tissue, meant to be within the abdominal cavity, is forced outwards through +defects in the muscular wall. +PART 11 | ABDOMINAL Abdominal hernia 1023 + + + + + + + + + + + + + + + +Figure 60.2 Divarification. + +ABDOMINAL HERNIA +A hernia is the bulging of part of the contents of the abdomi- +nal cavity through a weakness in the abdominal wall. + +Anatomical causes of abdominal +wall herniation +Despite the complex design of the abdominal wall, the only natural weaknesses caused by inadequate muscular strength are the lumbar triangles and the posterior wall of the inguinal canal (Figure 60.3). +Many structures pass into and out of the abdominal cavity creating weakness which can lead to hernia formation. The most common example is the inguinal canal in males, along which the testis descends from abdomen to scrotum at the time of birth. The testicular artery, veins and vas pass though this canal (the round ligament in females). The resultant weakness leads to an indirect or lateral-type inguinal hernia. In adult surgery, 80% of all hernia repairs are for inguinal her-nia. The evolutionary advantage of testicular descent must outweigh the disadvantage of a high risk of herniation. Other examples are: oesophagus ® hiatus hernia, femoral vessels ® femoral hernia, obturator nerve ® obturator hernia, sciatic +nerve ® sciatic hernia. + + + + + + + + + + + + + + + + + +Figure 60.3 Posterior wall defect. + +An inguinal hernia (indirect) also occurs through the developmental failure of the processus vaginalis to close. As the testis descends, it pulls a tube of peritoneum along with it. This tube should naturally fibrose and become obliterated but often it fails to fibrose and allows a hernia to form. Recent studies have shown that calcitonin gene-related peptide and hepatocyte growth factor influence the closure of the pro-cessus, raising the possibility of a hormonal cause of hernia development. +Failure of normal development may lead to weakness of the abdominal wall. Examples are diaphragmatic, umbilical and epigastric hernias. Muscles which should unite during development fail to form strong unions with hernia develop-ment at birth or in later life. +Herniation at the umbilicus has both components, i.e. weakness due to structures passing through the abdominal wall in fetal life and developmental failure of closure. +The risk of inguinal hernia is related to the anatomical shape of the pelvis and is higher in patients having a wider and shorter pelvis. +Weakness of abdominal muscles may be the result of sharp trauma. Most commonly, this results from abdominal surgery but also occurs after stabbing. A surgical scar, even with perfect wound healing, has only 70% of the initial muscle strength. This loss of strength can result in hernia-tion in at least 10% of surgical incisions. Smaller laparoscopic port-site incisions have a hernia rate of 1%. Increasing use of this surgical approach should lead to a fall in the incidence of incisional hernia. +Muscle damage by blunt trauma or tearing of the abdomi-nal muscles requires exceptional force and is rare. +The sudden presence of a mass in the rectus muscle may be a rectus sheath haematoma, occasionally due to trauma but nowadays more often due to excessive anticoagulation therapy. +Primary muscle pathology and neurological conditions can lead to muscle weakness and occasionally present to the +surgeon as a ‘hernia’. + + +Summary box 60.1 + +Causes of hernia +●● Basic design weakness +●● Weakness due to structures entering and leaving the abdomen +●● Developmental failures +●● Genetic weakness of collagen ●● Sharp and blunt trauma +●● Weakness due to ageing and pregnancy ●● Primary neurological and muscle diseases ●● ? Excessive intra-abdominal pressure + + +Pathophysiology of hernia formation +A normal abdominal wall has sufficient strength to resist high abdominal pressure and prevent herniation of con- +tent. Herniation has been attributed to high pressures from +PART 11 | ABDOMINAL +, +1024 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +constipation, prostatic symptoms, excessive coughing in respiratory disease and obesity. However it has been shown that hernia is no more common in Olympic weight lifters than the general population, suggesting that high pressure is not a major factor in causing a hernia. Many patients will first notice a hernia after excessive straining. +There is good evidence that hernia is a ‘collagen disease’ and due to an inherited imbalance in the types of collagen. This is supported by histological evidence and relationships between hernia and other diseases related to collagen, such as aortic aneurysm. +Hernia development is more common in pregnancy due to hormonally induced laxity of pelvic ligaments. It is also more common in elderly people due to degenerative weak-ness of muscles and fibrous tissue. A recent Swedish report has shown that inguinal hernia is less common in obese patients, with hernia risk being negatively related to body mass index (BMI), contrary to widespread belief. Hernia is more common +in smokers. + +Common principles in abdominal +hernia +An abdominal wall hernia has two essential components, a defect in the wall and content, i.e. tissue that has been forced outwards through the defect. The weakness may be entirely in muscle, such as an incisional hernia. It may also be in fascia, similar to an epigastric hernia through the linea alba. It may have a bony component, such as a femoral hernia. The weak-ness in the wall is usually the narrowest part of the hernia which expands into the subcutaneous fat outside the muscle. The defect varies in size and may be very small or indeed very large. The nature of the defect is important to understanding the risk of hernia complications. A small defect with rigid walls traps the content and prevents it from freely moving in and out of the defect, increasing the risk of complications. The content of the hernia may be tissue from the extra-peritoneal space alone, such as fat within an epigastric her-nia or urinary bladder in a direct inguinal hernia. However, if such a hernia enlarges then peritoneum may also be pulled into the hernia secondarily along with intraperitoneal struc-tures such as bowel or omentum; a good example is a ‘sliding +type’ of inguinal hernia. +More commonly, when peritoneum is lying immediately deep to the abdominal wall weakness, pressure forces the peri-toneum through the defect and into the subcutaneous tissues. This ‘sac’ of peritoneum allows bowel and omentum to pass through the defect. In most cases, the intraperitoneal organs can move freely in and out of the hernia, a ‘reducible’ hernia, but if adhesions form or the defect is small, bowel can become trapped and unable to return to the main peritoneal cavity, an ‘irreducible’ hernia, with high risk of further complications. The narrowest part of the sac, at the abdominal wall defect, is called the ‘neck of the sac’. +When tissue is trapped inside a hernia it is in a confined +space. The narrow neck acts as a constriction ring impeding + +venous return and increasing pressure within the hernia. Resulting tension leads to pain and tenderness. If the hernia contains bowel then it may become ‘obstructed’, partially or totally. If the pressure rises sufficiently, arterial blood is not able to enter the hernia and the contents become ischaemic and may infarct. The hernia is then said to have ‘strangu-lated’. The wall of the bowel perforates, releasing infected, toxic bowel content into the tissues and ultimately back into the peritoneal cavity. The risk of strangulation is highest in hernias that have a small neck of rigid tissue, leading first to irreducibility and on to strangulation. The term ‘incarcerated’ is not clearly defined and used to imply a hernia that is irre- +ducible and developing towards strangulation. + + +Summary box 60.2 + +Types of hernia by complexity +●● Occult – not detectable clinically; may cause severe pain ●● Reducible – a swelling that appears and disappears +●● Irreducible – a swelling that cannot be replaced in the abdomen, high risk of complications +●● Strangulated – painful swelling with vascular compromise, requires urgent surgery +●● Infarcted – when contents of the hernia have become gangrenous, high mortality + + +In a special circumstance (Richter’s hernia) only part of the bowel wall enters the hernia. It may be small and difficult or even impossible to detect clinically. Bowel obstruction may not be present but the bowel wall may still become necrotic and perforate with life-threatening consequences. Femoral hernia may present in this way often with diagnostic delay and high risk to the patient (Figure 60.4). +An interstitial hernia occurs when the hernia extends between the layers of muscle and not directly through them. This is typical of a spigelian hernia (see below under Spigelian hernia). +An internal hernia is a term used when adhesions form within the peritoneal cavity. leading to abnormal pockets into which bowel can enter and become trapped. As there is no defect within the abdominal wall, the term ‘hernia’ is +confusing. + + + + + + + + + + + +Figure 60.4 Diagrammatic representation of gangrenous Richter’s hernia from a case of strangulated femoral hernia. + + + +August Gottlieb Richter, 1742–1812, lecturer in Surgery, Göttingen, Germany described this form of hernia in 1777. +PART 11 | ABDOMINAL Abdominal hernia 1025 + + + +Clinical history and diagnosis in +hernia cases +Patients are usually aware of a lump on the abdominal wall under the skin. Self-diagnosis is common. The hernia is usu-ally painless but patients may complain of an aching or heavy feeling. Sharp, intermittent pains suggest pinching of tissue. Severe pain should alert the surgeon to a high risk of stran-gulation. One should determine whether the hernia reduces spontaneously or needs to be helped. The patient should be asked about symptoms that might suggest bowel obstruction. +It is important to know if this is a primary hernia or whether it is a recurrence after previous surgery. Recurrent hernia is more difficult to treat and may require a different surgical approach. +General questions about the cardiac and respiratory sys-tems are necessary to assess a patient’s anaesthetic risk. +In a man with a groin hernia, history of prostatic symp-toms indicates a high risk of postoperative urinary retention. +Intake of anticoagulants such as warfarin is important because this impacts on future surgery. Many hernia opera-tions can be performed as a day case or single overnight stay, so that suitability for such treatment needs to be assessed, including home support, distance from the hospital, mobility +levels, etc. + + +Examination for hernia +The patient should be examined lying down initially and then standing as this will usually increase hernia size. In some cases no hernia will be apparent with the patient lying. The patient is asked to cough, when an occult hernia may appear. Divarification is best seen by asking a supine patient to simply lift his head off the pillow. +The overlying skin is usually of normal colour. If bruis-ing is present this may suggest venous engorgement of the content. If there is overlying cellulitis then hernia content is strangulating and the case should be treated as an emergency. +In most cases a cough impulse is felt. Gentle pressure is applied to the lump and the patient is asked to cough. If an impulse is felt this is due to increased abdominal pressure being transmitted into the hernia. In cases where the neck is tight and the hernia irreducible there may be no cough impulse. This can lead to failure of diagnosis and is typical of femoral hernia where lack of an impulse leads the clinician to misdiagnose a lymph node. Cough impulse can also occur in a saphena varix (see Chapter 57), which may be referred to a surgeon as a suspected inguinal hernia. It is not unusual for a patient to describe an intermittent swelling but the surgeon finds nothing on examination. This is due to muscle tighten-ing in an anxious patient. +If, on lying, the hernia does not reduce spontaneously, the surgeon asks the patient to attempt reduction because he may be well practised in this task although the surgeon might cause unnecessary discomfort. If neither the patient nor the surgeon can reduce the hernia then treatment is more urgent. An irreducible hernia may influence the decision between +open and laparoscopic surgery. With the hernia reduced, the + + +Summary box 60.3 + +Checks +●● Reducibility +●● Cough impulse ●● Tenderness +●● Overlying skin colour changes +●● Multiple defects/contralateral side ●● Signs of previous repair +●● Scrotal content for groin hernia ●● Associated pathology + + +Summary box 60.4 + +Examination +●● A swelling with a cough impulse is not necessarily a hernia ●● A swelling with no cough impulse may still be a hernia + + +surgeon assesses the size, rigidity and number of defects. Mul- +tiple defects may be present in incisional hernia. + +Investigations for hernia +For most hernias, no specific investigation is required, the diagnosis being made on clinical examination. However, the patient may have symptoms suggesting a hernia, but no her-nia is found, or have a swelling suggestive of hernia but with clinical uncertainty. It is important to be certain that any symptoms described are due to a hernia and not to coexisting pathology. There may also be a requirement for more detailed information than can be found by examination alone. A plain radiograph of the abdomen is of little value (Figure 60.5) although a hiatus hernia and diaphragmatic hernia may be seen on a chest radiograph. An ultrasound scan may be help-ful in cases of irreducible hernia, where the differential diag-nosis includes a mass or fluid collection, or when the nature of the hernia content is in doubt. Ultrasonography is very useful in the early postoperative period when a haematoma +or seroma may develop, and be difficult to distinguish from + + + + + + + + + + + + + + + +Figure 60.5 A radiograph showing spiral tacks causing chronic pain after transabdominal preperitoneal repair. +PART 11 | ABDOMINAL +1026 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +an early recurrence. Ultrasonography is non-invasive and low cost but operator dependent. +Computed tomography is helpful in complex incisional hernia, determining the number and size of muscle defects, identifying the content, giving some indication of presence of adhesions and excluding other intra-abdominal pathology such as ascites, occult malignancy and portal hypertension. +Contrast barium radiology is occasionally useful in the absence of CT. Contrast may also be injected directly into the peritoneum, a herniagram, to identify an occult sac, espe-cially in occult inguinal hernia. Magnetic resonance imaging (MRI) can help in the diagnosis of sportsman’s groin where pain is the presenting feature and the surgeon needs to distin-guish an occult hernia from an orthopaedic injury. +Laparoscopy itself may be used. In incisional hernia, initial laparoscopy may determine that a laparoscopic approach is fea-sible or not depending on the extent of adhesions. In inguinal hernia repair by the transabdominal route, initial laparoscopy can determine the presence of an occult contralateral hernia +which has been described in up to 20% of patients. + +Summary box 60.5 + +Investigations +●● Plain radiograph – of little value +●● Ultrasound scan – low cost, operator dependent ●● CT scan – incisional hernia +●● MRI – good in sportsman’s groin with pain +●● Contrast radiology – especially for inguinal hernia +●● Laparoscopy – useful to identify occult inguinal hernia + + +Management principles +An abdominal wall hernia does not necessarily require repair. A patient may request surgery for relief of symptoms of dis-comfort, for cosmesis or to establish the diagnosis when in doubt. The surgeon should recommend repair when compli-cations are likely, the most worrying being strangulation with bowel obstruction and bowel infarction. All cases of femoral hernia, with high risk of strangulation, should be repaired sur-gically. Any case of irreducible hernia, especially where there is pain and tenderness, should be offered repair unless coex-isting medical factors place the patient at very high risk from surgery or anaesthesia. Increasing difficulty in reduction and increasing size are indications for surgery. Surgery should be offered to younger adult patients as symptoms and complica- +tions are likely over time. + + +Summary box 60.6 + +Management +●● Not all hernias require surgical repair +●● Small hernias can be more dangerous than large +●● Pain, tenderness and skin colour changes imply high risk of strangulation +●● Femoral hernia should always be repaired + +In reality, most patients with a hernia should be offered repair. In elderly people, if the hernia is asymptomatic, small in size, can be reduced easily and is not causing anxiety, then observation alone should be sufficient. This policy, called ‘watchful waiting’, has been studied in asymptomatic ingui-nal hernia. One study reported such a policy to be safe but a second study was abandoned when a small number of patients developed strangulation. A truss can be used to control a her-nia but few surgeons would recommend this approach. Small paraumbilical hernias are often seen. They cause few symp-toms and usually contain fat or omentum with a very low risk of complications. +Large incisional hernias, particularly recurrent, present a major problem. Surgical repair is a complex procedure with significant risk of complications and later recurrence. When the neck is wide, the risk of strangulation is low. In obese and elderly patients, these risks may outweigh the benefits of sur-gery and it is common for surgeons to adopt a conservative approach. +Any patient who presents with acute pain in a hernia, par-ticularly if it is irreducible, should be offered surgery. Often, in a patient with an irreducible hernia, after admission to hos-pital and adequate analgesia, the hernia will reduce due to muscle relaxation. The likelihood of similar episodes is very high and surgery should be recommended at this admission or +soon after. + + +Surgical approaches to hernia All surgical repairs follow the same basic principles: +1 reduction of the hernia content into the abdominal cavity with removal of any non-viable tissue and bowel repair if necessary; +2 excision and closure of a peritoneal sac if present or replac-ing it deep to the muscles; +3 reapproximation of the walls of the neck of the hernia if possible; +4 permanent reinforcement of the abdominal wall defect +with sutures or mesh. +Reduction of hernia content is essential for a successful repair. It is rare for a surgeon to fail to reduce the hernia but extensive dissection can lead to bowel injury sometimes requiring bowel resection with subsequent risks of infection and bowel anastomotic complications. +, +Excision and closure of the peritoneal sac are ideal but not essential. During laparoscopic repair of incisional hernia, surgeons will often leave the sac in situ after reducing the hernia contents, and simply fix a mesh over the neck to pre-vent recurrence. There is risk of fluid formation within the sac (seroma). This is a common complication in all forms of hernia repair. In lateral (indirect) inguinal hernia, most surgeons excise the peritoneal sac but some leading experts recommend that it be dissected from surrounding tissue and simply pushed back through the deep inguinal ring. In lapa-roscopic repair of inguinal hernias, surgeons simply pull the sac back into the abdominal cavity from within and do not +excise it. +PART 11 | ABDOMINAL Abdominal hernia 1027 + + + +Closure of the abdominal wall defect is ideal but may not be possible when the defect is large or tissues are rigid. Plas-tic surgical techniques have been developed to ‘borrow’ tissue from elsewhere in order to cover large muscle defects, but usu-ally at the cost of leaving a weak area elsewhere. Over the past 20–30 years, surgeons have realised that simple closure of a hernia defect by sutures alone leads to a high recurrence rate. Additional reinforcement of the defect with a non-absorbable mesh is now widely practised in most hernia repairs and evidence has shown that recurrence rates have improved but recurrence still remains a problem. There is some evidence that mesh repair delays but does not prevent recurrence. With improved surgical techniques and new meshes it is hoped that recurrence after surgery will fall fur-ther. Mesh repair has become so important in hernia surgery that some understanding of mesh technology is essential for +the modern surgeon. + +Mesh in hernia repair +The term ‘mesh’ refers to prosthetic material, either a net or a flat sheet, which is used to strengthen a hernia repair. Mesh +can be used: + + +(a) + + + + + + + + + + + + + + + + +(b) + +● to bridge a defect: the mesh is simply fixed over the defect as a tension-free patch; +● to plug a defect: a plug of mesh is pushed into the defect; ● to augment a repair: the defect is closed with sutures and +the mesh added for reinforcement. + +A well-placed mesh should have good overlap around all margins of the defect, at least 2cm but up to 5cm if possible. Suturing a mesh edge to edge into the defect (inlay), with no overlap, is not recommended. Mesh plug repairs have gained some popularity in small defects especially where overlap is hard to achieve. Plug operations are fast but plugs can form a dense ‘meshoma’ of plug and collagen. Other complications include migration, erosion into adjacent organs, fistula forma- +tion and chronic pain. +Mesh types +The wide array of meshes available can be classified as follows. + +GROSS STRUCTURE +Net meshes are woven or knitted. Flat sheets are not porous but can be perforated with multiple holes. Net meshes allow fibrous tissue in growth between the strands and becoming adherent and integrated into host tissues within a few months. Initial fixation of the mesh is by glue, sutures or staples which may be absorbable. In laparoscopic inguinal hernia, no fixa-tion is required at all as friction is sufficient to hold the mesh. ‘Sheet’ meshes do not allow host tissue in growth but become encapsulated by fibrous tissue. They always require strong, +non-absorbable fixation to prevent mesh migration. + +SYNTHETIC MESH +Most meshes used today are synthetic polymers of polypropy-lene, polyester or polytetrafluoroethylene (PTFE) (Figure 60.6). They are non-absorbable and provoke little tissue +reaction. Polypropylene makes a strong monofilament mesh. + + + + + + + +Figure 60.6 (a) Polypropylene mesh in totally extraperitoneal inguinal hernia repair and (b) polyester mesh in a paraumbilical hernia repair. + +It does not have any antibacterial properties but its hydropho-bic nature and monofilament microstructure impede bacterial in-growth. Polyester is a braided filament mesh. This struc-ture may allow infection to take hold, aided by its hydrophilic property. However, this property also allows rapid vascular and cellular infiltration within the fibrils, aiding host immune responses to infection and providing a stronger host–tissue interface. PTFE meshes are flat sheets and as a result do not allow any tissue in-growth. They are used as a non-adhesive +barrier between tissue layers. + +WEIGHT AND POROSITY +Synthetic meshes are very strong and early meshes were much stronger than a human abdominal wall, so they are consid-ered as ‘over-engineered’. All meshes provoke a fibrous reac-tion. More dense or heavyweight meshes provoke a greater reaction, leading to collagen contraction and stiffening. The term ‘mesh shrinkage’ is often used to describe a progressive decrease in size of a mesh over time. It is due to natural con- +traction of fibrous tissue embedded in the mesh, reducing the +PART 11 | ABDOMINAL +1028 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +area of mesh itself. This can lead to tissue tension and pain, a common complication of mesh repair. It can also lead to hernia recurrence if the mesh no longer covers the defect. Meshes can shrink by up to 50% and, in occasional cases, even more. Meshes with thinner strands and larger spaces between them, ‘lightweight, large-pore meshes’, are preferred because they have better tissue integration, less shrinkage, more flexibility and improved comfort. +The terms ‘light’, ‘medium’ and ‘heavy’ are not precisely defined but meshes <40g/m2 are generally referred to as light +and meshes >80g/m2 as heavy. + +BIOLOGICAL MESH +There are ‘biological meshes’ that are sheets of sterilised, decellularised, non-immunogenic connective tissue. They derive from human or animal dermis, bovine pericardium or porcine intestinal submucosa. They provide a ‘scaffold’ to encourage neovascular in-growth and new collagen deposi-tion. Host enzymes eventually break down the biological implant, which is replaced and remodelled with ‘normal’ host fibrous tissue. The rates of enzymatic degradation and collagen deposition vary between products and also depend on the local environment of the mesh. In the presence of infection, some biological meshes rapidly break down and weaken before remodelling can occur. Others remain strong, their labyrinthine microstructure allowing vascular in-growth to aid infection resistance. The choice of biological mesh depends on the clinical situation for which it is to be used. +They are expensive. + +ABSORBABLE MESHES +There are also synthetic absorbable meshes, such as those made from polyglycolic acid fibre. They are used in temporary abdominal wall closure and to buttress sutured repairs. They have no current role in hernia repair because they absorb and +induce minimal collagen deposition. + +TISSUE-SEPARATING MESHES +Most meshes induce fibrosis and, if placed within the perito- +neal cavity, promote unwanted adhesions. New meshes have + +been designed for intraperitoneal use. Most of these have very different surfaces, one being sticky and one slippery. Good adherence and host–tissue in-growth is required on the pari-etal (muscle) side of the mesh, but the opposite (bowel) side needs to prevent adhesions to bowel. Usually one side of the mesh is coated by material that prevents adhesions (Figure 60.7), such as polycellulose, collagen, PTFE. A recent mesh made entirely of a sheet of condensed PTFE with multiple perforations can be used intraperitoneally because the perito-neum will grow in through its perforations whereas bowel will +not adhere to its inside. + + +Summary box 60.7 + +Mesh characteristics +●● Woven, knitted or sheet +●● Synthetic or biological – mainly synthetic +●● Light, medium or heavyweight – lightweight becoming more popular +●● Large pore, small pore – large pore causes less fibrosis and pain +●● Intraperitoneal use or not – non-adhesive mesh on one side ●● Non-absorbable or absorbable – mainly non-absorbable + + +Positioning the mesh +The strength of a mesh repair depends on host–tissue in-growth. Meshes should be placed on a firm, well-vascularised tissue bed with generous overlap of the defect. +The mesh can be placed: +● just outside the muscle in the subcutaneous space (onlay); ● within the defect (inlay) – only applies to mesh plugs in +small defects; +● between fascial layers in the abdominal wall (intraparietal or sublay); +● immediately extraperitoneally, against muscle or fascia (also sublay); +● intraperitoneally. +At open surgery all of these planes are used but laparo-scopic surgeons currently use only intraperitoneal or extraper- +itoneal planes (Figure 60.8). + + + + +Anterior rectus sheath +Subcutaneous space Linea alba Onlay space + + + + + +Posterior rectus sheath + +Rectus abdominis muscles +Sublay spaces + + + +Retromuscular space + +Extraperitoneal space + + +Figure 60.8 Diagrammatic representation of the various layers into Figure 60.7 Adhesions to mesh. which meshes are placed in ventral hernia repair. +PART 11 | ABDOMINAL Specific hernia types 1029 + + + +Limitations to the use of mesh +The presence of infection limits the use of mesh, particularly heavyweight types. If a mesh becomes infected then it often needs to be removed. Some infected meshes can be salvaged using a combination of debridement of non-incorporated mesh, appropriate antibiotics and modern vacuum-assisted dressings. +Meshes are expensive, especially those for intraperitoneal use, but prices are falling and there are reports of low-cost +solutions such as mosquito netting! + +SPECIFIC HERNIA TYPES Hernia sites are shown in Figure 60.9. + +Inguinal hernia +The inguinal hernia, often referred to as a ‘rupture’ by patients, is the most common hernia in men and women but much more common in men. There are two basic types that are fundamentally different in anatomy, causation and com-plications. However, they are anatomically very close to each other, surgical repair techniques are very similar and ultimate reinforcement of the weakened anatomy is identical, so they +are often referred to together as inguinal hernia. + + +Summary box 60.8 + +Inguinal hernia +●● Types – lateral (oblique, indirect); medial (direct), sliding ●● Origin – congenital or acquired +●● Anatomy – inguinal canal +●● Classification – latest European Hernia Society +●● Diagnosis – usually clinical but radiological in special circumstances +●● Surgery – open and laparoscopic + + +The congenital inguinal hernia is known as indirect, oblique or lateral whereas the acquired hernia is called direct or medial. There is a third ‘sliding’ hernia that is acquired but +is lateral in position (see below). + + + + +Epigastric Paraumbilical Umbilical + +Spigelian + +Basic anatomy of the inguinal canal +As the testis descends from the abdominal cavity to the scrotum in males it firsts passes through a defect called the deep inguinal ring in the transversalis fascia, just deep to the abdominal muscles. This ring lies midway between the anterior superior iliac spine and the pubic tubercle, approx-imately 2–3cm above the femoral artery pulse in the groin. The inferior epigastric vessels lie just medial to the deep inguinal ring, passing from the iliac vessels to rectus abdom-inis. Muscle fibres of the innermost two layers of the lateral abdominal wall, the transversus muscle and the internal oblique muscle, arch over the deep inguinal ring from lateral to medial before descending to become attached to the pubic tubercle. These two muscles fuse and become tendinous, hence this arch is referred to as the conjoint tendon. Below this arch there is no muscle but only transversalis fascia and external oblique aponeurosis, resulting in weakness (Figure 60.10). +The testis proceeds medially and downwards along the inguinal canal. Anterior to the canal is the aponeurosis of the external oblique muscle, the fibres of which run downwards and medially. The testis finally emerges through a V-shaped defect in the aponeurosis, the superficial inguinal ring, and descends into the scrotum. The inguinal canal is roofed by the conjoint tendon, its posterior wall is transversalis fascia, an anterior wall is the external oblique aponeurosis and a floor, which is also the external oblique, that rolls inwards at its lower margin and thickens to become the inguinal (Poupart’s) ligament. The inguinal canal in males contains the testicu-lar artery, veins, lymphatics and the vas deferens. In females, the round ligament descends through the canal to end in the vulva. Three important nerves, the ilioinguinal, the iliohypo-gastric and the genital branch of the genitofemoral nerve, also pass through the canal. +As the testis descends, a tube of peritoneum is pulled with the testis and wraps around it ultimately to form the tunica vaginalis. This peritoneal tube should obliterate, pos-sibly under hormonal control, but it commonly fails to fuse either in part or totally. As a result, bowel within the perito-neal cavity is able to pass inside the tube down towards the scrotum. Inguinal hernia in neonates and young children is always of this congenital type. However, in other patients, +the muscles around the deep inguinal ring can prevent a + + + + + +Superior lumbar +Inferior lumbar + + + + + + +Figure 60.9 Diagram to show the sites of abdominal wall hernias, common in red and rare in black. Incisional and parastomal hernias can be found at various sites. + + +Inguinal Femoral +Obturator + +Gluteal +Sciatic +PART 11 | ABDOMINAL +1030 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +Internal inguinal ring +Iliacus Inferior epigastric Femoral nerve vessels +Femoral artery External inguinal +Femoral vein ring +Sac of femoral Pubic tubercele +hernia +Sac of indirect inguinal hernia + +Sac of direct inguinal hernia + +Figure 60.10 The close relationships of direct inguinal, indirect inguinal and femoral hernias. + + + + + + + + + + + + + + + + + + + + +Figure 60.11 A huge scrotal hernia that has descended into the scrotum. The overlying skin has become gangrenous and sloughed away (courtesy of Dr Anupam Rai, Jabalpur, India). + + +hernia from developing until later in life, when, under the constant positive abdominal pressure, the deep inguinal ring and muscles are stretched and a hernia becomes apparent. As the hernia increases in size, the contents are directed down into the scrotum. These hernias can become massive and may be referred to as a scrotal hernia (Figure 60.11). +An indirect hernia is lateral because its origin is lateral to the inferior epigastric vessels. It is also oblique as the hernia passes obliquely from lateral to medial through the abdominal muscle layers. +The second type of inguinal hernia, referred to as direct or medial, is acquired. It is a result of stretching and weak-ening of the abdominal wall just medial to the inferior epi-gastric (IE) vessels. Looked at from within the abdominal cavity, there is a triangle referred to as Hasselbach’s triangle, the three sides of which are the IE vessels laterally, the lat-eral edge of rectus abdominis medially and the pubic bone +below (the iliopubic tract) (Figure 60.12). This area is weak + +because the abdominal wall here consists of only transver-salis fascia covered by the external oblique aponeurosis. A direct, medial hernia is more likely in elderly patients. It is broadly based and therefore unlikely to strangulate. The medially placed bladder can be pulled into a direct hernia (Figure 60.13). +The third type of inguinal hernia is referred to as a slid-ing hernia. This is also an acquired hernia due to weakening of the abdominal wall, but occurs at the deep inguinal ring lateral to the IE vessels. Retroperitoneal fatty tissue is pushed downwards along the inguinal canal. As more tissue enters the hernia, peritoneum is pulled with it, thus creating a sac. However, the sac has formed secondarily, distinguishing it from a classic indirect hernia. On the left side, sigmoid colon may be pulled into a sliding hernia and on the right side the caecum. Surgeons need extra caution during repair because the wall of the large bowel may not be covered by peritoneum and can be damaged. +Occasionally, both lateral and medial hernias are present +in the same patient (pantaloon hernia). + + + + + + + + + + + + + + + + + + +Figure 60.12 Laparoscopic view of the posterior inguinal region with hernia defects highlighted: yellow, medial inguinal; blue, lateral inguinal; green, femoral. + + + + + + + + + + + + + + + +Figure 60.13 This cystogram shows the urinary bladder, part of which has descended into a left direct inguinal hernia (arrows). +PART 11 | ABDOMINAL Specific hernia types 1031 + + + +Classification +Many surgeons over the past 100 years have attempted to classify inguinal (and femoral) hernias, including Casten, Halverson and McVay, Zollinger, Ponka, Gilbert and Nyhus. The European Hernia Society has recently suggested a simpli- +fied system of: +● primary or recurrent (P or R); +● lateral, medial or femoral (L, M or F); +● defect size in fingerbreadths assumed to be 1.5cm. +A primary, indirect, inguinal hernia with a 3-cm defect +size would be PL2. +Diagnosis of an inguinal hernia +In most cases, the diagnosis of an inguinal hernia is simple and patients often know their diagnosis because they are so common. Usually these hernias are reducible presenting as intermittent swellings, lying above and lateral to the pubic tubercle, with an associated cough impulse. Often the her-nia will reduce on lying and reappear on standing. With the patient lying down, the patient is asked to reduce the her-nia if it has not spontaneously reduced. If the patient cannot then the surgeon gently attempts to reduce the hernia. Once reduced, the surgeon identifies the bony landmarks of the anterosuperior iliac spine and pubic tubercle to landmark the deep inguinal ring at the mid-inguinal point. Gentle pressure is applied at this point and the patient asked to cough. If the hernia is controlled with pressure on the deep inguinal ring then it is likely to be indirect/lateral and if the hernia appears medial to this point then it is direct/medial. Other examina-tion techniques have been suggested but even experienced surgeons find it difficult to distinguish lateral and medial her- +nias with certainty (Figure 60.14). + + + + + + + + + + + + + + + + + + + + + + +Figure 60.14 Oblique left inguinal hernia that became apparent when the patient coughed and persisted until it was reduced when he lay down. + +Diagnostic difficulties +Confirmation of the diagnosis may not be possible when the patient describes an intermittent swelling but nothing is found on examination. Surgeons will often accept the diag-nosis on history alone but re-examination at a later date or investigation by ultrasound scan may be requested. +If an inguinal hernia becomes irreducible and tense there may be no cough impulse. Differential diagnosis would include a lymph node groin mass or an abdominal mass (Figure 60.15). Such cases require urgent investigation by either ultrasonography or CT. +Large scrotal hernias may be misdiagnosed as a hydrocele or other testicular swelling. The surgeon should be able to identify the upper limit of a scrotal swelling but a large scro-tal hernia has no upper limit because it extends back along the inguinal canal to the peritoneal cavity. In cases of doubt, ultrasonography should establish the diagnosis. +As inguinal hernia is so common, less-experienced clini-cians might suggest this diagnosis when referring cases of fem-oral hernia or spigelian hernia. Also patients with a saphena varix may present with a swelling that increases in size on standing and with a definite cough impulse and be misdiag-nosed as a hernia. The same can be true for a varicocele. +It is essential to examine the scrotal contents to exclude other pathologies and to check that the patient has two testes. It is important to examine the opposite side because contralateral hernia is common. Even if the contralateral side is weak, then bilateral repair should be recommended because the risk of contralateral recurrence is high. Of all patients 10% will present with bilateral inguinal hernias and up to 20% more will have an occult contralateral hernia on laparoscopic evaluation. A patient with a single hernia has a lifetime 33% risk of developing a hernia on the other side. Some surgeons have suggested that all patients should be offered bilateral repair, especially if laparoscopic surgery is planned, but this is not widespread practice at +present. + + + + + + + + + + + + + + + + + + + + +Figure 60.15 Malignant mass of nodes. +PART 11 | ABDOMINAL +1032 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +Investigations for inguinal hernia +Most cases require no diagnostic tests but ultrasonography, CT and MRI are occasionally used. A herniogram involves the injection of contrast into the peritoneal cavity followed by screening which shows the presence of a sac or asymmetri- +cal bulging of the inguinal anatomy. +Management of inguinal hernia +It is safe to recommend no active treatment in cases of early, asymptomatic, direct hernia, particularly in elderly patients who do not wish for surgical intervention. These patients should be warned to seek early advice if the hernia increases in size or becomes symptomatic. Surgical trusses are not rec-ommended but may be required for occasional patients who refuse any form of surgical intervention. +Elective surgery for inguinal hernia is a common and simple operation. It can be undertaken under local, regional or general anaesthesia with minimal risk, even in high-risk +patients. + +HERNIOTOMY +In children who have lateral hernias with a persistent pro-cessus, it is sufficient just to remove and close the sac. This is called a herniotomy. In adult surgery, herniotomy alone has a high recurrence rate and some form of muscle strengthening +is added (herniorrhaphy). + +OPEN SUTURE REPAIR +In 1890, Eduardo Bassini described suture repair for inguinal hernia (Figure 60.16). This was a massive leap forward and has been the basis of open repair for over 100 years. The sur-geon enters the inguinal canal by opening its anterior wall, the external oblique aponeurosis. The spermatic cord is dis-sected free and the presence of a lateral or a medial hernia +is confirmed. The sac of a lateral hernia is separated from + + + + +B +A + + + +G F +C E + +A + + + + + +Figure 60.16 Bassini’s original diagram. A, subcutaneous fat; B, external oblique; C, iliac vein; E, spermatic cord; F, nerves in inguinal canal; G, transversalis fascia. + +the cord, opened and any contents reduced. The sac is then sutured closed at its neck and excess sac removed. If there is a medial hernia then it is inverted and the transversalis fascia is suture plicated. Sutures are now placed between the conjoint tendon above and the inguinal ligament below, extending from the pubic tubercle to the deep inguinal ring. The poste-rior wall of the inguinal canal is thus strengthened. +Over 150 modifications to Bassini’s operation have been described with little or no benefit except for the Shouldice modification. In this operation, the transversalis fascia is opened by a central incision from the deep inguinal ring to the pubic tubercle and then closed to create a double-thick, two-layered posterior wall (double breasting). The exter-nal oblique is closed in similar fashion. Expert centres have reported lifetime failure rates of less than 2% after Shouldice repair but it is a technically demanding operation which, in general hands, gives results identical to Bassini’s repair. +Today, when a Bassini-type operation is done, most sur-geons use a continuous, non-absorbable nylon or polypropyl-ene suture which is darned between the conjoint tendon and inguinal ligament. This operation was described by Maloney, and recently published large randomised trials have reported excellent results when compared with mesh techniques. It is the most common operation performed in countries where mesh is too expensive. +Suture repair is still under development and, recently, Desarda has described an operation where a 1- to 2-cm strip of external oblique aponeurosis lying over the inguinal canal is isolated from the main muscle, but left attached both medi-ally and laterally. It is then sutured to the conjoint tendon and inguinal ligament, reinforcing the posterior wall of the inguinal canal. As the abdominal muscles contract, this strip of aponeurosis tightens to add further physiological support to the posterior wall. This operation is currently being +evaluated. + +OPEN FLAT MESH REPAIR +Synthetic mesh has been used since the 1950s to reinforce hernia repair, and in the 1980s Lichtenstein described a tension-free, simple, flat, polypropylene mesh repair for inguinal hernia (Figure 60.17). The initial part of the oper-ation is identical to Bassini’s. Once the hernia sac has been removed and any medial defect closed, a piece of mesh, measuring 8 × 15cm, is placed over the posterior wall, behind the spermatic cord, and is split to wrap around the spermatic cord at the deep inguinal ring. Loose sutures hold the mesh to the inguinal ligament and conjoint tendon. Two major advantages are claimed: lowered hernia recurrence rates and accelerated postoperative recovery. Randomised trials show that hernia recurrence within the first 2 years is lowered but acute pain scores are similar. Recent research comparing Lichtenstein’s repair with laparoscopic surgery has identified chronic pain as the most common complication of open flat mesh repair with rates reported as high as 20%. Nevertheless, today, Lichtenstein’s repair is the most common operation for +inguinal hernia in resource-rich countries. + + + +Eduardo Bassini, 1844–1924, Professor of Surgery, Padua, Italy described this method of herniorraphy in 1889. +, +PART 11 | ABDOMINAL Specific hernia types 1033 + + + + + + + + + + + + + + + + + + +Figure 60.17 Lichtenstein’s repair. + + + +OPEN PLUG/DEVICE/COMPLEX MESH REPAIR +Surgeons and industry have been highly creative, attempting to improve on simple flat mesh repair. A surgeon in Europe has over 200 different products and techniques from which to choose. Shaped mesh plugs have gained much attention, being simple to insert into the defect and requiring little if any fixation. However, they can become solid (meshoma) and also migrate. Meshes have been designed to be placed beneath the transversalis fascia. The surgeon introduces a finger through the deep inguinal ring and bluntly (and blindly) opens the preperitoneal space deep to the inguinal canal into which a mesh is inserted. A two-layered mesh (‘hernia system’), in which the inner layer is placed deep to transversalis fascia and the outer layer superficial to it, is also gaining popularity. To date, there is little evidence to show that any of these techniques is superior to Lichten- +stein’s operation. + +OPEN PREPERITONEAL REPAIR +This approach was first described by Annandale in 1880, but was largely discarded until the 1950s when Stoppa, a French surgeon, described it with mesh reconstruction. It is useful when multiple attempts at open standard surgery have failed and the hernia(s) keeps recurring. It may now be superseded by the totally extraperitoneal laparoscopic approach, which is modelled on Stoppa’s operation and first described by Ger, +also French. + +LAPAROSCOPIC INGUINAL HERNIA REPAIR +Two techniques are described and have been extensively studied in randomised trials. The totally extraperitoneal (TEP) approach is more widely used than the transabdominal preperitoneal (TAPP) approach. In both, the aim of surgery is to reduce the hernia and hernia sac within the abdomen, and then place a 10 × 15cm mesh just deep to the abdominal wall, extending across the midline into the retropubic space and 5cm lateral to the deep inguinal ring. The mesh covers Hasselbach’s triangle, the deep inguinal ring and the femoral canal. In TE , the surgeon is able to create a space just deep +P +to the abdominal muscles without entering the peritoneal + +cavity whereas, in TAPP, the surgeon enters the peritoneal cavity then incises the peritoneum above the hernia defects, and reflects it away from the muscles, essentially entering the same space as in TEP. Once the hernia has been reduced, an identical mesh is inserted and the peritoneum closed over the mesh (Figures 60.18 and 60.19). +Over 60 randomised trials have compared laparoscopic surgery with Lichtenstein’s repair. They show that, although the laparoscopic operation takes longer to perform, proven advantages are reduced pain both after surgery and up to 5 years later, more rapid return to full activity, and the reduced incidence of the wound complications of infection, bleeding and seroma. Laparoscopic surgery is of particular benefit in bilateral cases and in patients with hernia recurrence after open surgery. National statistics show that the proportion of cases performed laparoscopically is slowly rising, but all agree that there is a slow learning curve associated with these tech- +nically demanding operations. + + + + + + + + + + + + + + + + + +Figure 60.18 Right/medial direct hernia – laparoscopic view. Note the medial (direct) defect upper left, the inferior epigastric vessels upper right and the structures of the spermatic cord lower right. + + +C + + + +A +D + + +B + + + + + + + + +Figure 60.19 Right/lateral indirect hernia – laparoscopic view. A, arch of pubic bone; B, vas deferens and testicular vessels retracted medially; C, inferior epigastric vessels; D, deep inguinal ring (hernia defect). +PART 11 | ABDOMINAL +1034 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + + +Summary box 60.9 + +Operations for inguinal hernia ●● Herniotomy +●● Open suture repair Bassini Shouldice Desarda +●● Open flat mesh repair Lichtenstein +●● Open complex mesh repair Plugs +Hernia systems +●● Open preperitoneal repair Stoppa +●● Laparoscopic repair TEP +TAPP + + +EMERGENCY INGUINAL HERNIA SURGERY +Of inguinal hernia patients 95% present at clinics and only 5% as an emergency with a painful irreducible hernia that may progress to strangulation and possible bowel infarction. The morbidity and mortality of emergency inguinal hernia surgery are high and surgery needs to be performed rapidly in a well-resuscitated patient with adequate postoperative high dependency or intensive care if necessary. The principles of surgery are the same as in an elective setting. Open surgery is preferred when a hernia is irreducible or if there is any risk of bowel resection. Infection may complicate these cases but most surgeons would still use a lightweight, synthetic mesh +repair covered by appropriate antibiotics. + +COMPLICATIONS OF INGUINAL HERNIA SURGERY +Despite this being a common procedure and technically straightforward, postoperative complications are common. Immediate complications include bleeding (which may be due to accidental damage to the inferior epigastric or iliac vessels) and urinary retention that may require catheterisation. Occa-sional over-enthusiastic infusion of local anaesthetic may lead to femoral nerve blockade, the patient being unable to move a leg. This usually resolves over 12 hours but is alarming. +Over the next week, seroma formation and wound infec-tion may occur. Seroma is due to an excessive inflammatory response to sutures or mesh and cannot be prevented. In most cases the fluid resolves spontaneously but may require aspiration. After laparoscopic surgery, a seroma may be mis-diagnosed as an early recurrence. Wound infection is not uncommon. Many surgeons use routine prophylactic antibi-otics but recent studies suggest little benefit even when mesh is used. +In the longer term, hernia recurrence and chronic pain are the main concerns. No operation can be guaranteed to be recurrence free. Evidence shows that mesh repairs have lower recurrence rates than suture repairs, but there is no difference between the various mesh repairs and no difference between +open and laparoscopic surgery. There is very strong evidence + +that specialist hernia surgeons will have lower recurrence rates whatever technique they use. +Chronic pain, defined as pain present 3 months after surgery, is common after all forms of surgery. It is less common and less severe after laparoscopic surgery. Different types of pain have been described but the most severe is neuralgic pain due to nerve irritation. This may be the result of nerve injury at the time of operation or chronic irritation of nerves by suture material or mesh. Careful identification and protection of all three nerves passing along the inguinal canal reduces the incidence of neu-ralgic pain. This type of pain is also very uncommon after lap-aroscopic surgery that is performed at a deeper level away from the nerves. Some contribution to chronic pain may be due to the mesh, which can become embedded in a dense collagenous reaction with shrinkage. This causes tissue tension and rigidity. +Rarely, damage to the testicular artery can lead to tes-ticular infarction, perhaps the most serious complication of inguinal hernia surgery. There is no good evidence that her-nia surgery has an effect on male fertility despite extensive +study in this area. + +Summary box 60.10 + +Complications +●● Early pain, bleeding, urinary retention, anaesthetic related ●● Medium seroma, wound infection +●● Late chronic pain, testicular atrophy + + +Sportsman’s hernia +This specific entity is well described and presents as severe pain in the groin area, extending into the scrotum and upper thigh. It is almost entirely restricted to young men who play contact sports such as football and rugby. The pain can be debilitat-ing and prevent the patient from exercising. On examination there may be some tenderness in the region of the inguinal canal, over the pubic tubercle, and over the insertion of the thigh adductor muscles. Usually no hernia can be felt and only occasionally can a true inguinal hernia be found. +In most cases, the pain is due to an orthopaedic injury, such as adductor strain or pubic symphasis diastasis. However, some believe that it can be due to muscle tearing (Gilmore’s groin) or stretching of the posterior wall of the inguinal canal. Other causes of pain should be excluded, such as hip, pelvic or lumbar spinal disease and bladder/prostate problems. MRI is most likely to detect an orthopaedic problem but ultrasonog-raphy, herniography or even laparoscopy may be used. +There are many anecdotal reports of successful treatment using all types of inguinal hernia surgery, suture and mesh, open and laparoscopic, but no randomised trials. Hernia sur-gery should be a last resort and the patient warned of a signif- +icant risk of failure to cure the pain. + +Femoral hernia Basic anatomy +The iliac artery and vein pass below the inguinal ligament +to become the femoral vessels in the leg. The vein lies +PART 11 | ABDOMINAL Specific hernia types 1035 + + + +medially and the artery just lateral to the artery with the fem-oral nerve lateral to the artery. They are enclosed in a fibrous sheath. Just medial to the vein is a small space containing fat and some lymphatic tissue (node of Cloquet). It is this space which is exploited by a femoral hernia. The walls of a femoral hernia are the femoral vein laterally, the inguinal ligament anteriorly, the pelvic bone covered by the iliopectineal liga-ment (Astley Cooper’s) posteriorly and the lacunar ligament (Gimbernat’s) medially. This is a strong curved ligament with a sharp unyielding edge which impedes reduction of a femoral hernia (Figure 60.20). +The female pelvis has a different shape to the male, increasing the size of the femoral canal and the risk of hernia. In old age, the femoral defect increases and femoral hernia is commonly seen in low-weight, elderly women. There is a substantial risk of developing a femoral hernia after a sutured +inguinal hernia repair (Denmark Hernia Registry). + + +Summary box 60.11 + +Femoral hernia +●● Less common than inguinal hernia +●● It is more common in women than in men ●● Easily missed on examination +●● Of cases 50% present as an emergency with very high risk of strangulation + +Diagnosis of femoral hernia +Diagnostic error is common and often leads to delay in diag-nosis and treatment. The hernia appears below and lateral to the pubic tubercle and lies in the upper leg rather than in the lower abdomen. Inadequate exposure of this area during routine examination leads to failure to detect the hernia. The hernia often rapidly becomes irreducible and loses any cough impulse due to the tightness of the neck. It may only be 1–2cm in size and can easily be mistaken for a lymph node. As it increases in size, it is reflected superiorly and becomes difficult to distinguish from a medial direct hernia, which arises only a few centimetres above the femoral canal. A direct inguinal hernia leaves the abdominal cavity just above the inguinal +ligament and a femoral hernia just below (Figure 60.21). + + + +A + + + + + +B + + + +D + + + + + + + +C + + + + + + +Figure 60.20 Right femoral hernia – laparoscopic view. The slightly oblique inguinal ligament can be seen superolaterally above the defect. The external iliac vein is not seen. A, Inguinal ligament; B, lacunar ligament; C, arch of pubic bone; D, fatty tissue overlying iliac vessels. + +Figure 60.21 The patient has a left femoral and a right inguinal her-nia. + +Summary box 60.12 + +Differential diagnosis ●● Direct inguinal hernia ●● Lymph node +●● Saphena varix +●● Femoral artery aneurysm ●● Psoas abscess +●● Rupture of adductor longus with haematoma + + +Jules Germain Cloquet, 1790–1883, Professor of Anatomy and Surgery, Paris, France. +Sir Astley Paston Cooper, 1768–1841, surgeon, Guy’s Hospital, London, UK, received a baronetcy and 1000 guineas for successfully removing an infected wen from the head of King George IV at Brighton in 1821. +Manoel Louise Antonio don Gimbernat, 1734–1816, Professor of Anatomy, Barcelona, Spain and later Director of the Royal College of Surgeons in Spain. +PART 11 | ABDOMINAL +1036 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +Investigations +In routine cases, no specific investigations are required. However, if there is uncertainly then ultrasonography or CT should be requested. In the emergency patient, bowel obstruc-tion usually occurs and a plain radiograph is likely to show small bowel obstruction. All patients with unexplained small bowel obstruction should undergo careful examination for a femoral hernia. It is now common to perform CT scanning in cases of bowel obstruction primarily to exclude malignancy, but it can identify an obstructing femoral hernia missed by +clinicians. +Surgery for femoral hernia +There is no alternative to surgery for femoral hernia and it is wise to treat such cases with some urgency. There are three open approaches and appropriate cases can be managed lap- +aroscopically. + +LOW APPROACH (LOCKWOOD) +This is the simplest operation for a femoral hernia but suitable only when there is no risk of bowel resection. It can easily be performed under local anaesthesia. A transverse incision is made over the hernia. The sac of the hernia is opened and its contents reduced. The sac is also reduced and non-absorbable sutures are placed between the inguinal ligament above and the fascia overlying the bone below. A small incision can be made in the medial lacunar ligament to aid reduction but there may be an abnormal branch of the obturator artery just deep to it, which can bleed. The femoral vein, lateral to the hernia, needs to be protected. Some surgeons place a mesh +plug into the hernia defect for further reinforcement. + +THE INGUINAL APPROACH (LOTHEISSEN) +The initial incision is identical to that of Bassini’s or Lichten-stein’s operation into the inguinal canal. The spermatic cord (or round ligament) is mobilised and the transversalis fascia opened from deep inguinal ring to the pubic tubercle. A fem-oral hernia lies immediately below this incision and can be reduced by a combination of pulling from above and pushing from below. If necessary, the peritoneum can be opened to help with reduction. Once reduced, the neck of the hernia is closed with sutures or a mesh plug, protecting the iliac vein throughout. The layers are closed as for inguinal hernia and the surgeon may place a mesh into the inguinal canal to pro-tect against development of an inguinal hernia. +Some surgeons believe that exploration of the femoral canal to exclude a hernia should be a routine part of inguinal +hernia surgery but most surgeons do not do this. + +HIGH APPROACH (McEVEDY) +This more complex operation is ideal in the emergency situa- +tion where the risk of bowel strangulation is high. It requires + +regional or general anaesthesia. A horizontal incision (classi-cally vertical) is made in the lower abdomen centred at the lateral edge of the rectus muscle. The anterior rectus sheath is incised and the rectus muscle displaced medially. The surgeon proceeds deep to the muscle in the preperitoneal space. The femoral hernia is reduced and the sac opened to allow careful inspection of the bowel, and a decision made regarding the need for bowel resection. This is performed if necessary. In dubious cases, the bowel is replaced into the peritoneal cavity for 5 minutes and then re-examined. The femoral defect is then closed with sutures, mesh or plug. This approach allows a generous incision to be made in the peri-toneum, which aids inspection of the bowel and facilitates +bowel resection. + +LAPAROSCOPIC APPROACH +Both the TEP and TAPP approaches can be used for a femoral hernia and a standard mesh inserted. This is ideal for reduc-ible femoral hernias presenting electively, but not for emer- +gency cases or irreducible hernia. + +VENTRAL HERNIA +This term refers to hernias of the anterior abdominal wall. Inguinal and femoral hernias are not included even though they are ventral. Lumbar hernia is included despite being dor-solateral. The European Hernia Society classification (2009) distinguished primary ventral from incisional hernia but did not include parastomal hernia. We have included parastomal +hernia and traumatic hernia. + + + +Summary box 60.13 + +Ventral hernias +●● Umbilical–paraumbilical ●● Epigastric +●● Incisional ●● Parastomal ●● Spigelian ●● Lumbar +●● Traumatic + + +Umbilical hernia +The umbilical defect is present at birth but closes as the stump of the umbilical cord heals, usually within a week of birth. This process may be delayed, leading to the development of herniation in the neonatal period. The umbilical ring may +also stretch and reopen in adult life. + + + + + +Charles Barrett Lockwood, 1856–1914, surgeon, St Bartholomew’s Hospital, London, UK. +George Lotheissen, 1868–1941, surgeon, the Kaiser Franz Joseph Hospital, Vienna, Austria, described this operation in 1898. Peter George McEvedy, 1890–1951, surgeon, Ancoats Hospital, Manchester, UK. +PART 11 | ABDOMINAL +Ventral hernia 1037 + + + +Umbilical hernia in children +This common condition occurs in up to 10% of infants, with a higher incidence in premature babies. The hernia appears within a few weeks of birth and is often symptomless, but increases in size on crying and assumes a classic conical shape. Sexes are equally affected but the incidence in black infants is up to eight times higher than in white. Obstruction and/ or strangulation is extremely uncommon below the age of +3 years. + +TREATMENT +Conservative treatment is indicated under the age of 2 years when the hernia is symptomless. Parental reassurance is all that is necessary. Of hernias 95% will resolve spontaneously. If the hernia persists beyond the age of 2 years it is unlikely to +resolve and surgical repair is indicated. + +SURGERY +A small curved incision is made immediately below the umbi-licus. The neck of the sac is defined, opened and any contents are returned to the peritoneal cavity. The sac is closed and redundant sac excised. The defect in the linea alba is closed +with interrupted sutures. + + +Summary box 60.14 + +Umbilical hernia in children +●● Common in infants and most resolve spontaneously ●● Rarely strangulate + + +Umbilical hernia in adults +Conditions that cause stretching and thinning of the midline raphe (linea alba), such as pregnancy, obesity and liver disease with cirrhosis, predispose to reopening of the umbilical defect. In adults, the defect in the median raphe is immediately adja-cent to (most often above) the true umbilicus, although at oper-ation this is indistinguishable. The term ‘paraumbilical hernia’ is commonly used. The defect is rounded with a well-defined fibrous margin. Small umbilical hernias often contain extra-peritoneal fat or omentum. Larger hernias can contain small or large bowel but, even when very large, the neck of the sac is narrow compared with the volume of its contents. As a result, in adults, umbilical hernias that include bowel are prone to +become irreducible, obstructed and strangulated. + +CLINICAL FEATURES +Patients are commonly overweight with a thinned and atten-uated midline raphe. The bulge is typically slightly to one side of the umbilical depression, creating a crescent-shaped +appearance to the umbilicus (Figure 60.22). Women are + + + + + + + + + + + + + + +Figure 60.22 A small paraumbilical hernia. + + +affected more than men. Most patients complain of pain due to tissue tension or symptoms of intermittent bowel obstruc-tion. In large hernias, the overlying skin may become thinned, +stretched and develop dermatitis. + +TREATMENT +As a result of the high risk of strangulation, surgery should be advised in cases where the hernia contains bowel. Small hernias may be left alone if they are asymptomatic, but they may enlarge and require surgery at a later date. Surgery may +be performed open or laparoscopically. + +OPEN UMBILICAL HERNIA REPAIR +Very small defects less than 1cm in size may be closed with a simple figure-of-eight suture, or repaired by a darn tech-nique where a non-absorbable, monofilament suture is criss-crossed across the defect and anchored firmly to the fascia all around. +Defects up to 2cm in diameter may be sutured primar-ily with minimal tension, although, the larger the defect, the more tension and the more likely it is that mesh rein-forcement will be beneficial. The classic repair was described by Mayo. A transverse incision is made and the hernia sac dissected, opened and its content reduced. Any non-viable tissue is removed, sometimes involving bowel resection. The peritoneum is closed. The defect in the anterior rectus sheath is extended laterally on both sides and elevated to create an upper and lower flap. The lower flap is then inserted beneath the upper flap and sutured to it, with the upper flap being brought downwards over it so that the tissue is two layered (double breasted). Non-absorbable sutures are used. There is often a large subcutaneous space. A suction drain is placed to reduce the risk of seroma and haematoma. The skin is closed but stretched or redundant skin may need to be excised (apronectomy) to achieve a better cosmetic result. Today, with modern suture materials, surgeons simply close the ante- +rior sheath in a single layer. + + + + +William James Mayo, 1861–1939, surgeon, the Mayo Clinic, Rochester, MN, USA, described this operation in 1901. He and his brother Charles Horace Mayo (1865–1939) joined their father’s private practice in Rochester. This practice became the Mayo Clinic. Their father William Worrall Mayo was born in Manchester, UK in 1819. +PART 11 | ABDOMINAL +1038 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + + + + + + + + + + + + + + + + +Figure 60.23 A massive paraumbilical hernia – operative view. + + +For defects larger than 2cm in diameter, mesh repair is recommended (Figure 60.23). The mesh may be placed in +one of several anatomical planes: +● Within the peritoneal cavity – a tissue separating mesh is placed through the defect and spread out on the under-side of the abdominal wall and fixed to it, ideally with an overlap of 5cm in each direction. This is a quick repair but requires the use of expensive mesh. +● In the retromuscular space – the linea alba is opened ver-tically and both left and right posterior rectus sheaths are incised 1cm to the side of the midline exposing the rectus muscle. The posterior sheaths are sutured together and the muscles elevated away from the sheath to develop the ret-romuscular space into which a sheet of mesh is placed and fixed by sutures. The mesh should overlap the midline by 5cm laterally and the umbilicus vertically. It should there-fore be a minimum diameter of 10cm. A drain may be placed deep to the linea alba. This is a very secure repair but requires extensive dissection. +● In the extraperitoneal space – it is difficult, but possible, to develop the plane below the posterior rectus sheath, just outside the peritoneum. Care must be taken to avoid ‘button-holing’ the peritoneum because it is thin and frag-ile. Mesh can then be tucked into in this space, ensuring a good overlap as before. Ideally, the linea alba is closed over the mesh but, if this is not possible, a flap of peritoneal sac can be used to cover the mesh. This is a good repair, but, if the peritoneum is extensively damaged during the dissection, it will have to be abandoned in favour of an alternative technique. +● In the subcutaneous plane – this is the simplest technique, called an onlay mesh. The peritoneal sac and contents are dealt with as above. An attempt is made to close linea alba vertically with sutures and a disc of mesh is placed on the anterior rectus sheath and sutured to it. The mesh is lying +in the subcutaneous space and is prone to infection. + +LAPAROSCOPIC UMBILICAL HERNIA REPAIR +Three ports are placed laterally on the abdominal wall, usu- +ally on the left side unless adhesions from previous surgery + +Figure 60.24 Paraumbilical defect – laparoscopic view. + + +are likely. The contents of the hernia are reduced by traction and external pressure. The falciform ligament above and the median umbilical fold below may need to be taken down if they interfere with mesh placement. A disc of non-adherent mesh, designed for intraperitoneal use, is introduced and posi-tioned on the under surface of the abdominal wall, centred on the defect. It is then fixed to the peritoneum and posterior rectus sheaths using staples, tacks or sutures. This is a simple and secure repair, which achieves generous overlap without surgical damage to umbilicus and surrounding fascia (Fig-ure 60.24). However, it requires specialised equipment and expensive tissue-separating mesh. Intraperitoneal meshes can cause severe pain lasting for 24–48 hours after surgery which +can mimic peritonitis. + + +EMERGENCY REPAIR OF UMBILICAL HERNIA +Incarceration, bowel obstruction and strangulation are fre-quent because of the narrow neck and the fibrous edge of the defect in the midline raphe. Delay to surgery can lead to gangrene of the omentum or bowel. Large hernias are often multiloculated and there may be strangulated bowel in one component when other areas are clinically soft and a +non-tender hernia. + + +SURGERY +In cases of simple incarceration without clinical evidence of strangulation, repair may be attempted laparoscopically but reduction of the contents can be very difficult if the hernia contains bowel. Most emergency repairs are performed by open surgery. In the presence of established strangulation it is unwise to place mesh at all because of the risk of infection, so an open sutured repair should be performed, accepting a high risk of later recurrence. Alternatively, a two-stage repair could be planned: the hernia contents being dealt with initially with little attempt made to close the defect and then subsequent definitive mesh repair once sepsis has +been controlled. +PART 11 | ABDOMINAL +Ventral hernia 1039 + + + +Epigastric hernia +These hernias arise through the midline raphe (linea alba) any where between the xiphoid process and the umbilicus, usually midway. When close to the umbilicus they are called supraumbilical hernias. Epigastric hernias begin with a trans-verse split in the midline raphe so, in contrast to umbilical hernias, the defect is elliptical. It has been hypothesised that the defect occurs at the site where small blood vessels pierce the linea alba or, more likely, that it arises at weaknesses due to abnormal decussation of aponeurotic fibres related to heavy physical activity (Figure 60.25). +Epigastric hernia defects are usually less than 1cm in max-imum diameter and commonly contain only extraperitoneal fat, which gradually enlarges, spreading in the subcutaneous plane to resemble the shape of a mushroom. When very large they may contain a peritoneal sac but rarely any bowel. More than one hernia may be present. The most common cause of ‘recurrence’ is failure to identify a second defect at the time +of original repair. +Clinical features +The patients are often fit, healthy men aged between 25 and 40 years. These hernias can be very painful even when the swelling is the size of a pea, due to the fatty contents becom-ing nipped sufficiently to produce partial strangulation. The pain may mimic that of a peptic ulcer but symptoms should not be ascribed to the hernia until gastrointestinal pathology has been excluded. A soft midline swelling can often be felt more easily than seen. It may be locally tender. It is unlikely to be reducible because of the narrow neck. It may resemble a +lipoma. A cough impulse may or may not be felt. +Treatment +Very small epigastric hernias have been known to disappear spontaneousl , probably due to infarction of the fat. Small-to-moderate-sized hernias without a peritoneal sac are not inherently dangerous and surgery should be offered only if the +y +hernia is sufficiently symptomatic. + + + + + + + + + + + + + + + + + + +Figure 60.25 Epigastric hernia – external view. + + +Surgery +This may be done by open or laparoscopic surgery. At open surgery, a vertical or transverse incision is made over the swell-ing and down to the linea alba. Protruding extraperitoneal fat can simply be pushed back through the defect or excised. Often a small vessel is present in the hernia content that can cause troublesome bleeding. The defect in the linea alba is closed with non-absorbable sutures in adults and absorbable sutures in children. In larger hernias and when a peritoneal sac is present, the surgical approach is similar to an umbilical mesh repair. +Laparoscopic repair is very similar to that for umbilical hernia except that the defect is hidden behind the falciform ligament which must be taken down from the undersurface of the abdominal wall. The margins of the defect must be clearly exposed and the fatty contents reduced before the mesh is placed. Simply placing a mesh under the linea midline may not in fact remove the hernia when its contents are extraper- +itoneal fat. + + +Incisional hernia +These arise through a defect in the musculofascial layers of the abdominal wall in the region of a postoperative scar. Thus +they may appear anywhere on the abdominal surface. +Incidence and aetiology +Incisional hernias have been reported in 10–50% of laparot-omy incisions and 1–5% of laparoscopic port-site incisions. Factors predisposing to their development are patient factors (obesity, general poor healing due to malnutrition, immuno-suppression or steroid therapy, chronic cough, cancer), wound factors (poor quality tissues, wound infection) and surgical factors (inappropriate suture material, incorrect suture placement). +An incisional hernia usually starts as disruption of the musculofascial layers of a wound in the early postoperative period. Often the event passes unnoticed if the overlying skin wound has healed securely. Many incisional hernias may be preventable with the use of good surgical technique. The clas- +sic sign of wound disruption is a serosanguineous discharge. +Clinical features +These hernias commonly appear as a localised swelling involving a small portion of the scar but may present as a diffuse bulging of the whole length of the incision (Figure 60.26). There may be several discrete hernias along the length of the incision and unsuspected defects are often found at surgery (Figure 60.27). Incisional hernias tend to increase steadily in size with time. The skin overlying large hernias may become thin and atrophic so that peristalsis may be seen in the underlying intestine. Vascular damage to skin may lead to dermatitis. Attacks of partial intestinal obstruc-tion are common because there are usually coexisting inter-nal adhesions. Strangulation is less frequent and most likely to occur when the fibrous defect is small and the sac is large. Most incisional hernias are broad-necked and carry a low risk +of strangulation. +PART 11 | ABDOMINAL +1040 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + + + + + + + + + + + + + + + + + +Figure 60.26 A large multilocular incisional hernia. + + + + + + + + + + + + + + + + + +Figure 60.27 Multiple defects seen during laparoscopic operation. + + +Summary box 60.15 + +Incisional hernia +●● Incidence 10–50% after surgery +●● Causation due to patient, wound and surgeon factors ●● Wide variation in size +●● Often multiple defects within the same scar +●● Obstruction is common but strangulation is rare ●● Open and laparoscopic repairs possible + + +Treatment +Asymptomatic incisional hernias may not require treatment at all. The wearing of an abdominal binder or belt may pre- +vent the hernia from increasing in size. +Principles of surgery +For most incisional hernias, surgery is relatively straightfor-ward and both open and laparoscopic options are available. A +number of principles apply, irrespective of the technique used. + +The repair should cover the whole length of the previous incision. Approximation of the musculofascial layers should be done with minimal tension and prosthetic mesh should be used to reduce the risk of recurrence. Mesh may be contrain-dicated in a contaminated field, e.g. bowel injury during the dissection but, in a clean-contaminated field, such as after an elective bowel resection, mesh may be used if placed in a dif-ferent anatomical plane to the contamination, such as in the extraperitoneal/retromuscular space. Appropriate systemic +antibiotics should be used. + +OPEN REPAIR +Simple suture techniques without the use of prosthetic mesh for reinforcement, even with layered closure such as in Mayo, ‘keel’ or da Silva repairs, are not recommended today because of the high risk of recurrence. However, they may be the only option in the presence of gross contamination such as peri-tonitis. +The previous incision is opened along its full length to reveal any clinically unsuspected defects. The hernial sac, its neck and the margins of the defect are fully exposed. The sac can be opened, contents reduced, local adhesions divided and any redundant sac excised to allow safe reclosure of the peri-toneum. +Mesh can be placed in one of several planes as for umbil-ical hernia repair. The simplest approach is an onlay mesh but increasingly the retromuscular sublay repair is preferred +by expert surgeons and is described below. + +RETROMUSCULAR SUBLAY MESH REPAIR +Vertical incisions are made through the fascia surrounding the rectus abdominis muscles so that the muscle can be sep-arated and elevated from the posterior rectus sheath below If possible, the medial edges of the posterior rectus sheath edges are sutured together with a continuous suture. In very large defects this may not be possible and below the arcuate line where the posterior sheath is deficient, being peritoneum and transversalis fascia only. In the case of transverse inci-sion, where the defect extends lateral to the rectus sheath, internal oblique and transversus abdominis muscles form the posterior layer. A sheet of lightweight, large-pore, prosthetic, elastic mesh is then laid between this posterior rectus sheath and belly/bellies of the rectus muscle. It is fixed to the sheath by interrupted sutures. The mesh must be large enough to ensure a 5-cm overlap of the underlying fascial defect in all directions. Careful haemostasis and meticulous asepsis are essential during this operation. The anterior rectus sheaths are then sutured together over the mesh so that, ideally, the mesh is completely covered by muscle and fascia and is not lying in the subcutaneous plane. Redundant skin may need to be excised. The risk of postoperative serous fluid collections is +. +reduced by suction drainage. + +LAPAROSCOPIC REPAIR +Incisional hernias are increasingly being repaired by laparo-scopic mesh techniques. Laparoscopy and division of adhe-sions is initially performed. Hernia contents are reduced and the fibrous margins of the hernia defect(s) are exposed. Often +the falciform ligament and median umbilical fold need to be +PART 11 | ABDOMINAL +Ventral hernia 1041 + + + +taken down. Some surgeons prefer to suture close the muscle defects first and then reinforce with mesh. Others simply fix the mesh under the defect with adequate overlap. The use of a tissue-separating mesh is essential. Various techniques have been described to size and then position the mesh accurately. The mesh is fixed to the abdominal wall by staples or trans-fascial sutures which pass through all muscle layers to hold the mesh. +In the presence of dense peritoneal adhesions, the lapa-roscopic surgeon needs to take great care because injury to the bowel is possible and may not be recognised. Diathermy is not used. If occult bowel injury does occur it can lead to postoperative peritonitis, which is an extremely dangerous +complication. +Management of the very large incisional hernia +Very large incisional hernias often require careful thought before treatment begins. If the volume of the sac is more than 25% of the volume of the abdominal cavity (and this can be calculated from CT images), then there are likely to be issues of loss of abdominal domain when the hernia is repaired. The contents of the hernia, which have been outside the abdominal cavity for a long time, will not fit back inside or, if they do, it will result in high tension. High intra-abdominal pressure can lead to visceral compression and pulmonary complications due to impaired diaphragmatic movement. A tight abdomen can lead to wound breakdown and failure of the repair. +echniques to overcome the potential loss of abdominal domain include preoperative abdominal expansion with pro-gressive preoperative pneumoperitoneum over several weeks, resection of the omentum and/or colon at the time of repair, the use of prosthetic mesh to span the uncloseable gap in the musculofascial layer, or the use of musculofascial advance-ment or transpositional flaps to achieve closure. +T +Even if loss of domain is not a concern, large defects can still be very difficult to close and the same special techniques may need to be used to avoid producing excessive tension in the repair. Ramirez’s component separation technique, which incorporates relaxing incisions in the external oblique apo-neurosis and/or the posterior sheath, is very useful because this enables either the anterior or the posterior component of the rectus sheath to be drawn together. It may then be rein-forced with a mesh. +Patients with poor quality or redundant skin may benefit from a wedge excision of skin and fat (lipectomy) to improve the abdominal contour postoperatively. Repair of these very large hernias is highly specialised surgery and is best done in +specialist centres. +Reducing the risk of incisional hernia +The incidence of incisional hernia may be reduced by improving the patient’s general condition preoperatively where possible, e.g. weight loss for obesity or improving nutri-tional state for malnutrition. Closing the fascial layers with +non-absorbable, or very slowly absorbable, sutures of adequate + +gauge is important. Traditional teaching was that sutures should be 1 cm deep and 1 cm apart. Recent work has shown that lower incisional hernia rates and reduced infection rates are gained when smaller and closer bites are used with a 2/0 suture rather than traditional heavier materials. +There is no evidence that interrupted sutures are better or worse than continuous. However, if continuous suturing is used, the tissue bites must not be too near the fascial edge or pulled too tight because they may cut out. It has also been confirmed that the optimal ratio of suture length to wound length is 4:1 (Jenkins’ rule). If less length than this is used, the suture bites are too far apart or too tight and the converse applies if more length than this is used. +Drains should be brought out through separate incisions and not through the wound itself because this leads to hernia formation. +Recent reports have suggested that placement of a pro-phylactic mesh in patients at high risk of hernia formation will substantially reduce that risk. This has been reported in obese patients undergoing bariatric surgery and also to pre-vent parastomal herniation, which occurs in up to 50% of +patients. + +Spigelian hernia +These hernias are uncommon although are probably under-diagnosed. They affect men and women equally and can occur at any age, but are most common in elderly people. They arise through a defect in the spigelian fascia, which is the aponeurosis of transversus abdominis. Often these hernias advance through the internal oblique as well and spread out deep to the external oblique aponeurosis. The spigelian fascia extends between the transversus muscle and the lateral bor-der of the rectus sheath from the costal margin to the groin, where it blends into the conjoint tendon. Most spigelian hernias appear below the level of the umbilicus near the edge of the rectus sheath, but they can be found anywhere +along the spigelian line (Figure 60.28). There is a common + + + + + + + + + + + + + + + + + +Figure 60.28 Spigelian hernia. + + +Adrian Van der Spieghel (Spigelius), 1578–1625, Professor of Anatomy, Padua, Italy +. +PART 11 | ABDOMINAL +1042 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +misconception that they protrude below the arcuate line as a result of deficiency of the posterior rectus sheath at that level, but in fact the defect is almost always above the arcuate line. In young patients they usually contain extraperitoneal fat only but in older patients there is often a peritoneal sac and they can become very large indeed. +They have also been described in infants and may be con-genital, reflecting incomplete differentiation of the mesen- +chymal layers within the abdominal wall. +Clinical features +Young patients usually present with intermittent pain, due to pinching of the fat, similar to an epigastric hernia. A lump may or may not be palpable because the fatty hernia is small and the overlying external oblique is intact. Older patients gener-ally present with a reducible swelling at the edge of the rectus sheath and may have symptoms of intermittent obstruction. The diagnosis should be suspected because of the location of the symptoms and is confirmed by CT. Ultrasonography has the advantage that it can be performed in the upright patient +because no defect may be visible with the patient lying down. +Treatment +Surgery is recommended because the narrow and fibrous neck predisposes to strangulation. Surgery can be open or laparo-scopic. At open surgery a skin crease is made over the hernia, but no abnormality will be seen until the external oblique is opened. The sac and contents are dealt with and the small defect in the spigelian fascia is repaired by suture or mesh laid deep to the external oblique aponeurosis. The plane of the mesh can be extended medially into the posterior rectus sheath if required. The external oblique aponeurosis is closed over the mesh. +Laparoscopy is useful if no sac is palpable, but, in young patients with a hernia containing only extraperitoneal fat, no hernia will be seen from within the peritoneum. In such cases, the peritoneum can be incised and the extraperitoneal plane explored for the small defect, which can then be closed by either suture or mesh. When an intraperitoneal sac is present, laparoscopic repair can be performed using either the intra- +peritoneal onlay of mesh (IPOM) or TAPP technique. + +Summary box 60.16 + +Spigelian hernia ●● Rare +●● Often misdiagnosed +●● High risk of complications + + +Lumbar hernia +Most primary lumbar hernias occur through the inferior lum-bar triangle of Petit bounded below by the crest of the ilium, laterally by the external oblique muscle and medially by latis-simus dorsi (Figure 60.29). Less commonly, the sac comes +through the superior lumbar triangle, which is bounded by the + + + + + + + + + + + + + + + + + + + + + + + +Figure 60.29 Inferior lumbar hernia, which contained caecum, appendix and small bowel. Note the filarial skin rash on the buttocks (courtesy of VJ Hartfield, formerly of south-east Nigeria). + + +twelfth rib above, medially by sacrospinalis and laterally by the posterior border of the internal oblique muscle. Primary lumbar hernias are rare, but may be mimicked by incisional hernias arising through flank incisions for renal operations, or through incisions for bone grafts harvested from the iliac +crest. +Differential diagnosis +A lumbar hernia must be distinguished from: +● a lipoma; +● a cold (tuberculous) abscess pointing to this position; +● a pseudo-hernia due to local muscular paralysis. Lumbar pseudo-hernia can result from any interference with the nerve supply of the affected muscles, the most common cause being injury to the subcostal nerve during a kidney +operation. +Treatment +The natural history is for these hernias to increase in size and surgery is recommended. Lumbar hernias can be approached by open or laparoscopic surgery. The defects can be difficult to close with sutures and mesh is recommended. +The TAPP laparoscopic approach is gaining popularity. With the patient in a semilateral position ports are inserted well away from the defect. The peritoneum is incised above the hernia and dissected back to expose the muscle defect. The content, often extraperitoneal fat, is reduced and a mesh fixed with ample overlap. The peritoneum can then be resu- +tured or tacked back to cover the mesh. + + +Jean Louis Petit, 1674–1750, Director of the Academie de Chirurgie, Paris, France. +PART 11 | ABDOMINAL +Ventral hernia 1043 + + + +Lumbar incisional hernias can be approached in the same way but large ones, especially if there is a component of neuropathic muscle atrophy causing a diffuse bulge (pseudo-hernia), can be very difficult and muscle-flap double breasting +with mesh reinforcement may be required. + +Parastomal hernia +When surgeons create a stoma, such as a colostomy or ileos-tomy, they are effectively creating a hernia by bringing bowel out through the abdominal wall. The muscle defect created tends to increase in size over time and can ultimately lead to massive herniation around the stoma. The rate of parastomal hernia is over 50%. For patients, it is very difficult to manage a stoma that is lying adjacent to or atop a large hernia. Stoma appliance bags fit poorly leading to leakage. +The ideal surgical solution for the patient is to rejoin the bowel and remove the stoma altogether, but this is not always possible. The stoma may be re-sited but further recur-rence is likely. Various open suture and mesh techniques have been described to repair parastomal hernia but failure rates are high. Meshes are best placed in the retromuscular space. Laparoscopic repair is also possible using a large mesh with a central hole. It can be positioned around the bowel onto the parietal peritoneum. +Recent reports (Millbourne et al.) have described the use of prophylactic mesh insertion at the time of formation of the stoma. A lightweight, polypropylene mesh is inserted in the retromuscular space so that the bowel passes through a hole in the mesh centre. Using this technique, parastomal hernia +rates have been reduced significantly. + +Traumatic hernia +These hernias arise through non-anatomical defects caused +by injury. They can be classified into three types: +1 Hernias through abdominal stab wound sites. These are effectively incisional hernias. +2 Hernias protruding through splits or tears in the abdomi-nal muscles after blunt trauma. +3 Abdominal bulging secondary to muscle atrophy that occurs as a result of nerve injury or other traumatic dener-vation. Akin to the lumbar pseudo-hernia seen after open nephrectomy, these can arise after a chest injury with +damage to the intercostal nerves. +Clinical features +Traumatic hernias present as any other hernia. The key to the aetiology is in the history and the non-anatomical location +of the hernia. +Treatment +Surgery may be justified if the hernia is sufficiently symp-tomatic, or if investigations suggest a narrow neck and hence a risk of obstruction or strangulation. Stab wound +traumatic hernias are straightforward to repair using open or + + +laparoscopic techniques as for other ventral hernias. Diffuse abdominal bulges are more difficult to correct and require some form of plication of the stretched musculofascial layer with mesh reinforcement to prevent further bulging in the +future. Some bulging may persist, however. + + +Rare external hernias Perineal hernia +This type of hernia is very rare and includes: +● postoperative hernia through a perineal scar, which may occur after excision of the rectum; +● median sliding perineal hernia, which is a complete pro-lapse of the rectum; +● anterolateral perineal hernia, which occurs in women and presents as a swelling of the labium majus; +● posterolateral perineal hernia, which passes through the +levator ani to enter the ischiorectal fossa. + +TREATMENT +A combined abdominoperineal operation is generally the most satisfactory for the last two types of hernia. The hernia is exposed by an incision directly over it. The sac is opened and its contents are reduced. The sac is cleared from sur-rounding structures and the wound closed. With the patient in semi-Trendelenburg position, either laparoscopically or at open surgery, the abdomen is opened and the mouth of the sac exposed. The sac is inverted, ligated and excised, and the pelvic floor repaired by muscle apposition and, if indicated, +buttressing of the repair with prosthetic mesh. +Obturator hernia +Obturator hernia, which passes through the obturator canal, occurs six times more frequently in women than in men. Most patients are aged >60 years. The swelling is liable to be overlooked because it is covered by pectineus. It seldom causes a definite swelling in Scarpa’s triangle, but, if the limb is flexed, abducted and rotated outwards, the hernia some-times becomes apparent. The leg is usually kept in a semi-flexed position and movement increases the pain. In more than 50% of cases of strangulated obturator hernia, pain is referred along the obturator nerve by its geniculate branch to the knee. On vaginal or rectal examination the hernia can sometimes be felt as a tender swelling in the region of the obturator foramen. +These hernias have often undergone strangulation, fre- +quently of the Richter type, by the time of presentation. + +TREATMENT +Surgery is indicated. The diagnosis is rarely made preoper-atively and so it is often approached through a laparotomy incision. The full Trendelenburg position is adopted. The con-stricting agent is the obturator fascia, which can be stretched by inserting the operator’s index finger, or suitable forceps, +through the gap in the fascia. Content is reduced. If incision + + +Antonio Scarpa, 1747–1832, Professor of Anatomy, Pavia, Italy +. +PART 11 | ABDOMINAL +1044 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +of the fascia is required, it is made parallel to the obturator vessels and nerve. The contents of the sac are dealt with in a standard manner. The defect cannot simply be closed because one margin is bone and the obturator nerve and vessels run through it. It is best closed using a mesh plug. In the absence of mesh or in an infected field, the broad ligament can be used as a plug. +Laparoscopic TAPP repair may also be performed again using a mesh. To avoid nerve injury, glue can be used to fix a +mesh over the defect. +Gluteal and sciatic hernias +Both of these hernias are very rare. A gluteal hernia passes through the greater sciatic foramen, either above or below piriformis. A sciatic hernia passes through the lesser sciatic foramen. Differential diagnosis must be made between these +conditions and: +● a lipoma or other soft tissue tumour beneath gluteus max-imus; +● a tuberculous abscess; +● a gluteal aneurysm. +All doubtful swellings in this situation can be charac-terised with CT scanning but, if in doubt, they should be +explored by operation. + + +UMBILICAL CONDITIONS IN THE +ADULT +Chronic infection +Chronic infection occurs in the umbilical area, particularly in patients with poor hygiene. It may also occur in obese people and when a paraumbilical hernia is present. It can be due to a plug of keratin causing chronic irritation. It is often encoun-tered during elective surgery and may complicate the insertion of a laparoscope port at the umbilicus. A range of bacteria and fungi can be involved. Occasionally, a rapid-onset, superficial cellulitis occurs even after minor surgery in this region. It is normally a streptococcus and can be treated with penicillin or other appropriate antibiotic. Pre-existing infection should be +treated before surgery where possible. + +Chronic fistula +Patients may present with a persistent discharge from the umbilical area. This may be due to simple, superficial infec- +tion or possibly an infected epidermoid cyst within the + +umbilicus. However, it may also be due to a fistulous connec-tion to deeper structures. +In normal patients, the umbilicus is connected to the liver, bladder and gynaecological organs by various ligaments. Diseases of these organs, such as infection or malignancy, can extend along these ligaments to appear at the umbilicus as a mass or fistulous discharge. +Chronic fistula may be a complication of umbilical her-nia repair due to chronic infection of a mesh or around non-absorbable suture material. In most cases this problem arises soon after surgery but occasionally a chronic infection can occur months or even years after an operation. Antibiotics may help but most commonly the synthetic suture or mesh will need to be removed with a risk of recurrence of the hernia. +In fetal life the umbilicus was also connected to the gut by the vitellointestinal duct. In most patients this duct becomes totally obliterated and vanishes. The bowel end of the duct may persist as Meckel’s diverticulum. More rarely, the umbil-ical end persists, leading to chronic discharge. If an abnormal connection between bowel and umbilicus persists, then this +band can act as a cause of adhesional intestinal obstruction. + +Patent urachus +A connection between the urinary bladder and umbilicus usu-ally presents in later life. This is due to increased pressure in the bladder as a result of obstruction from conditions such as prostatic hypertrophy. The cause of obstruction should be dealt with initially, but if the problem persists then surgical +excision of the patent urachus might be considered. + +Malignancy at the umbilicus +Primary squamous carcinoma may occur. If tumour presents at the umbilicus it is most probably due to spread from the internal organs along internal ligaments, e.g. from the liver along the falciform ligament. A malignant mass at the umbi-licus is called a Sister Joseph’s nodule. It usually indicates very advanced malignant disease and surgery probably has little to +offer (Figure 60.30). + +GENERAL INFECTION OF THE +ABDOMINAL WALL +The skin of the abdominal wall, similar to all skin, is prone to develop superficial infection that may be spontaneous, due to minor trauma or infection of skin lesions such as an +epidermoid cyst. Although antibiotics will suffice in most + + + + +Johann Frederick Meckel (the Younger), 1781–1833, Professor of Anatomy and Surgery, Halle, Germany, described his diverticulum in 1809. +Alexis Littre, 1658–1726, surgeon and lecturer in anatomy, Paris, France, described Meckel’s diverticulum in a hernial sac in 1700, 81 years before Meckel was born. +The neoplastic nodule sited at the umbilicus is known as Sister Joseph’s nodule. Sister Mary Joseph made the observation that her patients with terminal cancer sometimes developed a red papular lesion in the umbilicus. She and William Mayo published this observation in 1928. However, it was Hamilton Bailey who coined the term ‘Sister Mary Joseph’s nodule’ in 1949. +Sister Mary Joseph (nee Julia Dempsey), Nursing Superintendent, St Mary’s Hospital, which became the Mayo Clinic, Rochester MN, USA. +, +PART 11 | ABDOMINAL General infection of the abdominal wall 1045 + + +immunocompromised patients but can occasionally occur in healthy patients. Rapid diagnosis and aggressive surgical +debridement treatment are the key to success. + + + + + + + + + + + + + +Figure 60.30 Secondary nodule at the umbilicus – Sister Joseph’s nodule. + + +patients, if an abscess develops then surgical drainage may be required. +The close proximity of bowel and bowel organisms opens the abdominal wall to attack from a wide range of highly virulent bacteria. Most commonly, these are released during abdominal surgery such as appendicectomy and hence the +need for appropriate antibiotic prophylactic cover. + +Synergistic gangrene +This rare condition is due to the synergistic action of non-haemolytic streptococci and staphylococci causing rapid tissue necrosis and overwhelming systemic infection (Figure 60.31). It requires immediate administration of high-dose, broad-spectrum, powerful antibiotics in combination with early debridement of any non-viable tissue. Hyperbaric oxy-gen therapy has been advocated. +Other forms of severe abdominal wall infections occur, generally known as necrotising fasciitis (e.g. Fournier’s gangrene). All of these conditions have a high associated +morbidity and mortality. They occur in debilitated and + + + + + + + + + + + + + + + + +Figure 60.31 Bacterial synergistic gangrene of the chest and abdominal wall. The area has become gangrenous and looks like suede leather. Beware of amoebiasis cutis. + +Cutaneous fistula +Due to the thickness of the abdominal wall, it is rare for abdominal inflammatory conditions to discharge sponta-neously through the wall to the skin. Chronic intraperitoneal abscesses arising after occult bowel perforation, appendicitis, diverticulitis and cholecystitis are the most likely sources. CT will locate the internal abscess and suggest the likely origin. Treatment today is usually by CT- or ultrasound-guided drain-age but the surgeon may be called on to remove the source organ, e.g. gall bladder. +Malignancy in its later stages can occasionally erode through the abdominal wall. +Crohn’s disease also has a tendency to fistulate into adja- +cent organs and may develop an enterocutaneous fistula. + +Abdominal compartment syndrome +Surgeons are increasingly aware of the harmful effect of high intra-abdominal pressures that can occur in severe intra-abdominal sepsis, such as pancreatitis and also aortic aneurysm rupture. High pressure leads to reduced blood flow and tissue ischaemia, which contributes to multiorgan failure. Although the abdominal wall has elasticity, if intra-abdominal volume increases due to fluid, gas, pus, tissue oedema, etc., then max-imal capacity may be reached and pressure rises to a critical level. Tension-releasing incisions, equivalent to a fasciotomy, have been suggested, although this is not widely practised. +In some cases, after surgery for severe intraperitoneal sepsis, the surgeon cannot close the abdomen and may leave the incision wide open, covering abdominal contents with mesh or a saline-soaked dressing, planning to return at a +future date to close the defect. This is called a laparostomy. + +Neoplasms of the abdominal wall +As the abdominal wall is composed of muscle, fascia and bone, benign and malignant tumours can arise from each, +although these are rare. +Desmoid tumour +This is usually considered by pathologists to be a hamartoma and is more common in women. Some, however, believe it to be a fibroma and possibly the result of repeated trauma. Des-moids have been reported in familial adenomatous polyposis (FAP). Histologically, they contain plasmoidal cell masses resembling giant cells. They undergo central myxomatous change. Surgical excision with a wide margin is required to +prevent recurrence, which is a frequent problem. +Fibrosarcoma +These tumours can occur anywhere in the body They are generally highly malignant and respond poorly to both radio-and chemotherapy. Wide excision will often require plastic +. +surgical reconstruction. +PART 11 | ABDOMINAL +1046 CHAPTER 60 Abdominal wall, hernia and umbilicus + + + +FURTHER READING +Flum DR, Horvath K, Koepsell T. Have outcomes of incisional hernia repair improved with time? A population-based study. Ann Surg 2003; 237: 129–35. +Kingsnorth AN, LeBlanc KA. Management of abdominal hernias, 4th edn. London: Edward Arnold, 2013. +Millbourn D, Cengiz Y, Israelsson LA. Effect of stitch length on wound complications after closure of midline incisions. Arch Surg 2009; 144: 1056–9. + +WEBSITE ADDRESSES +Classification of groin hernia: www.herniaweb.org/documents/EHS_ groin_hernia_classification.pdf +Guidelines for management of groin hernia: www.herniaweb.org/ documents/EHS_Guidelines.pdf +European classification of primary and incisional abdominal wall her-nias: www.ncbi.nlm.nih.gov/pmc/articles/PMC2719726/ +NICE guidelines for laparoscopic inguinal hernia repair: guidance.nice. org.uk/TA83 +SIGN guidelines for antibiotics in surgery (including hernia): www.sign. ac.uk/pdf/sign104.pdf +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love +61 +Chapter + + + +The peritoneum, omentum, mesentery and retroperitoneal space + + + + + +Learning objectives +To recognise and understand: +• The causes and complications of localised and generalised peritonitis + + + +• The causes and pathophysiology of ascites +• The pathophysiology and complications of adhesion formation + + + +• The clinical features of peritonitis and intraperitoneal abscess +• The principles of surgical management in patients with peritonitis and intraperitoneal abscess + +• The spectrum of mesenteric and retroperitoneal conditions + + + + + +ANATOMY AND PHYSIOLOGY Embryology +The peritoneal cavity, mesenteries and omentum have an anatomical complexity that can perhaps be truly understood only with surgical experience. Nevertheless, an understand-ing of the geometric alterations occurring during early gas-trointestinal (GI) morphogenesis (regionalisation, elongation and coiling) of the derivatives of the endoderm (E = epithe-lium) and the visceral mesoderm (M = muscle and most of the rest) along with the later fusion of adjacent layers of perito-neum will give an appreciation of how the adult disposition +is as it is. + +Adult arrangement and functions +The peritoneal cavity is the largest cavity in the body, the surface area of its lining membrane (2m2 in an adult) being nearly equal to that of the skin. The peritoneal membrane is composed of flattened polyhedral cells (mesothelium), one layer thick, resting on a thin layer of fibroelastic tissue. This membrane is conveniently divided into two parts – the vis-ceral peritoneum surrounding the viscera and the parietal peritoneum lining the other surfaces of the cavity. Beneath the peritoneum, supported by a small amount of areolar tissue, lies a network of lymphatic vessels and rich plexus of capillary blood vessels from which all absorption and exudation must occur. In health, only a few millilitres of peritoneal fluid are found in the peritoneal cavity. The fluid is pale yellow, some-what viscid, and contains lymphocytes and other leukocytes; it lubricates the viscera, allowing easy movement and peri- +stalsis. The parietal portion is richly supplied with nerves and, + + +when irritated, causes severe pain that is accurately localised to the affected area. The visceral peritoneum, in contrast, is poorly supplied with nerves (these being situated around blood vessels) and its irritation causes pain that is usually poorly localised to the midline. +The peritoneum has a number of functions. + + +Summary box 61.1 + +Functions of the peritoneum In health +●● Visceral lubrication +●● Fluid and particulate absorption + +In disease +●● Pain perception (mainly parietal) +●● Inflammatory and immune responses ●● Fibrinolytic activity + + +The peritoneum has the capacity to absorb large volumes of fluid: this ability is used during peritoneal dialysis in the treatment of renal failure. However, the peritoneum can also produce large volumes of fluid (ascites) and an inflamma-tory exudate when injured (peritonitis). During expiration, intra-abdominal pressure is reduced and peritoneal fluid, aided by capillary attraction, travels in an upward direction towards the diaphragm. Experimental evidence shows that particulate matter and bacteria are absorbed within a few minutes into the lymphatic network through a number of ‘pores’ within the diaphragmatic peritoneum. The circulation +of peritoneal fluids may be responsible for the occurrence of +PART 11 | ABDOMINAL +y +1048 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + +abscesses distant from primary disease. When parietal perito-neal defects are created, healing occurs not from the edges but by the development of new mesothelial cells throughout the surface of the defect. In this way, large defects heal as rapidly +as small defects. + +SCOPE OF DISEASE +The peritoneum, mesentery and omentum may be the site of a variety of conditions that reflect their relationship to other anatomical structures or, in some instances, their primary +functions. + + +Summary box 61.2 + +Scope of disease Intraperitoneal disease ●● Peritonitis +Primary Secondary +●● Abscess ●● Ascites +Transudate Exudate +●● Tumours Primary Secondary +●● Adhesions + +Omental disease ●● Hernia +●● Adhesions ●● Torsion +●● Neoplasia + +Mesenteric disease ●● Trauma +●● Ischaemia +●● Inflammation ●● Cysts +●● Neoplasia + +Retroperitoneal disease +●● Chronic inflammation/fibrosis ●● Abscess +●● Tumours + + +PERITONITIS +Peritonitis is simply defined as inflammation of the perito-neum and may be localised or generalised. Most cases of peri-tonitis are caused by an invasion of the peritoneal cavity by bacteria, so that, when the term ‘peritonitis’ is used without qualification, acute bacterial peritonitis is often implied. In +this instance, free fluid spills into the peritoneal cavity and + +circulates, largely directed by the normal peritoneal attach-ments and gravit . For example, spillage from a perforated peptic ulcer may run down the right paracolic gutter, leading to presentation with pain in the right iliac fossa (Valentino’s syndrome). Even in patients with non-bacterial peritonitis (e.g. acute pancreatitis, intraperitoneal rupture of the blad-der or haemoperitoneum), the peritoneum often becomes infected by transmural spread of organisms from the bowel. Such translocation is a feature of the systemic inflammatory response on the bowel and it is not long (often a matter of hours) before a bacterial peritonitis develops. Most duodenal and gastric perforations are initially sterile for up to several +hours before becoming secondarily infected. + + +Summary box 61.3 + +Causes of peritoneal inflammation +●● Bacterial, gastrointestinal and non-gastrointestinal ●● Chemical, e.g. bile, barium +●● Allergic, e.g. starch peritonitis +●● Traumatic, e.g. operative handling +●● Ischaemia, e.g. strangulated bowel, vascular occlusion ●● Miscellaneous, e.g. familial Mediterranean fever + + +Although acute bacterial peritonitis most commonly arises from a perforation of a viscus of the alimentary tract, other routes of infection can include the female genital tract and exogenous contamination. There are also less common forms in which the aetiology is a primary ‘spontaneous’ peritonitis, in which a pure infection with streptococcal, pneumococcal +or haemophilus bacteria occurs. + + +Summary box 61.4 + +Paths to peritoneal infection +●● Gastrointestinal perforation, e.g. perforated ulcer, appendix, diverticulum +●● Transmural translocation (no perforation), e.g. pancreatitis, ischaemic bowel, primary bacterial peritonitis +●● Exogenous contamination, e.g. drains, open surgery, trauma, peritoneal dialysis +●● Female genital tract infection, e.g. pelvic inflammatory disease ●● Haematogenous spread (rare), e.g. septicaemia + + +Microbiology +Bacteria from the gastrointestinal tract +The number of bacteria within the lumen of the gastroin-testinal tract is normally low until the distal small bowel is reached. However, disease leading to stasis and overgrowth (e.g. obstruction, and chronic and acute motility distur- +bances) may increase proximal colonisation. The biliary + + +Rudolph Valentino, Italian actor, died during surgery for this in New York in 1926, aged 31 years. +PART 11 | ABDOMINAL Peritonitis 1049 + + + +and pancreatic tracts are also normally free from bacteria, although they may be infected in disease, e.g. gallstones. Peri-toneal infection is usually caused by two or more bacterial strains. Gram-negative bacteria contain endotoxins (lipo-polysaccharides) in their cell walls that have multiple toxic effects on the host, primarily by causing the release of tumour necrosis factor (TNF) from host leukocytes. Systemic absorp-tion of endotoxin may produce endotoxic shock, with hypo-tension and impaired tissue perfusion. Other bacteria such as Clostridium perfringens produce harmful exotoxins. Bacteroides spp. are commonly found in peritonitis. These gram-negative, non-sporing organisms, although predominant in the lower intestine, often escape detection because they are strictly anaerobic and slow to grow on culture media unless there is an adequate carbon dioxide tension in the anaerobic apparatus (Gillespie). In many laboratories, the culture is discarded if there is no growth in 48 hours. These organisms are resistant to penicillin and streptomycin but sensitive to metronidazole, clindamycin, Augmentin (amoxicillin plus clavulanic acid) and cephalosporin compounds. Since the widespread use of metronidazole (Flagyl), bacteroides infections have greatly +diminished. +Non-gastrointestinal causes of peritonitis +Pelvic infection via the fallopian tubes is responsible for a high proportion of ‘non-gastrointestinal’ infections. The most common offending organisms are Chlamydia spp. and gono-cocci. These organisms lead to a thinning of cervical mucus and allow bacteria from the vagina into the uterus and ovi-ducts, causing infection and inflammation. A variant of tran-speritoneal spread of such organisms is perihepatitis which can cause scar tissue to form on Glisson’s capsule, a thin layer of connective tissue surrounding the liver (Fitz-Hugh– +Curtis syndrome). Other bacterial variants that are discussed + + +Summary box 61.5 + +Microorganisms in peritonitis +Gastrointestinal source + +separately include tuberculosis and other mycobacterial strains and those causing primary peritonitis (pneumococci, staphylococci and streptoccoci). Fungal peritonitis is uncom- +mon but may complicate severely ill patients. + +Localised peritonitis +Anatomical and pathological factors may favour the localisa- +tion of peritonitis. +Anatomical +The greater sac of the peritoneum is divided into (1) the sub-phrenic spaces, (2) the pelvis and (3) the peritoneal cavity proper. The last is divided into a supracolic and an infracolic compartment by the transverse colon and transverse meso-colon, which deters the spread of infection from one to the other. When the supracolic compartment overflows, as is often the case when a peptic ulcer perforates, it does so over the colon into the infracolic compartment or by way of the right +paracolic gutter to the right iliac fossa and hence to the pelvis. +Pathological +The clinical course is determined in part by the manner in which adhesions form around the affected organ. Inflamed peritoneum loses its glistening appearance and becomes red-dened and velvet . Flakes of fibrin appear and cause loops of intestine to become adherent to each other and to the pari-etes. There is an outpouring of serous inflammatory exudate rich in leukocytes and plasma proteins that soon becomes turbid; if localisation occurs, the turbid fluid becomes frank pus. Peristalsis is retarded in affected bowel and this helps to prevent distribution of the infection. The greater omentum, by enveloping and becoming adherent to inflamed structures, often forms a substantial barrier to the spread of infection +y +(see below). + +Diffuse (generalised) peritonitis +A number of factors may favour the development of diffuse +peritonitis: + + + +●● Escherichia coli ●● Streptococci +●● Enterococci +●● Bacteroides spp. ●● Clostridium spp. +●● Klebsiella pneumoniae + +Other sources +●● Chlamydia trachomatis ●● Neisseria gonorrhoeae ●● Haemolytic streptococci ●● Staphylococci +●● Streptococcus pneumoniae +●● Mycobacterium tuberculosis and other species +●● Fungal infections + + +● Speed of peritoneal contamination is a prime factor. If an inflamed appendix or other hollow viscus perforates before localisation has taken place, there will be an efflux of con-tents into the peritoneal cavity, which may spread over a large area almost instantaneously. Perforation proximal to an obstruction or from sudden anastomotic separation is associated with severe generalised peritonitis and a high mortality rate. +● Stimulation of peristalsis by the ingestion of food or even water hinders localisation. Violent peristalsis occasioned by the administration of a purgative or an enema may cause the widespread distribution of an infection that would otherwise have remained localised. +● The virulence of the infecting organism may be so great as +to render the localisation of infection difficult or impossible. + + +William Alexander Gillespie, formerly Professor of Clinical Bacteriology, University of Bristol, Bristol, UK. +Named after the two physicians, Thomas Fitz-Hugh, Jr and Arthur Hale Curtis who first reported this condition in 1934 and 1930, respectively. +PART 11 | ABDOMINAL +1050 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + +● Young children have a small omentum, which is less effec-tive in localising infection. +● Disruption of localised collections may occur with inju-dicious handling, e.g. appendix mass or pericolic abscess. +● Deficient natural resistance (‘immune deficiency’) may result from use of drugs (e.g. steroids), disease [e.g. acquired +immune deficiency syndrome (AIDS)] or old age. +With appropriate treatment, localised peritonitis usually resolves; in about 20% of cases, an abscess follows. Infre-quently, localised peritonitis becomes diffuse. Conversely, in favourable circumstances, diffuse peritonitis can become localised, most frequently in the pelvis or at multiple sites +within the abdominal cavity. + + +Clinical features Localised peritonitis +The initial symptoms and signs of localised peritonitis are those of the underlying condition – usually visceral inflam-mation (hence abdominal pain, specific GI symptoms + mal-aise, anorexia and nausea). When the peritoneum becomes inflamed, abdominal pain will worsen and in general tempera-ture and pulse rate will rise. The pathognomonic signs are localised guarding (involuntary abdominal wall contraction to protect the viscus from the examining hand), a positive ‘release’ sign (rebound tenderness) and sometimes rigidity (involuntary constant contraction of the abdominal wall over the inflamed parietes). If inflammation arises under the dia-phragm, shoulder-tip (‘phrenic’) pain may be felt as the pain is referred to the C5 dermatome. In cases of pelvic peritonitis arising from an inflamed appendix in the pelvic position or from salpingitis, the abdominal signs are often slight; there may be deep tenderness of one or both lower quadrants alone, but a rectal or vaginal examination reveals marked tenderness +of the pelvic peritoneum. +Diffuse (generalised) peritonitis EARLY +Abdominal pain is severe and made worse by moving or breathing. It is first experienced at the site of the original lesion and spreads outwards from this point. The patient usu-ally lies still. enderness and generalised guarding are found on palpation, when the peritonitis affects the anterior abdom-inal wall. Infrequent bowel sounds may still be heard for a few hours but they cease with the onset of paralytic ileus. Pulse and temperature rise in accord with degree of inflammation +T +and infection. + +LATE +If resolution or localisation of generalised peritonitis does not occur, the abdomen will become rigid (generalised rigidity). Distension is common and bowel sounds are absent. Circu-latory failure ensues, with cold, clammy extremities, sunken eyes, dry tongue, thready (irregular) pulse, and drawn and +anxious face (hippocratic facies – Figure 61.1). The patient + + + + + + + +Figure 61.1 The hippocratic facies in terminal diffuse peritonitis. + + +finally lapses into unconsciousness. With early diagnosis and adequate treatment, this condition is rarely seen in modern +surgical practice. + + +Summary box 61.6 + +Clinical features of peritonitis +●● Abdominal pain, worse on movement, coughing and deep respiration +●● Constitutional upset: anorexia, malaise, fever, lassitude ●● GI upset: nausea ± vomiting +●● Pyrexia (may be absent) ●● Raised pulse rate +●● Tenderness ± guarding/rigidity/rebound of abdominal wall +●● Pain/tenderness on rectal/vaginal examination (pelvic peritonitis) +●● Absent or reduced bowel sounds +●● ‘Septic shock’ (systemic inflammatory response syndrome [SIRS] and multi-organ dysfunction syndrome [MODS]) in later stages + + +Diagnostic aids +Investigations may elucidate a doubtful diagnosis, but the importance of a careful history and repeated examination +must not be forgotten. +Bedside +● Urine dipstix for urinary tract infection. +● ECG if diagnostic doubt (as to cause of abdominal pain) +or cardiac history. + + +Hippocrates of Kos, Greek physician and surgeon, and by common consent ‘the father of medicine’, was born on the island of Kos, off Turkey +, +about 460bc and probably died in 375bc. +PART 11 | ABDOMINAL Peritonitis 1051 + + + +Bloods +● Baseline urea and electrolytes (U&Es) for treatment. ● Full blood count for white cell count (WCC). +● Serum amylase estimation may establish the diagnosis of acute pancreatitis provided that it is remembered that moderately raised values are frequently found following other abdominal catastrophes and operations, e.g. perfor-ated duodenal ulcer +. +● Group and save may be taken as an adjunct to impending +surgery. + +Imaging +● Erect chest radiograph to demonstrate free subdiaphrag-matic gas (Figure 61.2a). +● A supine radiograph of the abdomen may confirm the presence of dilated gas-filled loops of bowel (consistent with a paralytic ileus), and occasionally show other gas-filled structures that may aid diagnosis, e.g. biliary tree; the +faecal pattern may act as a guide to colonic disease (absent + +in sites of significant inflammation, e.g. diverticulitis). In the patient who is too ill for an ‘erect’ film, a lateral decu-bitus film can show gas beneath the abdominal wall (if CT unavailable). +● Multiplanar computed tomography (CT) is increasingly used to identify the cause of peritonitis (Figures 61.2b and 61.3) and may also influence management decisions, e.g. surgical strategy. There is an abundance of recent published evidence to support its use in managing acute abdominal pain. +● Ultrasonography has undoubted value in certain situa-tions such as pelvic peritonitis in women and localised +right upper quadrant peritonism. + + + +(a) + + + + + + +Figure 61.3 Acute pancreatitis seen on computed tomography scanning with swelling of the gland and surrounding inflammatory changes (courtesy of Dr J Healy, Chelsea and Westminster Hospital, London, UK). + + +Invasive +● In the era of access to high-quality CT scanning, perito- +neal diagnostic aspiration has little residual value. + + + + + +(b) + + + + + + + + + + + + + +Figure 61.2 (a) Gas under the diaphragm in a patient with free per-foration and peritonitis; (b) representative axial image through the pelvis of the same patient showing perforated sigmoid diverticular disease. + +Management +General care of the patient +The care of critically ill surgical patients is described in detail in Chapters 2, 17 and 20. Nutritional support is covered in Chapter 19, and anaesthesia and pain relief in Chapter 18. Depending on degree (localised/generalised), duration and +severity, patients will require some or all of the following. + +CORRECTION OF FLUID LOSS AND CIRCULATING VOLUME +Patients are frequently hypovolaemic with electrolyte distur-bances. The plasma volume must be restored and electrolyte concentrations corrected. Fluid balance should be monitored and pre-existent and ongoing losses corrected. Special mea-sures may be needed for cardiac, pulmonary and renal support, especially if septic shock is present (see Chapter 2), including central venous pressure monitoring in patients with concur- +rent disease. +PART 11 | ABDOMINAL +1052 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + +URINARY CATHETERISATION ± GASTROINTESTINAL DECOMPRESSION +A urinary catheter will give a guide to central perfusion and will be required if abdominal surgery is to proceed. A nasogastric tube is commonly passed to allow drainage ± aspiration until paralytic ileus has resolved. + +ANTIBIOTIC THERAPY +Administration of parenteral broad-spectrum (aerobic and anaerobic) antibiotics prevents the multiplication of bacteria and the release of endotoxins. + +ANALGESIA +The patient should be nursed in the sitting-up position and must be relieved of pain before and after the operation. If appropriate expertise is available, epidural infusion may provide excellent analgesia. Freedom from pain allows early mobilisation and adequate physiotherapy in the postoperative period, which helps to prevent basal pulmonary collapse, deep vein thrombosis and pulmonary embolism. + +Specific treatment of the cause +Although difficult to generalise, in patients in whom specific treatment has not been guided by CT scanning, early surgi-cal intervention is to be preferred to a ‘wait and see’ policy assuming that the patient is fit for anaesthesia and that resus-citation has resulted in a satisfactory restitution of normal body physiology. This rule is particularly true for previously healthy patients and those with postoperative peritonitis. More caution is of course required in patients at high opera-tive risk because of comorbidity or advanced age. +, +In those patients with a preoperative diagnosis, if the cause of peritonitis is amenable to surgery, an operation must be carried out as soon as the patient is fit. This is usually within a few hours. In peritonitis caused by pancreatitis or salpingitis, or in cases of primary peritonitis of streptococcal or pneumococcal origin, non-surgical treatment is preferred, provided that the diagnosis can be made with confidence. It is beyond the remit of this chapter to cover specific causes of peritonitis and their treatment, be it by an open or a laparo-scopic approach. However, in general, surgery is directed to removing (or diverting) the cause and subsequent adequate peritoneal lavage ± drainage. In operations for generalised + + +Summary box 61.7 + +Management of peritonitis General care of patient +●● Correction of fluid and electrolyte imbalance +●● Insertion of nasogastric drainage tube and urinary catheter ●● Broad-spectrum antibiotic therapy +●● Analgesia +●● Vital system support + +Surgical treatment of cause when appropriate ●● Remove or divert cause +●● Peritoneal lavage ± drainage + +peritonitis it is essential that, after the cause has been dealt with, the whole peritoneal cavity be explored with the sucker and, if necessary, mopped dry until all seropurulent exudate has been removed. The use of a large volume of saline (typi-cally 3 litres) containing dissolved antiseptic or antibiotic has +been shown to be effective. + +Prognosis and complications +With modern treatment, diffuse peritonitis carries a mortal-ity rate of about 10%, reflecting the degree and duration of peritoneal contamination, age and fitness of the patient, and the nature of the underlying cause. Paralytic ileus is covered in detail in Chapter 71, and abscess formation and adhesions +are covered below. + + +Summary box 61.8 + +Systemic complications of peritonitis ●● Septic shock +●● Systemic inflammatory response syndrome ●● Multi-organ dysfunction syndrome +●● Death + +Abdominal complications of peritonitis ●● Paralytic ileus +●● Residual or recurrent abscess/inflammatory mass ●● Portal pyaemia/liver abscess +●● Adhesional small bowel obstruction + + + +SPECIAL FORMS OF +PERITONITIS Bile peritonitis +Unless there is reason to suspect that the biliary tract was damaged during surgery or the patient has proven acute cho-lecystitis, it is improbable that bile will be thought of as a cause of peritonitis until the abdomen has been opened. +Unless the bile has extravasated slowly and the collection becomes shut off from the general peritoneal cavity, there are symptoms (often severe pain) and signs of diffuse peritonitis. After a few hours a tinge of jaundice is not unusual. Laparot-omy (or laparoscopy) should be undertaken with evacuation of the bile and peritoneal lavage. The source of bile leakage should be identified and treated accordingly. Infected bile is more lethal than sterile bile. A ‘blown’ duodenal stump should be drained because it is too oedematous to repair, but sometimes it can be covered by a jejunal patch. The patient is often jaundiced from absorption of peritoneal bile, but the surgeon must ensure that the abdomen is not closed until any obstruction to a major bile duct has been either excluded or relieved. Bile leaks after cholecystectomy or liver trauma may be dealt with by percutaneous (ultrasound-guided) drainage and endoscopic biliary stenting to reduce bile duct pressure. +The drain is removed when dry and the stent at 4–6 weeks. +PART 11 | ABDOMINAL +Special forms of peritonitis 1053 + + + + +Summary box 61.9 + +Causes of bile peritonitis +●● Perforated gall bladder secondary to inflammation or obstruction (especially empyema) +●● Post-cholecystectomy: Cystic duct stump leakage +Leakage from an accessory duct in the gall-bladder bed Bile duct injury +-tube drain dislodgement (or tract rupture on removal) ●● Following other operations/procedures: +T +Duodenal injury +Leaking duodenal stump post gastrectomy Leaking biliary–enteric anastomosis +Leakage around percutaneously placed biliary drains ●● Blunt or penetrating hepatobiliary or duodenal trauma + + +Spontaneous bacterial peritonitis +Spontaneous bacterial peritonitis (SBP; sometimes called pri-mary bacterial peritonitis) is an acute bacterial infection of ascitic fluid. It can occur in children and adults and can in theory occur as a complication of any disease state that pro-duces the clinical syndrome of ascites. In practice, it is rare except in patients with cirrhosis and ascites, affecting 1.5– 3.5% of outpatients and approximately 10% of inpatients. +Clinical features usually include local symptoms and/or signs of peritonitis, GI upset (secondary to ileus, e.g. nausea and vomiting), signs of systemic inflammation (hyper- or hypothermia, chills, tachycardia and tachypnoea ± signs of septic shock), worsening liver and renal function, hepatic encephalopathy and GI bleeding. It should, however, be noted that evolving infection may be asymptomatic, espe-cially in outpatients. +The diagnosis is made by paracentesis, and this should be considered in patients with cirrhosis and ascites even when there is a low index of suspicion. Some guidelines recom-mend diagnostic paracentesis in all patients with cirrhosis and ascites on hospital admission. The diagnosis is made by an increased neutrophil of 250/mm3 as determined by count-ing centrifuged ascitic fluid. Ascites culture is negative in as many as 60% of patients with clinical manifestations of SBP and increased ascitic neutrophil count. When culture is pos-itive (40% of cases), the most common pathogens include gram-negative bacteria, usually E. coli, and gram-positive cocci (mainly streptococci and enterococci). +Empirical treatment of SBP must be initiated immedi-ately after diagnosis before the results of culture have been received. Although the choice of antibiotic may vary inter-nationally, a third-generation cephalosporin, e.g. cefotaxime, is the best investigated, and it avoids the renal toxicity of aminoglycosides. Alternatives are amoxicillin/clavulanic acid and quinolones such as ciprofloxacin. Complications of SBP, e.g. septic shock, GI bleeding and hypoalbuminaemia, should be managed accordingly. +When first described, the mortality rate of SBP exceeded 90%, but it has been reduced to approximately 20% with +early diagnosis and treatment. + +Primary pneumococcal peritonitis +This may complicate nephrotic syndrome or cirrhosis in children. Otherwise healthy children, particularly girls aged between 3 and 9 years, may also be affected, and it is likely that the route of infection is sometimes via the vagina and fallopian tubes. At other times, and always in boys, the infec-tion is blood borne and secondary to respiratory tract or mid-dle-ear disease. The prevalence of pneumococcal peritonitis has declined greatly and the condition is now rare. In brief, the clinical onset is sudden, with pain usually localised to the lower half of the abdomen. The temperature is raised to 39°C or more and there is usually frequent vomiting. After 24–48 hours, profuse diarrhoea is characteristic. There is usually increased frequency of micturition. The last two symptoms are caused by severe pelvic peritonitis. On examination, peri-tonism is usually diffuse but less prominent than in most cases of a perforated viscus, leading to peritonitis. +A leukocytosis of ≥30000/µL, with approximately 90% polymorphs, suggests pneumococcal peritonitis rather than another cause, e.g. appendicitis. After starting antibiotic therapy and correcting dehydration and electrolyte imbal-ance, early surgery is required unless spontaneous infection of pre-existing ascites is strongly suspected, in which case a diagnostic peritoneal tap is useful. Laparotomy or laparoscopy may be used. Should the exudate be odourless and sticky, the diagnosis of pneumococcal peritonitis is practically certain, but it is essential to perform a careful exploration to exclude other pathology. Assuming that no other cause for the perito-nitis is discovered, some of the exudate is aspirated and sent to the laboratory for microscopy, culture and sensitivity tests. Thorough peritoneal lavage is carried out and the incision closed. Antibiotics and fluid replacement therapy are contin-ued and recovery is usual. +Other organisms are now known to cause some cases of primary peritonitis in children, including Haemophilus spp., group A streptococci and a few gram-negative bacteria. Underlying pathology (including an intravaginal foreign body in girls) must always be excluded before primary peritonitis can be diagnosed with certainty. Idiopathic streptococcal and +staphylococcal peritonitis can also occur in adults. + +Tuberculous peritonitis +Intra-abdominal tuberculosis (TB) is very common in resource-poor countries where all general surgeons are familiar with its presentation and management. The incidence is, how-ever, also rising in resource-rich countries as a consequence of migration and immunosuppression where Mycobacterium avium-intracellulare is becoming increasingly prevalent with the widespread increase in human immunodeficiency virus (HIV) co-infection. The abdomen is involved in 11% of patients with extrapulmonary TB and includes intraperito-neal, GI tract and solid organ disease forms, with TB perito-nitis being a common site-specific variant (ileocaecal is the most common site of involvement). Although still uncom-mon, TB peritonitis requires some specific mention because it is often diagnosed late in the course of the disease, resulting +in undue patient morbidity and mortality. +PART 11 | ABDOMINAL +1054 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + +(a) (c) + + + + + + + + + + + + + + + + + +(b) + + + + + + + + + +Figure 61.4 (a) Plain chest radiograph from a 55-year-old man showing miliary tuberculosis (TB); (b, c) representative computed tomography images from the same patient showing gross ascites, nodular stranding in the omentum and mesentery, as well as nodular enhancement of the peritoneum – TB peritonitis (courtesy of Dr S Burke, Homerton University Foundation Trust, London). + + + + + +Tuberculosis can spread to the peritoneum through the GI tract (typically the ileocaecal region) via mesenteric lymph nodes or directly from the blood, usually from the ‘miliary’ (Figure 61.4a), but occasionally from the ‘cavitating’ form of pulmonary TB, lymph and the fallopian tubes; 50–83% of patients with abdominal TB can be expected to have perito-neal involvement. Clinical or subclinical ascites is reported in virtually all patients with TB peritonitis and is frequently a presenting feature. In the most common form of the disease, ascites may be localised or generalised throughout the perito-neal cavity. Multiple tubercle deposits appear on both layers of the peritoneum. Diagnosis is via abdominal ultrasonogra-phy or CT to detect ascites and lymphadenopathy ± diffuse thickening of the peritoneum, mesentery and/or omentum (Figure 61.4b, c). Ascitic fluid is typically a straw-coloured exudate (protein >25–30g/L) with white cells >500mm3 and lymphocytes >40%. Unfortunately diagnostic smears for acid-fast bacilli are diagnostic in less than 3% of patients, and culture may take up to 4–8 weeks with no guarantee of a pos- +itive result. Laparoscopy and peritoneal biopsy may thus be + +helpful to couple typical appearances with histology. TB man-agement is principally supportive (nutrition and hydration) and medical (systemic antituberculous therapy, noting that multiple-drug resistance may be higher for abdominal than for pulmonary TB), although surgery may be required for specific +complications such as intestinal obstruction. + + + +Summary box 61.10 + +Tuberculous peritonitis +●● Acute (may be clinically indistinguishable from acute bacterial peritonitis) and chronic forms +●● Abdominal pain, sweats, malaise and weight loss are frequent ●● Ascites common, may be loculated +●● Caseating peritoneal nodules are common – distinguish from metastatic carcinoma and fat necrosis of pancreatitis +●● Intestinal obstruction may respond to anti-tuberculous treatment without surgery +PART 11 | ABDOMINAL Intraperitoneal abscess 1055 + + + +Familial Mediterranean fever +(periodic peritonitis) +Familial Mediterranean fever (periodic peritonitis) is char-acterised by abdominal pain and tenderness, mild pyrexia, polymorphonuclear leukocytosis and, occasionally, pain in the thorax and joints. The duration of an attack is 24–72 hours, when it is followed by complete remission, but exac-erbations recur at regular intervals. Most of the patients have undergone appendicectomy in childhood. This disease, often familial, is limited principally to Arab, Armenian and Jewish populations; other races are occasionally affected. Mutations in the MEFV (Mediterranean fever) gene appear to cause the disease. This gene produces a protein called pyrin, which is expressed mostly in neutrophils; however, the exact function of pyrin is not known. +Usually, children are affected but it is not rare for the dis-ease to make its first appearance in early adult life, with cases in women outnumbering those in men by two to one. Excep-tionally, the disease becomes manifest in patients aged >40 years. At surgery, which may be necessary to exclude other causes (but should be avoided if possible), the peritoneum is inflamed, particularly in the vicinity of the spleen and the gall bladder. There is no evidence that the interior of these organs is abnormal. Colchicine therapy is used during attacks and to +prevent recurrent attacks. + +INTRAPERITONEAL ABSCESS +Following intraperitoneal sepsis (usually manifest first as local or diffuse peritonitis), the anatomy of the peritoneal cavity is such that with the influence of gravity (depending on patient position – sitting or supine), abscess development usually occupies one of a number of specific abdominal or pelvic sites. In general, the symptoms and signs of a purulent collec-tion may be vague and consist of nothing more than lassi-tude, anorexia and malaise, pyrexia (often low grade), mild tachycardia and localised tenderness. Certain sites have more specific clinical features. Larger abscesses will give rise to the picture of swinging pyrexia and pulse and a palpable mass. +Blood tests will reveal elevated inflammatory markers. + + +Summary box 61.11 + +Clinical features of an abdominal/pelvic abscess Symptoms +●● Malaise, lethargy – failure to recover from surgery as expected ●● Anorexia and weight loss +●● Sweats ± rigors +●● Abdominal/pelvic pain +●● Symptoms from local irritation, e.g. shoulder tip/hiccoughs (subphrenic), diarrhoea and mucus (pelvic), nausea and vomiting (any upper abdominal) + +Signs +●● Increased temperature and pulse ± swinging pyrexia +●● Localised abdominal tenderness ± mass (including on pelvic exam) + +Pelvic abscess +The pelvis is the most common site of abscess formation because the vermiform appendix is often pelvic in position and the fallopian tubes are also frequent sites of infection. A pelvic abscess can also occur as a sequel to any case of diffuse peritonitis, and is common after anastomotic leakage follow- +ing colorectal surgery. +Clinical features +The most characteristic symptoms are of pelvic pain, diar-rhoea and the passage of mucus in the stools. Rectal examina-tion reveals a bulging of the anterior rectal wall, which, when +the abscess is ripe, becomes softly cystic. +Investigation and management +Left to nature, a proportion of these abscesses burst into the rectum, after which the patient almost always recovers rapidly. If this does not occur, the abscess should be drained deliberately. In women, vaginal drainage through the poste-rior fornix is often chosen. In other cases, when the abscess is definitely pointing into the rectum, rectal drainage (Figure 61.5) is employed. If any uncertainty exists, the presence of pus should be confirmed by ultrasonography or CT scanning (Figure 61.6). Laparotomy is almost never necessary and rec-tal drainage of a pelvic abscess is far preferable to suprapubic drainage, which risks exposing the general peritoneal cavity to infection. It is, however, increasingly common to insert drainage tubes percutaneously, e.g. via the buttock or via the +vagina or rectum under CT guidance. + +Abdominal abscess Anatomy +The complicated arrangement of the peritoneum results in the formation of four intraperitoneal spaces in which pus may +commonly collect (Figure 61.7). +LEFT SUBPHRENIC SPACE +This is bounded above by the diaphragm and behind by the left triangular ligament and the left lobe of the liver, the gas- +trohepatic omentum and the anterior surface of the stomach. + + + + + + + + + + + + + + + +Figure 61.5 Opening a pelvic abscess into the rectum. +PART 11 | ABDOMINAL +1056 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + + + + + + + + + + + + + + + + + + +Figure 61.6 A pelvic abscess seen on CT scanning (courtesy of Dr J Healy, Chelsea and Westminster Hospital, London, UK). + + +(a) Liver 1 Stomach + +3 2 + +Pancreas +4 + +Adhesions Spleen Kidney + +(b) +3 1 + +Liver +2 + + +4 Duodenum + + +Figure 61.7 (a) Intraperitoneal abscesses on transverse section: 1, the left subphrenic space; 2, left subhepatic space/lesser sac; 3, right subphrenic space; 4, right subhepatic space. (b) Intraperitoneal abscesses on sagittal section: 1, left subphrenic; 2, left subhepatic/ lesser sac; 3, right subphrenic; 4, right subhepatic. + +To the right is the falciform ligament and to the left the spleen, gastrosplenic omentum and diaphragm. The common cause of an abscess here is an operation on the stomach, the tail of the +pancreas, the spleen or the splenic flexure of the colon. + +LEFT SUBHEPATIC SPACE/LESSER SAC +The most common cause of infection here is complicated +acute pancreatitis. In practice, a perforated gastric ulcer rarely + +causes a collection here because the potential space is oblit- +erated by adhesions. + +RIGHT SUBPHRENIC SPACE +This space lies between the right lobe of the liver and the diaphragm. It is limited posteriorly by the anterior layer of the coronary and the right triangular ligaments, and to the left by the falciform ligament. Common causes of abscess here are perforating cholecystitis, a perforated duodenal ulcer and a duodenal cap ‘blow-out’ following gastrectomy and +appendicitis. + +RIGHT SUBHEPATIC SPACE +This lies transversely beneath the right lobe of the liver in Rutherford Morison’s pouch. It is bounded on the right by the right lobe of the liver and the diaphragm. To the left is situated the foramen of Winslow and below this lies the duo-denum. In front are the liver and the gall bladder, and behind are the upper part of the right kidney and the diaphragm. The space is bounded above by the liver and below by the trans-verse colon and hepatic flexure. It is the deepest space of the four and the most common site of a subphrenic abscess, which usually arises from appendicitis, cholecystitis, a perforated +duodenal ulcer or following upper abdominal surgery. +Clinical features +The symptoms and signs of subphrenic infection are fre-quently non-specific and it is well to remember the follow-ing aphorism: ‘pus somewhere, pus nowhere, pus under the diaphragm’. A common history is that, when some infective focus in the abdominal cavity has been dealt with, the condi-tion of the patient improves temporarily but, after an interval of a few days or weeks, symptoms of toxaemia reappear. The condition of the patient steadily, and often rapidly, deterio-rates. Sweating, wasting and anorexia are present. There is sometimes epigastric fullness and pain, or pain in the shoulder on the affected side, because of irritation of sensory fibres in the phrenic nerve, referred along the descending branches of the cervical plexus. Persistent hiccoughs may be a presenting symptom. A swinging pyrexia is usually present. If the abscess is anterior, abdominal examination will reveal some tender-ness, rigidity or even a palpable swelling. Sometimes the liver +is displaced downwards but more often it is fixed by adhesions. +Investigation and management +Examination of the chest and plain radiograph are import-ant because, in most cases, collapse of the lung or evidence of basal effusion or even an empyema is evident. The modern management of an abscess is by radiological diagnosis using ultrasound or CT guidance (Figures 61.8), followed by drain-age. The same tube can be used to instil antibiotic solutions or irrigate the abscess cavity if necessary. In some instances, monitoring may be appropriate by either clinically mark-ing out limits on the abdominal wall (if palpable) with daily examination. However, more commonly, repeat ultrasonogra- +phy or CT scanning will be required. Radiolabelled white cell + + +James Rutherford Morison, 1853–1939, Professor of Surgery, the University of Durham, Durham, UK. Jacob Benignus Winslow, 1669–1760, Professor of Anatomy, Physic and Surgery Paris, France. +, +PART 11 | ABDOMINAL +Ascites 1057 + + +(a) (b) + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 61.8 Computed tomography scans of a subphrenic abscess secondary to gallbladder empyema – (a) coronal and (b) sagittal sections (courtesy of Dr Tim Fotheringham, Consultant Interventional Radiologist, Bart’s Health NHS Trust, London, UK). + + + +scanning may occasionally prove helpful when other imaging techniques have failed. In most cases, with the aid of percuta-neous drainage and antibiotic treatment, the abscess or mass gradually reduces in size until, finally, it is undetectable. Open drainage of an intraperitoneal collection is thus now uncom-mon but may be necessary. If a swelling can be detected in the subcostal region or in the loin, an incision is made over the site of maximum tenderness or over any area where oedema or redness is discovered. Cautious blunt finger exploration can then be used to avoid dissemination of pus into the peritoneal or pleural cavities and minimise the risk of an intestinal fistula. When the cavity is reached, all of the fibrinous loculi must be broken down with the finger and one or two drainage tubes fully inserted. These drains are withdrawn gradually over the next 10 days, and the closure of the cavity can be checked by +sinograms or scanning. Appropriate antibiotics are also given. + +ASCITES +Ascites is defined as an accumulation of excess serous fluid +within the peritoneal cavity. + +Pathophysiology +The balanced effects of plasma and peritoneal colloid osmotic and hydrostatic pressures determine the exchange of fluid between the capillaries and the peritoneal fluid. Protein-rich fluid enters the peritoneal cavity when capillary permeability is increased, as in peritonitis and carcinomatosis peritonei. Capillary pressure may be increased because of generalised water retention, cardiac failure, constrictive pericarditis or vena cava obstruction. Capillary pressure is raised selectively in the portal venous system in the Budd–Chiari syndrome, cirrhosis of the liver or extrahepatic portal venous obstruc-tion. Plasma colloid osmotic pressure may be lowered in patients with reduced nutritional intake, diminished intesti-nal absorption, abnormal protein losses or defective protein synthesis, such as occurs in cirrhosis. Peritoneal lymphatic drainage may be impaired, resulting in the accumulation of +protein-rich fluid. + + + +George Budd, 1808–1882, Professor of Medicine, King’s College Hospital, London, UK, described this syndrome in 1845. +Hans Chiari, 1851–1916, Professor of Pathological Anatomy, Strasbourg, Germany (Strasbourg was returned to France in 1918 at the end of the First World War). He gave his account of this condition in 1898. +PART 11 | ABDOMINAL +1058 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + + +Summary box 61.12 + +Causes of ascites Transudates (protein <25 g/L) +●● Low plasma protein concentrations: Malnutrition +Nephrotic syndrome Protein-losing enteropathy +●● High central venous pressure: Congestive cardiac failure +●● Portal hypertension Portal vein thrombosis Cirrhosis + +Exudates (protein >25 g/L) ●● Peritoneal malignancy +●● Tuberculous peritonitis +●● Budd–Chiari syndrome (hepatic vein occlusion or thrombosis) ●● Pancreatic ascites +●● Chylous ascites ●● Meigs’ syndrome + + + +Clinical features +Ascites can usually be recognised clinically only when the amount of fluid present exceeds 1.5 L depending on body hab-itus: in obese individuals a greater quantity than this is neces-sary before there is clear evidence. The abdomen is distended evenly with fullness of the flanks, which are dull to percus-sion. Usually, shifting dullness is present but, when there is a very large accumulation of fluid, this sign is absent. In such cases, on flicking the abdominal wall, a characteristic fluid thrill is transmitted from one side to the other. In women, ascites must be differentiated from an enormous ovarian cyst. Congestive heart failure results in increased venous pres-sure in the vena cava and consequent obstruction to the venous outflow from the liver. The ascitic fluid is light yellow and of low specific gravity, about 1.010, with a low protein concentration (<25 g/L). Patients with constrictive pericardi-tis (Pick’s disease) have both peritoneal and pleural effusions because of engorgement of the venae cavae, consequent upon the diminished capacity of the right side of the heart. Asci-tes occurs with low plasma albumin concentrations, e.g. in patients with albuminuria or starvation. The ascites in this instance is caused by alterations in the osmotic pressure of the +capillary blood and has a low specific gravity. +In cirrhosis, now the most common cause of ascites inter-nationally, there is obstruction to the portal venous system, which is caused by obliterative fibrosis of the intrahepatic venous bed. In the Budd–Chiari syndrome (see Chapter 65), thrombosis or obstruction of the hepatic veins is responsible for obstruction to venous outflow from the liver. The asci-tes seen in patients with peritoneal metastases is caused by +excessive exudation of fluid and lymphatic blockage. The + +fluid is dark yellow and frequently blood stained. The spe-cific gravity, ≥1.020, and the protein content (>25g/L) are high. Microscopic examination often reveals cancer cells, especially if large quantities of fluid are ‘spun down’ to pro-duce a concentrated deposit for sampling. Rarely, ascites and pleural effusion are associated with solid fibromas of the ovary (Meigs’ syndrome). The effusions disappear when the tumour +is excised. +Investigation +In addition to relevant investigations to determine the under-lying cause, e.g. liver function tests (LFTs), cardiac function, ultrasonography and/or CT scanning (Figure 61.9) will determine much smaller quantities of ascites than possible clinically. These will often also diagnose aetiology, e.g. car-cinomatosis or liver disease. Ascitic aspiration or tap (below) is now most commonly performed under imaging guidance to minimise the risk of visceral injury. After the bladder has been emptied, puncture of the peritoneum is carried out under local anaesthetic using a moderately sized trocar and cannula. Alternatively, a peritoneal drain may be inserted. In cases where the effusion is caused by cardiac failure, the fluid must be evacuated slowly. Fluid is sent for microscopy/ cytology, culture, including mycobacteria, and analysis of pro-tein content and amylase. Unless other measures are taken the fluid soon reaccumulates, and repeated tappings remove +valuable protein. + + + + + + + + + + + + + + + +Figure 61.9 Computed tomography scan showing gross ascites. + + +Treatment +Treatment of the specific cause is undertaken whenever pos-sible, e.g. if portal venous pressure is raised, it may be possi-ble to lower it by treatment of the primary condition or by transjugular intrahepatic portosystemic shunt or transjugular intrahepatic portosystemic stent shunting (commonly abbre-viated as TIPS or TIPSS). These procedures have become increasingly popular since the 1980s and are performed by interventional radiologists under fluoroscopic guidance. +Access to the liver is gained via the internal jugular vein and + + +Joe Vincent Meigs, 1892–1963, Professor of Gynecology, Harvard University Medical School, Boston, MA, USA. Friedel Pick, 1867–1926, physician, Prague, the former Czechoslovakia, described this disease in 1896. +PART 11 | ABDOMINAL Tumours of the peritoneum 1059 + + + +a guidewire and introducer sheath facilitate cannulation of the hepatic vein. Once the catheter is in the hepatic vein, a wedge pressure is obtained to calculate the pressure gradient in the liver, and a special needle is advanced through the liver parenchyma to connect the hepatic to the portal vein. The channel for the shunt is created using an angioplasty balloon to dilate the needle tract and thence a stent placed through this track. Complications can occur, e.g. bleeding, liver injury and encephalopathy; however, in experienced hands the mor-tality rate is <1%. This procedure is thus, when possible, pref-erable to older surgical methods. +Dietary sodium restriction to 200 mg/day may be helpful, but diuretics are usually required (combination of spironolac-tone and furosemide). For patients failing to respond to such measures, therapeutic needle paracentesis can be performed. This is usually performed in the left of the right lower abdomi-nal quadrant after diagnostic needle aspiration, and using a ‘Z’ track approach to ensure that peritoneal and skin puncture do not overlie the same point. Serial large volume paracentesis (4–6 L/day and up to 8 L in one session) can be performed safely with colloid replacement, and it can be performed in patients with cirrhosis and deranged clotting. Guidelines rec-ommend albumin replacement after paracentesis to reduce complications. It is now also possible to leave an indwelling external drain for smaller-volume home paracentesis. In cases in which ascites accumulates rapidly after paracentesis and the patient is fit, drainage of the ascitic fluid by surgical peri-toneovenous shunting (e.g. LeVeen, Denver) was previously employed; however, these procedures have been essentially +abandoned due to their high rate of complications. + +Special cases Chylous ascites +In some patients the ascitic fluid appears milky because of an excess of chylomicrons (triglycerides). Most cases are asso-ciated with malignancy, usually lymphomas; other causes are cirrhosis, TB, filariasis, nephrotic syndrome, abdominal trauma (including surgery), constrictive pericarditis, sarcoid-osis and congenital lymphatic abnormality. The prognosis is poor unless the underlying condition can be cured. In addi-tion to other measures used to treat ascites, patients should be placed on a fat-free diet with medium-chain triglyceride +supplements. + +TUMOURS OF THE PERITONEUM Primary tumours +Primary tumours of the peritoneum are rare and in most cases take their origin not from the serous layer but from some adjacent structure, e.g. lipoma from appendices epiploicae, fibroma from connective tissue. Mesothelioma of the perito-neum is less frequent than in the pleural cavity but equally lethal. Asbestos is a recognised cause. It has a predilection +for the pelvic peritoneum. Chemocytotoxic agents are the + +mainstay of treatment. Desmoid tumours which have a relationship to the peritoneum are considered under familial +adenomatous polyposis (see Chapter 70). + +Secondary tumours Carcinomatosis peritonei +This is a common terminal event in many cases of carcinoma of the stomach, colon, ovary or other abdominal organs, and also of the breast and bronchus. The peritoneum, both parietal and visceral, is studded with secondary growths and the peritoneal cavity becomes filled with clear, straw-coloured or blood-stained ascitic fluid. +The main forms of peritoneal metastases are: +● discrete nodules – by far the most common variety; ● plaques varying in size and colour; +● diffuse adhesions – this form occurs at a late stage of the +disease and gives rise, sometimes, to a ‘frozen pelvis’. +Gravity probably determines the distribution of free malig-nant cells within the peritoneal cavity. Cells not caught in peritoneal folds gravitate into the pelvic pouches or into a hernial sac, the enlargement of which is occasionally the first indication of the condition. Implantation occurs also on the greater omentum, the appendices epiploicae and the inferior surface of the diaphragm. The main differential diagnosis is from tuberculous peritonitis (tubercles are greyish and trans-lucent and closely resemble the discrete nodules of peritoneal carcinomatosis). Investigation and treatment are as for under-lying malignancy; however, newer treatment options such as cytoreductive surgery, intraperitoneal chemotherapy and hyperthermic intraperitoneal chemotherapy (HIPEC) are now offered in specialist centres. The last, a highly concentrated, heated (41–42°C) chemotherapy treatment is delivered directly into the abdomen for 90 minutes after cytoreductive surgery, and is progressing to a standard of care in some coun- +tries for patients without disease outside the abdomen. +Pseudomyxoma peritonei +This rare condition occurs more frequently in women. The abdomen is filled with a yellow jelly, large quantities of which are often encysted. The condition is associated with muci-nous cystic tumours of the ovary and appendix. Recent stud-ies suggest that most cases arise from a primary appendiceal tumour with secondary implantation on to one or both ova-ries. It is often painless and there is frequently no impairment of general health. Pseudomyxoma peritonei does not give rise to extraperitoneal metastases but causes symptoms and complications due to tumour bulk. Although an abdomen distended with what seems to be fluid that cannot be made to shift should raise the possibility the diagnosis is more often suggested by ultrasonography and CT scanning, or made at operation. At laparotomy, masses of jelly are scooped out. The appendix, if present, should be excised together with any ovarian tumour. More definitive treatment can be achieved +, +by ‘complete cytoreduction’ (Sugarbaker technique) in which + + +Harry LeVeen, Professor of Surgery, the University of South Carolina, SC, USA. +Paul Sugarbaker, Director of Surgical Oncology at the Cancer Institute at Washington Hospital Center Washington DC, USA. +, +PART 11 | ABDOMINAL +1060 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + +the right hemicolon, spleen, gall bladder, and greater and lesser omentum are excised, along with stripping of the peri-toneum from the pelvis and diaphragm, and stripping of the tumour from the surface of the liver (the uterus and ovaries are also removed in women). Intraoperative heated chemo-therapy (using mitomycin C) follows and the whole proce-dure takes about 10 hours. There is international controversy about the effectiveness of this procedure although some large +series report good 5-year survival. + +Peritoneal inclusion cysts +These are usually caused by accumulation of ovarian fluid that is contained by peritoneal adhesions. The develop-ment of a peritoneal inclusion cyst thus depends on the presence of an active ovary and peritoneal adhesions. The normal peritoneum absorbs fluid easily. However, the absorp-tive capacity of the peritoneum is greatly diminished in the presence of mechanical injury, inflammation and peritoneal adhesions. +Peritoneal inclusion cysts occur only in premenopausal women with a history of pelvic or abdominal surgery. They range in size from several millimetres in diameter to bulky masses that may fill the entire pelvis and abdomen. Patholog-ically, the cyst results from non-neoplastic, reactive mesothe-lial proliferation. +Investigation includes the exclusion of ovarian tumour by blood tests and appropriate imaging (usually ultrasonography and magnetic resonance imaging [MRI]). Cysts may be man-aged expectantly or by hormonal modulation, e.g. oral contra-ceptives. Radiological drainage will give transient relief and may help diagnosis using cytology. Surgery can be performed +to remove adhesions but the risk of recurrence is 30–50%. + +Peritoneal loose bodies (peritoneal mice) +Peritoneal loose bodies (peritoneal mice) may be confused with a small tumour but almost never cause symptoms. One or more may be found in a hernial sac or in the pouch of Douglas. The loose body may come from an appendix epiplo-ica that has undergone axial rotation followed by necrosis of its pedicle and detachment, but they are also found in those who have subacute attacks of pancreatitis. These hyaline bodies attain the size of a pea or bean and contain saponified +fat surrounded by fibrin. + +ADHESIONS Pathophysiology +Adhesions are strands of fibrous tissue that form, usually as a result of surger , between surgically injured tissues. After injury, there is bleeding and an increase in vascular permeabil-ity with extravasation of fibrinogen-rich fluid from the injured surfaces forming a temporary fibrin matrix. An inflammatory response ensues with cell migration, release of cytokines and activation of the coagulation cascade. The activation of the coagulation system results in thrombin formation, which is necessary for the conversion of fibrinogen to fibrin. In the +y +absence of fibrinolysis, adhesions will form within 5–7 days as + +the matrix gradually becomes more organised with collagen secretion by fibroblasts. Fibrinolysis is therefore the key factor in determining whether an adhesion persists. This is governed by several cascades and activators that may account for inter-individual differences (see Chapter 3). Of great importance to the surgeon, however, is the fact that ischaemic tissue loses its ability to break down fibrin and inhibits fibrinolysis in +adjacent tissues. + +Complications +The most common adhesion-related problem is small bowel obstruction (SBO). Adhesions are the most frequent cause of SBO in resource-rich countries and are responsible for 60–70% of SBOs (see Chapter 71). Adhesions are also implicated as a major cause of secondary infertility (beyond the remit of this text). The relationship of adhesions to chronic abdominal and pelvic pain is contentious. Unguided division of adhesions has not been shown to reduce chronic abdominal pain in definitive RCTs although conscious pain mapping (laparoscopy under local anaesthesia) to direct lysis +may improve success rates. + +Prevention +As a result of the scale of the problem there has been signifi-cant research into ways of preventing postoperative adhesion formation. Minimising the production of ischaemic tissue by careful surgical technique, including meticulous control of bleeding, remain, however, the most critical concepts. The evolution of laparoscopic bowel surgery has been shown by collective review data to result in reduced adhesion-related readmissions for a number of abdominal and pelvic proce-dures, e.g. cholecystectomy, hysterectomy and colectomy. It should be noted, however, that only one randomised con-trolled trial (after Crohn’s resection) has ever shown a defin-itive effect, and this was not confirmed by evidence synthesis at Cochrane review. +The effect of a number of drugs including anti-inflammatory drugs such as aspirin and steroids, some hor-mones, anticlotting agents, antibiotics, vitamin E and even methylene blue have been investigated in adhesion preven-tion but have not achieved widespread use, because of either side effects or lack of consistent evidence of effectiveness. Many barrier methods of reducing adhesions have also been trialled. Adept® (4% icodextrin solution) is a solution applied inside the abdomen at the time of surgery; it has been shown to reduce the extent and severity of adhesion formation in animal models. It has also been used widely as a peritoneal dialysis solution for many years. Interceed TC7® is a mesh-like product (oxidised regenerated cellulose) which quickly forms a soft gelatinous mass around healing tissues and is absorbed within 2 weeks. It has been shown to significantly reduce the number of adhesions at the site where it is used. However, it is worth noting that a reduction in the number of adhesions in such studies does not necessarily equate to a reduction in adhesion-related problems in the future. In a +review of seven randomised trials looking at a similar +PART 11 | ABDOMINAL +The mesentery 1061 + + +barrier-type product (hyaluronic acid/carboxymethyl mem-brane), there was a significant reduction in the incidence, extent and severity of adhesions but no reduction in the incidence of intestinal obstruction or surgical interven-tion. Such barriers, when placed around bowel anasto-mosis, also led to a significant increase in the anastomotic leaks. For these reasons barrier approaches have not gained +popularity. + +Special forms of intraperitoneal +fibrosis +Sclerosing encapsulating peritonitis (a) (b) +Also known as abdominal cocoon syndrome, sclerosing encapsulating peritonitis (SEP) is described in patients as a complication of long-term peritoneal dialysis or portove-nous shunting. The peritoneal cavity becomes obliterated as a result of gross subserosal thickening by fibrosis, leading to bowel obstruction and other sequelae. Surgery should be +undertaken with trepidation and avoided if possible. + +Diffuse fibromatosis +This is a variant of intra-abdominal fibromatosis (IAF) and is actually a rare tumour characterised by an abnormal prolif-eration of myofibroblasts. Although non-metastasising, and said to be benign, it can nevertheless prove widely invasive, compressing and infiltrating surrounding tissues such as the bowel and mesentery with complications thereof. IAF is very rare within the general population but has a recognised asso- +ciation with familial adenomatous polyposis (FAP). + +THE OMENTUM +Rutherford Morison called the greater omentum ‘the abdom-inal policeman’. The greater omentum attempts, often suc-cessfully, to limit intraperitoneal infective and other noxious processes (Figure 61.10). For instance, an acutely inflamed appendix is often found wrapped in omentum, and this saves many patients from developing diffuse peritonitis. Some suf-ferers of herniae are also greatly indebted to this structure, because it often plugs the neck of a hernial sac and prevents a coil of intestine from entering and becoming strangulated. It can of course also be a cause of obstruction (acting as a large adhesion). The omentum is usually involved in tuberculous +peritonitis and carcinomatosis of the peritoneum. + +Torsion of the omentum +Torsion of the omentum is a rare emergency and consequently is seldom diagnosed correctly. It is usually mistaken for appen-dicitis with somewhat abnormal signs. It may be primary or secondary to adhesion of the omentum to an old focus of infec-tion or hernia. The patient is most frequently a middle-aged, obese man. A tender lump may be present in the abdomen. The blood supply having been jeopardised, the twisted mass sometimes becomes gangrenous, in which case bacterial peri-tonitis may follow. Treatment is surgical; the pedicle above +the twist is ligated securely and the mass removed. + + + + + +(c) + +Figure 61.10 The greater omentum: (a) normal; (b) in appendicitis; (c) in a (comparatively small) laceration of the spleen. + + + + +THE MESENTERY Mesenteric injury +A wound of the mesentery can follow severe abdominal con-tusion and is a cause of haemoperitoneum. More commonly, it is injured by a torsional force, so-called seatbelt syndrome. This occurs during a vehicular collision when a seatbelt is being worn with sudden deceleration resulting in a torn mes-entery. This possibility should be borne in mind, particularly as multiple injuries may distract attention from this injury (the management of abdominal trauma is covered in Chapter 27). Aside from control of any ongoing haemorrhage, associ- +ated ischaemic or ruptured gut will require resection. + +Ischaemia +Torsion of the mesentery is covered under midgut volvulus and volvulus of the small intestine (see Chapter 69). Embo-lism and thrombosis of mesenteric vessels leading to intestinal +ischaemia are also covered in Chapter 69. + +Inflammation +A number of somewhat miscellaneous conditions are best +described under this umbrella term. +PART 11 | ABDOMINAL +1062 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + +Acute, non-specific, ileocaecal, mesenteric adenitis +Non-specific mesenteric adenitis was so named to distinguish it from specific (tuberculous) mesenteric adenitis. It is now much more common than the tuberculous variety. The aeti-ology often remains unknown, although some cases are asso-ciated with yersinia infection of the ileum. In other cases, an unidentified virus is blamed. In about 25% of cases, a respira-tory infection precedes an attack of non-specific mesenteric adenitis. This self-limiting disease is never fatal but may be recurrent. Its significance thus mainly lies in its differential +diagnosis with appendicitis in children. + +DIAGNOSIS +During childhood, acute, non-specific mesenteric adenitis is a common condition. The typical history is one of short attacks of central abdominal pain lasting from 10 min to 30 min, commonly associated with vomiting. The patient sel-dom looks ill. In more than half the cases the temperature is elevated. Abdominal tenderness is poorly localised and, when present, shifting tenderness is a valuable sign for differenti-ating the condition from appendicitis. The neck, axillae and groins should be palpated for enlarged lymph nodes. There is often a leukocytosis of 10 000–12 000/µL (10–12 × 109/L) or more on the first day of the attack, but this falls on the +second day. + +TREATMENT +When the diagnosis can be made with assurance, bed rest and simple analgesia are the only treatment necessary. If, at a sec-ond examination a few hours later, acute appendicitis cannot be excluded, it is safer to perform either appendicectomy or diagnostic laparoscopy. If surgery is mistakenly undertaken, there is a small increase in the amount of peritoneal fluid. The ileocaecal mesenteric lymph nodes are enlarged, and can be seen and felt between the leaves of the mesentery. In very acute cases they are distinctly red, and many of them are the size of a walnut. The nodes nearest the attachment of the mesentery are the largest. They are not adherent to their peritoneal coats and, if a small incision is made through the +overlying peritoneum, a node is extruded easily. +Tuberculosis of the mesenteric lymph nodes +Tuberculous mesenteric lymphadenitis is considerably less common than acute non-specific lymphadenitis. Tubercle bacilli, usually, but not necessarily, bovine, are ingested and enter the mesenteric lymph nodes by way of Peyer’s patches. Sometimes only one lymph node is infected; usually there are several and occasionally massive involvement occurs. The presentation may be with abdominal pain (a rare differential for appendicitis) or with general constitutional symptoms (pyrexia, weight loss, etc.). Calcified lymph nodes may be demonstrated on a plain radiograph of the abdomen where they must be distinguished from other calcified lesions, e.g. +renal or ureteric stones. + +Misty mesentery +The term ‘misty mesentery’ indicates a pathological increase in mesenteric fat attenuation at CT (Figure 61.11). It is fre-quently observed on multidetector CT (MDCT) scans per-formed during daily clinical practice, and may be caused by various pathological conditions, including oedema, inflam-mation (especially in association with pancreatitis), haem-orrhage, neoplastic infiltration (especially otherwise occult lymphoma) or sclerosing mesenteritis. In patients with acute abdominal disease, misty mesentery may be considered a fea-ture of the underlying disease. Otherwise, it may represent an incidental finding on MDCT performed for other reasons. Follow-up scans may be required to dictate whether further investigation is required depending on progression or resolu-tion. It should be noted that the term ‘mesenteric panniculitis’ is frequently used synonymously with ‘misty mesentery’. Cor-rectly this term should be reserved for the mesenteric mani-festation of Weber–Christian disease; isolated lipodystrophy and mesenteric lipogranuloma. It is a very rare (200–300 cases reported worldwide) benign inflammatory or fibrotic change +in the mesentery of the bowel. + + + + + + + + + + + + + + + + + + + +Figure 61.11 Computed tomography scan showing ‘misty mesentery’ in a patient with probable peritoneal inflammation secondary to acute pancreatitis (courtesy of Dr K Patel, Homerton University Foundation Trust, London). + +Mesenteric cysts +Cysts may occur in the mesentery of either the small intes-tine (60%) or the colon (40%) and can be classified as the +following: +● Chylolymphatic ● Enterogenous +● Urogenital remnant (actually retroperitoneal but project into peritoneum) +● Dermoid. + + + +Johann Conrad Peyer, 1653–1712, Professor of Logic, Rhetoric and Medicine, Schaffhausen, Switzerland, described the lymph follicles in the intestine in 1677. +PART 11 | ABDOMINAL +The mesentery 1063 + + + +Pathology +CHYLO YMPHATIC CYST +L +Although all mesenteric cysts are rare, this is the most com-mon variety, probably arising in congenitally misplaced lym-phatic tissue that has no efferent communication with the lymphatic system (most frequently in the mesentery of the ileum). The thin wall of the cyst, which is composed of con-nective tissue lined by flat endothelium, is filled with clear lymph or, less frequently, with chyle, varying in consistency from watered milk to cream. Occasionally, the cyst attains a great size. More often unilocular than multilocular, a chy-lolymphatic cyst is almost invariably solitary, although there is an extremely rare variety in which myriads of cysts are found in the various mesenteries of the abdomen. A chylolymphatic cyst has a blood supply that is independent from that of the adjacent intestine and, thus, enucleation is possible without +the need for resection of gut. + +ENTEROGENOUS CYSTS +These are believed to be derived either from a diverticulum of the mesenteric border of the intestine that has become sequestrated from the intestinal canal during embryonic life, or from a duplication of the intestine (see Chapter 9). An enterogenous cyst has a thicker wall than a chylolymphatic cyst and it is lined by mucous membrane, which is sometimes ciliated. The content is mucinous and either colourless or yel-lowish brown as a result of past haemorrhage. The muscle in the wall of an enteric duplication cyst and the bowel with which it is in contact have a common blood supply; conse-quently, removal of the cyst always entails resection of the +related portion of intestine. + +UROGENITAL REMNANT +A cyst developing in the retroperitoneal space (see below) often attains very large dimensions and has first to be distin-guished from a large hydronephrosis. Even after the latter condition has been eliminated by scanning or urography, a retroperitoneal cyst can seldom be distinguished with cer-tainty from a retroperitoneal tumour until displayed at oper-ation. The cyst may be unilocular or multilocular. Many of these cysts are believed to be derived from a remnant of the +wolffian duct, in which case they are filled with clear fluid. +Clinical features +These are shown for mesenteric cysts in general in Summary +box 61.13. +Investigation and treatment +Ultrasonography and CT scanning will demonstrate the lesion and may allow diagnosis of cyst type (Figure 61.13a and b). There are no suitable medical therapies. The goal of surgi-cal therapy is complete excision of the mass. The preferred treatment of mesenteric cysts is enucleation, although bowel resection is frequently required to ensure that the remaining +bowel is viable. Bowel resection may be required in 50–60% + + +Summary box 61.13 + +Mesenteric cysts: clinical features +●● Cysts occur most commonly in adults with a mean age of 45 years +●● Twice as common in women as in me ●● Rare: incidence around 1 per 140 000 +●● Approximately a third of cases occur in children younger than 15 years +●● The mean age of children affected is 4.9 years +●● The most common presentation is of a painless abdominal swelling with characteristic physical signs +there is a fluctuant swelling near the umbilicus +the swelling moves freely in a plane at right angles to the attachment of the mesentery (Tillaux’s sign) (Figure 61.12) +there is a zone of resonance around the cyst +●● Other presentations are with recurrent attacks of abdominal pain with or without vomiting (pain resulting from recurring temporary impaction of a food bolus in a segment of +bowel narrowed by the cyst or possibly from torsion of the mesentery) and acute abdominal catastrophe, due to: +torsion of that portion of the mesentery containing the cyst +rupture of the cyst, often as a result of a comparatively trivial accident +haemorrhage into the cyst infection + + + + + + + + + + + +Attachment of the mesentery + + + + + + +Figure 61.12 A mesenteric cyst moves freely in the direction of the arrows, i.e. at right angles to the attachment of the mesentery (Tillaux’s sign). + + +of children with mesenteric cysts, whereas resection is nec-essary in about a third of adults (Figure 61.13c). If enucle-ation or resection is not possible because of either the size of the cyst or its location deep within the root of the mesentery, +the third option is partial excision with marsupialisation of + + +Kaspar Friedrich Wolff, 1733–1794, Professor of Anatomy and Physiology, St Petersburg, Russia, described the mesonephric duct and body in 1759. Paul Jules Tillaux, 1834–1904, surgeon, Paris, France. +PART 11 | ABDOMINAL +1064 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + +(a) + +the remaining cyst into the abdominal cavity. Approximately 10% of patients require this form of therapy. If marsupiali-sation is performed, the cyst lining should be sclerosed with 10% glucose solution, diathermy or tincture of iodine to min- +imise recurrence. Recurrence rates vary from 0% to 13%. + +DIFFERENTIAL DIAGNOSIS +The following, although not being mesenteric cysts in the +true meaning of the term, give rise to the same physical signs: + +● peritoneal inclusion cyst (see above); +● serosanguineous cyst, probably traumatic in origin although a history of an accident is seldom obtained; +● tuberculous abscess of the mesentery; +● hydatid cyst of the mesentery. + + +(b) Neoplasms of the mesentery +The mesentery is necessarily affected by local lymphatic spread of carcinoma arising from the peritoneal viscera. Other +benign and malignant tumours are less common. + + +Summary box 61.14 + +Mesenteric tumours Benign +●● Lipoma ●● Fibroma +●● Fibromyxoma + +(c) Malignant +●● Lymphoma +●● Secondary carcinoma + + +Tumours situated in the mesentery give rise to physical signs that are similar to those of a mesenteric cyst, the sole exception being that they sometimes feel solid. If indicated, a benign tumour of the mesentery is excised in the same way as an enterogenous mesenteric cyst, i.e. with resection of the adjacent intestine. A malignant tumour of the mes-entery requires biopsy confirmation and specific, usually non- +surgical, treatment, e.g. chemotherapy for lymphoma. + + + + + + + +Figure 61.13 A mesenteric enterogenous cyst in a 37-year-old woman presenting with an asymptomatic palpable abdominal mass: (a) ultrasound findings; (b) computed tomography findings; and (c) intraoperative findings requiring en masse small bowel resection (courtesy of Mr F Olagbaiye, Homerton University Foundation Trust, London). + +THE RETROPERITONEAL SPACE Retroperitoneal fibrosis +This is a relatively rare diagnosis characterised by the devel-opment of a flat grey/white plaque of tissue, which is found first in the low lumbar region but then spreads laterally and upwards to encase the common iliac vessels, ureters and aorta. Histological appearances vary from active inflammation with a high cellular content interspersed with bundles of collagen through to one of acellularity and mature fibrosis/calcification. +Its aetiology is obscure in most cases (idiopathic) being allied +PART 11 | ABDOMINAL +The retroperitoneal space 1065 + + + +Summary box 61.15 + +Causes of retroperitoneal fibrosis Benign +●● Idiopathic (Ormond’s disease) ●● Chronic inflammation +●● Extravasation of urine +●● Retroperitoneal irritation by leakage of blood or intestinal content +●● Aortic aneurysm (inflammatory type) ●● Trauma +●● Drugs (chemotherapeutic agents and previously methysergide) + +Malignant +●● Lymphoma +●● Carcinoid tumours +●● Secondary deposits (especially from carcinoma of stomach, colon, breast and prostate) + + +to other fibromatoses (others being Dupuytren’s contracture and Peyronie’s disease). In other patients the cause is known. +The clinical presentation may be one of ill-defined chronic backache or occur as a result of compromise to involved structures, e.g. lower limb or scrotal oedema second-ary to venous occlusion, or chronic renal failure secondary to ureteric obstruction. Treatment will be directed to the cause, the modification of disease activity when appropriate, e.g. immunomodulation with steroids, tamoxifen and restoration +of flow in affected structures, e.g. ureteric stenting. + +Retroperitoneal (psoas) abscess +The retroperitoneal space can also be a site for abscess formation, which for practical purposes is almost synonymous with psoas abscess. Psoas abscess is a relatively uncommon diagnosis, the true incidence of which is not well described. At the start of the twentieth century, psoas abscess was mainly caused by TB of the spine (Pott’s disease). With the decline of M. tuberculosis as a major pathogen in resource-rich countries, a psoas abscess was mostly found secondary to direct spread of infection from the inflamed ± perforated digestive or urinary tract. In recent years a primary psoas abscess due to haema-togenous spread from an occult source is more common, especially in immunocompromised and older patients, as well as in association with intravenous drug misuse. +Clinical presentation is with back pain, lassitude and fever. A swelling may point to the groin as it tracks along iliopsoas. Pain may be elicited by passive extension of the hip or a fixed flexion of the hip evident on inspection. +Radiological investigation is via CT scanning (Figure 61.14) and treatment usually by percutaneous CT-guided +drainage and appropriate antibiotic therapy. + + + + + + + + + + + + + + + + +Figure 61.14 Representative sagittal computed tomography recon-struction of a right-sided psoas abscess (courtesy of Dr K Patel, Homerton University Foundation Trust, London). + + +Retroperitoneal tumours +Although swellings in the retroperitoneum may include abscess, haematoma, cysts (see above) and spread of malig-nancy from retroperitoneal organs (kidney, ureter, adrenal), the term retroperitoneal tumour is usually confined to primary tumours arising in other tissues in this region e.g. muscles, fat, lymph nodes and nerves. The management of such tumours is now frequently by referral to a specialist centre and this should be done before biopsy which may compromise subse- +quent surgical cure. The two commonest are briefly described. +Retroperitoneal lipoma +The patient may seek advice on account of a swelling or because of indefinite abdominal pain. Women are more often affected. These swellings sometimes reach an immense size. Diagnosis is usually by ultrasonography and CT scanning. A retroperitoneal lipoma sometimes undergoes myxomatous degeneration, a complication that does not occur in a lipoma in any other part of the body. Moreover, a retroperitoneal lipoma is often malignant (liposarcoma) (see below) and may +increase rapidly in size. +Retroperitoneal sarcoma +Retroperitoneal sarcomas are rare tumours accounting for only 1–2% of all solid malignancies (10–20% of all sarco-mas are retroperitoneal). The peak incidence is in the fifth decade of life, although they can occur at almost any age. The most common types of retroperitoneal soft-tissue sarcomas in adults vary from study to study. However, in most studies, the +most frequently encountered cell types are: +● liposarcoma; +● leiomyosarcoma; +● malignant fibrous histiocytoma (MFH). + + +JK Ormond, American urologist. +Baron Guillaume Dupuytren, 1777–1835, Surgeon in Chief, Hôtel Dieu, Paris, France, described this condition in 1831. +François de la Peyronie, 1678–1747, Surgeon to King Louis XIV of France, and Founder of the Royal Academy of Surgery, Paris, France. +PART 11 | ABDOMINAL +1066 CHAPTER 61 The peritoneum, omentum, mesentery and retroperitoneal space + + + +CLINICAL FEATURES +Patients with sarcomas present late, because these tumours arise in the large potential spaces of the retroperitoneum and can grow very large without producing symptoms. More-over, when symptoms do occur, they are non-specific, such as abdominal pain and fullness, and are easily dismissed as being caused by other less serious processes. Retroperitoneal sarcomas are, therefore, usually very large at the time of pre- +sentation. + +INVESTIGATION +Detailed multiplanar imaging (CT + MRI) with reconstruc-tions is required not only for tumour detection, staging and surgical planning, but also for guiding percutaneous or surgi-cal biopsy of these tumours. Such biopsies have a greater role +than for other sarcomas. + +TREATMENT +The definitive treatment of primary retroperitoneal sarco-mas is surgical resection. Chemotherapy and radiotherapy without surgical debulking have rarely been beneficial, when used alone or in combination A multidisciplinary treatment approach with imaging review will be required when assessing operability (based on adjacency or involvement of vital struc-tures) and approach. Up to 75% of retroperitoneal sarcoma resections involve resection of at least one adjoining intra-abdominal visceral organ (commonly large or small bowel or kidney). The most common types of vascular involvement precluding resection are involvement of the proximal superior +mesenteric vessels or involvement of bilateral renal vessels. + +PROGNOSIS +In the vast majority of sarcomas, cell type has no impact on treatment and long-term survival. Survival rates are in gen- +eral poor, even after complete resection, being in the order of + +35–50% (excluding low-grade liposarcomas, which may fre- +quently be cured by resection). + +FURTHER READING +Ahmad G, Mackie FL, Iles DA et al. Fluid and pharmacological agents for adhesion prevention after gynaecological surgery. Cochrane Data-base Syst Rev 2014; 9(7): CD001298. +French FMF Consortium. A candidate gene for familial Mediterranean fever. Nat Genet 1997; 17(1): 25–31. +The International FMF Consortium. Ancient missense mutations in a new member of the RoRet gene family are likely to cause familial Mediterranean fever. Cell 1997; 90(4): 797–807. +Klaver CE, Musters GD, Bemelman WA et al. Adjuvant hyperthermic intraperitoneal chemotherapy (HIPEC) in patients with colon cancer at high risk of peritoneal carcinomatosis; the COLOPEC randomized multicentre trial. BMC Cancer 2015; 24;15: 428. +Lubner MG, Hinshaw JL, Pickhardt PJ. Primary malignant tumors of peritoneal and retroperitoneal origin: clinical and imaging features. Surg Oncol Clin N Am 2014; 23: 821–45 +Moore KP, Aithal GP. Guidelines on the management of ascites in cir-rhosis. Gut 2006; 55(Suppl 6): vi1–12. +Rasheed S, Zinicola R, Watson D, Bajwa A, McDonald PJ. Intra-abdominal and gastrointestinal tuberculosis. Colorectal Dis 2007; 9: 773–83. +Rudloff U, Langan RC, Mullinax JE et al. Impact of maximal cytore-ductive surgery plus regional heated intraperitoneal chemotherapy (HIPEC) on outcome of patients with peritoneal carcinomatosis of gastric origin: results of the GYMSSA trial. J Surg Oncol 2014; 110(3): 275–84. +Sadler TW (2010). Langman’s medical embryology (with CD-ROM), 11th edn. Baltimore, MD: Lippincott Williams & Wilkins. +Schnüriger B, Barmparas G, Branco BC, Lustenberger T, Inaba K, Deme-triades D. Prevention of postoperative peritoneal adhesions: a review of the literature. Am J Surg 2011; 201: 111–21. +Sugarbaker PH, Jablonski KA. Prognostic features of 51 colorectal and 130 appendiceal cancer patients with peritoneal carcinomatosis treated by cytoreductive surgery and intraperitoneal chemotherapy. Ann Surg 1995; 221(2): 124–32 +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love +62 +Chapter + + + + + +The oesophagus + + + + + + +Learning objectives +To understand: +• The anatomy and physiology of the oesophagus and their relationship to disease + + + +• The clinical features, investigations and treatment of benign and malignant disease with particular reference to the common adult disorders + + + + + +BACKGROUND Surgical anatomy +The oesophagus is a muscular tube, approximately 25 cm long, mainly occupying the posterior mediastinum and extending from the upper oesophageal sphincter (the cricopharyngeus muscle) in the neck to the junction with the cardia of the stomach. The musculature of the upper oesophagus, including the upper sphincter, is striated. This is followed by a transi-tional zone of both striated and smooth muscle with the pro-portion of the latter progressively increasing so that, in the lower half of the oesophagus, there is only smooth muscle. It is lined throughout with squamous epithelium. The parasym-pathetic nerve supply is mediated by branches of the vagus nerve that has synaptic connections to the myenteric (Auer-bach’s) plexus. Meissner’s submucosal plexus is sparse in the oesophagus. +The upper sphincter consists of powerful striated muscle. The lower sphincter is more subtle and created by the asym-metrical arrangement of muscle fibres in the distal oesoph-ageal wall just above the oesophagogastric junction. It is helpful to remember the distances 15, 25 and 40 cm for ana- +tomical location during endoscopy (Figure 62.1). + +Physiology +The main function of the oesophagus is to transfer food from the mouth to the stomach in a coordinated fashion. The ini-tial movement from the mouth is voluntary. The pharyngeal phase of swallowing involves sequential contraction of the oropharyngeal musculature, closure of the nasal and respi-ratory passages, cessation of breathing and opening of the +upper oesophageal sphincter. Beyond this level, swallowing + + +is involuntary. The body of the oesophagus propels the bolus through a relaxed lower oesophageal sphincter (LOS) into the stomach, taking air with it (Figure 62.2). This coordi-nated oesophageal wave which follows a conscious swallow is called primary peristalsis. It is under vagal control, although there are specific neurotransmitters that control the LOS. +The upper oesophageal sphincter is normally closed at rest and serves as a protective mechanism against regurgita-tion of oesophageal contents into the respiratory passages. It also serves to stop air entering the oesophagus other than the +small amount that enters during swallowing. + + +0 cm Incisor teeth + + + + +Cricopharyngeal 15 cm constriction + + +Aortic and bronchial constriction +25 cm + +Diaphragmatic and ‘sphincter’ constriction + +40 cm + + +Figure 62.1 Endoscopic landmarks. Distances are given from the incisor teeth. They vary slightly with the build of the individual. + + +Leopold Auerbach, 1828–1897, Professor of Neuropathology, Breslau, Germany (now Wroclaw, Poland), described the myenteric plexus in 1862. Georg Meissner, 1829–1905, Professor of Physiology, Göttingen, Germany described the submucosal plexus in 1852. +, +, +PART 11 | ABDOMINAL 1068 CHAPTER 62 The oesophagus +Oesophagus +Oesophagus +Oesophagus + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.2 A bolus of barium or food usually takes air with it into the stomach. + +The LOS is a zone of relatively high pressure that pre-vents gastric contents from refluxing into the lower oesoph-agus (Figure 62.3). In addition to opening in response to a primary peristaltic wave, the sphincter also relaxes to allow air to escape from the stomach and at the time of vomiting. A variety of factors influences sphincter tone, notably food, gastric distension, gastrointestinal (GI) hormones, drugs and smoking. The arrangement of muscle fibres, their differ-ential responses to specific neurotransmitters and the rela-tionship to diaphragmatic contraction all contribute to the action of the LOS. The presence of the physiological sphinc-ter was first demonstrated by Code using manometry with small balloons. Until recently LOS pressure was measured by water-perfused tubes, but the introduction of catheters containing multiple microtransducers has meant that this approach has been superseded by high-resolution manome-try. The normal LOS is 3–4 cm in length and has a pressure of 10–25 mmHg. +Manometry is also used to assess the speed and amplitude of oesophageal body contractions and ensure that peristalsis is propagated down the entire length of the oesophagus (Figure 62.4). Secondary peristalsis is the normal reflex preceded by a conscious swallow. It is worth remembering that most clear-ance swallows to neutralise refluxed gastric acid are, however, achieved by primary peristalsis, which carries saliva with its high bicarbonate content down to the lower oesophagus. Tertiary contractions are non-peristaltic waves that are infre-quent (<10%) during laboratory-based manometry, although readily detected if manometry is undertaken while the patient +eats a meal. + + + + + + + + + + + + + + + + +1 cm + + + + + + + +Anatomy of the sphincter +Stomach +pH +Stomach +PD +Pressure +Radiology Closed Open +Figure 62.3 Correlation between the radiological appearances of a barium column and the lower oesophageal sphincter open and closed. The three curves on the right, set up verticall , show the pH gradient, the mucosal potential difference (PD), marking the junction of squamous and columnar epithelium, and the high-pressure zone of the sphincter. +y +PART 11 | ABDOMINAL Background 1069 + +Spatiotemporal plot Line plots + + + + + + + + + + + + + + + + + + + +Figure 62.4 High-resolution manometry depicts oesophageal function from the pharynx to the stomach. The spatiotemporal plot presents the same information as presented in the ‘conventional’ line plots. Time is on the x-axis and distance from the nares is on the y-axis. Pressure is assigned a colour (legend right). The segmental anatomy of oesophagus is easy to appreciate (not seen on conventional manometry). The coordinated relaxation of the upper (UOS) and lower oesophageal sphincter (LOS) is obvious, as is the increasing pressure and duration of the peristaltic wave as it passes distally. The intrabolus pressure (IBP) between the peristaltic wave and oesophagogastric junction, and the pressure gradient across the gastro-oesophageal junction are visualized. These represent the force that drives bolus through the oesophagus and into the stomach. + + + + + +Symptoms + +Summary box 62.1 + +Symptoms of oesophageal disease +●● Difficulty in swallowing described as food or fluid sticking (oesophageal dysphagia): must rule out malignancy +●● Pain on swallowing (odynophagia): suggests inflammation and ulceration +●● Regurgitation or reflux (heartburn): common in gastro-oesophageal reflux disease +●● Chest pain: difficult to distinguish from cardiac pain + + +Dysphagia +Dysphagia is used to describe difficulty with swallowing. When there is a problem with swallowing in the voluntary (oral or pharyngeal) phases, patients will usually say that they cannot swallow properly, but they do not characteris-tically describe ‘food sticking’. Instead, when they try to ini-tiate a conscious swallow, food fails to enter the oesophagus, stays in the mouth or enters the airway, causing coughing or spluttering. Virtually all causes of this type of dysphagia are chronic neurological or muscular diseases. Oesophageal dys-phagia occurs in the involuntary phase and is characterised by a sensation of food sticking. The nature of this type of dysphagia is often informative with regard to a likely diagno-sis. Dysphagia may occur acutely or in a chronic fashion, can +affect solids and/or fluids, and be intermittent or progressive. + + +Although many patients point to a site of impaction, this is +unreliable. + +Odynophagia +Odynophagia refers to pain on swallowing. Patients with reflux oesophagitis often feel retrosternal discomfort within a few seconds of swallowing hot beverages, citrus drinks or alco-hol. Odynophagia is also a feature of infective oesophagitis +and may be particularly severe in chemical injury. + +Regurgitation and reflux +Regurgitation and reflux are often used synonymously. It is helpful to differentiate between them, although it is not always possible. Regurgitation should strictly refer to the return of oesophageal contents from above a functional or mechanical obstruction. Reflux is the passive return of gas-troduodenal contents to the mouth as part of the symptom-atology of gastro-oesophageal reflux disease (GORD). Loss of weight, anaemia, cachexia, change of voice due to refluxed material irritating the vocal folds, and cough or dyspnoea due to tracheal aspiration may all accompany regurgitation and/ +or reflux. + +Chest pain +Chest pain similar in character to angina pectoris may arise from an oesophageal cause, especially gastro-oesophageal reflux and motility disorders. Exercise-induced chest pain can +be due to reflux. +PART 11 | ABDOMINAL 1070 CHAPTER 62 The oesophagus + + + +Investigations Radiography +Contrast radiography has been somewhat overshadowed by endoscopy but remains a useful investigation to demonstrate changes in oesophageal diameter, anatomical distortion or abnormal motilit . An adequate barium swallow should be tailored to the problem under investigation. It may be help-ful to give a solid bolus (bread or marshmallow) if a motility disorder is suspected. Video recording is useful to allow subse-quent replay and detailed analysis. Barium radiology is, how-ever, inaccurate in the diagnosis of gastro-oesophageal reflux, unless reflux is gross, and should not be used for this purpose. Plain radiographs will show some foreign bodies. +y +Cross-sectional imaging by computed tomography (CT) is now an essential investigation in the assessment of neoplasms of the oesophagus and can be used in place of a contrast swal-low to demonstrate perforation. The role of CT and other +cancer-specific tests is described later. + + +of the oesophagus (Figure 62.5). Linear echoendoscopes produce a sectoral image in the line of the endoscope and are used to biopsy submucosal oesophageal lesions, mediasti-nal masses such as lymph nodes (Figure 62.6) or suspicious lesions that might lie outside a proposed surgical field. Radial scanners without optical components are available for pas-sage through narrow strictures over a guidewire, and catheter probes are available that can be passed down the endoscope +biopsy channel. +Oesophageal manometry +Manometry is now widely used to diagnose oesophageal motility disorders. Electronic microtransducers that are not influenced by changes in patient position during the test +have gradually supplanted perfusion systems. High-resolution + + + +Endoscopy +Endoscopy is necessary for the investigation of most oesoph-ageal conditions. It is required to view the inside of the oesophagus and the oesophagogastric junction, to obtain a biopsy or cytology specimen, for the removal of foreign bodies and to dilate strictures. Traditionally, there are two types of instrument available, the rigid oesophagoscope and the flexi-ble video endoscope, but the rigid instrument is now virtually obsolete. +For flexible video gastroduodenoscopy, general anaesthe-sia is not required; most examinations can be done on an out-patient basis, and the quality of the magnified image is superb. The technology associated with video endoscopy continues to improve. Magnification is a standard feature of the mod-ern endoscope and widely used in conjunction with agents that can be sprayed on to the mucosa, such as acetic acid to enhance mucosal detail. Novel techniques that rely on fluo-rescence and narrow band imaging to enhance visual contrast are becoming increasingly used for the identification of muco-sal abnormalities that are not easily seen with white light, e.g. in patients with Barrett’s oesophagus undergoing endoscopic surveillance. +As a matter of routine, the stomach and duodenum are examined as well as the oesophagus. If a stricture is encoun-tered, it may be helpful to dilate it to allow a complete inspec-tion of the upper GI tract, but this decision should be dictated by clinical circumstances and an appreciation of the perfora-tion risk, especially if the visual appearances are thought to +indicate neoplasia. + + + + + + + + + + + + + + + +Eus Histology + + +1 Mucosa + + +Deep +2 mucosa + +Muscularis mucosa + +3 Submucosa + + + +Endosonography +Endoscopic ultrasonography relies on a high-frequency (5–30 MHz) transducer located at the tip of the endoscope to provide highly detailed images of the layers of the oesoph-ageal wall and mediastinal structures close to the oesopha-gus. Radial echoendoscopes have a rotating transducer that creates a circular image with the endoscope in the centre, and this type of scanner is widely used to create diagnostic +transverse sectional images at right angles to the long axis + + + +Muscularis propia +4 + + + +5 Serosa + +Figure 62.5 Radial endosonography indicating wall layers as alter-nating hyper- and hypoechoic bands. +PART 11 | ABDOMINAL Foreign bodies in the oesophagus 1071 + + +events that cause symptoms. This may be important in the identification of patients with non-erosive reflux disease on +endoscopy. + + + + + + + + + + + + + +Figure 62.6 Linear endosonography with the needle traversing an ultrasound plane for guided biopsy. + + +manometry uses a multiple (up to 36) microtransducer cathe-ter with the results displayed as spatiotemporal plots, and this has now largely supplanted earlier manometry systems. This system provides comprehensive information about oesopha- +geal body function and associated behaviour of the LOS. +Twenty-four-hour pH and combined pH-impedance recording +Prolonged measurement of pH is now accepted as the most accurate method for the diagnosis of gastro-oesophageal reflux. It is particularly useful in patients with atypical reflux symptoms, those without endoscopic oesophagitis and when patients respond poorly to intensive medical therapy. A small pH probe is passed into the distal oesophagus and posi-tioned 5 cm above the upper margin of the LOS, as defined by manometry. The probe is connected to a miniature digital recorder that is worn on a belt and allows most normal activ-ities. Patients mark symptomatic events such as heartburn. A 24-hour recording period is usual, and the pH record is analysed by an automated computer program. An oesopha-geal pH <4 at the level of the pH electrode is conventionally considered the cut-off value and, in most oesophageal labo-ratories, the total time when pH is <4 in a 24-hour period does not exceed 4% in a healthy adult. Patterns of reflux and the correlation between symptoms and oesophageal pH <4 can be calculated. Most laboratories use a scoring system (Johnson–DeMeester) to create a numerical value, above which reflux is considered pathological. Radiotelemetry pH probes are also available that can be fixed to the oesophageal wall endoscopically without the need for a transnasal cathe-ter. The introduction of catheter assemblies that incorporate multiple electrodes along the length of the catheter, as well as a pH probe, has resulted in combined measurement of pH and electrical impedance. Impedance measurement differentiates anterograde from retrograde bolus transit and gas from liquid, +and provides a reliable measure of non-acidic or weakly acidic + +Therapeutic procedures Dilatation of strictures +Stricture dilatation is essentially undertaken for benign con-ditions and should be used with caution in the context of malignant disease. The risks associated with dilatation for malignant disease are discussed later. The advent of guide-wire-directed dilatation of the oesophagus in the 1970s was a major advance over earlier blind dilatation systems. Their use is now considered standard practice. There are many differ-ent designs, but essentially they are solid dilators of increas-ing diameter or inflatable balloons with rigid walls. To restore normal swallowing, a stricture should be dilated to at least 16mm in diameter, although this may need to be achieved over a series of procedures, depending on the perceived nature of the stricture. A guidewire is passed down the biopsy chan-nel of an endoscope and through the stricture under vision. If the stricture is long or tortuous, this should be undertaken under radiological guidance to ensure that the guidewire passes easily into the stomach. The endoscope is withdrawn, leaving the guidewire in place, and graduated dilators or a balloon dilator are passed over the guidewire, sometimes with radiographic screening for safety purposes. The dilatation of reflux-induced strictures is usually straightforward. These strictures are nearly always short and at the oesophagogastric junction, so that the stomach is visible through the narrowed segment. Radiological control is rarely needed. Conversely, distal oesophageal adenocarcinomas extending into the stom-ach are often soft, friable and tortuous to negotiate. Caustic strictures often occur high in the thorax and may be very long. Balloons for oesophageal use tend to have inflation diam-eters of 25–40 mm and may also be used for dilatation. Pneu-matic dilatation is widely used to disrupt the non-relaxing +LOS in achalasia. +Thermal recanalisation +Various types of laser (mainly Nd–YAG), bipolar diathermy, injection of absolute alcohol or argon-beam plasma coagula-tion have all been used successfully to ablate tissue in order to +recanalise the oesophagus. + +CONGENITAL ABNORMALITIES See Chapter 9. + +FOREIGN BODIES IN THE +OESOPHAGUS +All manner of foreign bodies have become arrested in the oesophagus (Figure 62.7). Button batteries may be a trouble- +some problem in children. The most common impacted + + +Lawrence F Johnson, gastroenterologist, Birmingham, AL, USA. Tom R DeMeester, Professor of Surgery, Los Angeles, CA, USA. +PART 11 | ABDOMINAL 1072 CHAPTER 62 The oesophagus + + +Summary box 62.2 + +Foreign bodies +●● The most common is a food bolus, which usually signifies underlying disease +●● It is usually possible to remove foreign bodies by flexible endoscopy +●● Beware of button batteries in the oesophagus + + + + + + + + +Figure 62.7 False teeth impacted in the oesophagus. (Note: modern dentures are usually radiolucent.) + + +material is food, and this usually occurs above a significant pathological lesion (Figure 62.8). Plain radiographs are often useful for foreign bodies, but modern denture materials are not always radiopaque. A contrast examination is not usually +required and only makes endoscopy more difficult. + +Foreign bodies that have become stuck in the oesophagus should be removed by flexible endoscopy using suitable grasp-ing forceps, a snare or a basket. If the object may injure the oesophagus on withdrawal, an overtube can be used, and the endoscope and object can be withdrawn into the overtube before removal. Button batteries can be a particular worry because they are difficult to grasp, and it is tempting to push them on into the stomach. However, an exhausted battery may rapidly corrode in the GI tract and is best extracted. A multiwire basket of the type used for gallstone retrieval nearly always works. An impacted food bolus will often break up and pass on if the patient is given fizzy drinks and confined to flu-ids for a short time. The cause of the impaction must then be investigated. If symptoms are severe or the bolus does not +pass, it can be extracted or broken up at endoscopy. + + +PERFORATION +Perforation of the oesophagus is usually iatrogenic (at thera-peutic endoscopy) or due to ‘barotrauma’ (spontaneous perforation). Many instrumental perforations can be man-aged conservatively, but spontaneous perforation is often a life-threatening condition that regularly requires surgical +intervention. + + +Summary box 62.3 + +Perforation of the oesophagus +●● Potentially lethal complication due to mediastinitis and septic shock +●● Numerous causes, but may be iatrogenic +●● Surgical emphysema is virtually pathognomonic +●● Treatment is urgent; it may be conservative or surgical, but requires specialised care + + + + + + + + + +Figure 62.8 An impacted meat bolus at the lower end of the oesoph-agus. This may be the first presentation of a benign stricture or a malignant tumour. + +Barotrauma (spontaneous +perforation, Boerhaave’s syndrome) +This occurs classically when a person vomits against a closed glottis. The pressure in the oesophagus increases rapidly, and the oesophagus bursts at its weakest point in the lower third, sending a stream of material into the mediastinum and often the pleural cavity as well. The condition was first identified +by Boerhaave, who reported the case of a grand admiral of + + +Hermann Boerhaave, 1668–1738, Professor of Medicine and Botany, the University of Leiden, the Netherlands, creator of the modern method of clinical teaching. +PART 11 | ABDOMINAL Perforation 1073 + + + +the Dutch fleet who was a glutton and practised autoemesis. Boerhaave’s syndrome is the most serious type of perforation because of the large volume of material that is released under pressure. This causes rapid chemical irritation in the medi-astinum and pleura followed by infection if untreated. Baro-trauma has also been described in relation to other pressure events when the patient strains against a closed glottis (e.g. +defecation, labour, weight-lifting). +Diagnosis of spontaneous perforation +The clinical history is usually of severe pain in the chest or upper abdomen after a meal or a bout of drinking. Associated shortness of breath is common. Many cases are misdiagnosed as myocardial infarction, perforated peptic ulcer or pancrea-titis if the pain is confined to the upper abdomen. There may be a surprising amount of rigidity on examination of the upper abdomen, even in the absence of any peritoneal contamination. +The diagnosis can usually be suspected from the history and associated clinical features. A chest radiograph is often confirmatory with air in the mediastinum, pleura or peri-toneum. Pleural effusion occurs rapidly either as a result of free communication with the pleural space or as a reaction to adjacent inflammation in the mediastinum. A contrast swallow or CT is nearly always required to guide management +(Figure 62.9). + +Pathological perforation +Free perforation of ulcers or tumours of the oesophagus into the pleural space is rare. Erosion into an adjacent structure with fistula formation is more common. Aerodigestive fis-tula is most common and usually encountered in primary malignant disease of the oesophagus or bronchus. Coughing on eating and signs of aspiration pneumonitis may allow the problem to be recognised at a time when intervention may be appropriate and feasible. Covering the communication with a self-expanding metal stent is the usual solution. Erosion into +a major vascular structure is invariably fatal. + + + + + + + + + + + + + + + + + +Figure 62.9 Computed tomography scan showing the site of perfor-ation in the lower oesophagus. + +Penetrating injury +Perforation by knives and bullets is uncommon, even in war, because the oesophagus is a relatively small target surrounded +by other vital organs. + +Foreign bodies +The oesophagus may be perforated during removal of a for-eign body but, occasionally, an object that has been left in the +oesophagus for several days will erode through the wall. + +Instrumental perforation +Instrumentation is by far the most common cause of perfor-ation. Modern instrumentation is remarkably safe, but perfor- +ation remains a risk that should never be forgotten. + + +Summary box 62.4 + +Instrumental perforation +●● Prevention of perforation is better than cure + + +Perforation related to diagnostic upper GI endoscopy is unusual with an estimated frequency of about 1:4000 exam-inations. Perforation can occur in the pharynx or oesopha-gus, usually at sites of pathology or when the endoscope is passed blindly. A number of patient-related factors are asso-ciated with increased risk, including large anterior cervical osteophytes, the presence of a pharyngeal pouch and mechan-ical causes of obstruction. Perforation may follow biopsy of a malignant tumour. +Patients undergoing therapeutic endoscopy have a perfor-ation risk that is at least 10 times greater than those undergo-ing diagnostic endoscopy. The oesophagus may be perforated by guidewires, graduated dilators or balloons, or during the placement of self-expanding stents. The risk is considerably +higher in patients with malignancy. +Diagnosis of instrumental perforation +In most cases, a combination of technical difficulties and an interventional procedure should lead to a high index of suspi-cion. History and physical signs may be useful pointers to the site of perforation. +Cervical perforation may result in pain localised to the neck, hoarseness, painful neck movements and subcutaneous emphysema. Intrathoracic and intra-abdominal perforations, which are more common, can give rise to immediate symp-toms and signs either during or at the end of the procedure, including chest pain, haemodynamic instability, oxygen desat-uration or visual evidence of perforation. Within the first 24 hours, patients may additionally complain of abdominal pain or respiratory difficulties. There may be evidence of subcuta-neous emphysema, pneumothorax or hydropneumothorax. In some patients, the diagnosis may be missed and recognised only at a late stage beyond 24 hours, as unexplained pyrexia, +systemic sepsis or the development of a clinical fistula. +PART 11 | ABDOMINAL 1074 CHAPTER 62 The oesophagus + + + +Prompt and thorough investigation is the key to manage-ment. Careful endoscopic assessment at the end of any pro-cedure combined with a chest radiograph will identify many cases of perforation immediately. If not recognised immedi-ately, then early and late suspected perforations should be assessed by CT. A pre-contrast scan easily demonstrates air outside the GI tract, while oral contrast localises the site of the oesophageal defect and accurately delineates specific fluid collections. Where CT is not available, a water-soluble con-trast swallow should be performed although this is not always reliable in disclosing a small contained leak. If a water-soluble study is negative and clinical suspicion is high, a dilute bar- +ium swallow should be considered. + +Treatment of oesophageal +perforations +Perforation of the oesophagus usually leads to mediastinitis. The loose areolar tissues of the posterior mediastinum allow a rapid spread of GI contents. The aim of treatment is to limit mediastinal contamination and prevent or deal with infection. Operative repair deals with the injury directly, but imposes risks of its own; non-operative treatment aims to limit the effects of mediastinitis and provide an environment in which healing can take place. +The decision between surgical and non-surgical manage- +ment rests on four factors. These are: +1 the site of the perforation (cervical vs thoracoabdominal oesophagus); +2 the event causing the perforation (spontaneous vs instru-mental); +3 underlying pathology (benign or malignant); +4 the status of the oesophagus before the perforation (fasted +and empty vs obstructed with a stagnant residue). +It follows that most perforations that can be managed non-surgically occur in the context of small instrumental per-forations of a clean oesophagus without obstruction, where leakage is likely to be confined to the nearby mediastinum at worst (Table 62.1). +Instrumental perforations in the cervical oesophagus are usually small and can nearly always be managed conserva-tively. The development of a local abscess is an indication for cervical drainage, preventing the extension of sepsis into the +mediastinum. + + + +TABLE 62.1 Management options in perforation of the oesophagus. +Factors that favour non-surgical Factors that favour management surgical repair +Small septic load Large septic load Minimal cardiovascular upset Septic shock Perforation confined to mediastinum Pleura breached Perforation by flexible endoscope Boerhaave’s syndrome +Perforation of cervical oesophagus Perforation of abdominal oesophagus + +The conservative management of an instrumental perfor-ation in the thoracoabdominal parts of the oesophagus can be undertaken when the perforation is detected early and before oral alimentation. General guidelines for non-surgical +management include: +● pain that is readily controlled with opiates; +● absence of crepitus, diffuse mediastinal gas, hydropneu-mothorax or pneumoperitoneum; +● mediastinal containment of the perforation with no evi-dence of widespread extravasation of contrast material; +● no evidence of on-going luminal obstruction or a retained +foreign body. +In addition, conservative management might be appropri-ate in patients who have remained clinically stable despite diagnostic delay. The principles of non-interventional man-agement involve hyperalimentation, preferably by an enteral route, nasogastric suction and broad-spectrum intravenous antibiotics. +Surgical management is required whenever patients: +● are unstable with sepsis or shock; +● have evidence of a heavily contaminated mediastinum, pleural space or peritoneum; +● have widespread intrapleural or intraperitoneal extravasa- +tion of contrast material. +Ongoing luminal obstruction (often related to malig-nancy) in a frail patient considered unfit for major surgery can be dealt with by placement of a covered self-expanding stent. Expanding metal stents should be used with caution in patients with benign disease because they cause significant tissue reaction and some designs are impossible to remove at a later date. Biodegradable and removable stents may be used alone or as a bridge to later definitive treatment where perfor-ation accompanies obstruction. +For patients requiring surgery, the choice is from direct repair, the deliberate creation of an external fistula or, rarely, oesophageal resection with a view to delayed reconstruction. Direct repair is preferred by many surgeons if the perforation is recognised early (within the first 4–6 hours), and the extent of mediastinal and pleural contamination is small. After 12 hours, the tissues become swollen and friable and less suitable for direct suture. The hole in the mucosa is always bigger than the hole in the muscle, and the muscle should be incised to see the mucosal edges clearly. It is essential that there should be no obstruction distal to the repair. A variety of local tissues (gastric fundus, pericardium, intercostal muscle) have been used to buttress such repairs. +Primary repair is inadvisable with late presentation and in the presence of widespread mediastinal and pleural contamina-tion. These patients tend to be more ill as a result of the delay, and the aim of treatment should be to achieve wide drainage with the creation of a controlled fistula and distal enteral feed-ing. This can usually be achieved by placing a T-tube into the oesophagus along with appropriately located drains and a feed-ing jejunostomy. In unusual circumstances, e.g. with extensive necrosis after corrosive ingestion, emergency oesophagectomy may be necessary. Oesophagostomy and gastrostomy should be +performed with a view to delayed reconstruction. +PART 11 | ABDOMINAL Corrosive injury 1075 + + + +MALLORY–WEISS SYNDROME +Forceful vomiting may produce a mucosal tear at the cardia rather than a full perforation. The mechanism of injury is different. In Boerhaave’s syndrome, vomiting occurs against a closed glottis and pressure builds up in the oesophagus. In Mallory–Weiss syndrome, vigorous vomiting produces a ver-tical split in the gastric mucosa, immediately below the squa-mocolumnar junction at the cardia in 90% of cases. In only 10% is the tear in the oesophagus (Figure 62.10). The con-dition presents with haematemesis. Usually, the bleeding is not severe, but endoscopic injection therapy may be required for the occasional, severe case. Surgery is rarely required. There are two other injuries to the oesophagus that lie within the spectrum of the mucosal tear of Mallory–Weiss and the full-thickness tear of Boerhaave. Intramural rupture produces a dissection within the oesophageal wall that causes severe chest pain, often with odynophagia. It is best diagnosed by contrast radiology. Intramural haematoma is seen most often in elderly patients on anticoagulants or patients with coagu-lation disorders, and usually follows an episode of vomiting. Large haematomas causing dysphagia can occur, extending from the cardia up to the carina. The diagnosis is readily made on endoscopy. Both intramural rupture and intramural hae-matoma can be managed conservatively. Symptoms usually resolve in 7–14 days, and oral intake can be reinstituted as +soon as symptoms allow. + +CORROSIVE INJURY +Corrosives such as sodium hydroxide (lye, caustic soda) or sul-phuric acid may be taken in an attempted suicide. Accidental ingestion occurs in children and when corrosives are stored in bottles labelled as beverages. All can cause severe dam-age to the mouth, pharynx, larynx, oesophagus and stomach. The type of agent, its concentration and the volume ingested largely determine the extent of damage. In general, alkalis are relatively odourless and tasteless, making them more likely to be ingested in large volume. Alkalis cause liquefaction, saponification of fats, dehydration and thrombosis of blood +vessels which usually leads to fibrous scarring. Acids cause + + + + + + + + + + + + +Figure 62.10 The endoscopic appearance of a mucosal tear at the cardia (Mallory–Weiss). + + +coagulative necrosis with eschar formation, and this coagu-lant may limit penetration to deeper layers of the oesopha-geal wall. Acids also cause more gastric damage than alkalis because of the induction of intense pylorospasm with pooling in the antrum. +Symptoms and signs are notoriously unreliable in predict-ing the severity of injury. The key to management is early endoscopy by an experienced endoscopist to inspect the whole of the oesophagus and stomach (Figure 62.11). Deep ulcers and the recognition of a grey or black eschar signify the most severe lesions with the greatest risk of perforation. Minor injuries with only oedema of the mucosa resolve rap-idly with no late sequelae. These patients can safely be fed. With more severe injuries, a feeding jejunostomy may be appropriate until the patient can swallow saliva satisfactorily. The widespread use of broad-spectrum antibiotics and ste-roids is not supported by evidence. +Regular endoscopic examinations are the best way to assess stricture development (Figure 62.12). Significant stric-ture formation occurs in about 50% of patients with exten-sive mucosal damage (Figure 62.13). The role and timing of +repeat endoscopies with or without dilatation in such patients + + + + + + + + + + + + + + + + + +Figure 62.11 Acute caustic burn in the haemorrhagic phase. + + + + + + + + + + + + + +Figure 62.12 The late result of a caustic alkali burn with a high oesophageal stricture. + + + +George Kenneth Mallory, b.1926, Professor of Pathology, Boston University Boston, MA, USA. +, +Soma Weiss, 1898–1942, Professor of Medicine, Harvard University Medical School, Boston, MA, USA. +PART 11 | ABDOMINAL 1076 CHAPTER 62 The oesophagus + + +an adequate drink. Acute injury presents with dysphagia and odynophagia, which may be severe. The inflammation usu-ally resolves within 2–3 weeks, and no specific treatment is required apart from appropriate nutritional support. A stric- +ture may follow. + + + + + + + + + + + + + + + + + + + + + +Figure 62.13 Caustic or lye stricture with marked stenosis high in the body of the oesophagus. The strictures are frequently multiple and difficult to dilate unless treated energetically at an early stage. + + +remain controversial. Other than the need for emergency sur-gery for bleeding or perforation, elective oesophageal resec-tion should be deferred for at least 3 months until the fibrotic phase has been established. Oesophageal replacement is usu-ally required for very long or multiple strictures. Resection can be difficult because of perioesophageal inflammation in these patients. Because of associated gastric damage, colon may have to be used as the replacement conduit. +There is also controversy with regard to the risk of devel-oping carcinoma in the damaged oesophagus and stomach and how this might influence management. The lifetime risk is certainly less than 5%. Some surgeons advocate resection and replacement, whereas others believe that oesophageal bypass and endoscopic surveillance are preferable, because removal of the badly damaged oesophagus from a scarred mediastinum +can be hazardous. + + +Summary box 62.5 + +Corrosive injury +●● Skilled early endoscopy is mandatory + + +DRUG-INDUCED INJURY +Many medications, such as antibiotics and potassium prepar-ations, are potentially damaging to the oesophagus, because +tablets may remain for a long time, especially if taken without + +GASTRO-OESOPHAGEAL +REFLUX DISEASE Aetiology +Normal competence of the gastro-oesophageal junction is maintained by the LOS. This is influenced by both its physiological function and its anatomical location relative to the diaphragm and the oesophageal hiatus. In normal circum-stances, the LOS transiently relaxes as a coordinated part of swallowing, as a means of allowing vomiting to occur and in response to stretching of the gastric fundus, particularly after a meal to allow swallowed air to be vented. Most episodes of physiological reflux occur during postprandial transient LOS relaxations (TLOSRs). In the early stages of GORD, most pathological reflux occurs as a result of an increased number of TLOSRs rather than a persistent fall in overall sphincter pressure. In more severe GORD, LOS pressure tends to be generally low, and this loss of sphincter function seems to be made worse if there is loss of an adequate length of intra-abdominal oesophagus. +The absence of an intra-abdominal length of oesoph-agus results in a sliding hiatus hernia. The normal conden-sation of peritoneal fascia over the lower oesophagus (the phreno-oesophageal ligament) is weak, and the crural open-ing widens, allowing the upper stomach to slide up through the hiatus. The loss of the normal anatomical configura-tion exacerbates reflux, although sliding hiatus hernia alone should not be viewed as the cause of reflux. Sliding hiatus hernia is associated with GORD and may make it worse but, as long as the LOS remains competent, pathological GORD does not occur. Many GORD sufferers do not have a her-nia, and many of those with a hernia do not have GORD. It should be noted that rolling or paraoesophageal hiatus her-nia is a quite different and potentially dangerous condition (see below). A proportion of patients have a rolling hernia and symptomatic GORD or a mixed hernia with both sliding and rolling components. Reflux oesophagitis that is visible endoscopically is a complication of GORD and occurs in a minority of sufferers overall, but in around 40% of patients referred to hospital. +In western societies, GORD is the most common con-dition affecting the upper GI tract. This is partly due to the declining incidence of peptic ulcer as the incidence of infection with Helicobacter pylori has reduced as a result of improved socioeconomic conditions, along with a rising inci-dence of GORD in the last 30 years. The cause of the increase is unclear, but may be due in part to increasing obesity. The strong association between GORD, obesity and the paral-lel rise in the incidence of adenocarcinoma of the oesoph-agus represents a major health challenge for most western +countries. +PART 11 | ABDOMINAL +Gastro-oesophageal reflux disease 1077 + + + +Clinical features +The classic triad of symptoms is retrosternal burning pain (heartburn), epigastric pain (sometimes radiating through to the back) and regurgitation. Most patients do not experience all three. Symptoms are often provoked by food, particularly those that delay gastric emptying (e.g. fats, spicy foods). As the condition becomes more severe, gastric juice may reflux to the mouth and produce an unpleasant taste, often described as ‘acid’ or ‘bitter’. Heartburn and regurgitation can be brought on by stooping or exercise. A proportion of patients have odynophagia with hot beverages, citrus drinks or alcohol. Patients with nocturnal reflux and those who reflux food to the mouth nearly always have severe GORD. Some patients present with less typical symptoms such as angina-like chest pain, pulmonary or laryngeal symptoms. Dysphagia is usually a sign that a stricture has occurred, but may be caused by an associated motility disorder. +As GORD is such a common disorder, it should always be the first thought when a patient presents with oesophageal symptoms that are unusual or that defy diagnosis after a series +of investigations. + +Diagnosis +In most cases, the diagnosis is assumed rather than proven, and treatment is empirical. Investigation is required only when the diagnosis is in doubt, when the patient does not respond to a proton pump inhibitor (PPI) or if dysphagia is present. The most appropriate examination is endoscopy with biopsy. If the typical appearance of reflux oesophagitis, peptic stric-ture or Barrett’s oesophagus is seen, the diagnosis is clinched, but visible oesophagitis is not always present, even in patients selected as above. This is compounded in clinical practice by the widespread use of PPIs, which cause rapid healing of early mucosal lesions. Many patients will have received such treat-ment before referral. The endoscopic appearances of the nor-mal oesophagus, hiatus hernia, oesophagitis and stricture are shown in Figures 62.14–62.20. It is worth remembering that the correlation between symptoms and endoscopic appear-ances is poor. On the other hand, there is a strong correlation between worsening endoscopic appearances and the duration of oesophageal acidification on pH testing. +In patients with atypical or persistent symptoms despite therapy, oesophageal manometry and 24-hour oesophageal pH recording (ideally with impedance measurement) may be +justified to establish the diagnosis and guide management. + + +Summary box 62.6 + +Diagnostic measurement in GORD +●● 24-hour pH recording is the ‘gold standard’ for diagnosis of GORD +●● Length and pressure of the LOS are important + + + + + + + + + + + + + + + + + +Figure 62.14 The endoscopic appearance of the normal squamous mucosa in the body of the oesophagus. + + + +(a) + + + + + + + + + + + + + + + +(b) + + + + + + + + + + + + + + + +Figure 62.15 The normal lower oesophageal sphincter: (a) open; (b) closed. + + + + +Norman Rupert Barrett, 1903–1979, surgeon, St Thomas’ Hospital, London, UK. +PART 11 | ABDOMINAL 1078 CHAPTER 62 The oesophagus + + + + + + + + + + + + + + +Figure 62.16 The squamocolumnar junction is clearly seen in the lower oesophagus with a normal sharp demarcation. + + + + + + + + + + + + + + +Figure 62.17 Sliding hiatus hernia. The diaphragm can be seen con-stricting the upper stomach. + + + + + + + + + + + +Figure 62.18 Reflux oesophagitis. + + + + + + + + + + + + + +Figure 62.19 Benign stricture with active oesophagitis (left) and healed with columnar epithelium (right). + + +Figure 62.20 Ulceration associated with a benign peptic stricture. + + +As a matter of routine, proton pump inhibitors (PPIs) are stopped 1 week before oesophageal pH recording, but acid secretion is sometimes reduced for 2 weeks or more, and this can necessitate repeat examination after a prolonged interval without a PPI. Manometry and pH recording are essential in patients being considered for antireflux surgery. Although the main purpose of the test is objectively to quantify the extent of reflux disease, it is also used to rule out a diagnosis of acha-lasia. In the early stages of achalasia, chest pain can dominate the clinical picture and, when associated with intermittent swallowing problems and non-specific symptoms, it is easy to see how a clinical diagnosis of GORD might be made. Patients with achalasia can also have an abnormal pH study as a result of fermentation of food residue in a dilated oesoph-agus. Usually, the form of the pH trace is different from that of GORD, with slow undulations of pH rather than rapid bursts of reflux, but the complete absence of peristalsis on manome-try is pathognomonic of achalasia. +A CT scan gives the best appreciation of gastro-oesoph-ageal anatomy. This may be important in the context of sur-gery for rolling or mixed hiatus hernias, but it is unimportant +in most patients with GORD. + +Management of uncomplicated +GORD +Medical management +Most sufferers from GORD do not consult a doctor and do not need to do so. They self-medicate with over-the-counter medicines such as simple antacids, antacid–alginate prepara-tions, H2-receptor antagonists or PPIs. Consultation is more likely when symptoms are severe, prolonged and unrespon-sive to the above treatments. Simple measures that are often neglected include advice about weight loss, smoking, exces-sive consumption of alcohol, tea or coffee, the avoidance of large meals late at night and a modest degree of head-up tilt of the bed. Tilting the bed has been shown to have an effect that is similar to taking an H2-receptor antagonist. The common practice of using additional pillows has no significant effect. +PPIs are the most effective drug treatment for GORD. +Indeed, they are so effective that, once started, patients are +PART 11 | ABDOMINAL +Gastro-oesophageal reflux disease 1079 + + + +very reluctant to stop taking them. Given an adequate dose for 8 weeks, most patients have a rapid improvement in symp-toms (within a few days), and more than 90% can expect full mucosal healing at the end of this time. For this reason, a policy of ‘step-down’ medical treatment is advocated based on the general advice outlined above and a standard dose of a PPI given for 8 weeks. At the end of that time, the dose of PPI is reduced to that which keeps the patient free of symptoms, and this might even mean the cessation of PPI treatment. As most patients do not make major lifestyle changes and as PPIs are so effective, many remain on long-term treatment. For the minority who do not respond adequately to a standard dose, a trial at an increased dose or the addition of an H2-receptor antagonist is recommended. If unsuccessful, these patients should be formally investigated. +PPI therapy is also important in patients with reflux-induced strictures, resulting in significant prolongation of the intervals between endoscopic dilatation. As yet, fears that chronic acid suppression might have serious long-term side +effects including the risk of gastric cancer seem unwarranted. + +Surgery +Strictly speaking, the need for surgery should have been reduced as medication has improved so much. Paradoxically, the number of antireflux operations has remained relatively constant and may even be increasing. This is probably due partly to increased patient expectations and partly to the advent of minimal access surgery, which has improved the +acceptability of procedures. + +operation (5–10%) and side effects such as dysphagia, gas bloat or abdominal discomfort (10%). With current surgical techniques, 85–90% of patients should be satisfied with the result of an antireflux operation. Patients who are asymptomatic on a PPI need a careful discussion of the risk side of the equation. Those who are symptomatic on a PPI need a careful clinical review to make sure that they will benefit from an operation. Reasons for failure on a PPI include ‘volume’ reflux (a good indication for surgery), a ‘hermit’ lifestyle in which the least deviation from lifestyle rules leads to symptoms (a good indication), psychological distress with intolerance of minor symptoms (a poor indication; these patients are likely to be dissatisfied with surgery), poor compliance (a good indication if the reason for poor compliance is the side effects of treatment, otherwise a bad indication) and misdiagnosis of +GORD. + +WHICH SURGERY? +There are many operations for GORD, but they are virtu-ally all based on the creation of an intra-abdominal segment of oesophagus, crural repair and some form of wrap of the upper stomach (fundoplication) around the intra-abdominal oesophagus. The contribution of each component to opera-tive success is widely debated, but it is clear that operations that fail to address all three components have inferior suc-cess rates. The major types of anti-reflux operation were all developed in the 1950s (Figure 62.21). When performed correctly, nearly all are effective. Most randomised trials do +not show a clear advantage for any one operation over the + + +Endoscopic treatments +A number of endoscopic treatments have been tried in the last 10 years that attempt to augment a failing LOS. These involve endoscopic suturing devices that plicate gastric mucosa just below the cardia to accentuate the angle of His, radiofrequency ablation (RFA) to the level of the sphinc-ter and the injection of submucosal polymers into the lower +oesophagus. The procedures have generally been applied (a) (b) to patients with only small hiatus hernias or none at all, so +only a small proportion of patients who present to hospitals are suitable. Although most methods produce some tempo-rary improvement in symptoms and objective assessments of reflux, failure rates at 1 year are high and can be over 50%. There are few large series that have reported long-term out-comes. Only RFA has been shown to be effective for as long as 10 years when around two-thirds patients remain off PPI +medication. + + +Surgical treatments +The indication for surgery in uncomplicated GORD is essentially patient choice. The risks and possible benefits need to be discussed in detail. Risks include a small mortality +rate (0.1–0.5%, depending on patient selection), failed + + +(c) (d) +Figure 62.21 Various operations for the surgical correction of gastro-oesophageal reflux disease. (a) The original Allison repair of hiatus hernia (this is ineffective and is no longer done); (b) Nissen fundoplication; (c) Hill procedure; (d) Belsey mark IV operation. + + + +Philip Rowland Allison, 1907–1974, Professor of Surgery, Oxford University, Oxford, UK. Rudolph Nissen, 1896–1981, Professor of Surgery, Istanbul, T rkey, and later at Basle, Switzerland. Lucius Davis Hill, surgeon, the Mason Clinic, Seattle, MN, USA. +u +Ronald Herbert Robert Belsey, d.2007, thoracic surgeon, Frenchay Hospital, Bristol, UK. +PART 11 | ABDOMINAL 1080 CHAPTER 62 The oesophagus + + + +others, although one meta-analysis has come down in favour of posterior partial fundoplication over total fundoplication when performed laparoscopically. +Total fundoplication (Nissen) tends to be associated with slightly more short-term dysphagia but is the most durable repair in terms of long-term reflux control. Partial fundopli-cation, whether performed posteriorly (Toupet) or anteriorly (Dor, Watson), has fewer short-term side effects, although this is sometimes at the expense of a slightly higher long-term fail-ure rate. One disadvantage of total fundoplication is the cre-ation of an overcompetent cardia, resulting in the ‘gas bloat’ syndrome in which belching is impossible. The stomach fills with air, the patient feels very full after small meals and passes excessive flatus. This does not seem to occur with partial fun-doplication. The problem has been largely overcome by the ‘floppy’ Nissen technique in which the fundoplication is loose around the oesophagus and is kept short in length. Although the other short-term side effects of fundoplication usually resolve within 3 months of surgery, this is rarely the case for gas bloat. The problem is best remedied by conversion to a partial fundoplication. +As with primary surgery, various revisional procedures have been described, usually for recurrent reflux or persistent dyspha-gia. For most patients, recurrent reflux relates to anatomical fail-ure, so the solution is a revisional fundoplication. The results of surgery for recurrent reflux tend to be better than those for dys-phagia, because the latter problem has many causes (too tight a wrap, slipping of the wrap, hiatal fibrosis) A very small pro-portion of patients may undergo more than two operations to correct recurrent reflux or unacceptable side effects. Revisional surgery carries a lower chance of success and, in some patients, local revision becomes technically impossible. The final resort is antrectomy with a Roux-en-Y reconstruction. This reduces gastric acid secretion and diverts bile and pancreatic secretions away from the stomach. Thus, the volume of potential refluxate in the stomach is reduced and, because of its changed composi-tion, it is less damaging to the oesophagus. +For many years, the relative merits of thoracic and abdom-inal approaches were hotly debated. The introduction of mini-mal access surgery has made this debate obsolete. Most antireflux +operations are now done with a laparoscopic approach. + +LAPAROSCOPIC FUNDOPLICATION +Five cannulae are inserted in the upper abdomen (Figure 62.22). The cardia and lower oesophagus are separated from the diaphragmatic hiatus. An appropriate length of oesoph-agus is mobilised in the mediastinum. The fundus may be mobilised by dividing the short gastric vessels that tether the fundus to the spleen, although some surgeons feel that this is unnecessary. The hiatus is narrowed by sutures placed behind the oesophagus. In total (Nissen) fundoplication, the fundus is drawn behind the oesophagus and then sutured to itself in front of the oesophagus (Figure 62.23a). In partial fundopli-cation, the fundus is drawn either behind or in front of the +oesophagus and sutured to it on each side, leaving a strip of + + + + + + + + + + + +Figure 62.22 Laparoscope cannula sites for laparoscopic fundo-plication. + + + + + + + + + +(a) (b) +Figure 62.23 (a) Total (Nissen) fundoplication; (b) partial fundopli-cation (Toupet). + +exposed oesophagus either at the front (Figure 62.23b) or at the back. Robotic fundoplication has been described but comparisons with laparoscopic surgery do not indicate any +clear clinical benefits. + +Complications of GORD Stricture +Reflux-induced strictures (see Figure 62.20) occur mainly in late middle-aged and elderly people, but they may present in children. It is important to distinguish a benign reflux-induced stricture from a carcinoma. This is not usually difficult on the basis of location (immediately above the oesophagogastric junction), length (only about 1–2 cm) and smooth mucosa, but sometimes a cancer spreads under the oesophageal mucosa at its upper margin, producing a benign-looking stricture. +Peptic strictures generally respond well to dilatation and long-term treatment with a PPI. As most patients are elderly, antireflux surgery is not usually considered. However, it is an alternative to long-term PPI treatment, just as in uncompli-cated GORD in younger and fitter patients. Most patients do +not require anything other than a standard operation. + + +Summary box 62.7 + +Peptic stricture +●● Day-case dilatation and PPI for peptic stricture + + + +AM Toupet, French surgeon. +Cesar Roux, 1857–1934, Professor of Surgery and Gynaecology, Lausanne, Switzerland, described this method of forming a jejunal conduit in 1908. +PART 11 | ABDOMINAL +Gastro-oesophageal reflux disease 1081 + + +Oesophageal shortening (a) +The issue of oesophageal shortening continues to provoke debate. There can be no doubt that, in the presence of a large sliding hiatus hernia, the oesophagus is short, but this does not necessarily mean that, with mobilisation from the medi-astinum, it cannot easily be restored to its normal length. The extent to which severe inflammation in the wall of the oesophagus causes fibrosis and real shortening is less clear. If a good segment of intra-abdominal oesophagus cannot be restored without tension, a Collis gastroplasty should be performed (Figure 62.24). This produces a neo-oesophagus around which a fundoplication can be done (Collis–Nissen +operation). (b) + + + +Summary box 62.8 + +GORD +●● Is due to loss of competence of the LOS and is extremely common +●● May be associated with a hiatus hernia, which may be sliding or, less commonly, rolling (paraoesophageal) +●● The most common symptoms are heartburn, epigastric discomfort and regurgitation, often made worse by stooping and lying +●● Achalasia and GORD are diagnostically easily confused Dysphagia may occur, but a neoplasm must be excluded +Diagnosis and treatment can be instituted on clinical grounds +●● Endoscopy may be required and 24-hour pH is the ‘gold standard’ +Management is primarily medical (PPIs being the most effective), but surgery may be required; laparoscopic fundoplication is the most popular technique +Stricture may develop in time + + +Barrett’s oesophagus (columnar- +lined lower oesophagus) +Barrett’s oesophagus is a metaplastic change in the lining mucosa of the oesophagus in response to chronic gastro-oesophageal reflux (Figure 62.25). Many of these patients +do not have particularly severe symptoms, although they do + + + + + + + + + +Figure 62.25 Barrett’s oesophagus with (a) proximal migration of the squamocolumnar junction and (b) a view of the distal oesoph-agus. + +have the most abnormal pH profiles. This adaptive response involves a mosaic of cell types, probably beginning as a sim-ple columnar epithelium which becomes ‘specialised’ with time. The hallmark of ‘specialised’ Barrett’s epithelium is the presence of mucus-secreting goblet cells (intestinal metapla-sia). One of the great mysteries of GORD is why some people develop oesophagitis and others develop Barrett’s oesophagus, often without significant oesophagitis. In Barrett’s oesopha-gus, the junction between squamous oesophageal mucosa and gastric mucosa moves proximally. It may be difficult to distin-guish a Barrett’s oesophagus from a tubular, sliding hiatus her-nia during endoscopy, as the two often coexist (Figure 62.26) +or where the visible Barrett’s segment is very short. The key + +(a) (b) + + + + + + + + + +Figure 62.24 Collis gastroplasty to produce a neo-oesophagus around which a Nissen fundoplication is done. The operation may be performed by a laparoscopic as well as an open approach, using cir-cular and linear staplers. + + +Figure 62.26 (a) The interrelationship of the lower oesophageal sphincte , the squamocolumnar junction and the diaphragm in sliding hiatus hernia. (b) Barrett’s oesophagus and sliding hernia. +r + + + +John Leigh Collis, 1911–2003, Professor of Thoracic Surgery, the University of Birmingham, Birmingham, UK. +PART 11 | ABDOMINAL 1082 CHAPTER 62 The oesophagus + + + +(a) + +is where the gastric mucosal folds end along with recognition of the mucosal vascular pattern. The mucosa in the body of the stomach has longitudinal folds; the columnar lining of Barrett’s oesophagus is smooth. The lower oesophagus is char-acterised by palisade vessels that run longitudinally and are easily seen through the lower oesophageal mucosa. Strictures can occur in Barrett’s oesophagus and nearly always appear at the new squamocolumnar junction (Figure 62.27). Rarely, a stricture may occur in the columnar segment after healing of a Barrett’s ulcer (Figure 62.28). Although intestinal metapla-sia and length of the Barrett’s segment are important risk fac-tors for the development of carcinoma, neither represents an essential feature for cancer development. The risk of transfor-mation to cancer is probably no more than 0.5% per patient per year, which is about 25 times that of the general popula- +tion (Figures 62.29 and 62.30). + + + +Summary box 62.9 + +Barrett’s oesophagus +●● Intestinal metaplasia is a risk factor for the development of adenocarcinoma +●● Do not confuse Barrett’s ulcer with oesophagitis + + + + + +(b) + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.27 The radiological appearance of a midoesophageal stricture in (a), a patient with Barrett’s oesophagus and in (b) a normal lumen after dilatation. + +Patients who are found to have Barrett’s oesophagus may be submitted to regular surveillance endoscopy with multi-ple biopsies in the hope of finding dysplasia or in situ cancer, rather than allowing invasive cancer to develop and cause symptoms. There is no general agreement about the bene-fits of surveillance endoscopy, or its ideal frequency. Endos-copy at 2-year intervals is probably adequate, provided that no dysplasia has been detected. A significant problem is that +the incidence of Barrett’s oesophagus in the community is + + + + + + + + + + + + + + + + + + + + + + +Figure 62.28 Barretts ulcer in the columnar cell-lined oesophagus. +’ +PART 11 | ABDOMINAL Paraoesophageal (‘rolling’) hiatus hernia 1083 + + + + + + + + + + + + + + + + + +Figure 62.29 The macroscopic appearances of an adenocarcinoma in Barrett’s oesophagus. + +Barrett’s mucosa in the hope of eliminating the risk of can-cer development. Laser, photodynamic therapy, argon-beam plasma coagulation, RFA and endoscopic mucosal resection (EMR) have all been used. Until recently, it was felt that intervention should only be offered in patients with high-grade dysplasia, but recent evidence suggests that patients with low-grade dysplasia also have a substantial risk of pro-gressing to high-grade dysplasia of around 9% per year. For this reason endoscopic treatment is considered appropriate for patients with low-grade dysplasia when confirmed by a second independent histopathologist. EMR of dysplastic areas followed by RFA seems to be the most popular approach and, together with PPI treatment, these endoscopic methods can result in a neosquamous lining. Fears of buried glands that +might give rise to cancer seem unfounded. + +PARAOESOPHAGEAL +(‘ROLLING’) HIATUS HERNIA +True paraoesophageal hernias in which the cardia remains in its normal anatomical position are rare. The vast majority of rolling hernias are mixed hernias in which the cardia is dis-placed into the chest and the greater curve of the stomach rolls into the mediastinum (Figure 62.31). Sometimes, the whole of the stomach lies in the chest (Figure 62.32). Colon or small intestine may sometimes lie in the hernia sac. The hernia is most common in elderly people, but may occur in young fit people. As the stomach rolls up into the chest, there +is always an element of rotation (volvulus). + + + + + + +Figure 62.30 Endoscopic view of carcinoma in Barrett’s oesopha-gus. + + +estimated to be at least 10 times the incidence discovered in dyspeptic patients referred for endoscopy. Thus, adenocar-cinoma in Barrett’s oesophagus often presents with invasive cancer without any preceding reflux symptoms. +Until recently, Barrett’s oesophagus was not diagnosed until there was at least 3cm of columnar epithelium in the distal oesophagus. With the better appreciation of the impor-tance of intestinal metaplasia, Barrett’ oesophagus may be diagnosed if there is any intestinal metaplasia in the oesopha-gus. The relative risk of cancer rises with increasing length of +s +abnormal mucosa. The following terms are widely used: +● classic Barrett’s (≥3cm columnar epithelium); +● short-segment Barrett’s (<3cm of columnar epithelium); ● cardia metaplasia (intestinal metaplasia at the oesoph- +agogastric junction without any macroscopic change at +endoscopy). + +When Barrett’s oesophagus is discovered, the treatment is that of the underlying GORD. There has been consider- +able interest in recent years in endoscopic methods of treating + + +Figure 62.31 A paraoesophageal hernia showing the gastro-oesophageal junction just above the diaphragm and the fundus along-side the oesophagus, compressing the lumen. +PART 11 | ABDOMINAL 1084 CHAPTER 62 The oesophagus + + + + + + + + + + + + + + + + + + + + + +Figure 62.32 A huge paraoesophageal hernia with an upside-down stomach and the pylorus just below the hiatus. + + + +Summary box 62.10 + +‘Rolling’ hiatus hernia +●● Potentially dangerous, because of volvulus + + + +The symptoms of rolling hernia are mostly due to twist-ing and distortion of the oesophagus and stomach. Dyspha-gia is common. Chest pain may occur from distension of an obstructed stomach. Classically, the pain is relieved by a loud belch. Symptoms of GORD are variable. Strangulation, gas-tric perforation and gangrene can occur. Emergency presenta-tion with any of these complications carries high mortality on account of a combination of late diagnosis, generally elderly patients with comorbid diseases and the complexity of surgery involved. +The hernia may be visible on a plain radiograph of the chest as a gas bubble, often with a fluid level behind the heart (Figure 62.33). A CT scan with oral contrast is the best method of diagnosis, highlighting the gastric anatomy but also identifying other structures involved in the hernia. The endoscopic appearances may be confusing, especially in large hernias when it is easy to become disoriented. +Symptomatic rolling hernias nearly always require surgi-cal repair because they are potentially dangerous. The risk of an asymptomatic patient developing a significant problem when a rolling hiatus hernia is discovered incidentally has probably been underestimated in the past. The annual risk is probably greater than the historical estimate of 1%. Patients who present as an emergency with acute chest pain may be treated initially by nasogastric tube, to relieve the distension that causes the pain, followed by surgical repair. Endoscopy is useful if nasogastric intubation is unsuccessful. If the pain is not relieved or perforation is suspected, immediate operation +is mandatory. + +Figure 62.33 A gas bubble seen on a plain chest radiograph, show-ing the fundus of the stomach in the chest (courtesy of Dr Stephen Ellis, Bart’s and the London NHS Trust). + +Emergency surgery needs to be tailored to the problem encountered and the fitness of the patient. Elective surgery involves reduction of the hernia, excision of the sac, reduction of the crural defect and some form of retention of the stom-ach in the abdomen. Some surgeons perform a fundoplication, arguing that this is a very effective means of maintaining reduc-tion and that it deals with the associated GORD. Others argue that fundoplication should be done only if reflux can be con-clusively demonstrated beforehand. Surprisingly, both philos-ophies achieve good results. Laparoscopic repair has recently become popular. Full anatomical repair of a large rolling hernia can be difficult using this approach and requires considerable expertise. Secure closure of the hiatal defect can be a problem, +and some surgeons advocate mesh to reinforce the repair. + +NEOPLASMS OF THE +OESOPHAGUS Benign tumours +Benign tumours of the oesophagus are relatively rare. True papillomas, adenomas and hyperplastic polyps do occur, but most ‘benign’ tumours are not epithelial in origin and arise from other layers of the oesophageal wall (GI stromal tumour [GIST], lipoma, granular cell tumour). Most benign oesophageal tumours are small and asymptomatic, and even a large benign tumour may cause only mild symptoms (Figure 62.34). The most important point in their management is usually to carry out an adequate number of biopsies to prove beyond reasonable doubt that the lesion is not malignant +(Figure 62.35). + +Malignant tumours +Non-epithelial primary malignancies are also rare, as is malig-nant melanoma. Secondary malignancies rarely involve the +oesophagus, with the exception of bronchogenic carcinoma +PART 11 | ABDOMINAL Neoplasms of the oesophagus 1085 + + +by direct invasion of either the primary and/or contiguous +lymph nodes. + +Carcinoma of the oesophagus +Cancer of the oesophagus is the sixth most common cancer in the world. In general, it is a disease of mid to late adulthood, with a poor survival rate. Only 5–10% of those diagnosed will +survive for 5 years. + +Summary box 62.11 + +Carcinoma of the oesophagus +●● Squamous cell usually affects the upper two-thirds; adenocarcinoma usually affects the lower third +●● Common aetiological factors are tobacco and alcohol (squamous cell), GORD and obesity (adenocarcinoma) +●● The incidence of adenocarcinoma is increasing +●● Lymph node involvement is a bad prognostic factor +●● Dysphagia is the most common presenting symptom, but is a late feature +●● Accurate pretreatment staging is essential in patients thought to be fit to undergo ‘curative’ treatment + + + + + + +Figure 62.34 Classic appearance of a large oesophageal gastro-intestinal stromal tumour on barium swallow. + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.35 An intraluminal polyp that proved to be a leiomyosar-coma. + +Pathology and aetiology +Squamous cell cancer (Figures 62.36 and 62.37) and adeno-carcinoma (Figures 62.38 and 62.39) are the most common +types. Squamous cell carcinoma generally affects the upper + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.36 The classic appearances of a midoesophageal prolif-erative squamous cell carcinoma. +PART 11 | ABDOMINAL 1086 CHAPTER 62 The oesophagus + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.38 Adenocarcinoma of the lower oesophagus, spreading upwards from the cardia. + +Figure 62.37 Squamous cell carcinoma of the oesophagus produc-ing an irregular stricture with shouldered margins. + + + + +two-thirds of the oesophagus and adenocarcinoma the lower third. Worldwide, squamous cell cancer is most common, but adenocarcinoma predominates in the west and is increasing +in incidence. +Geographical variation in oesophageal cancer +The incidence of oesophageal cancer varies more than that of any other cancer. Squamous cell cancer is endemic in the Transkei region of South Africa and in the Asian ‘cancer belt’, which extends across the middle of Asia from the shores of the Caspian Sea (in northern Iran) to China. The highest incidence in the world is in Linxian in Henan province in China, where it is the most common single cause of death, with more than 100 cases per 100 000 population per annum. The cause of the disease in the endemic areas is not known. Although there is evidence of genetic susceptibility across much of central Asia, a variety of environmental factors along with nutritional deficiencies are probably involved. In Linx-ian, supplementation of the diet with b-carotene, vitamin E and selenium has been shown to reduce the incidence. +Away from the endemic areas, tobacco and alcohol are +major factors in the occurrence of squamous cancer. Incidence + + + + + + + + + + + + + + +Figure 62.39 Computed tomography scan showing a primary tumour of the lower oesophagus. + + +rates vary from less than 5:100 000 in white people in the USA to 26.5:100 000 in some regions of France. +In many western countries, the incidence of squamous cell cancer has fallen or remained static, but the incidence of adenocarcinoma of the oesophagus has increased dramat- +ically since the mid-1970s at a rate of 5–10% per annum. +PART 11 | ABDOMINAL Neoplasms of the oesophagus 1087 + + + +The change is greater than that of any other neoplasm in this time. Adenocarcinoma now accounts for 60–75% of all oesophageal cancers in several countries. The reason for this change is not understood. A similar rate of increase in GORD over the same period, which mirrors an increase in obesity in the west, is likely to be an important factor, particularly through the link to Barrett’s oesophagus. Obese, white men in their 60s represent the highest-risk group. There has also been an increase in the incidence of carcinoma of the cardia of the stomach, which suggests that cancer of the cardia and adenocarcinoma of the oesophagus may share common aeti-ological factors. With a falling incidence of cancer in the rest of the stomach, more than 60% of all upper GI cancers in the west involve the cardia or distal oesophagus. +Both adenocarcinomas and squamous cell carcinomas tend to disseminate early. Sadly, the classic presenting symptoms of dysphagia, regurgitation and weight loss are often absent until the primary tumour has become advanced, and so the tumour is often well established before the diagnosis is made. Tumours can spread in three ways: invasion directly through the oesophageal wall, via the lymphatics or in the bloodstream. Direct spread occurs both laterally, through the component layers of the oesophageal wall, and longitudinally within the oesophageal wall. Longitudinal spread is mainly via the sub-mucosal lymphatic channels of the oesophagus. The pattern of lymphatic drainage is therefore not segmental, as in other parts of the GI tract. Consequently, the length of oesophagus involved by tumour is frequently much longer than the mac-roscopic length of the malignancy at the epithelial surface. Lymph node spread occurs commonly. Although the direc-tion of spread to regional lymphatics is predominantly caudal, the involvement of lymph nodes is potentially widespread and can also occur in a cranial direction. Any regional lymph node from the superior mediastinum to the coeliac axis and lesser curve of the stomach may be involved, regardless of the location of the primary lesion within the oesophagus. Hae-matogenous spread may involve a variety of different organs including the liver, lungs, brain and bones. Tumours arising from the intra-abdominal portion of the oesophagus may also +disseminate transperitoneally. +Clinical features +Most oesophageal neoplasms present with mechanical symp-toms, principally dysphagia, but sometimes also regurgitation, vomiting, odynophagia and weight loss. Clinical findings suggestive of advanced malignancy include recurrent laryn-geal nerve palsy, Horner’s syndrome, chronic spinal pain and diaphragmatic paralysis. Other factors making surgical cure unlikely include weight loss of more than 20% and loss of appetite. Cutaneous tumour metastases or enlarged supracla-vicular lymph nodes may be seen on clinical examination and indicate disseminated disease. Hoarseness due to recurrent laryngeal nerve palsy is a sign of advanced and incurable dis-ease. Palpable lymphadenopathy in the neck is likewise a sign +of advanced disease. + +Patients with early disease may have non-specific dyspep-tic symptoms or a vague feeling of ‘something that is not quite right’ during swallowing. Some are diagnosed during endo-scopic surveillance of patients with Barrett’s oesophagus and, although this does identify patients with the earliest stages of disease, such programmes have little overall impact, because most patients with Barrett’s oesophagus are unknown to the medical profession and make their first presentation with a symptomatic, and therefore usually locally advanced, oesoph-ageal cancer. The widespread use of endoscopy as a diagnostic tool does, nevertheless, provide an opportunity for early diag-nosis (Figure 62.40). Biopsies should be taken of all lesions in the oesophagus (Figures 62.41 and 62.42), no matter how trivial they appear and irrespective of the indication for the +examination. + + + + + +(a) + + + + + + + + + + + +(b) + + + + + + + + + + + +(c) + + + + + + + +Figure 62.40 Carcinoma in situ showing the varied presentations: (a) occult form; (b) erythroplakia; (c) leukoplakia. The right-hand pic-tures in (a) and (b) demonstrate the use of vital staining with methy-lene blue. + + + + +Johann Friedrich Horner, 1831–1886, Professor of Ophthalmology, Zurich, Switzerland, described this syndrome in 1869. +PART 11 | ABDOMINAL 1088 CHAPTER 62 The oesophagus + + + + + + + + + + + + + + + + + + +Figure 62.41 Endoscopic appearances of a midoesophageal squa-mous cell carcinoma. + +is that only the mucosal surface can be studied and biopsied. Other investigations are therefore usually required to define the extent of local or distant spread. The improved image res-olution of modern endoscopes and novel techniques involv-ing magnification and the use of dyes to enhance surface +detail may lead to more early lesions being recognised. + +GENERAL ASSESSMENT AND STAGING +Once the initial diagnosis of a malignant oesophageal neo-plasm has been made, patients should be assessed first in terms of their general health and fitness for potential therapies. Their preferences should also be considered. Most potentially curative therapies include radical surgery, although definitive chemoradiotherapy is an alternative in squamous cell car-cinoma. Patients who are unfit for, or who do not wish to contemplate, radical treatments should not be investigated further, but should be diverted to appropriate palliative ther-apies, depending on the symptoms and current quality of life. Only those patients suitable for potentially curative therapies should proceed to staging investigations to rule out haematog-enous spread (CT scan) and then to assess locoregional stage (endoscopic ultrasonography [EUS] ± laparoscopy). This will distinguish between early (T1/T2, N0) and advanced lesions (T3/T4, N1) and indicate whether surgery alone or multimodal therapy is most appropriate. Where attempted cure is deemed possible, the aim should be to provide the best chance of cure while minimising procedural risks. In general, surgery alone should be reserved for patients with early dis-ease, and multimodal therapy should be used in patients with locally advanced disease, in whom the chance of cure by sur-gery alone is small (generally <20%). +The most widely used pathological staging system is that of the World Health Organization (tumour–nodes–metastasis TNM). +Table 62.2 shows the TNM system for oesophageal can-cer. Similar to all pathological systems, it relies on the nature +and extent of the surgery performed, e.g. performing more + + +TABLE 62.2 TNM staging scheme for oesophageal cancer. + + + + + + + + +Figure 62.42 Beware the differential diagnosis of infection, for what appears to be a tumour. This midoesophageal mass was actually tuberculosis. + +Investigation +Endoscopy is the first-line investigation for most patients. It provides an unrivalled direct view of the oesophageal mucosa and any lesion allowing its site and size to be documented. Cytology and/or histology specimens taken via the endoscope are crucial for accurate diagnosis. The combination of histol-ogy and cytology increases the diagnostic accuracy to more +than 95%. The chief limitation of conventional endoscopy + +Tis High-grade dysplasia +T1 Tumour invading lamina propria or submucosa T2 Tumour invading muscularis propria +T3 Tumour invading beyond muscularis propria +T4a Tumour invading adjacent structures (pleura, pericardium, diaphragm) +T4b Tumour invading adjacent structures (trachea, bone, aorta) N0 No lymph node metastases +N1 Lymph node metastases in 1–2 nodes N2 Lymph nodes metastases in 3–6 nodes +N3 Lymph node metastases in 7 or more lymph nodes M0 No distant metastases +M1 All other distant metastases +Stage 1A: T1N0M0; 1B: T2N0M0; 2A: T3N0M0; 2B: T1/2N0M0; 3A: T4aN0M0, T3N1M0, T1/2N2M0 +Stage 3B: T3N2M0; 3C: T4aN1/2M0, T4bN0–3M0, T1–4N3M0; 4T:1–4N1–3M1 +PART 11 | ABDOMINAL Neoplasms of the oesophagus 1089 + + + +extensive radical surgical lymphadenectomy provides a more accurate assessment of the ‘N’ stage. There is evidence that many patients described as N0 in the past were probably N1, a phenomenon described as stage migration. +Staging information may be gathered before the com-mencement of therapy, during therapy (e.g. at open operation) or following treatment (histology or postmortem examina-tion). The techniques commonly used to provide preopera-tive staging data are described in Figure 62.43, along with a +suggested algorithm. + +BLOOD TESTS +These are of limited value. Blood tests reveal nothing about local invasion or regional lymph node spread and, to date, no reliable tumour marker for oesophageal cancer has been isolated from peripheral blood. The presence of abnormal liver function tests (LFTs) may suggest the presence of liver metastases, but this is generally too insensitive to be diagnos-tic. Many patients with known liver metastases have normal LFTs. At best, abnormal LFTs only reinforce the clinical sus-picion of spread to the liver, and further imaging is usually +required to confirm the diagnosis. + +TRANSCUTANEOUS ULTRASONOGRAPHY +It is difficult to visualise mediastinal structures with transcu-taneous ultrasonography. With the relatively low-frequency sound waves used, good depth of tissue penetration is achieved at the expense of poor image resolution. In addition, the medi-astinal organs are surrounded by bone and air, which renders them largely inaccessible to external ultrasound. The technique is therefore used mainly to assess spread to the liver, the whole +of which can be clearly visualised by standard transcutaneous + +ultrasonography. Haematogenous spread can be more fully assessed by combining ultrasonography with chest radiography, +although this combination is less accurate than CT scanning. + +BRONCHOSCOPY +Many middle- and upper-third oesophageal carcinomas (and therefore usually squamous carcinomas) are sufficiently advanced at the time of diagnosis that the trachea or bron-chi are already involved (Figure 62.44). Bronchoscopy may reveal either impingement or invasion of the main airways in over 30% of new patients with cancers in the upper third of the oesophagus. In some cases, therefore, bronchoscopy alone +can confirm that the tumour is locally unresectable. + +LAPAROSCOPY +This is a useful technique for the diagnosis of intra-abdominal and hepatic metastases. It has the advantage of enabling tissue samples or peritoneal cytology to be obtained and is the only modality reliably able to detect peritoneal tumour seedlings (Figure 62.45). This is particularly important for tumours +arising from the intra-abdominal portion of the oesophagus, + + +Is the patient fit for surgery? + +Yes No Palliation + + +Haematogenous metastases? + +No Yes + + +Figure 62.44 Invasion into the posterior wall of the trachea from an Palliation oesophageal carcinoma. + + +Contiguous organ invasion + +No Yes Palliation + +Peritoneal spread? + +No Yes Palliation + +Lymph node metastases? + +No Yes + + + +Endoscopy for Multimodal therapy mucosal disease +or surgery alone +Figure 62.43 Algorithm for the management of oesophageal cancer. + + + +Figure 62.45 Adenocarcinoma of the cardia. Transcoelomic spread may occur with this type of lesion. +PART 11 | ABDOMINAL 1090 CHAPTER 62 The oesophagus + + + +cardia and where there is a potential communication between a full-thickness tumour and the peritoneal cavity, for instance +where there is a hiatus hernia. + +COMPUTED TOMOGRAPHY +Computed tomography (CT) from the neck to the pelvis with intravenous contrast is the modality most used to identify haematogenous metastases (Figure 62.46). Distant organs are easily seen and metastases within them visualised with high accuracy (94–100%). The normal thoracic oesophagus is easily demonstrated by CT scanning. The mediastinal fat planes are usually clearly imaged in healthy individuals, and any blurring or distortion of these images is a fairly reliable indicator of abnormality. In cachectic patients with dysphagia and malnutrition, the mediastinal fat plane may be virtually absent, making local invasion difficult to assess. Thin-slice CT permits structures such as lymph nodes to be adequately imaged, down to a minimum diameter of about 5mm. Smaller nodes cannot be reliably visualised, and it is not possible to distinguish between enlarged lymph nodes that have reactive changes only and metastatic nodes. Similarly, micrometasta- +ses within normal-sized nodes cannot be detected. + +MAGNETIC RESONANCE IMAGING +Magnetic resonance imaging (MRI) does not expose the patient to ionising radiation and needs no intravascular contrast medium, although intraoesophageal air or contrast media may help to assess wall thickness. Distant metastases to organs such as the liver are usually reliably identified by MRI but, at the moment, there do not seem to be additional +benefits over CT. + +ENDOSCOPIC ULTRASONOGRAPHY +After haematogenous spread, the two principal prognostic factors for oesophageal cancer are the depth of tumour pene-tration through the oesophageal wall and regional lymph +node spread. Although CT will detect distant metastasis, its + + + + + + + + + + + + + + + + + + + +Figure 62.46 Computed tomography scan demonstrating liver metastases. + +limited axial resolution precludes a reliable assessment of both the depth of wall penetration and lymph node involvement. Endoscopic ultrasonography (EUS) can determine the depth of spread of a malignant tumour through the oesophageal wall (T1–3), the invasion of adjacent organs (T4) and metastasis to lymph nodes (N0 or N1) (Figures 62.47–62.49). It can also detect contiguous spread downward into the cardia and more distant metastases to the left lobe of the liver. +EUS visualises the oesophageal wall as a multilayered structure. The layers represent ultrasound interfaces rather than true anatomical layers, but there is close enough cor-relation to allow accurate assessment of the depth of invasion through the oesophageal wall. Structures smaller than 5mm can be clearly seen, enabling very small nodes to be imaged. The EUS image morphology of such structures provides an additional means of distinguishing malignant from reactive or benign lymph nodes. For submucosal lesions, EUS can demonstrate the wall layer of origin of a lesion, suggesting the +likely histological type. + + + + + + + + + + + + + + + +Figure 62.47 Endosonography demonstrating an ‘early’ tumour. Note the preservation of the outer dark wall layer that represents the muscle coat. + + + + + + + + + + + + + + + + + + +Figure 62.48 Endosonography demonstrating an ‘advanced’ local tumour. Note the breach of the outer white line that represents the interface between the oesophageal wall and the mediastinum. +PART 11 | ABDOMINAL Neoplasms of the oesophagus 1091 + + + + + + + + + + + + +Figure 62.50 Positron emission tomography/computed tomogra-phy demonstrating a primary tumour and a distant metastatic node. + + + + +Figure 62.49 Endosonography demonstrating malignant nodes. These are usually large, hypoechoic and round compared with normal nodes. + + +Narrow EUS instruments are available for insertion over a guidewire to minimise the risk of technical failure, and lin-ear array echoendoscopes can be used to biopsy lesions that might signify incurability outside the wall of the GI tract which might be considered to lie outside the proposed field +of resection (usually lymph nodes outside a standard field). + +Positron emission tomography +with CT +Positron emission tomography (PET) with CT (PET-CT) in the context of cancer staging relies on the generally high metabolic activity (particularly in the glycolytic pathway) of tumours compared with normal tissues. The patient is given a small dose of the radiopharmaceutical agent [18F]fluorode-oxyglucose (FDG). This enters cells and is phosphorylated. FDG 6-phosphate cannot be metabolised further and, as it is a highly polar molecule, it cannot easily diffuse back out of the cell. After intravenous injection of FDG, it contin-uously accumulates in metabolically active cells. Primary oesophageal cancers are usually sufficiently active to be easily visible, and spatial resolution of positive PET areas occurs down to about 5–8 mm. When used in isolation, there are problems with the anatomical location of these areas. This has been significantly improved by combin-ing PET with CT (Figure 62.50), which covers the whole body. The main value of PET-CT is the identification of distant metastatic disease not seen on CT alone. Although there are wide variations between centres, a change in stage is frequently reported in around 15% of patients, but the timing of the scan in relation to the last date of treatment is important and probably accounts for differences in per-ceived effect. It has also been suggested that a reduction in PET activity after chemotherapy might be a way of predict-ing ‘responders’ to this approach, although the persistence of PET activity after a neoadjuvant treatment seems to be a +poor prognostic sign. + +Treatment of malignant tumours PRINCIPLES +At the time of diagnosis, around two-thirds of all patients with oesophageal cancer will already have incurable disease. The aim of palliative treatment is to overcome debilitating or distressing symptoms while maintaining the best quality of life possible for the patient. Some patients do not require spe-cific therapeutic interventions, but do need supportive care and appropriate liaison with community nursing and hospice care services. +As dysphagia is the predominant symptom in advanced oesophageal cancer, the principal aim of palliation is to restore adequate swallowing. A variety of methods is avail-able and, given the short life expectancy of most patients, it is important that the choice of treatment should be tailored to each individual. Tumour location and endoscopic appearance are important in this regard, as is the general condition of the patient. +Once oesophageal neoplasms reach the submucosal layer of the oesophagus, the tumour has access to the lymphatic system, meaning that, even at this early local stage, there is an incidence of nodal positivity for both squamous cell car-cinoma and adenocarcinomas of about 20%. The principle of oesophagectomy is to deal adequately with the local tumour in order to minimise the risk of local recurrence and achieve an adequate lymphadenectomy to reduce the risk of staging error. Although studies in Japan would indicate that more extensive lymphadenectomy is associated with better survival, this may simply reflect more accurate staging. A number of studies support the view that the proximal extent of resection should ideally be 10cm above the macroscopic tumour and 5cm distal. When such a margin cannot be achieved prox-imally, particularly with squamous cell carcinoma, there is evidence that postoperative radiotherapy can minimise local recurrence, although this does not improve survival. +Adenocarcinoma commonly involves the gastric cardia and may therefore extend into the fundus or down the lesser curve. Some degree of gastric excision is essential in order to achieve adequate local clearance and accomplish an appro-priate lymphadenectomy. Excision of contiguous structures, such as the crura, diaphragm and mediastinal pleura, needs to be considered as a method of creating negative resection +margins. +PART 11 | ABDOMINAL 1092 CHAPTER 62 The oesophagus + + + +The rarity of intramucosal cancer in symptomatic patients means that there are no randomised studies to compare dif-ferent approaches to this type of very early disease. EMR for these apparently early (T1a) lesions has become increasingly popular, providing either a cure or at least sufficient histo-logical information on which to base a further management strategy. +Surgery alone is best suited to patients with disease con-fined to the oesophagus (T1b, T2) without nodal metastasis (N0). The problem here is ensuring that patients are truly node negative. Despite careful preoperative investigation, these patients are not easily identified and so there is an argu-ment for neoadjuvant therapy followed by surgery. A cure rate for the patient who is truly node negative having surgery alone of between 50% and 80% needs to be balanced against adverse events related to neoadjuvant therapy, the likelihood of response and the potential impact on survival. Patients with more advanced stages of disease require either multi-modal approaches or entry into appropriate trials. +It is essential that oesophagectomy should be performed with a low hospital mortality and complication rate. Patient selection, volume and experience of the surgical team are all important. Preoperative risk analysis has shown that this can play a major part in reducing hospital mortality. There are really no circumstances in the western world in which surgery should be undertaken if it is not part of an overall treatment +plan aimed at cure. + + +Summary box 62.12 + +Treatment of carcinoma of the oesophagus +●● Radical oesophagectomy is the most important aspect of curative treatment +●● Neoadjuvant treatments before surgery may improve survival in a proportion of patients +●● Chemoradiotherapy alone may cure selected patients, particularly those with squamous cell cancers +●● Useful palliation may be achieved by chemo-/radiotherapy or endoscopic treatments + + +Treatments with curative intent SURGERY +Histological tumour type, location and the extent of the pro-posed lymphadenectomy all influence the surgical approach. This is largely an issue of surgical preference, although it should be recognised that a left thoracoabdominal approach is limited proximally by the aortic arch and should be avoided when the primary tumour is at or above this level. Similarly, transhiatal oesophagectomy is unsuitable for most patients with squamous cell carcinoma because a complete mediasti-nal lymphadenectomy is not easily achieved by this approach. The most widely practised approach in the west is the two- +phase Ivor Lewis (sometimes called Lewis–Tanner) operation + +(Figure 62.51), with an initial laparotomy and construction of a gastric tube, followed by a right thoracotomy to excise the tumour and create an oesophagogastric anastomosis. The closer this is placed to the apex of the thoracic cavity, the fewer problems there are with reflux disease. Three-phase oesophagectomy (McKeown) may be more appropriate for more proximal tumours in order to achieve better longitudi-nal clearance, although the additional distance gained is less than many surgeons believe. A third cervical incision also permits lymphadenectomy in this region. +The extent of lymphadenectomy is highly controversial. For squamous cell carcinoma, because a higher proportion of patients will have middle- and upper-third tumours in the thoracic oesophagus, the rationale behind a three-phase oper-ation with three-field lymphadenectomy is more understand-able, even though this approach has not been widely adopted in the west. For adenocarcinoma, the incidence of metastases in the neck is relatively low in the context of patients who would otherwise be curable. For this reason, two-phase oper-ations with two-field lymphadenectomy seem the most log-ical operations. Although two-field lymphadenectomy does not substantially increase surgical morbidity or mortality, the same cannot be said for more extended operations. +Minimal access techniques, pioneered in Australia by Gotley and Smithers and in North America by Luketich, have enjoyed increasing popularity, often combined with enhanced recovery programmes after surgery (ERAS). Hybrid (where at least one phase is performed as open surgery), total minimally invasive and robotic oesophagectomy all have their advocates, although there are no high-quality comparative studies and +little evidence to indicate clear superiority of one approach + + + + + + + + +(a) + + + + + + + + +(b) +Figure 62.51 The two usual approaches for surgery of the oesoph-agus are (a) the thoracoabdominal, which opens the abdominal and thoracic cavities together, and (b) the two-stage Ivor Lewis approach, in which the abdomen is opened first, closed and then the thoracot-omy is performed. In the McKeown operation, a third incision in the neck is made to complete the cervical anastomosis. + + +Ivor Lewis, 1895–1982, surgeon, the North Middlesex Hospital, London, UK, and later at Rhyl, North Wales, UK. Norman Cecil Tanner, 1906–1982, surgeon, Charing Cross Hospital, London, UK. +Kenneth Charles McKeown, 1912–1995, surgeon, Darlington Memorial Hospital, Darlington, County Durham, UK. +PART 11 | ABDOMINAL Neoplasms of the oesophagus 1093 + + + +over the others. In experienced hands, the open operation can be reproduced by less invasive approaches, without signif-icant compromise. Many groups have reported similar lymph node yields and rates of resection margin positivity with open and minimal access approaches. As yet, benefits seem to be confined to reduced wound pain and the absence of specific complications associated with long incisions. +Although many centres have reduced hospital mortality to low single figures after oesophagectomy, the complication rate remains high. At least a third of all patients will develop some significant complication after surgery. The most com-mon of these is respiratory, followed by anastomotic leakage, chylothorax and injury to the recurrent laryngeal nerves. The most common late problem is benign anastomotic stricture, which seems to be higher with cervical rather than with intrathoracic anastomoses, although the problem is usually easily dealt with by endoscopic dilatation. +Lesions of the cardia that do not involve the oesophagus to any significant extent may be dealt with by extended total gastrectomy to include the distal oesophagus, or by proximal +gastrectomy and distal oesophagectomy. + + +Summary box 62.13 + +Oesophagogastric surgery +●● Beware of satellite nodules proximal to the primary lesion +●● Carefully preserve the blood supply of the stomach, both venous and arterial +●● Right thoracic approach gives easy access to the oesophagus + + +TWO-PHASE OESOPHAGECTOMY (ABDOMEN AND RIGHT CHEST, IVOR LEWIS) +Mobilisation of the stomach must be done with care because it is essential to have a tension-free, well-vascularised stom-ach for transposition. The left gastric, short gastric and left gastroepiploic arteries are all divided. The viability of the transposed stomach mainly depends on the right gastroepip-loic and, to a lesser extent, the right gastric vessels. It should be noted that venous drainage is as important as arterial sup-ply, and it is essential to perform an accurate anatomical dis-section that preserves the right gastroepiploic vein as well as the artery. The stomach is divided to remove the cardia and the upper part of the lesser curve, including the whole of the left gastric artery and its associated lymph nodes. +The approach to the oesophagus through the right chest is straightforward, providing excellent access to the medias-tinum, the thoracic inlet and the hiatus. The azygos vein is divided, and the whole of the intrathoracic oesophagus can be mobilised along with the thoracic duct (which is ligated by most surgeons) and the mediastinal lymph nodes. The oesophagus is divided just below the thoracic inlet. As most +lesions are in the lower or middle third, this usually gives + +adequate proximal clearance of at least 5cm. Carcinomas of the upper thoracic oesophagus are almost always incurable at the time of diagnosis, and invasion of the trachea is common. If one of these lesions is resectable, it is essential to use an incision in the neck (McKeown or three-phase operation) and to resect more of the oesophagus than is customary in the operation of subtotal oesophagectomy. +Oesophagogastric anastomosis may be performed equally well by hand or stapler. Both methods require attention to detail. In experienced hands, clinical anastomotic leakage should be less than 10%. Most surgeons still prefer to keep patients nil by mouth for 5–7 days. Most centres have aban-doned the use of routine contrast swallows in patients who are clinically well. Conversely, aggressive investigation of a sus-pected leak is mandatory for any unexplained fever or clinical event. This should involve early endoscopy, which has been shown to be safe and is the most reliable method for identify-ing necrosis in any part of the replacement conduit and/or CT scan to resolve the situation adequately. +Postoperative nutritional support remains controversial. There is general agreement that parenteral feeding is associ-ated with more nosocomial infection, including pneumonia, than enteral feeding. It is also expensive. The simplicity of placing a modern feeding jejunostomy device at the time of +resection means that this is in routine use in many centres. + +TRANSHIATAL OESOPHAGECTOMY (WITHOUT THORACOTOMY) +This approach was popularised for cancer by Orringer, adapt-ing a technique developed in Brazil by Pinotti for the removal of chagasic megaoesophagus (see the section on achalasia, p. 1095). The stomach is mobilised through a midline abdom-inal incision or by laparoscopy, and the cervical oesophagus is mobilised through an incision in the neck. The diaphragm is then opened from the abdomen, and the posterior medi-astinum is entered. The lower oesophagus and the tumour are mobilised under direct vision, and the upper oesophagus is mobilised by blunt dissection. This approach can provide an adequate removal of the tumour and lymph nodes in the lower mediastinum, but it is not possible to remove the nodes in the middle or upper mediastinum. It may be a useful proce-dure for lesions of the lower oesophagus, but is hazardous for a middle third lesion that may be adherent to the bronchus or +to the azygos vein. + +NEOADJUVANT TREATMENTS WITH SURGERY +Apart from the earliest stages of disease, surgery alone pro-duces relatively few cures in either squamous cell carcinoma or adenocarcinoma patients. This led to a number of trials throughout the 1980s and 1990s to investigate the value of chemotherapy and surgery or chemoradiotherapy and sur-gery compared with surgery alone. Some studies relate only to squamous cell cancer, and many are open to criticism on +the grounds of trial design or patient numbers. Nevertheless, + + + +Mark Burton Orringer, surgeon, Ann Arbor, MI, USA. Walter Pinotti, Professor of Surgery, Sao Paulo, Brazil. +Carlos Justiniano Ribeiro Chagas, 1879–1934, Director of the Oswaldo Cruz Institute and Professor of Tropical Medicine, the University of Rio de Janeiro, Brazil. +PART 11 | ABDOMINAL 1094 CHAPTER 62 The oesophagus + + + +positive results in favour of neoadjuvant therapy for adenocarcinoma in two large studies as well as a limited meta-analysis indicated that it was no longer appropriate to consider surgery alone as the ‘gold standard’ treatment for most patients who are surgical candidates with adenocarci-noma. The Dutch trial (CROSS) that compared chemora-diotherapy and surgery versus surgery alone has provided the most convincing evidence so far of survival benefit for squa-mous cell carcinoma. The same study was initially unable to show a survival benefit for adenocarcinomas, although subse-quent analyses again suggested survival benefit of similar mag- +nitude to that seen with chemotherapy and surgery. + +GASTRO-OESOPHAGEAL REFLUX FOLLOWING OESOPHAGOGASTRIC RESECTION +Gastro-oesophageal reflux may be a major problem after any operation that involves resecting the cardia. Reflux may pres-ent with the typical symptoms of GORD or with a peptic stricture at the site of the anastomosis. However, the presen-tation may be different with a miserable patient who fails to thrive after the operation and who is then suspected of hav-ing recurrent cancer. This atypical presentation is particularly common after total gastrectomy with an inadequate recon- +struction that allows bile reflux. + + +Summary box 62.14 + +Postoesophagectomy +●● Reflux may be a problem after resection ●● Symptoms may be atypical +●● Reflux may be limited or avoided by subtotal oesophagectomy and gastric transposition high in the chest + + +Non-surgical treatments +Radiotherapy alone was widely used as a single-modality treat-ment for squamous cell carcinoma of the oesophagus until the late 1970s. The 5-year survival overall rate was 6%. As a result, multimodal approaches were adopted throughout the 1980s, initial trials indicating that similar long-term survival rates could be obtained with surgery. Subsequent randomised studies, essentially confined to patients with squamous cell carcinoma, have indicated significant survival advantages with chemoradiotherapy over radiotherapy alone. Although it is clear that chemoradiotherapy does offer a prospect of cure for patients who may not be fit for surgery, particularly in squamous cell carcinoma, the high rate of locoregional fail-ure has meant that surgery remains the mainstay of attempted curative treatments for both adenocarcinoma and squamous cell carcinoma in patients who have potentially resectable disease and are fit for oesophagectomy. In most western series, this represents about a third of patients with adenocarcinoma and a slightly lower percentage of patients with squamous cell carcinoma. There has been no high-quality randomised +comparison of the results of definitive radiotherapy versus + +chemoradiotherapy and surgical resection, and it is therefore impossible to make dogmatic statements about the relative +merits of each form of treatment. + +Summary box 62.15 + +Alternative therapeutic approaches +●● Chemoradiotherapy may be a useful alternative to surgery, especially in unfit patients + +Palliative treatment +Surgical resection and external beam radiotherapy may be used for palliation, but are not suitable when the expected survival is short, because most of the remainder of life will be spent recovering from the ‘treatment’. Surgical bypass is likewise too major a procedure for use in a patient with lim-ited life expectancy. A variety of relatively simple methods of palliation is now available that will produce worthwhile relief +of dysphagia with minimal disturbance to the patient. + +Summary box 62.16 + +Palliation +●● Palliation should be simple and effective + +Intubation has been used for many years after the inven-tion of the Souttar tube, which was made of coiled silver wire. A variety of rigid plastic or rubber tubes had been developed for placement under endoscopic and/or radiological control. The technology of intubation has now moved on with the development of various types of expanding metal stent (Fig-ure 62.52). These are also inserted under radiographic or endoscopic control. The stent is collapsed during insertion and released when it is in the correct position. Expanding stents produce a wider lumen for swallowing than rigid tubes. More importantly, it is not necessary to dilate the oesophagus to beyond 8 mm to insert the unexpanded stent through the tumour, so there is a lower risk of injury to the oesophagus. +Endoscopic laser treatment may be used to core a channel through the tumour. It is based on thermal tumour destruc-tion. It produces a worthwhile improvement in swallowing, but has the disadvantage that it has to be repeated every few weeks. Lasers may also be used to unblock a stent that has +become occluded by tumour overgrowth. Other endoscopic + + + + + + + + + +Figure 62.52 Expanding metal stents, covered and uncovered. + + + +Sir Henry Sessions Souttar, 1875–1964, surgeon, the London Hospital, London, UK. +PART 11 | ABDOMINAL +Motility disorders and diverticula 1095 + + + +methods include bipolar diathermy, argon-beam plasma coag-ulation and alcohol injection. +Brachytherapy is a method of delivering intraluminal radiation with a short penetration distance (hence the pre-fix ‘brachy’) to a tumour. An introduction system is inserted through the tumour, and the treatment is then delivered in a single session lasting approximately 20 minutes. The equip-ment is expensive to purchase, but running costs are low. +Although the above methods are suitable for patients with very advanced disease, elderly people and those with significant comorbidities that would make more aggressive strategies inappropriate, an increasing proportion of patients (particularly with adenocarcinoma) are being treated with platinum-based chemotherapy. In general, this leads to only a modest prolongation of survival but a better quality of life +than in those receiving an endoscopic treatment alone. + +Malignant tracheo-oesophageal +fistula +Malignant tracheo-oesophageal fistula is a sign of incurable disease. Some have advocated surgical bypass and oesopha-geal exclusion, but this is a major procedure. An expanding +metal stent is probably the best treatment. + +Post-cricoid carcinoma +Post-cricoid carcinoma is considered in Chapter 47. + +MOTILITY DISORDERS AND +DIVERTICULA +Oesophageal motility disorders +A motility disorder can be readily understood when a patient has dysphagia in the absence of a stricture, and a barium-impregnated food bolus is seen to stick in the oesophagus. If this can be correlated with a specific abnormality on oesophageal manometry, accepting that this is the cause of the patient’s symptoms may be straightforward. Unfortunately, this is often not the case. Pain, with or without a swallowing problem, is frequently the dominant symptom, and patients often undergo extensive hospital investigation before the oesophagus is considered as a source of symptoms. Symptoms are often intermittent, and the correlation between symptoms and test ‘abnormalities’ is poor. Confirmation of a specific motility disturbance is made by high-resolution manometry. Much harm may be done by inappropriate enthusiastic surgery for ill-defined conditions. It should also be remembered that oesophageal dysmotility may be only a feature of a general +disturbance in GI function. + + +Summary box 62.17 + +Oesophageal motility disorders +●● May be part of a more diffuse GI motility problem +●● May be associated with GORD + +Oesophageal motility disorders are currently best classified by incorporation of the Chicago classification developed for use with high-resolution manometry as shown in Tables 62.3 +and 62.4. + +Functional pain and the oesophagus +Pain that is assumed to arise from dysfunction of the GI tract may reflect abnormal motor activity, abnormal percep-tion or a combination of the two. There is evidence that all three exist. Very high-pressure uncoordinated contractions (‘spasm’) have been shown to correlate with pain. Distension of a balloon in the oesophagus indicates that some patients have a low threshold for the sensation of pain (visceral hyper-sensitivity), and this itself may reflect local or central neuro-nal dysfunction. In practice, the difficulty is in understanding the relative contributions of these elements, so that a logical +treatment might follow. + +Achalasia +Pathology and aetiology +Achalasia (Greek ‘failure to relax’) is uncommon, but mer-its prominence because it is reasonably understood and responds to treatment. It is due to loss of the ganglion cells in the myenteric (Auerbach’s) plexus, the cause of which is unknown. In South America, chronic infection with the parasite Trypanosoma cruzi causes Chagas’ disease, which +has marked clinical similarities to achalasia. Achalasia differs + + + + +TABLE 62.3 General classification of oesophageal motility disorders. +Disorders of the pharyngo-oesophageal junction +Neurological – stroke, motor neuron disease, multiple sclerosis, Parkinson’s disease +Myogenic – myasthenia, muscular dystrophy Pharyngo-oesophageal (Zenker’s) diverticulum Disorders of the body of the oesophagus Diffuse oesophageal spasm +Nutcracker oesophagus +Autoimmune disorders – especially systemic sclerosis (CREST) Reflux associated +Idiopathic Allergic +Eosinophilic oesophagitis +Non-specific oesophageal dysmotility Disorders of the lower oesophageal sphincter Achalasia +Incompetent lower sphincter (i.e. GORD) +CREST, calcinosis, Raynaud’s syndrome, (o)esophageal motility disorders, sclerodactyly and telangiectasia. +GORD, gastro-oesophageal reflux disease. +PART 11 | ABDOMINAL 1096 CHAPTER 62 The oesophagus + + +TABLE 62.4 The Chicago Classification of oesophageal motility v3.0. + +Achalasia and EGJ outflow obstruction Type I achalasia (classic achalasia) + + +Type II achalasia (with oesophageal compression) + + + +Type III achalasia (spastic achalasia) + + +EGJ outflow obstruction + +Major disorders of peristalsis Absent contractility + + + +Distal oesophageal spasm + +Hypercontractile oesophagus (jackhammer) + +Minor disorders of peristalsis Ineffective oesophageal motility (IEM) + + +Fragmented peristalsis +Normal oesophageal motility + +Criteria +Elevated median IRP (>15 mmHg), 100% failed peristalsis (DCI <100 mmHg·s·cm) Premature contractions with DCI values <450 mmHg·s·cm satisfy criteria for failed peristalsis +Elevated median IRP (>15 mmHg), 100% failed peristalsis, panoesophageal pressurisation with ≥20% of swallows +Contractions may be masked by oesophageal pressurisation and DCI should not be calculated +Elevated median IRP (>15 mmHg), no normal peristalsis, premature (spastic) contractions with DCI >450 mmHg·s·cm with ≥20% of swallows +May be mixed with panoesophageal pressurisation +Elevated median IRP (>15 mmHg), sufficient evidence of peristalsis such that criteria for types I–III achalasia are not met +(Not encountered in normal individuals) +Normal median IRP, 100% failed peristalsis +Achalasia should be considered when IRP values are borderline and when there is evidence of oesophageal pressurisation +Premature contractions with DCI values <450 mmHg·s·cm meet criteria for failed peristalsis +Normal median IRP, ≥20% premature contractions with DCI >450 mmHg·s·cm. Some normal peristalsis may be present +At least two swallows with DCI >8000 mmHg·s·cm Hypercontractility may involve, or even be localised to, the LES +(Characterised by contractile vigour and contraction pattern) +≥50% ineffective swallows +Ineffective swallows can be failed or weak (DCI <450 mmHg·s·cm) +Multiple repetitive swallow assessment may be helpful in determining peristaltic reserve +≥50% fragmented contractions with DCI >450 mmHg·s·cm +Not fulfilling any of the above classifications + +DCI, distal contractile integral (mmHg·s·cm); EGJ, o(e)sophaogastric junction; IRP, integrated relaxation pressure (mmHg); LES, lower o(e)sophageal sphincter. (From Kahrilas PJ et al. The Chicago Classification of esophageal motility disorders v3.0. Neurogastroenterology and Motility 2015; 27(2): 160–74, with kind permission, John Wiley & Sons.) + + +from Hirschsprung’s disease of the colon because the dilated oesophagus usually contains few ganglion cells, whereas the dilated colon contains normal ganglion cells proximal to a constricted, aganglionic segment. Histology of muscle speci-mens generally shows a reduction in the number of ganglion cells (and mainly inhibitory neurons) with a variable degree of chronic inflammation. When powerful non-peristaltic con-tractions are still present, perhaps representing an early stage of the disease, inflammation and neural fibrosis may be seen +with normal numbers of ganglion cells. + + + +Summary box 62.18 + +Achalasia +●● Is uncommon +●● Is due to selective loss of inhibitory neurons in the lower oesophagus +●● The causes dysphagia and carcinoma must be excluded +●● Treatment is by either endoscopic dilatation, or endoscopic or surgical myotomy + +The classic physiological abnormalities are a non-relax-ing LOS and absent peristalsis in the body of the oesophagus. The Chicago classification identifies three variants. Although most patients have almost no recognisable contractions (type I), some patients continue to exhibit pressurisation through-out the oesophagus (type 2), whereas in others the oesopha-gus is of normal calibre and exhibits high-pressure contractile (although non-peristaltic) activity (spastic oesophagus, vig-orous achalasia). In some patients, these uncoordinated con-tractions result in pain as much as a sense of food sticking. High-resolution manometry recognises these contraction patterns, which may be important in predicting the outcome of treatment. With time, the oesophagus dilates and contrac-tions disappear, so that the oesophagus empties mainly by the hydrostatic pressure of its contents. This is nearly always incomplete, leaving residual food and fluid. The gas bub-ble in the stomach is frequently absent, as no bolus with its accompanying normal gas passes through the sphincter. The ‘megaoesophagus’ becomes tortuous with a persistent reten-tion oesophagitis due to fermentation of food residues (Figure 62.53), and this may account for the increased incidence of +carcinoma of the oesophagus + + +Harald Hirschsprung, 1831–1916, physician, the Queen Louise Hospital for Children, and Professor of Paediatrics, Copenhagen, Denmark, described congenital megacolon in 1888. +PART 11 | ABDOMINAL +Motility disorders and diverticula 1097 + + +(a) (b) (c) + + + + + + + + + + + + + + + + + + + + + +Figure 62.53 Achalasia of the oesophagus. (a) Barium swallow showing the smooth outline of the stricture, which narrows to a point at its lower end. (b) Tortuosity and sigmoid appearance of the lower oesophagus. (c) Mediastinal shadow due to a large, fluid-filled oesophagus. + + + + + +Summary box 62.19 + +Lower oesophageal stricture +●● Beware pseudoachalasia; look for tumour + + + +Pseudoachalasia is an achalasia-like disorder that is usually produced by adenocarcinoma of the cardia (Figure 62.54), but has also been described in relation to benign tumours at this level. It has been presumed that the inability of the sphincter to relax is linked to the loss of body peristal-sis, although other cancers outside the oesophagus (bronchus, +pancreas) have also been associated with pseudoachalasia. +Clinical features +The disease is most common in middle life, but can occur at any age. It typically presents with dysphagia, although pain (often mistaken for reflux) is common in the early stages. Patients often present late and, having had relatively mild symptoms, remain untreated for many years. Regurgitation is frequent, and there may be overspill into the trachea, espe- +cially at night. +Diagnosis +Achalasia may be suspected at endoscopy by finding a tight cardia and food residue in the oesophagus. Barium radiology may show hold-up in the distal oesophagus, abnormal con-tractions in the oesophageal body and a tapering stricture in +the distal oesophagus, often described as a ‘bird’s beak’ (see + + + + + + + + + + + + + + + + + + + + + + +Figure 62.54 Almost achalasia, but note the irregularity of the taper which indicates carcinoma of the cardia. +, +PART 11 | ABDOMINAL 1098 CHAPTER 62 The oesophagus + + +Swallow UOS relaxation Swallow + + + + + + + + + + + + + + + + +No LOS relaxation Common cavity pressure rise + +No LOS relaxation on e-sleeve (bold brown line plot) + + +Figure 62.55 High-resolution oesophageal manometry from a patient presenting with dysphagia and regurgitation. The swallow is followed by a common cavity rise in oesophageal pressure indicating filling. Lower oesophageal sphincter (LOS) relaxation is absent and there is a positive oesophagogastric pressure gradient. Upper oesophageal sphincter (UOS) relaxation shortly after the swallow was related to regurgitation of oesophageal contents. + + + + + + +Figure 62.53). The gastric gas bubble is usually absent. These typical features of well-developed achalasia are often absent, and endoscopy and radiology can be normal. A firm diagno-sis is established by high-resolution oesophageal manometry. Classically, the LOS does not relax completely on swallowing, there is no peristalsis and there is a raised resting pressure in the oesophagus (Figure 62.55). The LOS pressure may be +elevated, but is often normal. +Treatment +Alone among motility disorders, achalasia responds well to treatment. The two main methods are forceful dilatation of the cardia and laparoscopic cardiomyotomy (conventionally performed together with a partial fundoplication), although in recent years there has been growing interest in the use of peroral endoscopic myotomy (POEM). Comparative studies between pneumatic dilatation and surgical myotomy suggest equivalence in terms of safety effectiveness and cost when +, +considered over a number of years. + +PNEUMATIC DILATATION +This involves stretching the cardia with a balloon to disrupt the muscle and render it less competent. The treatment was first described by Plummer. Many varieties of balloon have been used but, nowadays, plastic balloons with a precisely controlled external diameter are used. If the pressure in the balloon is too high, the balloon is designed to split along its +length rather than expanding further. Balloons of 30–40 mm + + + + + + +Figure 62.56 Balloon dilator for the treatment of achalasia by force-ful dilatation. + +in diameter are available and are inserted over a guidewire (Figure 62.56). Perforation is the major complication. With a 30-mm balloon, the incidence of perforation should be less than 0.5%. The risk of perforation increases with bigger bal-loons, and they should be used cautiously for progressive dil-atation over a period of weeks. Forceful dilatation is curative in 75–85% of cases. The results are best in patients aged over +45 years. + + +Summary box 62.20 + +Achalasia +●● Beware perforation due to dilatation of achalasia ●● Beware postoperative reflux + + +HELLER’S MYOTOMY +This involves cutting the muscle of the lower oesopha- +gus and cardia (Figure 62.57). The major complication is + + + +Henry Stanley Plummer, 1874–1937, physician, the Mayo Clinic, Rochester, MN, USA. Ernst Heller, 1877–1964, surgeon, St George’s Krankenhaus, Leipzig, Germany +. +PART 11 | ABDOMINAL +Motility disorders and diverticula 1099 + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.57 Heller’s myotomy: the incision should not go too far on to the stomach. The lateral extent must enable the mucosa to pout out, to prevent the edges healing together. + + +gastro-oesophageal reflux, and most surgeons therefore add a partial anterior fundoplication (Heller–Dor operation). The procedure is ideally suited to a minimal access laparoscopic approach, and most surgeons use intraoperative endoscopy to judge the extent of the myotomy and to ensure that the nar-row segment is abolished. +It is successful in more than 90% of cases and may be used +after failed dilatation. + +ENDOSCOPIC MYOTOMY +This was introduced in Japan in 2009. It involves the creation of a submucosal tunnel in the midoesophagus that can be developed distally, allowing a long myotomy to be performed. The oesophageal mucosal defect is clipped shut at the end. Many cohort studies indicate that the procedure can be per-formed safely and early clinical results imply equivalence to +other approaches. + +BOTULINUM TOXIN +This is done by endoscopic injection into the LOS. It acts by interfering with cholinergic excitatory neural activity at the LOS. The effect is not permanent, and the injection usually has to be repeated after a few months. For this reason, its use +is restricted to elderly patients with other comorbidities. + +DRUGS +Drugs such as calcium channel antagonists have been used but are ineffective for long-term use. However, sublingual nifedipine may be useful for transient relief of symptoms if +definitive treatment is postponed. + +Other oesophageal motility +disorders +Disorders of the pharyngo-oesophageal junction +With the exception of Zenker’s diverticulum (see below), most patients with oropharyngeal dysphagia have generalised neu-rological or muscular disorders with pharyngeal involvement. A small number of patients who have sustained a cerebro-vascular accident benefit from myotomy of cricopharyngeus to alleviate pooling of saliva and nocturnal aspiration, but they should have good deglutition and phonation before this is performed. The operation is also effective in patients with +oculopharyngeal muscular dystrophy. +Disorders of the body of the oesophagus +The older terms, ‘diffuse oesophageal spasm’ and ‘jackhammer (nutcracker) oesophagus’, have been replaced in the Chicago classification. There are, nevertheless, patients with incoordi-nate contractions of the oesophagus who experience dyspha-gia and/or chest pain. The condition may be dramatic, with spastic pressures on manometry of 400–500mmHg, marked hypertrophy of the circular muscle and a corkscrew oesoph-agus on barium swallow (Figure 62.58). These abnormal contractions are more common in the distal two-thirds of the oesophageal body (Figure 62.59) and this may have some relevance to treatment. Making the diagnosis when chest pain is the only symptom may be difficult. Prolonged oesoph-ageal manometry that correlates episodes of chest pain with manometric abnormalities may establish the diagnosis. +There is no proven pharmacological or endoscopic treat-ment. Calcium channel antagonists, vasodilators and endo-scopic dilatation have only transient effects. Although the +severity and frequency of symptoms may be tolerated by most + + + + + + + + + + + + + + + + + + + + +Figure 62.58 Corkscrew oesophagus in diffuse oesophageal spasm. + + + +Friedrich Albert Zenker, 1825–1898, Professor of Pathology, Dresden, Germany +. +PART 11 | ABDOMINAL 1100 CHAPTER 62 The oesophagus + + +Swallow Swallow + + + + + + + + + + + + + + + + +LOS relaxation Repetitive LOS relaxation segmental spasm +Figure 62.59 High-resolution oesophageal manometry from a patient presenting with dysphagia and chest pain. The swallow is followed by simultaneous, repetitive contractions in the mid-distal smooth muscle oesophagus. Lower oesophageal sphincter (LOS) relaxation is preserved. Note the sequential simultaneous contractions first in the mid- and distal segments of the oesophagus and then in the LOS make it appear as if there is progressive peristalsis on the conventional line plots (dotted arrow). Repetitive contractions are seen clearly on both. + + + + + +patients, sometimes the combination of chest pain and dys-phagia is sufficiently severe that malnutrition begins. In these patients, extended oesophageal myotomy up to the aortic arch may be required. Surgical treatment is more successful in improving dysphagia than chest pain, and caution should be exercised in patients in whom chest pain is the only symptom. Jackhammer or nutcracker oesophagus is a condition with characteristic high-pressure manometric features (see Table 62.4). The correlation of manometric abnormalities with +symptoms remains poor. + +OESOPHAGEAL INVOLVEMENT IN AUTOIMMUNE DISEASE +Oesophageal involvement is mainly seen in systemic scle-rosis, but may be a feature of polymyositis, dermatomyosi-tis, systemic lupus erythematosus, polyarteritis nodosa and rheumatoid disease. Although most involve weak peristalsis, swallowing difficulties may be compounded by pharyngeal problems in the disorders that primarily affect skeletal mus-cle (e.g. polymyositis) or extraoesophageal problems such as involvement of the cricoarytenoid joint in rheumatoid dis-ease or dry mouth in Sjögren’s syndrome. In systemic scle-rosis, smooth muscle atrophy causes hypoperistalsis (Figure 62.60). The LOS is involved, leading to a loss of the antire-flux barrier. A wide range of symptoms can follow from mild to severe dysphagia accompanied by regurgitation and aspir-ation. Reflux can be severe and is exacerbated by weak acid clearance so that strictures can occur. There are no drugs that +specifically correct the motor disorder, and medical treatment + + + + + + + + + + + + + + + + + + + + +Figure 62.60 Advanced scleroderma of the oesophagus. The oesophagus dilates, and the lower oesophageal sphincter is widely incompetent. + + +is mainly directed at minimising reflux-induced damage with PPIs. A small number of patients may require anti-reflux sur-gery. +Eosinophilic oesophagitis is a disorder that occurs in +children and adults either alone or as a manifestation of + + + +Henrik Samuel Conrad Sjögren, 1899–1986, Professor of Ophthalmology, Gothenburg, Sweden. +PART 11 | ABDOMINAL +Motility disorders and diverticula 1101 + + + +eosinophilic gastroenteritis. It is characterised by eosinophilic infiltration of the oesophageal wall, presumably of allergic or idiopathic origin. The most common presenting symptom is dysphagia, and more than half have some history of atopy. The oesophagus often seems narrow and friable on endoscopy and may include mucosal rings. The most important feature is the development of deep ulcers, leading to stricture develop-ment, especially in the proximal oesophagus. The diagnosis is established by endoscopic biopsy. +Elimination diets, and topical and systemic steroids all seem to be helpful in the short term, but there is scant infor-mation on the long-term impact of any particular approach. Immunotherapy directed against interleukin (IL)-5, which has a major role in eosinophil recruitment, seems to be a promising innovative approach. Although endoscopic dil-atation has been recommended, this can create deep ulcers and further scarring, so should be used with caution and only +when the above therapies fail. +Pharyngeal and oesophageal diverticula +Most oesophageal diverticula are pulsion diverticula that develop at a site of weakness as a result of chronic pressure against an obstruction. Symptoms are mostly caused by the underlying disorder unless the diverticulum is particularly +large. Traction diverticula (Figure 62.61) are much less + + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.61 Midoesophageal traction diverticulum with the mouth facing downwards. + +common. They are mostly a consequence of chronic granu-lomatous disease affecting the tracheobronchial lymph nodes due to tuberculosis, atypical mycobacteria or histoplasmosis. Fibrotic healing of the lymph nodes exerts traction on the oesophageal wall and produces a focal outpouching which is usually small and has a conical shape. There may be asso-ciated broncholithiasis, and additional complications may occur, such as aerodigestive fistulation (Figure 62.62) and bleeding. +Zenker’s diverticulum (pharyngeal pouch) is not really an oesophageal diverticulum as it protrudes posteriorly above the cricopharyngeal sphincter through the natural weak point (the dehiscence of Killian) between the oblique and horizontal (cricopharyngeus) fibres of the inferior pharyngeal constrictor (Figures 62.63 and 62.64). The exact mecha-nism that leads to its formation is unknown, but it involves loss of the coordination between pharyngeal contraction and opening of the upper sphincter. When the diverticulum is small, symptoms largely reflect this incoordination with pre-dominantly pharyngeal dysphagia. As the pouch enlarges, it tends to fill with food on eating, and the fundus descends into the mediastinum. This leads to halitosis and oesopha-geal dysphagia. Treatment can be undertaken endoscopically with a linear cutting stapler to divide the septum between +the diverticulum and the upper oesophagus, producing a + + + + + + + + + + + + + + + + + + + + + + + + + + + + +Figure 62.62 Midoesophageal diverticulum with a trachea-oeso-phageal fistula. + + + +Gustav Killian, 1860–1921, Professor of Laryngology at Freiburg, and later at Berlin, Germany. +PART 11 | ABDOMINAL 1102 CHAPTER 62 The oesophagus + + +(a) + + + + + + + + + + + + + + + + + + + + + + + + + + + +(b) + + + + + + + + + + + + + + + + + + +Figure 62.63 The typical appearances of: (a) a small pharyngeal pouch with a prominent cricopharyngeal impression and ‘stream-ing’ of barium, indicating partial obstruction; and (b) a large pouch extending behind the oesophagus towards the thoracic inlet. + + +diverticulo-oesophagostomy, or can be done by open sur-gery involving pouch excision, pouch suspension (diverticu-lopexy) and/or myotomy of cricopharyngeus. All techniques +have good results. + +Richard Schatzki, 1901–1992, American radiologist. + +Figure 62.64 The endoscopic appearance of the mouth of a pharyn-geal pouch posterior to the normal opening (left) of the oesophagus. + +Midoesophageal diverticula are usually small pulsion diverticula of no particular consequence. The underlying motility disorder does not usually require treatment. Some pulsion diverticula may fistulate into the trachea (see Figure 62.62), but this is more common with traction diverticula in granulomatous disease. +Epiphrenic diverticula are pulsion diverticula situated in the lower oesophagus above the diaphragm (Figure 62.65). They may be quite large, but cause surprisingly few symptoms. They again probably reflect some loss of coordination between an incoming pressure wave and appropriate relaxation of the LOS. This needs to be acknowledged in the surgical manage-ment of the patient. The diverticulum, in isolation, should not be assumed to account for a patient’s illness just because it looks dramatic on a radiograph. Large diverticula may be excised, and this should be combined with a myotomy from the site of the diverticulum down to the cardia to relieve +functional obstruction. + +Summary box 62.21 + +Oesophageal diverticula +●● Diverticula are indicators of a motor disorder and not necessarily the cause of symptoms + +Diffuse intramural pseudodiverticulosis is a rare condi-tion in which there are multiple tiny outpouchings from the lumen of the oesophagus. The pseudodiverticula are dilated excretory ducts of oesophageal sebaceous glands. It is ques-tionable whether the condition produces any symptoms in its +own right. + +OTHER NON-NEOPLASTIC +CONDITIONS Schatzki’s ring +Schatzki’s ring is a circular ring in the distal oesophagus +(Figure 62.66), usually at the squamocolumnar junction. The +PART 11 | ABDOMINAL +Other non-neoplastic conditions 1103 + + +(a) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +(b) + +Figure 62.66 Schatzki’s ring, a thin submucosal web completely encircling the whole of the lumen, usually situated at the squamo-columnar junction. + + + +cause is obscure, but there is a strong association with reflux disease. The core of the ring consists of variable amounts of fibrous tissue and cellular infiltrate. Most rings are incidental findings. Some are associated with dysphagia and respond to +dilatation in conjunction with medical antireflux therapy. + + + + + + + + + + + + + + + +Figure 62.65 Epiphrenic diverticulum proximal to the gastro-oesophageal sphincter. (a) Small and asymptomatic; (b) large, symp-tomatic and appearing as a gas-filled bubble on the chest radiograph. + +Oesophageal infections +Bacterial infection of the oesophagus is rare, but fungal and viral infections do occur. They are particularly important in immunocompromised patients. +Oesophagitis due to Candida albicans is relatively common in patients taking steroids (especially transplant recipients) or those undergoing cancer chemotherapy. It may present with dysphagia or odynophagia. There may be visible thrush in the throat. Endoscopy shows numerous white plaques that cannot be moved, unlike food residues (Figure 62.67). Biopsies are diagnostic. In severe cases, a barium swallow may show dramatic mucosal ulceration and irregularity that is surprisingly similar to the appearance of oesophageal vari-ces (Figures 62.68 and 62.69). Treatment is with a topical +antifungal agent. +PART 11 | ABDOMINAL 1104 CHAPTER 62 The oesophagus + + + + + + + + + + + + + + + + + + +Figure 62.67 Endoscopic appearance of oesophageal candidiasis. + + + + + + + + + + +Figure 62.69 Oesophageal varices with smooth outline of the filling defects. + +raised margins, usually in the upper half of the oesophagus. CMV infection may be apparent in graft-versus-host disease following bone marrow transplantation. It has a characteristic endoscopic appearance with a geographical, serpiginous bor- +der. In both cases, endoscopic biopsy is diagnostic. + + + + + + + + + + + + +Figure 62.68 Oesophageal candidiasis with shaggy appearance of mucosal defects. + + +Dysphagia and odynophagia can also be caused by herpes simplex virus and cytomegalovirus (CMV). With the former, there may be a history of a herpetic lesion on the lip some days +earlier, and endoscopy may reveal vesicles or small ulcers with + +Chagas’ disease +This condition is confined to South American countries, but is of interest because oesophageal symptoms occur that are similar to severe achalasia. It is caused by a protozoan, Trypanosoma cruzi, transmitted by an insect vector. Para-sites reach the bloodstream and, after a long latent period, there is damage particularly to cardiac and smooth muscle. Destruction of both Auerbach’s and Meissner’s plexus leads to +acquired megaoesophagus. + +Crohn’s disease +The oesophagus is not commonly affected by symptomatic Crohn’s disease. However, pathological studies indicate that it may be present in 20% of patients without symptoms. Symp-toms are often severe, and a diagnosis of reflux oesophagitis is +usually made on the basis of retrosternal pain and dysphagia. + + + +Burrill Bernard Crohn, 1884–1983, gastroenterologist, Mount Sinai Hospital, New York, NY USA, described regional ileitis in 1932. +, +PART 11 | ABDOMINAL +Further reading 1105 + + + +Endoscopy shows extensive oesophagitis that extends much further proximally than reflux oesophagitis. Biopsies may be diagnostic, but may show only non-specific inflammation. In severe cases, deep sinuses occur, and fistulation has been described. Crohn’s oesophagitis is said to respond poorly to medical treatment and, although balloon dilatation of stric-tures and surgical resection for multiple internal fistulae have both been described, these interventions should be used with +great caution. + +Plummer–Vinson syndrome +This is also called the Paterson–Kelly (or Paterson–Brown Kelly) syndrome or sideropenic dysphagia. The original descriptions are vague and poorly supported by evidence of a coherent syndrome. Dysphagia is said to occur because of the presence of a postcricoid web that is associated with iron deficiency anaemia, glossitis and koilonychia. The classic syndrome is rarely complete. Some patients may have oro-pharyngeal leukoplakia, and this may account for an alleged increased risk of developing hypopharyngeal cancer. +Webs certainly occur in the upper and middle oesophagus, usually without any kind of associated syndrome. They are nearly always thin diaphanous membranes identified coinci-dentally by contrast radiology. Even symptomatic webs that cause a degree of obstruction may be inadvertently ruptured +at endoscopy. Few require formal endoscopic dilatation. + +Vascular abnormalities affecting the +oesophagus +Several congenital vascular anomalies may produce dyspha- +gia by compression of the oesophagus. Classically, this results + +from an aberrant right subclavian artery (arteria lusoria). However, the oesophagus is more commonly compressed by vascular rings, such as a double aortic arch. Dysphagia occurs in only a minority of cases and usually presents early in child-hood, although it can occur in the late teens. Treatment is usually by division of the non-dominant component of +the ring. +In adults, acquired causes include aneurysm of the aorta, diffuse cardiac enlargement and pressure from the left com-mon carotid or vertebral arteries. It is rare that symptom +severity justifies surgical intervention. + +Mediastinal fibrosis +This rare condition can occur alone or together with retro-peritoneal fibrosis. The cause is unknown and, although the major consequences are usually cardiovascular as a result of caval compression, dysphagia can occur. The existence of irreparable cardiovascular problems usually precludes surgical +intervention on the oesophagus. + + +FURTHER READING +Castell DO, Richter JE. The esophagus, 5th edn. Oxford: Wiley-Blackwell, 2011. +Griffin SM, Raimes SA, Shenfine J. Oesophago-gastric surgery, 5th edn. Philadelphia, PA: Saunders Ltd, 2013. +Luketich JD. Master techniques in surgery: esophageal surgery.Philadelphia, PA: Wolters Kluwer, 2014. +Pleskow D, Erim T. Barrett’s esophagus: emerging evidence for improved clinical practice. Cambridge, MA: Academic Press, 2016. +Sharma P, Sampliner RE, Ilson D. Esophageal cancer and Barrett’s oesophagus, 3rd edn. Hoboken, NJ: John Wiley & Sons, 2015. +Takubo K. Pathology of the esophagus. New York: Springer, 2010. + + + + + + + + + + + + + + + + + + + + + +Henry Stanley Plummer, 1874–1936, American internist and endocrinologist. +Porter Paisley Vinson, 1890–1959, Physician, the Mayo Clinic, Rochester, MN, who later practised in Richmond, VA, USA. Adam Brown Kelly, 1865–1941, surgeon, Ear, Nose and Throat Department, the Royal V ctoria Infirmary, Glasgow, UK. Donald Rose Paterson, 1863–1939, surgeon, Ear, Nose and Throat Department, the Royal Infirmary Cardiff, UK. +i +, +Vinson, Kelly and Paterson all described this syndrome independently in 1919. +Bailey & Love +Bailey & Love +63 +Chapter + +Bailey & Love +Bailey & Love + +Bailey & Love +Bailey & Love + + + + + +Stomach and duodenum + + + + + +Learning objectives + +• To understand the gross and microscopic anatomy and pathophysiology of the stomach in relation to disease +• To be able to decide on the most appropriate techniques to use in the investigation of patients with complaints relating to the stomach and duodenum +• To understand the critical importance of gastritis and Helicobacter pylori in upper gastrointestinal disease + +• To be able to investigate and treat peptic ulcer disease and its complications +• To be able to recognise the presentation of gastric cancer and understand the principles involved in its treatment +• To know about the causes of duodenal obstruction and the presentation of duodenal tumours + + + + + + +INTRODUCTION +The function of the stomach is to act as a reservoir for ingested food. It also serves to break down foodstuffs mechanically and commence the processes of digestion before these products +are passed on into the duodenum. + +GROSS ANATOMY OF THE +STOMACH AND DUODENUM Blood supply + + +Splenic artery Left gastric artery +Right gastric artery Hepatic artery +Duodenal cap +Gastroduodenal artery + + +Vasa brevia + + + + + +Splenic artery + + + +Arteries +The stomach has an arterial supply on both lesser and greater curves (Figure 63.1). On the lesser curve, the left gastric artery, a branch of the coeliac axis, forms an anastomotic arcade with the right gastric artery, which arises from the common hepatic artery. Branches of the left gastric artery pass up towards the cardia. The gastroduodenal artery, which is also a branch of the hepatic artery, passes behind the first part of the duodenum, highly relevant with respect to the bleed-ing duodenal ulcer. Here it divides into the superior pancre-aticoduodenal artery and the right gastroepiploic artery. The superior pancreaticoduodenal artery supplies the duodenum and pancreatic head, and forms an anastomosis with the inferior pancreaticoduodenal artery, a branch of the superior mesenteric artery. The right gastroepiploic artery runs along the greater curvature of the stomach, eventually forming an anastomosis with the left gastroepiploic artery, a branch of the splenic artery. This vascular arcade, important for the construction of the gastric conduit in oesophageal resection, +is often variably incomplete. The fundus of the stomach is + + +Superior +pancreatico- Left gastro-duodenal artery epiploic +artery +Right gastroepiploic Gastrocolic +artery (greater) omentum +Figure 63.1 The arterial blood supply of the stomach. + + +supplied by the vasa brevia (or short gastric arteries), which +arise from near the termination of the splenic artery. +Veins +In general, the veins are equivalent to the arteries, those along the lesser curve ending in the portal vein and those on the greater curve joining via the splenic vein. On the lesser curve, the coronary vein is particularly important. It runs up the lesser curve towards the oesophagus and then passes left to right to join the portal vein. This vein becomes markedly +dilated in portal hypertension. +PART 11 | ABDOMINAL Microscopic anatomy of the stomach and duodenum 1107 + + + +Lymphatics +The lymphatics of the stomach are of considerable importance in the surgery of gastric cancer and are therefore described in +detail in that section. + +Nerves +As with the entire gastrointestinal tract, the stomach and duodenum possess both intrinsic and extrinsic nerve supplies. The intrinsic nerves exist principally in two plexuses, the myenteric plexus of Auerbach and the submucosal plexus of Meissner. Compared with the rest of the gut, the submucosal plexus of the stomach contains relatively few ganglionic cells, as does the myenteric plexus in the fundus. However, in the antrum the ganglia of the myenteric plexus are well devel-oped. The extrinsic supply is derived mainly from the vagus nerves (CN XI), fibres of which originate in the brainstem. The vagal plexus around the oesophagus condenses into bun-dles that pass through the oesophageal hiatus (Figure 63.2), the posterior bundle being usually identifiable as a large nerve trunk. Vagal fibres are both afferent (sensory) and efferent. The efferent fibres are involved in the receptive relaxation of the stomach and the stimulation of gastric motility, as well as having the well-known secretory function. The sympathetic +supply is derived mainly from the coeliac ganglia. + +MICROSCOPIC ANATOMY OF THE STOMACH AND +DUODENUM +The gastric epithelial cells are mucus producing and are turned over rapidl . In the pyloric part of the stomach, and +y +also the duodenum, mucus-secreting glands are found. Most + + +Anterior + + +of the specialised cells of the stomach (parietal and chief cells) are found in the gastric crypts (Figure 63.3). The stom- +ach also has numerous endocrine cells. + +Parietal cells +These are in the body (acid-secreting portion) of the stomach and line the gastric crypts, being more abundant distally. They are responsible for the production of hydrogen ions to form hydrochloric acid. The hydrogen ions are actively secreted by the proton pump, a hydrogen–potassium-ATPase (Sachs), which exchanges intraluminal potassium for hydrogen ions. The potassium ions enter the lumen of the crypts passively, but the hydrogen ions are pumped against an immense con- +centration gradient (1000000:1). + +Chief cells +These lie principally proximally in the gastric crypts and produce pepsinogen. Two forms of pepsinogen are described: pepsinogen I and pepsinogen II. Both are produced by the chief cell, but pepsinogen I is produced only in the stomach. The ratio between pepsinogens I and II in the serum decreases with gastric atrophy. Pepsinogen is activated in the stomach +to produce the digestive protease, pepsin. + +Endocrine cells +The stomach has numerous endocrine cells, which are critical to its function. In the gastric antrum, the mucosa contains G cells, which produce gastrin. Throughout the body of the stomach, enterochromaffin-like (ECL) cells are abundant and produce histamine, a key factor in driving gastric acid secre- +tion. In addition, there are large numbers of somatostatin- + + +Posterior + + +‘Criminal’ nerve of Grassi +Coelia +Hepatic branches + + + + +Anterior nerve of Latarjet + + + +Crow’s foot + +Gastric branch + + +Posterior nerve + + + + +Gastric branch + +Crow’s foot + + + +Figure 63.2 The anatomy of the anterior and posterior vagus nerves in relation to the stomach. + + +Leopold Auerbach, 1828–1897, Professor of Neuropathology, Breslau, Germany (now Wroclaw, Poland), described the myenteric plexus in 1862. George Meissner, 1829–1905, Professor of Physiology, Gottingen, Germany +. +George Sachs, Professor of Medicine, CURE, Los Angeles, CA, USA, discoverer of the proton pump. +PART 11 | ABDOMINAL +1108 CHAPTER 63 Stomach and duodenum + + +approximately that of plasma. Endocrine cells in the duode-num produce cholecystokinin, which stimulates the pancreas to produce trypsin and the gall bladder to contract. Secretin is also produced by the endocrine cells of the duodenum. This hormone inhibits gastric acid secretion and promotes produc- +tion of bicarbonate by the pancreas. + + +Summary box 63.1 + +The anatomy and physiology of the stomach +●● The stomach acts as a reservoir for food and commences the process of digestion +●● Gastric acid is produced by a proton pump in the parietal cells, which in turn is controlled by histamine acting on the H2-receptors +●● The histamine is produced by the endocrine gastric ECL cells in response to a number of factors, particularly gastrin and the vagus +●● Proton pump inhibitors abolish gastric acid production, whereas H2-receptor antagonists only markedly reduce it +●● The gastric mucous layer is essential to the integrity of the gastric mucosa + + + + + + + +Figure 63.3 The histological appearance of a gastric gland. The mucus-secreting cells are seen at the mucosal surface, the eosino-philic parietal cells superficially in the glands and the basophilic chief cells in the deepest layer. + + +producing D cells throughout the stomach, and somatostatin has a negative regulatory role. The peptides and neuropep- +tides produced in the stomach are discussed later. + +Duodenum +The duodenum is lined by a mucus-secreting columnar epi-thelium. In addition, Brunner’s glands lie beneath the mucosa and are similar to the pyloric glands in the pyloric part of the stomach. Endocrine cells in the duodenum produce cholecys- +tokinin and secretin. + +PHYSIOLOGY OF THE STOMACH +AND DUODENUM +The stomach mechanically breaks up ingested food and, together with the actions of acid and pepsin, forms chyme that passes into the duodenum. In contrast with the acidic environment of the stomach, the environment of the duo-denum is alkaline, due to the secretion of bicarbonate ions from both the pancreas and the duodenum. This neutra- +lises the acid chyme and adjusts the luminal osmolarity to + +Gastric acid secretion +The secretion of gastric acid and pepsin tends to run in parallel, although the understanding of the mechanisms of gastric acid secretion is considerably greater than that of pepsin. Numer-ous factors are involved to some degree in the production of gastric acid. These include neurotransmitters, neuropeptides and peptide hormones. This complexity need not detract from the fact that there are basic principles that are relatively easily understood (Figure 63.4). Hydrogen ions are produced by the parietal cell by the proton pump. Although numerous factors can act on the parietal cell, the most important of these is his-tamine, which acts via the H2-receptor. Histamine is produced, in turn, by the ECL cells of the stomach and acts in a para-crine (local) fashion on the parietal cells. These relationships explain why proton pump inhibitors can abolish gastric acid secretion, as they act on the final common pathway – hydrogen ion secretion. H2-receptor antagonists have profound effects on gastric acid secretion, but this is not insurmountable (Fig-ure 63.4). The ECL cell produces histamine in response to a number of stimuli that include the vagus nerve and gastrin. Gastrin is released by the G cells in response to the presence of the food in the stomach. The production of gastrin is inhib-ited by acid, creating a negative feedback loop. Various other peptides, including secretin, inhibit gastric acid secretion. +Classically, three phases of gastric secretion are described. The cephalic phase is mediated by vagal activity, secondary to sensory arousal as first demonstrated by Pavlov. The gas-tric phase is a response to food within the stomach, which is mediated principally, but not exclusively, by gastrin. In the intestinal phase, the presence of chyme in the duodenum and +small bowel inhibits gastric emptying, and the acidification + + +Johann Conrad Brunner, 1653–1729, Professor of Anatomy, Heidelberg, Germany and later at Strasbourg, France. Ivan Petrovich Pavlov, 1849–1936, Professor of Physiology, The Medico-Chirurgical Academy St Petersburg, Russia. +, +, +PART 11 | ABDOMINAL Physiology of the stomach and duodenum 1109 +, + + + + + + +Chemical in food + + + + + +G cell + +Release of gastrin + + +Distension of stomach M1 + + + + +Release of acetylcholine +HCI +Gastric ECL cell + +Release of histamine +G H1 M3 + +considered distinct in terms of their products. However, it is increasingly realised that there is enormous overlap within these systems. Many peptides recognised as hormones may also be produced by neurones, hence the term neuropeptides. The term ‘messenger’ can be used to describe all such products. +There are three conventional modes of action that overlap. +1 Endocrine. The messenger is secreted into the circulation, where it affects tissues that may be remote from the site of origin (Bayliss and Starling). +2 Paracrine. Messengers are produced locally and have local effects on tissues. Neurones and endocrine cells both act in this way. +3 Neurocrine (classical neurotransmitter). Messengers are produced by the neurone via the synaptic knob and pass +across the synaptic cleft to the target. + + + + + +Figure 63.4 The parietal cell in relation to the mechanism of gastric acid secretion. ECL, enterochromaffin-like; G, gastrin receptor; H, his-tamine receptor; HCl, hydrochloric acid; M, muscarinic receptor. + + +of the duodenum leads to the production of secretin, which inhibits gastric acid secretion, along with numerous other peptides originating from the gut. The stomach also possesses somatostatin-containing D cells. Somatostatin is released in response to a number of factors including acidification. This peptide acts probably on the G cell, the ECL cell and the +parietal cell itself to inhibit the production of acid. + +Gastric mucus and the gastric +mucosal barrier +The gastric mucous layer is essential to the integrity of the gas-tric mucosa. It is a viscid layer of mucopolysaccharides pro-duced by the mucus-producing cells of the stomach and the pyloric glands. Gastric mucus is an important physiological barrier to protect the gastric mucosa from mechanical damage, and also the effects of acid and pepsin. Its considerable buffer-ing capacity is enhanced by the presence of bicarbonate ions within the mucus. Many factors can lead to the breakdown of this gastric mucous barrier. These include bile, non-steroidal anti-inflammatory drugs (NSAIDs), alcohol, trauma and shock. Tonometry studies have shown that, of the entire gas-trointestinal tract, the stomach is the most sensitive to isch-aemia following a hypovolaemic insult and also the slowest to recover. This may explain the high incidence of stress ulcer- +ation in the stomach. + +Peptides and neuropeptides in the +stomach and duodenum +As with most of the gastrointestinal tract, the endocrine cells of the stomach produce peptide hormones and neuro- +transmitters. Previously, nerves and endocrine cells were + +Many peptide hormones act on the intrinsic nerve plexus of the gut (see later) and influence motility. Similarly neuro-peptides may influence the structure and function of the mucosa. Some of these peptides, neuropeptides and neuro-transmitters are shown in Table 63.1. The stomach is vital in the regulation of appetite and weight control via a com-bination of mechanical and hormonal mechanisms; these are +discussed further in following chapters. + +Gastroduodenal motor activity +The motility of the entire gastrointestinal tract is modulated to a large degree by its intrinsic nervous system. Critical in this process is the migrating motor complex (MMC). In the fasted state, and after food has cleared, in the small bowel there is a period of quiescence lasting in the region of 40 min-utes (phase I). There follows a series of waves of electrical and motor activity, also lasting for about 40 minutes, propagated from the fundus of the stomach in a caudal direction at a rate of about three per minute (phase II). These pass as far the pylorus, but not beyond. Duodenal slow waves are generated in the duodenum at a rate of about 10 per minute, which carry down the small bowel. The amplitude of these contractions increases to a maximum in phase III, which lasts for about 10 minutes. This 90-minute cycle of activity is then repeated. From the duodenum, the MMC moves distally at 5–10 cm/ min, reaching the terminal ileum after 1.5 hours. +Following a meal, the stomach exhibits receptive relax-ation, which lasts for a few seconds. Following this, adaptive relaxation occurs, which allows the proximal stomach to act as a reservoir. Most of the peristaltic activity is found in the distal stomach (the antral mill) and the proximal stomach demon-strates only tonic activity. The pylorus, which is most com-monly open, contracts with the peristaltic wave and allows only a few millilitres of chyme through at a time. The antral contraction against the closed sphincter is important in the milling activity of the stomach. Although the duodenum is capable of generating 10 waves per minute, after a meal it only contracts after an antral wave reaches the pylorus. The coor- +dination of the motility of the antrum, pylorus and duodenum + + +Sir William Maddock Bayliss, 1860–1924, physiologist, University College, London, UK. Ernest Henry Starling, 1866–1927, physiologist, University College, London, UK. +PART 11 | ABDOMINAL +1110 CHAPTER 63 Stomach and duodenum + + + +TABLE 63.1 Function and source of peptides and neuropeptides in the stomach. +Function Source Stimulate secretion +Gastrin G cells + +were fibreoptic (Hirschowitz), but now most use a solid-state camera mounted at the instrument’s tip (Figures 63.5 and 63.6). Other members of the endoscopy team see the image and this is useful when taking biopsies or performing interven- +tional techniques, and it also facilitates teaching and training. + + + +Histamine Acetylcholine +Gastrin-releasing peptide Cholecystokinin (CCK) Inhibit secretion Somatostatin +Secretin Enteroglucagon Prostaglandins Neurotensin GIP +PYY +Stimulate motility Acetylcholine +5-HT Histamine Substance P Substance K Motilin Gastrin Angiotensin +Inhibit motility Somatostatin VIP +Nitric oxide +Noradrenaline + +ECL cells (a) Neurones +Neurones and mucosa Duodenal endocrine cells + +D cells and neurones Duodenal endocrine cells Small intestinal endocrine cells Mucosa +Neurones +Duodenal and jejunal endocrine cells Small intestinal endocrine cells + +Neurones Neurones ECL cell +Neurones (b) Neurones +Neurones G cells + + +D cells and neurones Neurones +Neurones and smooth muscle +Neurones + + + +Encephalin Neurones Dopamine Neurones +ECL, enterochromaffin-like cells; GIP, gastric inhibitory polypeptide; PYY, peptide YY; VIP, vasoactive intestinal peptide. + + +means that only small quantities of food reach the small bowel at a time. It is important to consider that this control of gas-tric emptying can be abolished after gastric surgery leading to significant symptoms (discussed in later sections). Motility is influenced by numerous factors, including mechanical stim- +ulation and neuronal and endocrine influences (Table 63.1). + +INVESTIGATION OF THE +STOMACH AND DUODENUM Flexible endoscopy +Flexible endoscopy is the ‘gold standard’ investigation of the +upper gastrointestinal tract. The original flexible endoscopes + + +Figure 63.5 A video-gastroscope (courtesy of Keymed (Medical and Industrial Equipment Ltd)). (a) The camera stack; (b) the gastroscope and biopsy forceps in the working channel. + + +Summary box 63.2 + +The investigation of gastric disorders +●● Flexible endoscopy is the most commonly used and sensitive technique for investigating the stomach and duodenum +●● Great care needs to be exercised in performing endoscopy to avoid complications and missing important pathology +●● Axial imaging, particularly multislice CT, is useful in the staging of gastric cance , although it may be less sensitive in the detection of liver metastases than other modalities +r +●● Endoscopic ultrasound is the most sensitive technique in the evaluation of the ‘T’ stage of gastric cancer and in the assessment of duodenal tumours +●● Laparoscopy is very sensitive in detecting peritoneal metastases, and laparoscopic ultrasound provides an accurate evaluation of lymph node and liver metastases + + +Basil I Hirschowitz, Professor of Medicine, Birmingham, AL, USA. +PART 11 | ABDOMINAL Investigation of the stomach and duodenum 1111 + + + + + + + + + + + + + + + + + + + + + + + + +Figure 63.6 A view of the normal stomach during endoscopy (cour-tesy of GNJ Tytgat, Amsterdam, The Netherlands). + + +Flexible endoscopy is more sensitive than conventional radiology in the assessment of the majority of gastroduodenal conditions. This is particularly the case for peptic ulceration, gastritis and duodenitis. In upper gastrointestinal bleeding, endoscopy is far superior to any other investigation and offers the possibility of endoscopic therapy. In most circumstances it is the only investigation required. +Fibreoptic endoscopy is generally a safe investigation, but it is important that all personnel undertaking these proce-dures are adequately trained. Careless and rough handling of the endoscope during intubation of a patient may result in perforations of the pharynx and oesophagus. Any other part of the upper gastrointestinal tract may also be perforated. An inadequately performed endoscopy is also dangerous as a seri-ous condition may be overlooked. This is particularly the case in respect of early and curable gastric cancer, the appearances of which may often be extremely subtle and may be missed by inexperienced endoscopists. A more experienced endos-copist will have a higher index of suspicion for any muco-sal abnormalities and will take more biopsies. Spraying the mucosa with dye endoscopically may allow better discrimi-nation between normal and abnormal mucosa, so allowing a small cancer to be more easily seen. In the future, advances in technology may allow ‘optical biopsy’ to determine the nature of mucosal abnormalities in real time. +Upper gastrointestinal endoscopy can be performed with-out sedation, but when sedation is required incremental doses of a benzodiazepine are usually administered. Sedation is of particular concern in the case of gastrointestinal bleeding as it may have a more profound effect on the patient’s car-diovascular stability. It has now become the standard to use pulse oximetry to monitor patients during upper gastrointes- +tinal endoscopy, and nasal oxygen is often also administered. + +Buscopan is useful to abolish duodenal motility for examina-tions of the second and third parts of the duodenum. Exam-inations of this type are best carried out using a side-viewing endoscope such as is used for endoscopic retrograde cholan-giopancreatography (ERCP). +Some patients are relatively resistant to sedation with benzodiazepines, particularly those who are accustomed to drinking alcohol. Increasing the dose of benzodiazepines in these patients may not result in any useful sedation, but merely make the patient more restless and confused. Such patients are sometimes better endoscoped fully awake using a local anaesthetic throat spray and a narrow-gauge endoscope. Whatever the circumstances, it is important that resuscita-tion facilities are available including agents that reverse the effects of benzodiazepines, such as flumazenil. +The technology associated with upper gastrointestinal endoscopy is continuing to advance. Instruments which allow both endoscopy and endoluminal ultrasound to be performed simultaneously (see later) are used routinely. Bleeding from the stomach and duodenum can be treated with a number of haemostatic measures. These include injection with various substances, diathermy, heater probes, lasers and clips. These approaches appear to be useful in the treatment of bleeding ulcers, although there are few good controlled trials in this area. There is no good evidence that such interventional procedures at the moment work in patients who are bleeding from very large vessels, such as the gastroduodenal artery or splenic artery, although technology may overcome this prob- +lem in the future. + + +Contrast radiology +Upper gastrointestinal radiology is not used as much as in previous years, as endoscopy is a more sensitive investigation for most gastric problems. Computed tomography (CT) imag-ing with oral contrast has also replaced contrast radiology in many of the areas where anatomical information is sought, eg large hiatus hernias of the rolling type and chronic gastric volvulus. In these conditions it may be difficult for the endos-copist to determine exactly the anatomy or, indeed, negotiate +the deformity to see the distal stomach. + +Ultrasonography +Standard ultrasound imaging can be used to investigate the stomach, but used conventionally it is less sensitive than other modalities. In contrast, endoluminal ultrasound and laparo-scopic ultrasound are probably the most sensitive techniques available in the preoperative local staging of gastric cancer. In endoluminal ultrasound, the transducer is usually attached to the distal tip of the instrument. However, devices have been developed which may be passed down the biopsy channel, albeit with poorer image quality. Five layers (Figure 63.7) of the gastric wall may be identified on endoluminal ultra-sound and the depth of invasion of a tumour can be assessed with exquisite accuracy (90% accuracy for the ‘T’ component of the staging). Enlarged lymph nodes can also be identified +and the technique’s accuracy in this situation is about 80%. +PART 11 | ABDOMINAL +1112 CHAPTER 63 Stomach and duodenum + + + + + + + + + + + + + +Figure 63.7 Endoscopic ultrasound of the stomach. Five layers can be identified in the normal stomach. A gastric cancer is shown invad-ing the muscle of the gastric wall (courtesy of KeyMed (Medical and Industrial Equipment Ltd)). + + +Finally, it may be possible to identify liver metastases not seen on axial imaging. Laparoscopic ultrasound is also a very sensitive imaging modality to a large measure because of the laparoscopy itself (see below). It is one of the most sensitive methods of detecting liver metastases from gastric cancer. +An additional use of ultrasound is in the assessment of gastric emptying. Swallowed contrast is utilised, which is designed to be easily seen using an ultrasound transducer. The emptying of this contrast is then followed directly. The accu-racy of the technique is similar to that of radioisotope gastric +emptying studies (see below). + +CT scanning and magnetic +resonance imaging +The resolution of CT scanners is continuing to improve, and multislice CT is of increasing value in the investigation of the stomach, especially gastric malignancies (Figure 63.8). The presence of gastric wall thickening associated with a carci- +noma of any reasonable size can be easily detected by CT, but + +the investigation lacks sensitivity in detecting smaller lesions. It is much less accurate in ‘T’ staging than endoluminal ultra-sound. Lymph node enlargement can be detected and, based on the size and shape of the nodes, it is possible to be reason-ably accurate in detecting nodal involvement with tumour. However, as with all imaging techniques, it is limited. Micro-scopic tumour deposits in lymph nodes cannot be detected when the node is not enlarged and, in contrast, lymph nodes may undergo reactive enlargement but not contain tumour. These problems apply to all imaging techniques. +The detection of small liver metastases is improving, although in general terms metastases from gastric cancer are less easy to detect using CT than those, for instance, from colorectal cancer. This is because metastases from gastric can-cer may be of the same density as liver and may not handle the intravenous contrast any differently. At present, magnetic resonance imaging (MRI) scanning does not offer any specific advantage in assessing the stomach, although it has a higher sensitivity for the detection of gastric cancer liver metastases +than conventional CT imaging. + +CT/positron emission tomography +Positron emission tomography (PET) is a functional imaging technique which relies on the uptake of a tracer in most cases by metabolically active tumour tissue. Fluorodeoxyglucose (FDG) is the most commonly used tracer This tracer has a short half-life hence manufacture and use have to be care-fully coordinated. To be of value, anatomical and functional information need to be linked, hence PET/CT is now used universally. It is increasingly being used in the preoperative staging of gastro-oesophageal cancer as it will demonstrate occult spread which renders the patient surgically incurable in up to 10% of patients who would otherwise have under-gone major resections (Figure 63.9). PET/CT may also be used to determine the response to neoadjuvant chemotherapy in oesophagogastric malignancies although this is the subject +. +of ongoing studies. + + +Laparoscopy +This technique is routine in the assessment of patients with gastric cancer. Its particular value is in the detection of perito-neal disease, which is difficult by any other technique, unless the patient has ascites or bulky intraperitoneal deposits. Its main limitation is in the evaluation of posterior extension but other techniques are available to evaluate posterior invasion, especially CT and endoluminal ultrasound. Usually laparos-copy is combined with peritoneal cytology unless laparotomy +follows immediately. + + + + + + + +Figure 63.8 A computed tomography scan of the abdomen, show-ing a gastric cancer arising in the body of the stomach. + +Gastric emptying studies +These are useful in the study of gastric dysmotility problems, particularly those that follow gastric surgery. The principle of the examination is that radioisotope-labelled liquid and solid meals are ingested by the patient and the emptying of +the stomach is followed on a gamma camera. This allows the +PART 11 | ABDOMINAL Helicobacter pylori 1113 + + + + + + + + + + + + + + + + +Figure 63.9 Computed tomography/positron emission tomography of a patient with gastric cancer. The middle pair of images shows the primary tumour. The two images on the left show unsuspected liver metastases, whereas the two on the right show a left cervical node positive for metastases. + + + +proportion of activity in the remaining stomach to be assessed numerically, and it is possible to follow liquid and solid gastric +emptying independently (Figure 63.10). + + +Angiography +Angiography is used most commonly in the investigation of upper gastrointestinal bleeding that is not identified using endoscopy. Therapeutic embolisation may also be of value in the treatment of bleeding in patients in whom surgery is difficult or inadvisable. In expert centres embolisation now +replaces surgery in the majority of cases. + +HELICOBACTER PYLORI +Over the last 30 years, this organism has proved to be of over-whelming importance in the aetiology of a number of com-mon gastroduodenal diseases such as chronic gastritis, peptic ulceration and gastric cancer. The organism had unquestion-ably been observed by a number of workers since Bircher’s first description in 1874, but it was not until 1980 that Warren and Marshall, with enthusiasm but perhaps a lack of caution, ingested the organism to confirm that Koch’s postulates could be fulfilled with respect to the gastritis that they succeeded in causing in themselves. Eradication therapy was then +employed with mixed success, but both received the Nobel + + + + +(a) (b) 120 +Liquid 100 Solid + +80 +Proportion remaining + +60 + +40 + +20 + +0 +0 10 20 30 40 50 60 +Time (min) +Figure 63.10 Dual-phase solid and liquid gastric emptying. The use of two isotopic labels allows the liquid and solid phases of the emptying to be followed separately. (a) Image acquisition. (b) Gastric emptying curves in a normal individual showing typical lag period in the solid phase before linear emptying (courtesy of Dr V Lewington, Southampton, UK). + + +C.S. Warren, Perth, Australia, now Chicago, IL, USA, credited with awakening the current interest in Helicobacter. +Barry Marshall, Perth, Australia, now Charlottesville, VA, USA, credited with awakening the current interest in Helicobacter. +Robert Koch, 1843–1910, Professor of Hygiene and Bacteriology, Berlin, German . Stated his ‘postulates’ in 1882. The postulates define the conditions that must be met before an organism can be shown to be the causal agent for a particular disease. +y +PART 11 | ABDOMINAL +1114 CHAPTER 63 Stomach and duodenum + + + +Prize for Medicine and Physiology in 2005. The organism is spiral shaped and is fastidious in its requirements, being diffi-cult to culture outside the mucous layer of the stomach. +One of the characteristics of the organism is its ability to hydrolyse urea, resulting in the production of ammonia, a strong alkali. The effect of ammonia on the antral G cells is to cause the release of gastrin via the previously described negative-feedback loop. This is probably responsible for the modest, but inappropriate, hypergastrinaemia in patients with peptic ulcer disease, which, in turn, may result in gastric acid hypersecretion. The organism’s obligate urease activity is uti-lised by various tests used to detect the presence of the organ-ism, including the 13C and 14C breath tests and the CLO test (a commercially available urease test kit), which is performed on gastric biopsies. The organism can also be detected histo-logically (Figure 63.11), using the Giemsa or the Warthin– Starry stains, and cultured using appropriate media. Previous or current infection with the organism may also be detected serologically. Breath tests or faecal antigen tests are recom-mended for the pre-treatment diagnosis of H. pylori infection in the community. Less accurate, hospital-based serology tests have a place within the non-invasive test-and-treat strategy. +Infection with H. pylori leads to the disruption of the gas-tric mucous barrier by the enzymes produced by the organ-ism, and the inflammation induced in the gastric epithelium is the basis of many of the associated disease processes. The association of the organism with chronic (type B) gastritis is not in doubt. Some strains of H. pylori produce cytotoxins, notably the Cag A and Vac A products, and the production of cytotoxins seems to be associated with the ability of the organism to cause gastritis, peptic ulceration and cancer. The effect of the organism on the gastric epithelium is to incite a classical inflammatory response that involves the migration and degranulation of acute inflammatory cells, such as neu-trophils, and also the accumulation of chronic inflammatory +cells, such as macrophages and lymphocytes. + + + + + + + + + + + + + + + + +Figure 63.11 Antral mucosa showing colonisation with Helicobacter pylori (modified Giemsa stain). + +It is evident how H. pylori infection results in chronic gastritis and also how this may progress to gastric ulceration, but for a while it remained an enigma as to how the organism could be involved in duodenal ulceration, as the normal duodenum is not colonised. As mentioned above, the production of ammonia does increase the level of circulating gastrin and it has been shown subsequently that eradication of the organism in patients with duodenal ulcer disease will reduce the acid levels to normal. However, the overlap in gastric acid secretion between normal subjects and those with duodenal ulcers is considerable and the modestly increased acid levels in patients with Helicobacter-associated antral gastritis are insufficient to explain the aetiology of duodenal ulceration. +The explanation can probably be found in the phenome-non of duodenal gastric metaplasia. Gastric metaplasia is the normal response of the duodenal mucosa to excess acidity. It can be thought of in the same way as any other metaplasia in the gastrointestinal tract: an attempt by the mucosa to resist an injurious stimulus. Although normal duodenal mucosa cannot be infected with H. pylori, gastric metaplasia in the duodenum is commonly infected and this infection results in the same inflammatory process that is observed in the gastric mucosa. The result is duodenitis, which is almost certainly the precursor of duodenal ulceration. +Infection with H. pylori may be the most common human infection. The incidence of infection within a population increases with age, and in many populations infection rates of 80–90% are not unusual. Up to 50% of the world’s popu-lation may be infected with Helicobacter. It appears that most infection is acquired in childhood and the possibility of infec-tion is inversely related to socioeconomic group. The means of spread has not been identified, but the organism can occur in the faeces and faecal–oral spread seems most likely. The organism is not normally found in saliva or dental plaque. There is evidence in different environments and in different population groups that the manifestations of the infection may be different. Predominantly antral gastritis, which is commonly seen in resource-rich countries, results initially in increased levels of acid production and peptic ulcer disease, whereas gastritis affecting the body, common in resource-poor countries, may lead to hypochlorhydria and gastric neoplasia. It has been known since 1984 that Helicobacter infec-tion is amenable to treatment with antibiotics. The pro-found hypochlorhydria produced by proton pump inhibitors combined with antibiotics is also effective in eradicating the organism. Commonly used eradication regimes include a proton pump inhibitor and two antibiotics, such as met-ronidazole and amoxycillin. Very high eradication rates, in the region of 90%, can be achieved with combinations that include the antibiotic clarithromycin, although it may be that in the future antibiotic resistance will become a prob-lem. Reinfection following successful eradication appears rare (<0.5%) but incomplete eradication is a more important +clinical problem. + + +Gustav Giemsa, 1867–1948, bacteriologist who became Privatdozent in Chemotherapy at the University of Hamburg, Hamburg, Germany. Aldred Scott Warthin, 1866–1931, pathologist, the University of Michigan, USA. +Allen Chronister Starry, 1890–1973, pathologist, the University of Michigan, USA. +PART 11 | ABDOMINAL Gastritis 1115 + + + +At present, eradication therapy is recommended for patients with duodenal ulcer disease, but not for patients with non-ulcer dyspepsia or in asymptomatic patients who are infected. However, recent data show that a proportion of patients with non-ulcer dyspepsia do respond to treatment. H. pylori is now classed by the World Health Organisation as a class 1 carcinogen and it may be that further epidemiolog-ical studies on the risk of gastric cancer change the current +advice on treatment. + + +GASTRITIS +The great variety of names and classification systems used in gastritis is confusing. Thankfully, the understanding of gastri-tis has increased markedly following elucidation of the role of H. pylori in chronic gastritis and there is broad agreement that gastritis should be classified according to the underlying aeti-ology. Gastritis describes any histologically confirmed inflam-mation of the gastric mucosa. In most modern classification systems the amount of inflammatory infiltrate and the degree +of gastric atrophy will be included. + + +Summary box 63.3 + +Gastritis +●● The spiral bacterium Helicobacter pylori is critical in the development of gastritis, peptic ulceration and gastric cancer +●● Infection appears to be acquired mainly in childhood and the infection rate is inversely associated with socioeconomic status +●● Eradication, recommended specifically in patients with peptic ulcer disease, can be achieved in up to 90% of patients with a combination of a proton pump inhibitor and antibiotics, and reinfection is uncommon (<0.5%) +●● Erosive gastritis is usually related to the use of NSAIDs +●● Autoimmune gastritis is associated with the development of pernicious anaemia and gastric cancer + + + +Autoimmune gastritis +This is an autoimmune condition in which there are circulat-ing antibodies to the parietal cell. This results in the atrophy of the parietal cell mass, hence hypochlorhydria and ulti-mately achlorhydria. As intrinsic factor is also produced by the parietal cell there is malabsorption of vitamin B12, which, if untreated, may result in pernicious anaemia. The antrum is not affected and the hypochlorhydria leads to the production of high levels of gastrin from the antral G cells. This results in chronic hypergastrinaemia. This, in turn, results in hyper-trophy of the ECL cells in the body of the stomach, which are not affected by the autoimmune damage. Over time it is apparent that microadenomas develop in the ECL cells of the stomach, sometimes becoming identifiable tumour nodules. Very rarely, these tumours can become malignant. Patients with autoimmune gastritis are predisposed to the develop-ment of gastric cancer, and screening such patients endoscop- +ically may be appropriate. + +H. pylori gastritis +Previously described as type B gastritis, this affects the antrum, and it is these patients who are prone to peptic ulcer disease. Helicobacter-associated pangastritis is also a very common manifestation of infection, but gastritis affecting the corpus alone does not seem to be associated. However, there are some data to suggest that Helicobacter may be involved in the initiation of the process. Patients with pangastritis seem to be most prone to the development of gastric cancer. +Intestinal metaplasia is associated with chronic pangas-tritis with atrophy. Although intestinal metaplasia per se is common, intestinal metaplasia associated with dysplasia has significant malignant potential and, if this condition is iden- +tified, endoscopic screening may be appropriate. + + +Reflux gastritis +This is caused by enterogastric reflux and is particularly com-mon after gastric surgery. Its histological features are distinct from other types of gastritis. Although commonly seen after gastric surgery, it is occasionally found in patients with no previous surgical intervention or who have had a cholecys-tectomy. Bile chelating or prokinetic agents may be useful in treatment and as a temporising measure to avoid consider-ation of revisional surgery. Operation for the condition should +be reserved for the most severe cases. + + +Erosive gastritis +This is caused by agents that disturb the gastric muco-sal barrier; NSAIDs and alcohol are common causes. The NSAID-induced gastric lesion is associated with inhibition of the cyclo-oxygenase type 1 (COX-1) receptor enzyme, hence reducing the production of cytoprotective prostaglandins in the stomach. Many of the beneficial anti-inflammatory activities of NSAIDs are mediated by COX-2, and the use of specific COX-2 inhibitors reduces the incidence of these side effects. However, taken in the long term COX-2 inhibitors are associated with cardiovascular complications in common +with many NSAIDs. + + +Stress gastritis +This is a common sequel of serious illness or injury and is characterised by a reduction in the blood supply to superficial mucosa of the stomach. Although common, this is not usually recognised unless stress ulceration and bleeding supervene, in which case treatment can be extremely difficult. The condition also sometimes follows cardiopulmonary bypass. Prevention of the stress bleeding from the stomach is much easier than treating it, and hence the routine use of H2 antagonists with or without barrier agents, such as sucralfate, in patients who are on intensive care. These measures have been shown to reduce the incidence of bleeding from stress +ulceration. +PART 11 | ABDOMINAL +1116 CHAPTER 63 Stomach and duodenum + + + +Ménétrier’s disease +This is an unusual condition characterised by gross hyper-trophy of the gastric mucosal folds, mucus production and hypochlorhydria. The condition is premalignant and may present with hypoproteinaemia and anaemia. There is no treatment other than a gastrectomy. The disease seems to be caused by overexpression of transforming growth factor alpha (TGF-a). Like epidermal growth factor (EGF), this peptide also binds to the EGF receptor. The histological features of Ménétrier’s disease may be reproduced in transgenic mice +overexpressing TGF-a. + +Lymphocytic gastritis +This type of gastritis is seen rarely. It is characterised by the infiltration of the gastric mucosa by T cells and is probably associated with H. pylori infection. The pattern of inflam-mation resembles that seen in coeliac disease or lymphocytic +colitis. + +Other forms of gastritis +Eosinophilic gastritis appears to have an allergic basis, and is treated with steroids and cromoglycate. Granulomatous gas-tritis is seen rarely in Crohn’s disease and also may be associ-ated with tuberculosis. Acquired immunodeficiency syndrome (AIDS) gastritis is secondary to infection with cryptospiridio-sis. Phlegmonous gastritis is a rare bacterial infection of the stomach found in patients with severe intercurrent illness. It +is usually an agonal event. + +PEPTIC ULCER +Although the name ‘peptic’ ulcer suggests an association with pepsin, this is essentially unimportant as in the absence of acid, peptic ulcers do not occur. Nearly all peptic ulcers can be healed by using proton pump inhibitors, which can render a patient virtually achlorhydric. +Common sites for peptic ulcers are the first part of the duodenum and the lesser curve of the stomach, but they also occur on the stoma following gastric surgery, the oesophagus and even in a Meckel’s diverticulum, which contains ectopic gastric epithelium. In general, the ulcer occurs at a junction between different types of epithelia, the ulcer occurring in the epithelium least resistant to acid damage. +In the past, much distinction has been made between acute and chronic peptic ulcers, but this difference can sometimes be difficult to determine clinically. It is probably +best to consider that there is a spectrum of disease from the + + +Summary box 63.4 + +Peptic ulceration +●● Most peptic ulcers are caused by H. pylori or NSAIDs and changes in epidemiology mirror changes in these principal aetiological factors +●● Duodenal ulcers are more common than gastric ulcers, but the symptoms are indistinguishable +●● Gastric ulcers may become malignant and an ulcerated gastric cancer may mimic a benign ulcer +●● Gastric antisecretory agents and H. pylori eradication therapy are the mainstay of treatment, and elective surgery is very rarely performed +●● The long-term complications of peptic ulcer surgery may be difficult to treat +●● The common complications of peptic ulcers are perforation, bleeding and stenosis +●● The treatment of the perforated peptic ulcer is primarily surgical, although some patients may be managed conservatively + + + +superficial gastric and duodenal ulceration, frequently seen at endoscopy, to deep chronic penetrating ulcers. This does not minimise the importance of acute stress ulceration. These ulcers can both perforate and bleed. +For many years, the cause of peptic ulceration remained an enigma. When comparing groups of patients with duode-nal and pre-pyloric peptic ulcers with normal subjects, gastric acid levels are higher, but the overlap is very considerable. Patients with gastric ulceration have relatively normal levels of gastric acid secretion. As peptic ulceration will occur in the presence of very high acid levels, such as those found in patients with a gastrinoma (Zollinger–Ellison syndrome), and as all ulcers can be healed in the absence of acid, it is clear that acid is important. In patients with a gastrinoma it may be the only aetiological factor, but this is not the case in the majority of patients. As with many diseases, genetic factors may be involved to a limited degree and social stress has also been falsely implicated (Asher). +It is now widely accepted that infection with H. pylori and the consumption of NSAIDs are the most important factors in the development of peptic ulceration. In combination H. pylori and NSAIDs act synergistically to promote ulcer devel-opment and ulcer bleeding. Cigarette smoking predisposes to peptic ulceration and increases the relapse rate after treat-ment, with either gastric antisecretory agents or, in the past, elective surgery. Multiple other factors may be involved in transition between the superficial and the deep penetrating +chronic ulcer, but they are of lesser importance. + + +Pierre Ménétrier, 1859–1935, French physician. +Burrill Bernard Crohn, 1884–1956, gastroenterologist, Mount Sinai Hospital, New York, NY, USA. Described regional ileitis in 1932. +Johann Friedrich Meckel (The Younger), 1781–1833, Professor of Anatomy and Surgery, Halle, Germany. Described the diverticulum in 1809. Robert Milton Zollinger, 1903–1992, Professor of Surgery, Ohio State University, Columbus, OH, USA. +Edwin Homer Ellison, 1918–1970, Professor of Surgery, Marquette University, Milwaukee, WI, USA. Zollinger and Ellison described this condition in a joint paper in 1955 when they were both working at the Ohio State University. +Richard Asher, 1912–1969, physician, the Central Middlesex Hospital, London, UK. Ridiculed the concept that the stress of modern living caused peptic ulcer-ation by pointing out that the same claim was made for syphilis! It is an interesting coincidence that both diseases have strong aetiological associations with spiral organisms. +PART 11 | ABDOMINAL +Peptic ulcer 1117 + + + +Duodenal ulceration Incidence +There have been marked changes in the last two decades in the demography of patients presenting with duodenal ulceration in resource-rich countries. First, even before the introduction of H2-receptor antagonists, the incidence of duodenal ulcer-ation and the frequency of elective surgery for the condition were falling. This trend has continued and now, dyspeptic patients presenting with a duodenal ulcer at gastroscopy are uncommon. In part, this may relate to the widespread use of gastric antisecretory agents and H. pylori eradication therapy for patients with dyspepsia. Second, the peak incidence is now in a much older age group than previously and, although it is still more common in men, the difference is less marked. These changes mirror the changes, at least in part, in the epi-demiology of H. pylori infection. In Eastern Europe, the disease remains common and, from having been uncommon in some resource-poor nations, it is now observed more frequently. +Again, the relationship with H. pylori appears convincing. +Pathology +Most occur in the first part of the duodenum (Figures 63.12 and 63.13). A chronic ulcer penetrates the mucosa and into the muscle coat, leading to fibrosis. The fibrosis causes deformities such as pyloric stenosis. When an ulcer heals, a scar can be observed in the mucosa. Sometimes there may be more than one duodenal ulcer. The situation in which there is both a posterior and an anterior duodenal ulcer is referred to as ‘kissing ulcers’. Anteriorly placed ulcers tend to perfor-ate and, in contrast, posterior duodenal ulcers tend to bleed, sometimes by eroding into the gastroduodenal artery. Occa-sionally, the ulceration may be so extensive that the entire +duodenal cap is ulcerated and devoid of mucosa. With respect + + + + + + + + + + + + + + + + + + + + + + +Figure 63.12 Duodenal ulcer at gastroduodenoscopy (courtesy of Dr GNJ Tytgat, Amsterdam, The Netherlands). + + + + + + + + + + + + + + + + + + + + + + +Figure 63.13 Duodenal ulcer shown by barium meal. + +to the giant duodenal ulcer, malignancy in this region is so uncommon that under normal circumstances surgeons can be confident that they are dealing with benign disease, even though from external palpation it may not appear so. In the +stomach the situation is different. +Histopathology +Microscopically, destruction of the muscular coat is observed and the base of the ulcer is covered with granulation tissue, the arteries in this region showing the typical changes of end-arteritis obliterans. Sometimes the terminations of nerves can be seen among the fibrosis. The pathological appearances of the healing ulcer must be carefully interpreted as some of the epithelial downgrowths can be misinterpreted as invasion. This is unlikely to be important in duodenal ulcers when malignancy rarely, if ever, occurs, but it is much more import- +ant with gastric ulcers. + +Gastric ulcers Incidence +As with duodenal ulceration, H. pylori and NSAIDs are the important aetiological factors. Gastric ulceration is also asso-ciated with smoking; other factors are of lesser importance. +There are marked differences between the populations afflicted by chronic gastric ulceration compared with duodenal ulceration. First, gastric ulceration is substantially less com-mon than duodenal ulceration. The sex incidence is equal and the population with gastric ulcers tends to be older. It is more prevalent in low socioeconomic groups and is considerably +more common in resource-poor countries than in richer ones. +Pathology +This is essentially similar to that of a duodenal ulcer except +, +that gastric ulcers tend to be larger. Fibrosis, when it occurs, +PART 11 | ABDOMINAL +1118 CHAPTER 63 Stomach and duodenum + + + +may result in the now rarely seen hourglass contraction of the stomach. Large chronic ulcers may erode posteriorly into the pancreas and, on other occasions, into major vessels such as the splenic artery. Less commonly, they may erode into other organs such as the transverse colon. Chronic gastric ulcers are much more common on the lesser curve (especially at the incisura angularis; Figures 63.14 and 63.15) than on the greater curve and, even when high on the lesser curve, they tend to be at the boundary between the acid-secreting and the non-acid-secreting epithelia. With atrophy of parietal cell mass, non-acid-secreting epithelium migrates up the +lesser curvature. + + + + + + + + + + + + + + + + + + + + + +Figure 63.14 Benign incisural gastric ulcer shown at gastroscopy (courtesy of Dr GNJ Tytgat, Amsterdam, The Netherlands). + + + +(a) (b) + + + + + +Large gastric ulcer + + +Malignancy in gastric ulcers +Chronic duodenal ulcers are not associated with malignancy and, in contrast, gastric ulcers are. Widely varying estimates are made of the incidence of gastric malignancy in gastric ulcers. The reason for this is that the authors reporting such diverse incidences are describing different clinical situations. Two clinical extremes must be distinguished to understand this problem properly. First, there is the situation in which a benign chronic gastric ulcer undergoes malignant transforma-tion. This is known to happen, albeit rarely. The contrasting clinical extreme is the patient identified as having an ulcer in the stomach, either endoscopically or on contrast radiology, which is assessed as benign but biopsies reveal malignancy. In this situation the patient does not have, and probably never has had, chronic peptic ulceration in the stomach but has pre-sented with an ulcerated cancer. This situation is common, although whether a lesion found in the stomach is described as being benign or malignant on clinical grounds depends very much on the skill and experience of the endoscopist or radiologist. +It is fundamental that any gastric ulcer should be regarded as being malignant, no matter how classical the features of a benign gastric ulcer. Multiple biopsies should always be taken, perhaps as many as 10 well-targeted biopsies, before an ulcer can be tentatively accepted as being benign. Even then it is important that further biopsies are taken while the ulcer is healing and when healed. Modern antisecretory agents can frequently heal the ulceration associated with gastric cancer but, clearly, are ineffective in treating the malignancy itself. At operation, even experienced surgeons may have difficulty distinguishing between the gastric cancer and a benign ulcer. Operative strategies differ so radically that it is essential, if at all possible, that a confident diagnosis is made before opera-tion. If, at operation for perforation, it is determined that the ulcer is probably benign it should, nonetheless, be excised, in totality if possible, and submitted for histological examina-tion. It is not known whether a patient’s survival is compro-mised by this approach if the ulcer turns out to be malignant +on biopsy, as convincing data are not available. + +Other peptic ulcers +The pre-pyloric gastric ulcer was in the past difficult to treat, a problem overcome with the introduction of proton pump inhibitors. Pyloric channel ulcers are similar to duodenal ulcers. Both pre-pyloric and pyloric ulcers may be malignant, and biopsy is essential. Stomal ulcers occur after a gastroen-terostomy (now most commonly after bariatric surgery; see later) or a gastrectomy of the Billroth II type. The ulcer is +usually found on the jejunal side of the stoma. + + + + + + +Figure 63.15 Benign gastric ulcer shown on barium meal. (a) Radiograph; (b) diagrammatic outline. + +Clinical features of peptic ulcers +Although many textbooks try to create differences in the clinical feature of gastric and duodenal ulceration, detailed +analysis has shown that they cannot be differentiated on the + + +Christian Albert Theodor Billroth, 1829–1894, Professor of Surgery, V enna, Austria. +i +PART 11 | ABDOMINAL +Peptic ulcer 1119 + + + +basis of symptoms. The demographic characteristics of groups of patients with gastric and duodenal ulceration do differ but this does not allow discrimination. + +Pain +The pain is epigastric, often described as gnawing and may radiate to the back. Eating may sometimes relieve the dis-comfort. The pain is normally intermittent rather than intractable. + +Periodicity +One of the classical features of untreated peptic ulceration is periodicity. Symptoms may disappear for weeks or months to return again. This periodicity may be related to the sponta-neous healing of the ulcer. + +Vomiting +While this occurs, it is not a notable feature unless stenosis has occurred. + +Alteration in weight +Weight loss or, sometimes, weight gain may occur. Patients with gastric ulceration are often underweight but this may precede the occurrence of the ulcer. + +Bleeding +All peptic ulcers may bleed. The bleeding may be chronic and presentation with microcytic anaemia is not uncommon. All such patients should be investigated with endoscopy. Acute presentation with haematemesis and melaena is discussed later +. + +Clinical examination +Examination of the patient may reveal epigastric tender-ness but, except in extreme cases (for instance, gastric outlet obstruction), there is unlikely to be much else to find. + +Investigation of the patient with suspected peptic ulcer +Gastroduodenoscopy +This is the investigation of choice in the management of sus-pected peptic ulceration and in the hands of a well-trained operator is highly sensitive and specific. +In the stomach, any abnormal lesion should be multiply biopsied, and in the case of a suspected benign gastric ulcer numerous biopsies must be taken in order to exclude, as far as possible, the presence of a malignancy. Commonly biopsies of the antrum will be taken to see whether there is histological evidence of gastritis and a CLO test performed to determine the presence of H. pylori. A ‘U’ manoeuvre should be per-formed to exclude ulcers around the gastro-oesophageal junc-tion. This is important as the increasing incidence of cancer at the gastro-oesophageal junction requires that all mucosal abnormalities in this region should undergo multiple biopsy. +, + +Similarly, if a stoma is present, for instance after gastroen-terostomy or Billroth II gastrectomy, it is important to enter both afferent and efferent loops. Almost all stomal ulcers will be very close to the junction between the jejunal and gas-tric mucosa. Attention should be given to the pylorus to note whether there is any pre-pyloric or pyloric channel ulceration, and also whether it is deformed, which is often the case with chronic duodenal ulceration. In the duodenum, care must be taken to view all of the first part. It is not infrequent for an ulcer +to be just beyond the pylorus and easily overlooked. + +Treatment of peptic ulceration +The vast majority of uncomplicated peptic ulcers are treated medically. Surgical treatment of uncomplicated peptic ulcer-ation has decreased markedly since the 1960s and is now seldom performed in resource-rich countries. Surgical treat-ment was aimed principally at reducing gastric acid secretion and, in the case of gastric ulceration, removing the diseased mucosa. When originally devised, medical treatment also aimed to reduce gastric acid secretion, initially using the highly successful H2-receptor antagonist and, subsequently, proton pump inhibitors. This has now largely given way to +eradication therapy +. +Medical treatment +It is reasonable that a doctor managing a patient with an uncomplicated peptic ulcer should suggest modifications to the patient’s lifestyle, particularly the cessation of cigarette smoking. This advice is rarely followed and pharmacological +measures form the mainstay of treatment. +H -receptor antagonists and proton pump inhibitors +2 +H2-antagonists (Black) revolutionised the management of peptic ulceration. Most duodenal ulcers and gastric ulcers can be healed by a few weeks of treatment with these drugs provided that they are taken and absorbed. There remained, however, a group of patients who were relatively refractory to conventional doses of H2-receptor antagonists. This is largely now irrelevant as proton pump inhibitors can effectively ren-der a patient achlorhydric and all benign ulcers will heal using these drugs, the majority within 2 weeks. Symptom relief is impressively rapid, most patients being asymptomatic within a few days. Like H2-antagonists, proton pump inhibitors are safe and relatively devoid of serious side effects. The problem with all gastric antisecretory agents is that following cessation +of therapy relapse is almost universal. +Eradication therapy +Eradication therapy is now routinely given to patients with peptic ulceration, and this is described earlier in this chap-ter. Evidence suggests that if a patient has a peptic ulcer and H. pylori is the principal aetiological factor (essentially the patient not taking NSAIDs) then complete eradication of the +organism will cure the disease and reinfection as an adult is + + +Sir James Black, 1924–2002, Professor of Pharmacology, Kings College Hospital Medical School, London, UK, introduced beta-blockers and H2-receptor antagonists. He received the Nobel Prize for Physiology or Medicine in 1988. +’ +PART 11 | ABDOMINAL +1120 CHAPTER 63 Stomach and duodenum + + + +uncommon. Eradication therapy is therefore the mainstay of treatment for peptic ulceration. It is extremely economical by comparison with prolonged courses of antisecretory agents or surgery. It is also considerably safer than surgical treatment. +There are some patients with peptic ulcers in whom erad-ication therapy may not be appropriate and this includes patients with NSAID-associated ulcers. Such patients should avoid these drugs if possible and, if not, they should be co-prescribed with a potent antisecretory agent. Similarly, patients with stomal ulceration are not effectively treated with eradication therapy and require prolonged prescription of antisecretory agents. Patients with Zollinger–Ellison syn-drome should be treated in the long term with proton pump inhibitors unless the tumour can be adequately managed +by surgery. +Ulcers that fail to heal +The introduction of antisecretory agents and effective treat-ments for H. pylori have revolutionised the management of peptic ulcers. Despite these advances peptic ulceration fails to heal in a small minority of patients. Endoscopic re-evaluation should be regarded as mandatory to confirm healing of all gas-tric ulcers. Furthermore, endoscopy permits the differentiation between a refractory ulcer and persistent symptoms despite ulcer healing. The most common cause of failed healing is per-sistent H. pylori infection. Biopsies should be repeated at the time of endoscopy as false-negative results with breath tests may be expected soon after eradication therapy and serum antibody titres may not fall for 6 months after successful eradication. Failure of eradication is usually due to poor compliance or bac-terial resistance and bacteriological culture will guide further attempts at H. pylori eradication. The ingestion of NSAIDs should once again be addressed. A diagnosis of Zollinger-Ellison syndrome (described in detail later) should be suspected in H. pylori negative, non-NSAID-related peptic ulceration and serum gastrin levels should be measured. Very rarely, a recently described autoimmune IgG4-related phenomenon is +the cause of resistant and recurrent gastric ulceration. + + +from the duodenum. This has been achieved by diversion of the acid away from the duodenum, reducing the secre-tory potential of the stomach, or both. All of the operations devised achieved their aim to some extent, but with varying degrees of morbidity, mortality and postoperative side effects. There is now no role for acid-reducing operations in the rou-tine management of peptic ulcer disease but occasionally operations which involve gastrectomy have to be performed in the emergency situation. In addition many patients have had such operations performed and suffer from the sequelae. Hence it is important for the clinician to understand the ana-tomical and physiological consequences of surgery. The oper- +ations are described in historical sequence. +Billroth II gastrectomy +The first successful gastrectomy was performed by Billroth in January 1881, and Wolfler performed the first gastroen-terostomy in the same yea . The original Billroth operations consisted of a gastric resection with gastroduodenal anasto-mosis (Billroth I technique) (Figure 63.16). The Billroth II operation was devised more by accident than design (Figure 63.17). A gastroenterostomy (Figure 63.18) was performed on a gravely ill patient with a pyloric cancer, who was not expected to survive. Contrary to expectations, the patient improved and the stomach distal to the anastomosis was resected. It soon became evident that the use of gastrojeju-nal anastomosis after gastric resection could be safer and eas-ier than the Billroth I procedure, and it became popular and effective in the surgical treatment of duodenal ulcer. Because of its disadvantages, such as higher operative mortality and morbidity, it has not been used for many years in the patient +r +with an uncomplicated ulcer, but it is still used occasionally + + +Surgical treatment of uncomplicated +peptic ulceration +From its peak in the 1960s, the incidence of surgery for uncom-plicated peptic ulceration has fallen markedly, to the extent that peptic ulcer surgery is now of little more than historical interest. A description of operations used in the treatment of peptic ulcers is still necessary because surgery is occasionally employed for the complicated ulcer and, in addition, many patients are left suffering from the consequences of the more +destructive operations. + + +Operations for duodenal ulceration Duodenal ulcer surgery (rationale) +Procedures devised for the treatment of duodenal ulcers have +the common aim of excluding the damaging effects of acid + + + + +Figure 63.16 Billroth I gastrectomy. The lower half of the stomach is removed and the cut stomach anastomosed to the first part of the duodenum. + + +Anton Wolfler, 1850–1917, Professor of Surgery, Prague, Czechoslovakia. +PART 11 | ABDOMINAL +Peptic ulcer 1121 + + + + + + + + + + + + + + + + + + + + + +Figure 63.17 Billroth II. Two-thirds of the stomach is removed, the duodenal stump is closed and the stomach anastomosed to the jeju-num. + +a catheter brought out through the abdominal wall. Follow-ing resection, the distal end of the stomach is narrowed by the closure of the lesser curve aspect of the remnant. The greater curve aspect is then anastomosed, usually in a retro-colic fashion, to the jejunum, leaving as short an afferent loop as feasible (Figure 63.17). Even when well performed, this procedure has an operative mortality rate of a few per cent and morbidity is not unusual. A common cause of morbidity is leakage from the duodenal stump, which is particularly associated with kinking of the afferent loop. Leakage from the gastrojejunal anastomosis is unusual unless it is under tension or the stomach has been devascularised during the mobilisa-tion. The incidence of side effects following gastrectomy is considerable, as shown in Table 63.2. Recurrence of the ulcer at the stoma is uncommon but can occur, especially as this procedure is traditionally not combined with the vagotomy. In the majority of expert centres a Roux-en-Y reconstruction rather than the Billroth II procedure is performed because +postoperative function is better. + +TABLE 63.2 Operative mortality, side effects and incidence of recurrence following duodenal ulcer operations. + + + +Operation + + +Gastrectomy +Gastroenterostomy alone +Truncal vagotomy and drainage +Selective vagotomy and drainage +Highly selective vagotomy +Truncal vagotomy and antrectomy + + +Operative mortality (%) + +<1–2 +<1 + +<1 + +<1 + +<0.2 + +<1 + + +Significant side effects (%) +<20–40 +<10–20 + +<10–20 + +<10–20 + +<5 + +<10–20 + +Recurrent ulceration (%) +1–4 +50 + +2–7 + +5–10 + +2–10 + +1 + + + + + +Figure 63.18 Gastroenterostomy. The jejunum is anastomosed to the posterior, dependent, wall of the stomach. + + +in the treatment of a complicated ulcer with a ‘difficult’ duo-denum. In Billroth II gastrectomy, or its close relation Pólya gastrectomy, the antrum and distal body of the stomach are mobilised by opening the greater and lesser omentum and dividing the gastroepiploic arteries, right gastric artery and the left gastric artery arcade at the limit of the resection. The duodenum is closed off either by suture or using staples, sometimes with difficulty in patients with a very deformed duodenum. Various techniques are available to close the difficult duodenum and, in extremis, a catheter may be placed +in the duodenal stump, the duodenum closed around it and + +Gastrojejunostomy +Because of the potential for mortality after gastrectomy, the use of gastrojejunostomy alone in the treatment of duodenal ulceration was developed (Figure 63.18). Reflux of alkali from the small bowel into the stomach reduced duodenal acid exposure and was often successful in healing the ulcer. How-ever, because the jejunal loop was exposed directly to gastric acid, stomal ulceration was extremely common, hence the +procedure in isolation was ineffective. +Truncal vagotomy and drainage +Truncal vagotomy was first introduced in 1943 by Dragstedt and, for many years, combined with drainage, was the main-stay of treatment of duodenal ulceration (Figure 63.19). The principle of the operation is that section of the vagus nerves, +which are critically involved in the secretion of gastric acid, + + +Eugen (Jeno) Alexander Pólya, 1876–1944, surgeon, St Stephen’s Hospital, Budapest, Hungary +. +César Roux, 1857–1934, Professor of Surgery and Gynaecology, Lausanne, Switzerland. Described this method of forming a jejunal conduit in 1908. Lester Reynold Dragstedt, 1893–1975, Professor of Surgery, Chicago, IL, USA. +PART 11 | ABDOMINAL +1122 CHAPTER 63 Stomach and duodenum + + +(a) (b) (c) + + + + + + + + + + +Figure 63.19 Truncal vagotomy: (a) division of the anterior vagus; (b) mobilisation of the oesophagus; (c) division of the posterior vagus. + + + +reduces the maximal acid output by approximately 50%. Because the vagal nerves are motor to the stomach, dener-vation of the antropyloroduodenal segment results in gastric stasis in a substantial proportion of patients on whom truncal vagotomy alone is performed. This was first noted by Dragst-edt, who did not perform a drainage procedure when he first introduced the operation. The most popular drainage proce-dure is the Heineke–Mikulicz pyloroplasty (Figure 63.20). It is simple to perform and involves the longitudinal section of the pyloric ring. The incision is closed transversely. Gas-trojejunostomy (Figure 63.18) was the alternative drainage procedure to pyloroplasty. This is performed through open-ing the lesser sac and performing an anastomosis between the most dependent part of the antrum and the first jejunal loop. An isoperistaltic anastomosis was most commonly performed. The operation of truncal vagotomy and drainage is substan-tially safer than gastrectomy (Table 63.2). However, the side effects of surgery are, in fact, little different from those that +follow gastrectomy. +Highly selective vagotomy +In 1968 Johnston and Amdrup independently devised the operation of highly selective vagotomy in which only the parie- +tal cell mass of the stomach was denervated (Figure 63.21). + + + + + + + + + + + + + + +Figure 63.20 Pyroplasty. + + +6–8 cm + + + + + + + + + + + + +7 cm + + +Figure 63.21 Highly selective vagotomy. The anterior and posterior vagus nerves are preserved but all branches to the fundus and body of the stomach are divided. + + +This proved to be the most satisfactory operation for duodenal ulceration, with a low incidence of side effects and acceptable recurrence rates when performed to a high technical stan-dard. This operation became the gold standard for operations on duodenal ulceration in the 1970s. The operative mortal-ity was lower than any other definitive operation for duode-nal ulceration, in all probability because the gastrointestinal tract was not opened during this procedure. The unpleasant effects of peptic ulcer surgery were largely avoided, although loss of receptive relaxation of the stomach did occur, leading to epigastric fullness and sometimes mild dumping. However, the severe symptoms that occur after other more destructive gastric operations did not occur. It is often said that recurrent ulceration is the Achilles’ heel of this operation although, when performed well, recurrence was no more common than after truncal vagotomy. The operation disappeared from rou-tine use with the advent of antisecretory agents and eradica- +tion therapy. + + +Walther Hermann Heineke, 1834–1901, surgeon, Erlangen, Germany. +Johann von Mikulicz-Radecki, 1850–1905, Professor of Surgery, Breslau, Germany (now Wroclaw, Poland). David Johnston, contemporary, Professor of Surgery University of Leeds, Leeds, UK. +, +Eric Amdrup, 1923–1998, Professor of Surgery, Aarhus, Denmark. +PART 11 | ABDOMINAL +Peptic ulcer 1123 + + + +Truncal vagotomy and antrectomy +For completeness, this operation should be mentioned as at one stage it was popular in the USA. In addition to a trun-cal vagotomy, the antrum of the stomach is removed, thus removing the source of gastrin, and the gastric remnant is joined to the duodenum. The recurrence rates after this pro-cedure are exceedingly low. However, the operative mortality is higher than after vagotomy and drainage (Table 63.2) and +the incidence of unpleasant side effects is similar. + +Operations for gastric ulcer +In contrast with duodenal ulcer surgery, when the principal objective was to reduce duodenal acid exposure, in gastric ulceration the diseased tissue is usually removed as well. This has the advantage that malignancy can then be confidently excluded. As with duodenal ulceration such surgery is not +now performed except for complications of gastric ulcer. +Billroth I gastrectomy +This was the standard operation (Figure 63.16) for gastric ulceration until medical treatments became prevalent. The distal stomach is mobilised and resected in the same way as in the Billroth II gastrectomy. This resection should include the ulcer that is usually situated on the lesser curve. The cut edge of the remnant is then partially closed from the lesser curve aspect, leaving a stoma at the greater curve aspect, which should be similar in size to the duodenum. Reconstruction may be facilitated by mobilising the duodenum using Kocher’s manoeuvre. The incidence of recurrent ulceration after this operation is low, but it carries with it the morbidity and mor-tality associated with any gastric resection. This operation is +virtually never performed now. + +Sequelae of peptic ulcer surgery +There are a number of sequelae of peptic ulcer surgery, which include recurrent ulceration, small stomach syndrome, bilious vomiting, early and late dumping, diarrhoea and malignant transformation. These sequelae principally follow from the more destructive operations that are now seldom performed. Howeve , a substantial number of patients suffer from side effects from operations undertaken in the past. Approxi-mately 30% of patients can expect to suffer a degree of dys-function following peptic ulcer surgery (Table 63.2) and, in +r +about 5% of such patients, the symptoms will be intractable. +Recurrent ulceration +Although mentioned first, this is by far the easiest problem to treat. Just as all peptic ulcers will heal with potent antisecre-tory agents, so will ulcers that are recurrent after ulcer surgery. As with other peptic ulcers, recurrent ulcers may pres- +ent with complications, particularly bleeding and perfora-tion. In this respect, the complication of gastrojejunal colic fistula requires a particular mention. In this rare condition, +the anastomotic ulcer penetrates into the transverse colon. + +Patients suffer from diarrhoea that is severe and follows every meal. They have foul breath and may vomit formed faeces. Severe weight loss and dehydration are rapid in onset, and for this reason the condition may be mistaken for malignancy. The major factor producing the nutritional disturbance is the severe contamination of the jejunum with colonic bacteria. A number of imaging techniques can be used to detect the fistula, most commonly CT with oral contrast or indeed a bar-ium enema. Endoscopy may not convincingly demonstrate the fistula and, in about one-half of such cases, the barium meal will not reveal the problem. The treatment of gastro-colic fistula consists of first correcting the dehydration and +malnutrition and then performing revisional surgery. + +Small stomach syndrome +Early satiety follows most ulcer operations to some degree, including highly selective vagotomy. In this latter circum-stance, although there is no anatomical disturbance of the stomach there is loss of receptive relaxation. Fortunately, this problem does tend to get better with time and revisional sur- +gery is not necessary. + +Bile vomiting +Bile vomiting can occur after any form of vagotomy with drainage or gastrectomy. Commonly, the patient presents with vomiting a mixture of food and bile or sometimes some bile alone after a meal. Often eating will precipitate abdom-inal pain and reflux symptoms are common. Bile chelating agents can be tried but are usually ineffective. In intractable cases, revisional surgery may be indicated. The nature of that revisional surgery depends very much on the original opera-tion. Following gastrectomy, Roux-en-Y diversion is probably the best treatment. In patients with a gastroenterostomy, this can be taken down and, in most circumstances a small pylo-roplasty can be performed. In patients with a pyloroplasty, reconstruction of the pylorus has been attempted but, in gen-eral terms, the results of this operation have been rather poor. Antrectomy and Roux-en-Y reconstruction may be the better +option. + +Early and late dumping +Although considered together because the symptoms are sim-ilar, early and late dumping have different aetiologies. A com-mon feature, however, is early rapid gastric emptying. Many +patients have both early and late dumping. + +EARLY DUMPING +Early dumping consists of abdominal and vasomotor symp-toms that are found in about 10% of patients following gas-trectomy or vagotomy and drainage. It also affects a small percentage of patients following highly selective vagotomy due to the loss of receptive relaxation of the stomach. The small bowel is filled with foodstuffs from the stomach, which have a high osmotic load, and this leads to the sequestration of fluid from the circulation into the gastrointestinal tract. +This can be observed by the rise in the packed cell volume + + +Emil Theodor Kocher, 1841–1917, Professor of Surgery, Berne, Switzerland. +PART 11 | ABDOMINAL +1124 CHAPTER 63 Stomach and duodenum + + + +while the symptoms are present. All of the symptoms shown in Table 63.3 can be related to this effect on the gut and the circulation. +The principal treatment is dietary manipulation. Small, regular meals based on fat and protein are best, and avoid-ing fluids with a high carbohydrate content also helps. Fortu-nately, following operation, the syndrome tends to improve with time. For some reason, however, there is a group of patients who suffer intractable dumping regardless of any of these measures. The somatostatin analogue octreotide given before meals has been shown to be useful in some individuals and the long-acting preparation may also be useful. However, this treatment can lead to the development of gallstones and it does not help the diarrhoea from which many patients with dumping also suffer. +Revisional surgery may be occasionally required. In patients with a gastroenterostomy, the drainage may be taken down or, in the case of a pyloroplasty, repaired. Alternatively, antrectomy with Roux-en-Y reconstruction is often effective, although the procedure is of greater magnitude; following gas- +trectomy, it is the revisional procedure of choice. + +TABLE 63.3 Features of early and late dumping. Early Late +Incidence 5–10% 5% +Relation to meals Almost immediate Second hour after meal +Durations of attack 30–40 minutes 30–40 minutes +Relief Lying down Food +Aggravated by More food Exercise +Precipitating factor Food, especially As early dumping carbohydrate-rich +and wet +Major symptoms Epigastric fullness, Tremor, faintness, sweating, light- prostration headedness, +tachycardia, colic, sometimes diarrhoea + +LATE DUMPING +This is reactive hypoglycaemia. The carbohydrate load in the small bowel causes a rise in the plasma glucose, which, in turn, causes insulin levels to rise, causing a secondary hypoglycae-mia. This can be easily demonstrated by serial measurements of blood glucose in a patient following a test meal. The treat-ment is essentially the same as for early dumping. Octreotide +is very effective in dealing with this problem. +Post-vagotomy diarrhoea +This can be the most devastating symptom to afflict patients having peptic ulcer surgery. Most patients will suffer some looseness of bowel action to some degree (with the exception of highly selective vagotomy) but, in about 5%, it may be intractable. Despite much investigation, the precise aetiology of the problem is uncertain. It is related, to some degree, to rapid gastric emptying. In all probability, the denervation of +the upper gastrointestinal tract as a result of the vagotomy is + +also important. Exaggerated gastrointestinal peptide responses may also aggravate the condition. +The diarrhoea in post-vagotomy patients may take several forms. It may be severe and explosive, the patient experienc-ing a considerable degree of urgency. The patients sometimes describe the diarrhoea as feeling like passing boiling water. At the other extreme, some patients only have minor episodes of diarrhoea, which are not as directly related to food. +Many authors regard diarrhoea and dumping as being essentially the same problem. However, many patients with severe diarrhoea do not have any of the other symptoms of dumping and likewise some patients with dumping do not experience any significant diarrhoea. +The condition is difficult to treat. The patient should be managed as for early dumping and antidiarrhoeal preparations may be of some value. Octreotide is not effective in this con-dition and the results of revisional surgery are too unpredict- +able to make this an attractive treatment option. +Malignant transformation +Many large studies now confirm that operations such as gas-trectomy or vagotomy and drainage are independent risk fac-tors for the development of gastric cancer. The increased risk appears to be approximately four times that of the control population. +It is not difficult to understand the increased incidence of gastric cancer, as bile reflux gastritis, intestinal metaplasia and gastric cancer are linked. The lag phase between opera-tion and the development of malignancy is at least 10 years. Highly selective vagotomy does not seem to be associated +with an increased incidence of gastric cancer in the long term. +Nutritional consequences +Nutritional disorders are more common after gastrectomy than after vagotomy and drainage. Weight loss is common after gastrectomy and the patient may, in fact, never return to their original weight. Nutritional advice advising the taking of small meals is often more useful. Anaemia may be due to either iron or vitamin B12 deficiency. +Iron-deficiency anaemia occurs after both gastrectomy and vagotomy and drainage and is probably multifactorial in origin. Reduced iron absorption is probably the most import-ant factor, although the loss of blood from the gastric mucosa may also be important. Vitamin B12 deficiency is prone to occur after total gastrectomy. However, because of the very large vitamin B12 stores that most patients have, this may be very late in occurring. Vitamin B12 supplementation after total gastrectomy is essential. Rarely, vitamin B12 deficiency may occur after lesser forms of gastrectomy. In such patients the cause is probably a combination of reduced intrinsic factor production and also the fact that some patients have bacterial colonisation, which results in the destruction of the vitamin B12 in the gut. +Bone disease is seen principally after gastrectomy and mainly in women. The condition is essentially indistinguish-able from the osteoporosis commonly seen in post-meno-pausal women. It is only the frequency and magnitude of the disorder that distinguish it. Treatment is with dietary supple- +mentation, calcium and vitamin D, and exercise. +PART 11 | ABDOMINAL +Peptic ulcer 1125 + + + +Gallstones +The development of gallstones is strongly associated with truncal vagotomy. Following truncal vagotomy the biliary tree, as well as the stomach, is denervated, leading to stasis and hence stone formation. Patients developing symptomatic gallstones will require cholecystectomy. However, this may induce or worsen other postpeptic ulcer surgery syndromes +, +such as bilious vomiting and postvagotomy diarrhoea. + +The complications of peptic +ulceration +The common complications of peptic ulcer are perforation, bleeding and stenosis. Bleeding and stenosis are considered +below in the relevant sections. +Perforated peptic ulcer EPIDEMIOLOGY +Despite the widespread use of gastric antisecretory agents and eradication therapy, the incidence of perforated peptic ulcer has changed little. However, there has been a con-siderable change in the epidemiology of perforated peptic ulcer in resource-rich countries over the last two decades. Previously, most patients were middle aged, with a ratio of 2:1 of male:female. With time there has been a steady increase in the age of the patients suffering this complication and an increase in the numbers of females, such that perforations now occur most commonly in elderly female patients. NSAIDs +appear to be responsible for most of these perforations. + +CLINICAL FEATURES +The classical presentation of perforated duodenal ulcer is instantly recognisable (Figure 63.22). The patient, who may have a history of peptic ulceration, develops sudden-onset severe generalised abdominal pain due to the irritant effect of gastric acid on the peritoneum. Although the contents of an acid-producing stomach are relatively low in bacterial +load, bacterial peritonitis supervenes over a few hours, usually + + + + + + + + + + + + + + +Figure 63.22 A sketch of Mr Hamilton Bailey watching for abdom-inal movement on respiration. In the case of a classically presenting perforated ulcer, the abdominal movement is restricted or absent. + +accompanied by a deterioration in the patient’s condition. Initially, the patient may be shocked with a tachycardia but a pyrexia is not usually observed until some hours after the event. The abdomen exhibits a board-like rigidity and the patient is disinclined to move because of the pain. The abdo-men does not move with respiration. Patients with this form of presentation need an operation, without which the patient will deteriorate with a septic peritonitis. +This classical presentation of the perforated peptic ulcer is observed less commonly than in the past. Very frequently the elderly patient who is taking NSAIDs will have a less dramatic presentation, perhaps because of the use of potent anti-inflammatory drugs (steroids). The board-like rigidity seen in the abdomen of younger patients may also not be observed and a higher index of suspicion is necessary to make the correct diagnosis. In other patients, the leak from the ulcer may not be massive. They may present only with pain in the epigastrium and right iliac fossa as the fluid may track down the right paracolic gutter. Sometimes perforations will seal owing to the inflammatory response and adhesion within the abdominal cavity, and so the perforation may be self-limiting. All of these factors may combine to make the diagnosis of perforated peptic ulcer difficult. +By far the most common site of perforation is the anterior aspect of the duodenum. However, the anterior or incisural gastric ulcer may perforate and, in addition, gastric ulcers may perforate into the lesser sac, which can be particularly difficult +to diagnose. These patients may not have obvious peritonitis. + +INVESTIGATIONS +An erect plain chest radiograph will reveal free gas under the diaphragm in excess of 50% of cases with perforated peptic ulcer (Figure 63.23) but CT imaging is more accurate (see below). All patients should have serum amylase performed, as distinguishing between peptic ulcer, perforation and pancre-atitis can be difficult. Measuring the serum amylase, however, may not remove the diagnostic difficulty. It can be elevated following perforation of a peptic ulcer although, fortunately, +the levels are not usually as high as the levels commonly seen + + + + + + + + + + + + + + + +Figure 63.23 Plain abdominal radiograph of a perforated ulcer, showing air under the diaphragm. + + + +Henry Hamilton Bailey, 1894–1961, surgeon, The Royal Northern Hospital, London, UK. +PART 11 | ABDOMINAL +1126 CHAPTER 63 Stomach and duodenum + + + +in acute pancreatitis. Several other investigations are useful if doubt remains. A CT scan will normally be diagnostic in +both conditions. + +TREATMENT +The initial priorities are resuscitation and analgesia. Analge-sia should not be withheld for fear of removing the signs of an intra-abdominal catastrophe. In fact, adequate analgesia makes the clinical signs more obvious. It is important, how-ever, to titrate the analgesic dose. Following resuscitation, the treatment is principally surgical. Laparotomy is performed, usually through an upper midline incision if the diagnosis of perforated peptic ulcer can be made with confidence. This is not always possible and hence it may be better to place a small incision around the umbilicus to localise the perfora-tion with more certainty. Alternatively, laparoscopy may be used. The most important component of the operation is a thorough peritoneal toilet to remove all of the fluid and food debris. If the perforation is in the duodenum it can usually be closed by several well-placed sutures, closing the ulcer in a transverse direction as with a pyloroplasty. It is important that sufficient tissue is taken in the suture to allow the edges to be approximated, and the sutures should not be tied so tight that they tear out. It is common to place an omental patch over the perforation in the hope of enhancing the chances of the leak sealing. If the perforation is difficult to close primarily it is frequently possible to seal the leak with an omental patch alone, and many surgeons now employ this strategy for all per-forations. When securing the omental patch it is important not to tie the sutures too tight so as to obliterate the omental blood supply. Gastric ulcers should, if possible, be excised and closed, so that malignancy can be excluded. Occasionally a patient is seen who has a massive duodenal or gastric perfor-ation such that simple closure is impossible; in these patients a distal gastrectomy with Roux-en-Y reconstruction is the procedure of choice. +All patients should be treated with systemic antibiotics in addition to a thorough peritoneal lavage. In the past, many surgeons performed definitive procedures such as either trun-cal vagotomy and pyloroplasty or, more recently and proba-bly more successfully, highly selective vagotomy during the course of an operation for a perforation. Studies show that in well-selected patients and in expert hands this is a very safe strategy. However, nowadays, surgery is confined to first-aid measures most commonly, and the peptic ulcer is treated med-ically as described earlier in this chapter. Following operation, gastric antisecretory agents should be started immediately. H. pylori eradication is mandatory. +Perforated peptic ulcers can often be managed by mini-mally invasive techniques if the expertise is available. The principles of operation are, however, the same; thorough peritoneal toilet is performed and the perforation is closed by intracorporeal suturing. Whatever technique is used, it is important that the stomach is kept empty postoperatively by nasogastric suction, and that gastric antisecretory agents are +commenced to promote healing in the residual ulcer. + +A great deal has been written about the conservative man-agement of perforated ulcer. Some writers say that virtually all patients can be managed conservatively, whereas most sur-geons have difficulty in understanding how a patient who is ill with widespread peritonitis and who has food debris widely distributed through the abdominal cavity will improve with-out an operation. However, undoubtedly, there are patients who have small leaks from a perforated peptic ulcer and rela-tively mild peritoneal contamination, who may be managed with intravenous fluids, nasogastric suction and antibiotics. These patients are in the minority. A number of factors have been associated with poor outcome after perforated peptic +ulcer, including: +● delay in diagnosis (>24 hours); ● medical comorbidities; +● shock; +● increasing age (>75). +There is little evidence to advocate the conservative man-agement of patients who exhibit any of these characteristics. +Patients who have suffered one perforation may suffer another one. Therefore, they should be managed aggressively to ensure that this does not happen. Lifelong treatment with proton pump inhibitors is a reasonable option especially in +those who have to continue with NSAID treatment. + +HAEMATEMESIS AND MELAENA +Upper gastrointestinal haemorrhage remains a major medical problem with an incidence over 100/100 000 per year in west-ern practice that increases with increasing age. Haemorrhage is strongly associated with NSAID use. Despite improvements in diagnosis and the proliferation in treatment modalities over the last few decades, an in-hospital mortality of 5–10% can be expected. This rises to 33% when bleeding is first observed in patients who are hospitalised for other reasons. In patients in whom the cause of bleeding can be found, the most com-mon causes are peptic ulcer, erosions, Mallory–Weiss tear and bleeding oesophageal varices (Table 63.4). +Whatever the cause, the principles of management are identical. First, the patient should be adequately resusci-tated and, following this, the patient should be investigated urgently to determine the cause of the bleeding. Only then should treatment of a definitive nature be instituted. For any significant gastrointestinal bleed, intravenous access should be established and, for those with severe bleeding, central venous pressure monitoring should be set up and bladder cath-eterisation performed. Blood should be cross-matched and the patient transfused as clinically indicated, usually when >30% of blood volume has been lost. There is no evidence for the use of intravenous proton pump inhibitors prior to endoscopy. As a general rule, most gastrointestinal bleeding will stop, albeit temporarily, but there are sometimes instances when this is not the case. In these circumstances, resuscitation, diagnosis and treatment should be carried out simultaneously. +There are occasions when life-saving manoeuvres have to + + +George Kenneth Mallory, b.1926, Professor of Pathology, Boston, MA, USA. Soma Weiss, 1899–1942, Professor of Medicine, Boston, MA, USA. +PART 11 | ABDOMINAL Haematemesis and melaena 1127 + + + +TABLE 63.4 Causes of upper gastrointestinal bleeding. Condition % Ulcers 60 Oesophageal 6 +Gastric 21 Duodenal 33 Erosions 26 Oesophageal 13 Gastric 9 Duodenal 4 Mallory–Weiss tear 4 Oesophageal varices 4 +Tumour 0.5 Vascular lesions, e.g. Dieulafoy’s disease 0.5 Others 5 + + +be undertaken without the benefit of an absolute diagnosis. For instance, in patients with known oesophageal varices and uncontrollable bleeding, a Sengstaken–Blakemore tube may be inserted before an endoscopy has been carried out. This practice is not to be encouraged, except in extremis. In some patients, bleeding is secondary to a coagulopathy. The most important current causes of this are liver disease and inadequately controlled warfarin therapy. In these circum-stances the coagulopathy should be corrected, if possible, with fresh-frozen plasma or concentrated clotting factors. +Upper gastrointestinal endoscopy should be carried out by an experienced operator as soon as practicable after the patient has been stabilised. In patients in whom the bleed-ing is relatively mild, endoscopy may be carried out on the morning after admission; this is usually guided by local policy. In all cases of severe bleeding it should be carried out imme-diately. A number of scoring systems have been advocated for the assessment of rebleeding and death after upper gastroin-testinal haemorrhage. Perhaps the most useful of these is the Rockall score. This can be used in a pre-endoscopy format to stratify patients to safe early discharge and postendoscopy it +can relatively accurately predict rebleeding and death. + +Bleeding peptic ulcers +The epidemiology of bleeding peptic ulcers exactly mirrors that of perforated ulcers. In recent years, the population affected has become much older and the bleeding is com-monly associated with the ingestion of NSAIDs. Diagnosis can normally be made endoscopically, although occasionally the nature of the blood loss precludes accurately identifying the lesion. However, the more experienced the endoscopist, +the less likely this is to be a problem. + +Medical and minimally interventional treatments +Medical treatment has limited efficacy. All patients are com-monly started on either an H2-antagonist or a proton pump antagonist, and recent evidence confirms the benefit of pro-ton pump inhibitor administration to prevent rebleeding after endoscopy. Furthermore, meta-analysis of studies suggests that tranexamic acid, an inhibitor of fibrinolysis, may reduce overall mortality +. +Therapeutic endoscopy can achieve haemostasis in approximately 70% of cases, with the best evidence support-ing a combination of adrenaline injection with heater probe and/or clips. Therapeutic endoscopy will probably never be effective in patients who are bleeding from large vessels and with which the majority of the mortality is associated. +In patients where the source of bleeding cannot be iden-tified or in those who rebleed after endoscopy, angiography with transcatheter embolisation may offer a valuable alter-native to surgery in expert centres. The risk of significant ischaemia following embolisation is low because of the rich collateral blood supply of the stomach and duodenum. The surgeon should be mindful that rescue surgery after failed embolisation is associated with poor outcome and it may be +advantageous to proceed directly to surgery. + +Surgical treatment +Criteria for surgery are well worked out. A patient who continues to bleed requires surgical treatment. The same applies to a significant rebleed. The only exception applies in expert centres with 24-hour interventional radiology and experience of angiographic embolisation where attempts may be made to arrest bleeding and avoid surgery. The surgical team should care for these patients and an operation should not be delayed if any concerns remain. Patients with a visible vessel in the ulcer base, a spurting vessel or an ulcer with a clot in the base are statistically likely to require surgical treatment to stop the bleeding. Elderly and unfit patients are more likely to die as a result of bleeding than younger patients. Ironically, they should have early surgery. A patient who has required more than six units of blood in general needs surgical treatment. +The aim of the operation is to stop the bleeding. The advent of endoscopy has greatly helped in the management of upper gastrointestinal bleeding as a surgeon can usually be confident about the site of bleeding prior to operation. The most common site of bleeding from a peptic ulcer is the duodenum. In tackling this, it is essential that the duodenum is fully mobilised. This should be done before the duodenum is opened as it makes the ulcer much more accessible and also allows the surgeon’s hand to be placed behind the gastroduodenal artery, which is commonly the source of major bleeding. Following mobilisation, the duodenum, +and usually the pylorus, is opened longitudinally as in a + + + +Robert William Sengstaken, b.1923, surgeon, Garden City, NY, USA. +Arthur Hendley Blakemore, 1897–1970, Associate Professor of Surgery, the Columbia College of Physicians and Surgeons, New York, NY, USA. Sengstaken and Blakemore originally described the tube in 1950. +Timothy A Rockall, contemporary, Professor of Surgery University of Surrey, UK. +, +PART 11 | ABDOMINAL +1128 CHAPTER 63 Stomach and duodenum + + + +pyloroplasty. This allows good access to the ulcer, which is usually found posteriorly or superiorly. Accurate haemostasis is important and can be achieved initially by direct pressure. It is the vessel within the ulcer that is bleeding and this should be controlled using well-placed sutures on a small round-bodied needle that under-run the vessel. The placing of more and more inaccurately positioned sutures is counter-productive. Following under-running, it is often possible to close the mucosa over the ulcer. The pyloroplasty is then closed with interrupted sutures in a transverse direction as in the usual fashion. In a giant ulcer the first part of the duodenum may be destroyed making primary closure impossible. In this circumstance one should proceed to distal gastrectomy with Roux-en-Y reconstruction. The duodenal stump may then be closed using the Nissen technique with T-tube drainage. +The principles of management of bleeding gastric ulcers are essentially the same. The stomach is opened at an appropriate position anteriorly and the vessel in the ulcer under-run. If the ulcer is not excised then a biopsy of the edge needs to be taken to exclude malignant transformation. Sometimes the bleeding is from the splenic artery and if there is a lot of fibrosis present then the operation may be challenging. However, most patients can be managed by conservative surgery. Gastrectomy for bleeding has been widely practised in the past, but is associated with a high perioperative mortality even if the incidence of recurrent bleeding is less. +Bearing in mind that most patients nowadays are elderly and unfit, the minimum surgery that stops the bleeding is probably optimal (damage control surgery). Acid can be inhibited by pharmacological means and appropriate eradication therapy will prevent ulcer recurrence. Definitive acid-lowering surgery is not now required. Patients on long- +term NSAIDs can be managed as outlined earlier. + +Stress ulceration +This commonly occurs in patients with major injury or ill-ness, who have undergone major surgery or who have major comorbidity. Many such patients are found in intensive care units. There seems little doubt that the incidence of this problem has reduced in recent years due to the widespread use of prophylaxis. Acid inhibition and the nasogastric or oral administration of sucralfate has been shown to reduce the incidence of stress ulceration. There is no doubt that it is far better to prevent this condition than to try to treat it once it occurs. Endoscopic means of treating stress ulceration may be ineffective and operation may be required. The principles of +management are the same as for the chronic ulcer. + +Gastric erosions +Erosive gastritis has a variety of causes, especially NSAIDs. +Fortunately, most such bleeding settles spontaneously, but + +when it does not it can be a major problem to treat. In general terms, although there is a diffuse erosive gastritis, there is one (or more) specific lesion that has a significant-sized vessel within it. This should be dealt with appropriately, preferably +endoscopically, but sometimes surgery is necessary. + +Mallory–Weiss tear +This is a longitudinal tear at the gastro-oesophageal junc-tion, which is induced by repetitive and strenuous vomiting. Doubtless, many such lesions occur and do not cause bleed-ing. When it is a cause of haematemesis, the lesion may often be missed as it can be difficult to see as it is just below the gastro-oesophageal junction, a position that can be difficult for the inexperienced endoscopist. Occasionally these lesions continue to bleed and require surgical treatment. Often the situation arises in which the surgeon does not have guidance from the endoscopists as regards the site of bleeding, and a high index of suspicion in such circumstances is important. The experienced surgeon will perform on-table endoscopy prior to embarking on surgery. The stomach is opened by longitudinal gastrotomy and the upper section is carefully inspected. It is normally possible to palpate the longitudinal mucosal tear with a little induration at the edges, which gives a clue to the lesion’s location. Under-running is all that is +required. + +Dieulafoy’s disease +This is essentially a gastric arterial venous malformation that has a characteristic histological appearance. Bleeding due to this malformation is one of the most difficult causes of upper gastrointestinal bleeding to treat. The lesion itself is covered by normal mucosa and, when not bleeding, it may be invis-ible. If it can be seen while bleeding, all that may be visible is profuse bleeding coming from an area of apparently normal mucosa. If this occurs, the cause is instantly recognisable. If the lesion can be identified endoscopically there are various means of dealing with it, including injection of sclerosant and endoscopic clips. If it is identified at operation then only a local excision is necessary. Occasionally, a lesion is only rec-ognised after gastrectomy and sometimes not even then. The pathologist, as well as the endoscopist, may have difficulty in +finding it. + +Tumours +All of the gastric tumours described below may present with chronic or acute upper gastrointestinal bleeding. Bleeding is not normally torrential but can be unremitting. Gastric stro-mal tumours commonly present with bleeding and have a characteristic appearance, as the mucosa breaks down over the tumour in the gastric wall (Figure 63.24). Whatever the +nature, the tumours should be dealt with as appropriate. + + + +Rudolph Nissen, 1896–1981, Professor of Surgery, Istanbul, T rkey, and later at Basle, Switzerland. Georges Dieulafoy, 1839–1911, physician, Paris, France. +u +PART 11 | ABDOMINAL +Gastric outlet obstruction 1129 + + +not always massive, although it can be. Very often there is nothing much to distinguish between the bleeding from the aortic enteric fistula and any other recurrent upper gastroin-testinal bleeding. The vast majority of patients will have had an aortic graft and, in the absence of this, the diagnosis is unlikely. However, it is occasionally seen in patients with an untreated aortic aneurysm. A well-performed CT scan will commonly allow the diagnosis to be made with certainty. The condition should be managed by an expert vascular surgeon as, whether secondary or primary, the morbidity and mortality +are high. + + + + + + + + + +Figure 63.24 Smooth muscle tumour of the stomach, with ulcer-ation. + + +Portal hypertension and portal +gastropathy +The management of bleeding gastric varices is very challeng-ing. Fortunately, most bleeding from varices is oesophageal and this is much more amenable to sclerotherapy, banding and balloon tamponade. Gastric varices may also be injected, although this is technically more difficult. Banding can also be used, again with difficulty. The gastric balloon of the Sengstaken–Blakemore tube can be used to arrest the hae-morrhage if it is occurring from the fundus of the stomach or gastro-oesophageal junction. Octreotide is a somatostatin analogue that reduces portal pressure in patients with varices, and trials suggest that it is of value in arresting haemorrhage in these patients, although its overall effect on mortality remains in doubt. Glypressin is also said to be of use. +Most surgeons prefer to avoid acute surgery on bleeding varices as, in contrast with elective operations for portal hypertension, acute shunts are attended by considerable oper-ative mortality. For this reason the acute TIPSS procedure (transjugular intrahepatic portosystemic shunt), which is described in Chapter 65, can be an extremely useful, although +technically demanding, procedure. + +Portal gastropathy +Portal gastropathy is essentially the same disease process as described above. The mucosa is affected by the increased portal pressure and may exude blood, even in the absence of +well-developed visible varices. The treatment is as above. + +Aortic enteric fistula +This diagnosis should be considered in any patient with hae-matemesis and melaena that cannot be otherwise explained. +Contrary to expectation, the bleeding from such patients is + +GASTRIC OUTLET +OBSTRUCTION +The two common causes of gastric outlet obstruction are gastric cancer (see below) and pyloric stenosis secondary to peptic ulceration. Previously, the latter was more common. Now, with the decrease in the incidence of peptic ulceration and the advent of potent medical treatments, gastric outlet obstruction should be considered malignant until proven otherwise, at least in resource-rich countries. +The term ‘pyloric stenosis’ is normally a misnomer. The stenosis is seldom at the pylorus. Commonly, when the con-dition is due to underlying peptic ulcer disease, the stenosis is found in the first part of the duodenum, the most common site for a peptic ulcer. True pyloric stenosis can occur due to fibrosis around a pyloric channel ulcer. However, in recent years the most common cause of gastric outlet obstruction has been gastric cancer. In this circumstance the metabolic con-sequences may be somewhat different from those of benign pyloric stenosis because of the relative hypochlorhydria found +in patients with gastric cancer. + + +Summary box 63.5 + +Gastric outlet obstruction +●● Gastric outlet obstruction is most commonly associated with longstanding peptic ulcer disease and gastric cancer +●● The metabolic abnormality of hypochloraemic alkalosis is usually only seen with peptic ulcer disease and should be treated with isotonic saline with potassium +●● Endoscopic biopsy is essential to determine whether the cause of the problem is malignancy +●● Aggressive medical therapy for peptic ulcer disease often leads to resolution +●● Endoscopic dilatation of the gastric outlet may be effective in less severe cases of benign stenosis +●● Operation is frequently required, with a drainage procedure being performed for benign disease and appropriate resectional surgery if malignant + + +Clinical features +In benign gastric outlet obstruction there is usually a long history of peptic ulcer disease. Nowadays, as most patients with peptic ulcer symptoms are treated medically, it is easy +to understand why the condition is becoming much less +PART 11 | ABDOMINAL +1130 CHAPTER 63 Stomach and duodenum + + + +common. In some patients the pain may become unremitting and in other cases it may largely disappear. The vomitus is characteristically unpleasant in nature and is totally lacking in bile. Very often it is possible to recognise foodstuff taken several days previously. The patient commonly complains of losing weight, and appears unwell and dehydrated. When examining the patient, it may be possible to see the distended stomach and a succussion splash may be audible on shaking +the patient’s abdomen. + +Metabolic effects +These are most interesting, as the metabolic consequences of benign pyloric stenosis are unique. The vomiting of hydro-chloric acid results in hypochloraemic alkalosis. Initially the sodium and potassium may be relatively normal. However, as dehydration progresses, more profound metabolic abnor-malities arise, partly related to renal dysfunction. Initially, the urine has a low chloride and high bicarbonate content, reflecting the primary metabolic abnormality. This bicarbon-ate is excreted along with sodium, and so with time the patient becomes progressively hyponatraemic and more profoundly dehydrated. Because of the dehydration, a phase of sodium retention follows and potassium and hydrogen are excreted in preference. This results in the urine becoming paradoxically acidic and hypokalaemia ensues. Alkalosis leads to a lowering +in the circulating ionised calcium, and tetany can occur. + +Management +Treating the patient involves correcting the metabolic abnor-mality and dealing with the mechanical problem. The patient should be rehydrated with intravenous isotonic saline with potassium supplementation. Replacing the sodium chloride and water allows the kidney to correct the acid–base abnor-mality. Following rehydration, it may become obvious that the patient is also anaemic, the haemoglobin being spuriously high on presentation. +It is notable that the metabolic abnormalities may be less if the obstruction is due to malignancy, as the acid–base dis-turbance is less pronounced. +The stomach should be emptied using a wide-bore gastric tube. A large nasogastric tube may not be sufficiently large to deal with the contents of the stomach, and it may be neces-sary to pass an orogastric tube and lavage the stomach until it is completely emptied. This then allows investigation of the patient with endoscopy and contrast radiology. Biopsy of the area around the pylorus is essential to exclude malignancy. The patient should also have a gastric antisecretory agent, initially given intravenously to ensure absorption. +Early cases may settle with conservative treatment, pre-sumably as the oedema around the ulcer diminishes as the ulcer is healed. Traditionally, severe cases are treated surgi-cally, usually with a gastroenterostomy rather than a pyloro-plasty. Endoscopic treatment with balloon dilatation has been practised and may be most useful in early cases. However, this treatment is not devoid of problems. Dilating the duodenal stenosis may result in perforation. The dilatation may have +to be performed several times and may not be successful in + +the long term. Occasionally duodenal stent insertion will be +considered in specialist centres. + +Other causes of gastric outlet +obstruction +Adult pyloric stenosis +This is a rare condition and its relationship to the childhood condition is unclear, although some patients have a long his-tory of problems with gastric emptying. It is commonly treated +by pyloroplasty rather than pyloromyotomy. +Pyloric mucosal diaphragm +The origin of this rare condition is unknown. It usually does not become apparent until middle life. When found, simple +excision of the mucosal diaphragm is all that is required. + +GASTRIC POLYPS +A number of conditions manifest as gastric polyps. Their main importance is that they may actually represent early gastric cancer. Biopsy is essential. +The most common type of gastric polyp is metaplastic. These are associated with H. pylori infection and regress following eradication therapy. Inflammatory polyps are also common. Fundic gland polyps deserve particular attention. They seem to be associated with the use of proton pump inhibitors and are also found in patients with familial polyp-osis. None of the above polypoid lesions has proven malig-nant potential. True adenomas have malignant potential and should be removed, but they account for only 10% of polyp-oid lesions. Gastric carcinoids arising from the ECL cells are seen in patients with pernicious anaemia and usually appear +as small polyps. + +GASTRIC CANCER +Carcinoma of the stomach is a major cause of cancer mortal-ity worldwide. Its prognosis tends to be poor, with cure rates little better than 5–10%, although better results are obtained in Japan, where the disease is common. Gastric cancer is actu-ally an eminently curable disease provided that it is detected at an appropriate stage and treated adequately. It rarely dis-seminates widely before it has involved the lymph nodes and, therefore, there is an opportunity to cure the disease prior to dissemination. Early diagnosis is therefore the key to suc-cess with this disease. Unfortunately, the late presentation of many cases is the cause of the poor overall survival figures. The only treatment modality able to cure the disease is resec- +tional surgery. + +Incidence +There are marked variations in the incidence of gastric cancer worldwide. In the UK it is approximately 15 per 100000 per year, in the USA 10 per 100000 per year and in Eastern Europe 40 per 100000 per year. In Japan, the disease is much +more common, with an incidence of approximately 70 per +PART 11 | ABDOMINAL +Gastric cancer 1131 + + + + +Summary box 63.6 + +Gastric cancer +●● Gastric cancer is one of the most common causes of cancer death in the world +●● The outlook is generally poor, owing to the advanced stage of the tumour at presentation +●● Better results are obtained in Japan, which has a high population incidence, screening programmes and a high-quality surgical treatment +●● The aetiology of gastric cancer is multifactorial, but H. pylori is an important factor for distal but not proximal gastric cancer +●● Early gastric cancer is associated with very high cure rates +●● Gastric cancer can be classified into intestinal and diffuse types, the latter having a worse prognosis +●● In western resource-rich countries, proximal gastric cancer is now more common than distal cancer and is usually of the diffuse type +●● Spread may be by lymphatics, blood, transcoelomic or direct, but distant metastases are uncommon in the absence of lymph node involvement +●● The treatment of curable cases is by radical surgery and removal of the second tier of nodes (around the principal arterial trunks) may be advantageous +●● Gastric cancer is chemosensitive and chemotherapy improves survival in patients having surgery for the condition and in advanced disease + + +100000 per year, and there are small geographical areas in China where the incidence is double that in Japan. These underlying epidemiological data make it clear that this is an environmental disease. In general, men are more affected by the disease than women and, as with most solid organ malignancies, the incidence increases with age. +At present, marked changes are being observed in resource-rich countries in terms of the incidence and site of gastric cancer and the population affected, changes that to date have not been observed in Japan. First, the incidence of gastric cancer is continuing to fall at about 1% per year. This reduction exclusively affects carcinoma arising in the body and distal stomach. In contrast, there appears to be an increase in the incidence of carcinoma in the proximal stomach, particularly the oesophagogastric junction. Carci-noma of the distal stomach and body of the stomach is most common in low socioeconomic groups, whereas the increase in proximal gastric cancer seems to affect principally higher socioeconomic groups. Proximal gastric cancer does not seem to be associated with H. pylori infection, in contrast with car- +cinoma of the body and distal stomach. + +Aetiology +Gastric cancer is a multifactorial disease (Correa). Epidemi-ological studies point to a role for H. pylori, although there is argument about how important this factor is. Studies reveal a +correlation between the incidence of gastric cancer in various + +populations and the prevalence of H. pylori infection, but other factors are also important. There is insufficient evi-dence at the moment to support eradication programmes in asymptomatic patients who are infected with Helicobacter, with a view to reducing the population incidence of gastric cancer. However, clinical trials may subsequently change this view. As mentioned above, Helicobacter seems to be princi-pally associated with carcinoma of the body, stomach and distal stomach rather than the proximal stomach. As Heli-cobacter is associated with gastritis, gastric atrophy and intes-tinal metaplasia, the association with malignancy is perhaps not surprising. +Several other risk factors have been identified as being important in the aetiology of gastric cancer. Patients with pernicious anaemia and gastric atrophy are at increased risk, as are those with gastric polyps. Patients who have had pep-tic ulcer surgery, particularly those who have had drainage procedures such as Billroth II or Polya gastrectomy, gastroen-terostomy or pyloroplasty, are at approximately four times the average risk. Presumably duodenogastric reflux and reflux gas-tritis are related to the increased risk of malignancy in these patients. Intestinal metaplasia is a risk factor. Carcinoma is associated with cigarette smoking and dust ingestion from a variety of industrial processes. Diet appears to be important, as illustrated by the often quoted example of the change in the incidence of gastric cancer in Japanese families living in the USA. The high incidence of gastric cancer in some pockets in China is probably environmental and probably diet related. The ingestion of substances such as spirits may induce gastritis and, in the long term, cancer. Excessive salt intake, deficiency of antioxidants and exposure to N-nitroso compounds are also related. The aetiology of proximal gastric cancer remains an enigma. It is not associated with Helico-bacter but is associated with obesity and higher socioeconomic status. Genetic factors are also important but imperfectly elu- +cidated (see below). + +Clinical features +The features of advanced gastric cancer are usually obvious. However, curable gastric cancer has no specific features to dis-tinguish it symptomatically from benign dyspepsia. The key to improving the outcome of gastric cancer is early diagnosis and, although in Japan there is a screening programme, most curable cases are picked up by the liberal use of gastroscopy in patients with dyspepsia. In western resource-rich countries it is much more difficult as the population incidence is much lower. Hence the cost effectiveness of performing gastroscopy for mild dyspeptic symptoms is low. However a high index of suspicion is necessary as only endoscoping patients with symptoms of advanced cancer is unlikely to be beneficial as such patients are not surgically curable. It is important to note that gastric antisecretory agents will improve the symptoms of gastric cancer so the disease should be excluded, preferably +before therapy is started. + + + +P. Correa, contemporary, pathologist, New Orleans, LA, USA. Produced a cogent hypothesis to explain the development of the intestinal type of gastric cancer. +PART 11 | ABDOMINAL +1132 CHAPTER 63 Stomach and duodenum + + + +In advanced cancer, early satiety, bloating, distension and vomiting may occur. The tumour frequently bleeds, resulting in iron deficiency anaemia. Obstruction leads to dysphagia, epigastric fullness or vomiting. Weight loss can be profound. With pyloric involvement the presentation may be of gastric outlet obstruction, although the alkalosis is usually less pro-nounced or absent compared with when duodenal ulceration leads to obstruction. In recent years, gastric outlet obstruction is more commonly associated with malignancy than benign disease. Non-metastatic effects of malignancy are seen, par-ticularly thrombophlebitis (Trousseau’s sign) and deep venous thrombosis. These features result from the effects of the tumour on thrombotic and haemostatic mechanisms. + +Site +The proximal stomach is now the most common site for gastric cancer in resource-rich western countries. Because so many malignancies occur at the oesophageal–gastric junction, and because the lower oesophagus is also a very common site of adenocarcinoma, it is artificial to separate the stomach from the oesophagus. Therefore, it is best to consider the whole of the upper gastrointestinal tract from the cricopharyngeus to the pylorus. The incidence of cancer at these various sites is shown in Figure 63.25. It can be seen that just under 60% of all of the malignancies occurring in the oesophagus and stomach occur in proximity to the oesophagogastric junction. Adenocarcinoma at this site has doubled in incidence in the UK over the last 30 years. This high prevalence of proximal gastric cancer is not seen in Japan, where distal cancer still predominates, as it does in most of the rest of the world. + + + + +Upper oesophagus 2% + +Pathology +The most useful clinicopathological classification of gastric cancer is the Lauren classification. In this system there are principally two forms of gastric cancer: intestinal gastric can-cer and diffuse gastric cancer (often with signet ring cells). In intestinal gastric cancer, the tumour resembles a carcinoma elsewhere in the tubular gastrointestinal tract and forms poly-poid tumours or ulcers. It probably arises in areas of intestinal metaplasia. In contrast, diffuse gastric cancer infiltrates deeply into the stomach without forming obvious mass lesions, but spreads widely in the gastric wall. Not surprisingly, this has a much worse prognosis. A small proportion of gastric cancers are of mixed morphology. +Gastric cancer can be divided into early gastric cancer and advanced gastric cancer. Early gastric cancer is defined as cancer limited to the mucosa and submucosa with or without lymph node involvement (T1, any N). The classification is shown in Figures 63.26 and 63.27. This can be either pro-truding, superficial or excavated in the Japanese classification. This type of cancer is eminently curable, and even early gas-tric cancers associated with lymph node involvement have 5-year survival rates in the region of 90%. In Japan, approxi-mately one-third of gastric cancers diagnosed are in this stage. However, in the UK it is uncommon to detect gastric cancers at this stage. A number of reasons probably still account for this. First, because gastric cancer is less common in the UK, dyspeptic patients are not always referred for endoscopy at an appropriate stage. Second, endoscopists are unfamiliar with the appearances of early gastric cancer and in all probability many such cases are missed. +Advanced gastric cancer involves the muscularis. Its mac-roscopic appearances have been classified by Bormann into four types (Figures 63.28 and 63.29). Types III and IV are +commonly incurable. + + + +Mid-oesophagus 6% Type I Type IIa + +Lower oesophagus 22% + + +OG junction 18% Cardia 17% + +Approximately 60% +Type IIb Type IIc + + + +Body 15% +Type III + + +Type III and IIc + + + +Pylorus 7% Antrum 13% + +Figure 63.25 The incidence of cancer in the various parts of the upper gastrointestinal tract in the UK. OG, oesophagogastric. + + +Cancer + +Figure 63.26 Early gastric cancer, Japanese classification. + + + + +Armand Trousseau, 1801–1867, physician, Hôtel Dieu, Paris, France. This sign led him to suspect himself of having gastric cancer. He actually had pancreatic cancer, which was diagnosed at postmortem. +Pekka A. Lauren, contemporary, pathologist, University of T rku, Finland. +u +PART 11 | ABDOMINAL +Gastric cancer 1133 + + +(a) (b) + + + + + + + +(c) + + + + + + + + + + + + + + + + +Figure 63.27 Early gastric cancer: (a) type I; (b) type IIa; (c) type III (courtesy of Dr GNJ Tytgat, Amsterdam, The Netherlands). + + + + +Type I Type II + + + + + + +Type III Type IV + + + + +Cancer +Figure 63.28 Bormann classification of advanced gastric cancer. + + + + + + +The molecular pathology of gastric +cancer +Our understanding of the molecular pathology of gastric cancer has been revolutionised by the application of next-generation sequencing platforms to the disease. In a recent landmark paper The Cancer Genome Atlas (TCGA) group described four molecular subtypes of gastric cancer: Epstein– +Barr virus positive, microsatellite instable, genomically stable + + +and chromosomal instability. Recognition of these sub-groups and their underlying common gene mutations and driver events is leading to the development of targeted therapies including immunotherapies. Similar genetic classifications are now available for other tumours of the gastrointestinal tract meaning that novel treatments can be applied across tumour types. A range of mutations in genes related to genome integ-rity (e.g. BRCA2, TP53, ARID1A), chromatin remodelling +(e.g. SMARCA1, CHD3&4) and cell–cell adhesion and +PART 11 | ABDOMINAL +. +1134 CHAPTER 63 Stomach and duodenum + + +(a) (b) + + + + + + + + + + + + + + + + + + + + + +(c) (d) + + + + + + + + + + + + + + + + + + + + +Figure 63.29 Advanced gastric cancer: (a) type I; (b) type II; (c) type III; (d) type IV (linitus plastica) (courtesy of Dr GNJ Tytgat, Amsterdam, The Netherlands). + + + +motility (e.g. RHOA, CDH1) have been described in gas-tric cancer. In addition, cell signalling pathways commonly mutated in other solid organ tumours are commonly found perturbed in gastric cancer For example, the Wnt path-way may be amenable to specific small molecule inhibitors. Unsurprisingly, gastric cancer exhibits a range of mutations in receptor tyrosine kinases and PI3K/MAPK signalling. +The rapid development of sequencing technologies, includ-ing single-cell platforms and the development of real-time +sequencing, offers the promise of precision therapy for gastric + +cancer in the future, but currently treatment is still based on +surgery with or without conventional chemo/radiotherapy. + + +Staging +The International Union Against Cancer (UICC) staging system is shown in Table 63.5. Important changes have been made in the seventh edition of the TNM staging system that +are worthy of discussion. In an attempt to reflect the current +PART 11 | ABDOMINAL +Gastric cancer 1135 + + +TABLE 63.5 UICC staging of gastric cancer. Spread of carcinoma of the stomach + +T1 + + + +T2 +T3 +T4a +T4b +N0 +N1 +N2 +N3a +N3b +M0 +M1 + +Staging +IA +IB +T2 +IIA +T2 +T3 +IIB +T2 +T3 +T4a +IIIA +T3 +T4a +IIIB +T4a + +Tumour involves lamina propria, submucosa +T1a lamina propria +T1b submucosa +Tumour invades muscularis propria +Tumour involves subserosa +Tumour perforates serosa +Tumour invades adjacent organs +No lymph nodes +Metastasis in 1–2 regional nodes +Metastasis in 3-6 regional nodes +Metastasis in 7–15 regional nodes +Metastasis in more than 15 regional nodes +No distant metastasis +Distant metastasis (this includes peritoneum and distant lymph nodes) + + +T1 N0 M0 +T1 N1 M0 +N0 M0 +T1 N2 M0 +N1 M0 +N0 M0 +T1 N3 M0 +N2 M0 +N1 M0 +N0 M0 +T2 N3 M0 +N2 M0 +N1 M0 +T3 N3 M0 +N2 M0 + + +No better example of the various modes by which carcinoma spreads can be given than the case of stomach cancer. It is important to note that this distant spread is unusual before the disease spreads locally, and distant metastases are uncom-mon in the absence of lymph node metastases. The intestinal and diffuse types of gastric cancer spread differently. The dif-fuse type spreads via the submucosal and subserosal lymphatic +plexus and it penetrates the gastric wall at an early stage. +Direct spread +The tumour penetrates the muscularis, serosa and ultimately +adjacent organs such as the pancreas, colon and liver. +Lymphatic spread +This is by both permeation and emboli to the affected tiers (see below) of nodes. This may be extensive, the tumour even appearing in the supraclavicular nodes (Troisier’s sign). Unlike malignancies such as breast cancer, nodal involve- +ment does not imply systemic dissemination. +Blood-borne metastases +These occur first to the liver and subsequently to other organs, including lung and bone. This is uncommon in the absence +of nodal disease. +Transperitoneal spread +This is a common mode of spread once the tumour has reached the serosa of the stomach and indicates incurability. Tumours can manifest anywhere in the peritoneal cavity and com-monly give rise to ascites. Advanced peritoneal disease may be palpated either abdominally or rectally as a tumour ‘shelf’. The ovaries may sometimes be the sole site of transcoelomic spread (Krukenberg’s tumours). Tumour may spread via the abdominal cavity to the umbilicus (Sister Joseph’s nodule). Transperitoneal spread of gastric cancer can be detected most +effectively by laparoscopy and cytology. + + + +T4b N0-1 M0 +IIIC T4a N3 M0 + + +Lymphatic drainage of the stomach + + + +T4b N2-3 M0 +IV Any T Any N M1 + + +evidence base and to improve outcome prediction for individ-ual patients, all gastric tumours whose epicentre is within 5cm of the gastro-oesophageal junction and which extend into the oesophagus are now classified according to the oesophageal system. Tumours whose epicentre is within 5cm of the gastro-oesophageal junction but which do not extend into the oesophagus, and all other gastric cancers, are staged using the revised gastric staging system. In addition, any tumour that +perforates the serosa is now classified as T4 disease. + +Understanding the lymphatic drainage of the stomach is the key to comprehending the radical surgery of gastric cancer. The lymphatics of the antrum drain into the right gastric lymph node superiorly, and right gastroepiploic and subpyloric lymph nodes inferiorly. The lymphatics of the pylorus drain into the right gastric suprapyloric nodes superiorly and the subpyloric lymph nodes situated around the gastroduodenal artery inferiorly. The efferent lymphatics from suprapyloric lymph nodes converge on the para-aortic nodes around the coeliac axis, whereas the efferent lymphatics from the sub-pyloric lymph nodes pass up to the main superior mesenteric lymph nodes situated around the origin of the superior mes- +enteric artery. The lymphatic vessels related to the cardiac + + +Charles Emile Troisier, 1844–1919, Professor of Pathology, Paris, France. +Friedrich Ernst Krukenberg, 1870–1946, ophthalmologist, Halle, Germany, wrote a classic paper on malignant tumours of the ovary in 1896. +Sister Mary Joseph (Julia Dempsey), 1856–1939, Nursing Superintendent, St Mary’s Hospital, Rochester MN, USA, noted the presence of an umbilical nodule in many patients who proved to have advanced gastric cancer and pointed this out to William Mayo while she was working as theatre sister at the Mayo Clinic. +, +PART 11 | ABDOMINAL +1136 CHAPTER 63 Stomach and duodenum + + + +orifice of the stomach communicate freely with those of the oesophagus. +The prognosis of operable cases of carcinoma of the stomach depends on whether or not there is histological evidence of regional lymph node involvement. Retrograde (downwards) spread may occur if the upper lymphatics are blocked. In Japan, the lymph node dissection is highly advanced and the Japanese Research Society for Gastric Cancer has assigned a number to each lymph node station to aid the pathological staging (Figure 63.30). Many centres in resource-rich countries now perform surgery that involves a radical lymphadenectomy but, in other centres, both the +staging and surgery are less developed. + +Operability +It is important that patients with incurable disease are not subjected to radical surgery that cannot help them, hence the value of CT/PET and preoperative laparoscopy. Unequivo-cal evidence of incurability is haematogenous metastases, involvement of the distant peritoneum, N4 nodal disease +and disease beyond the N4 nodes, and fixation to structures + +18 2 +11 10 1 + + +12 8 9 +7 +5 13 16 3 + +4 6 + +4 + + + +(a) + + + +4 + + +10 + +17 +11 + +6 + + +14 + + + +(b) 15 + +Figure 63.30 Lymphatic drainage of the stomach and nodal stations by the Japanese classification: (a) the anterior view of the stomach; (b) the posterior view. + +that cannot be removed. It is important to note that involve-ment of another organ per se does not imply incurability, pro-vided that it can be removed. Controversies with respect to operability include N3 nodal involvement and involvement of the adjacent peritoneum, performed in Japan but seldom elsewhere. Curative resection should be considered on the remaining patients. +Most operable patients should have neoadjuvant chemo- +therapy as described below as this improves survival. + +Total gastrectomy +This is best performed through a long upper midline incision. The stomach is removed en bloc, including the tissues of the entire greater omentum and lesser omentum (Figure 63.31). In commencing the operation, the transverse colon is com-pletely separated from the greater omentum. The dissection may then be commenced proximally or, more usually dis-tally. The subpyloric nodes are dissected and the first part of the duodenum is divided, usually with a surgical stapler. The hepatic nodes are dissected down to clear the hepatic artery; this dissection also includes the suprapyloric nodes. The right gastric artery is taken on the hepatic artery. The lymph node dissection is continued to the origin of the left gastric artery, which is divided flush with its origin. The dissection is con-tinued along the splenic artery, taking all of the nodes at the superior aspect of the pancreas and accessible nodes in the splenic hilum. Separation of the stomach from the spleen, if this organ is not going to be removed, is carried out and this then allows access to the nodal tissues around the upper stomach and oesophagogastric junction. The oesophagus can then be divided at an appropriate point using a combination of stay sutures and a soft non-crushing clamp, usually of the right-angled variety. It is important that the resection mar-gins are well clear of the tumour (>5cm). Involvement of either proximal or distal resection margin carries an appalling prognosis and, if in doubt, frozen section should be performed. There is some controversy regarding the management of the spleen and distal pancreas in this procedure and this is dis-cussed below. +, +Gastrointestinal continuity is reconstituted by means of a Roux loop. Other methods of reconstruction should be dis-couraged because of poor functional results. The alimentary limb of the Roux loop should be at least 50cm long to avoid bile reflux oesophagitis. The simplest means of effecting the oesophagojejunostomy is to place a purse string in the cut end of the oesophagus and, using a circular stapler introduced through the blind end of the Roux loop, staple the end of the oesophagus onto the side of the Roux loop. The blind open end of the Roux loop may then be closed either with sutures or, alternatively, with a linear stapler. The anastomosis can also be fashioned end to end. The Roux loop may be placed in either an anticolic or retrocolic position. The jejunojejunos-tomy is undertaken at a convenient point in the usual fashion (end to side, Figure 63.32). +There remains some controversy about the extent of the lymphadenectomy required for the optimal treatment of curable gastric cancer. In Japan, at least a D2 gastrectomy +(removal of the second tier of nodes) is performed on all +PART 11 | ABDOMINAL +Gastric cancer 1137 + + + + + + + + + + + + + + +(a) (b) (c) + + + + + + + + + + + + +(d) (e) (f) + +Figure 63.31 Radical total gastrectomy: (a) dissection of omentum off the transverse colon; (b) exposure of the lesser sac; (c) splenectomy; (d) division and oversewing of the duodenum; (e) dissection of the left gastric artery nodes (group 17); (f) mobilisation of the oesophagus. + + + + + + + + + + + + + + + + + + + + +Figure 63.32 Oesophagojejunostomy Roux-en-Y. + + +operable gastric cancer and some centres are practicing more radical surgery (D3 and even D4 resections). Certainly the results of surgical treatment stage for stage in Japan are much better than commonly reported in western resource-rich countries, and the Japanese contention is that the difference +, +is principally related to the staging and the quality of the + +surgery. It is observed that the physical proportions of the average Japanese patient favour the performance of more radical procedures compared with the average patient in western countries. However, radical lymphadenectomies above D2 have not been subjected to any randomised con-trolled trials. In the UK and Europe, randomised trials have been set up to compare D1 and D2 gastrectomy, but the results are difficult to interpret. One of the problems relates to stan-dardisation of the operation. Overall, it seems that the onco-logical outcome may be better following a D2 gastrectomy, but this operation is associated with higher levels of morbid-ity and perioperative mortality. It is clear that most of this morbidity and mortality relates to the removal of the spleen with or without the distal pancreas. The traditional radical gastrectomy removes the spleen and distal pancreas en bloc with the stomach and, although this is indeed an adequate means of performing clearance of the lymph nodes around the splenic artery, there now seems little doubt that adding this substantially increases the complication rate. The Japanese D2 gastrectomy will commonly preserve spleen and pancreas and this practice has been widely adopted by specialist centres in western countries. +The differentiation between a D1 and a D2 operation depends upon the tiers of nodes removed. Different tiers need to be removed depending on the positions of primary tumour and this is outlined in Table 63.6. In general, a D1 resec- +tion involves the removal of the perigastric nodes and a D2 +PART 11 | ABDOMINAL +1138 CHAPTER 63 Stomach and duodenum + + +TABLE 63.6 The lymph node stations (see Figure 63.30) that need to be removed in a D1 (N1 nodes removed) or a D2 (N2 nodes removed) resection. + + +LN number + +1 +2 +3 +4sa +4sb +4d +5 +6 +7 +8a +9 +10 +11 +19 +20 +110 +111 + + + +Right cardia +Left cardial +Lesser curve +Short gastric +Left gastroepiploic +Right gastroepiploic +Suprapyloric +Infrapyloric +Left gastric artery +Anterior hepatic artery +Coeliac artery +Splenic hilum +Splenic artery +Infradiaphragmatic +Oesophageal hiatus +Lower oesophagus +Supradiaphragmatic + +Site of cancer +Antrum Middle Cardia +N2 N1 N1 +N1 N1 +N1 N1 N1 +N1 N1 N1 +N1 N1 N1 +N1 N1 N2 +N1 N1 N2 +N1 N1 N2 +N2 N2 N2 +N2 N2 N2 +N2 N2 N2 +N2 N2 +N2 N2 + + +N2 \ No newline at end of file