The infrarenal aorta and iliac arteries are among the most common sites of chronic atherosclerotic occlusive disease in patients with symptomatic arterial insufficiency of the lower extremities. Proper management of patients with aortoiliac disease requires an understanding of the various clinical presentations, the typical patterns of commonly associated infrainguinal arteriosclerotic disease, the incidence of coexistent cardiopulmonary disease, and the variety of surgical techniques available for therapeutic intervention. Proper patient selection, with careful history taking and physical examination, well accepted and standardized indications for surgery, appropriate preoperative testing and perioperative monitoring techniques, and use of appropriate procedures for each individual patient will usually result in a favourable clinical result with a low risk.

The initial manifestation of aortoiliac occlusive disease is intermittent claudication of the lower extremities, usually the buttock, hip, thigh, and calf muscle groups. The claudication is often more disabling than that associated with isolated femoropopliteal disease due to the larger number of muscle groups affected in more proximal occlusive disease. Some patients with aortoiliac disease, particularly those with disease affecting both inflow and outflow vessels, will present with claudicatory symptoms isolated to the calf. In addition to intermittent claudication, male patients with aortoiliac disease may present with the classic triad of diminished femoral pulses, lower extremity claudication, and impotence, known as the Leriche syndrome.

The symptoms experienced by an individual patient depend upon the distribution and severity of the occlusive process (Fig. 1) 240. When disease is confined to the aortic bifurcation and the common iliac arteries (type 1 disease), limb-threatening ischaemia is rare and the symptoms are limited to claudication. Numerous pathways of collateral circulation are often found in patients with type I disease, and these may result in only mild to moderate claudication, even when the aortoiliac segment is completely occluded. Haemodynamic compromise of the distal aortic segment may result in the recruitment of stem collaterals from the intercostal and lumbar arteries to anastomose with re-entry collaterals of the iliolumbar, gluteal, deep circumflex iliac, and epigastric arteries. Visceral collaterals involving the left colic branch of the superior mesenteric artery may reconstitute flow in the hypogastric artery via the marginal artery of Drummond and the haemorrhoidal plexus. Complete occlusions of the external iliac arteries may be compensated for by collateral flow through the gluteal branches of the hypogastric artery to the circumflex femoral branch of the profunda femoral artery (‘cruciate anastomosis’).

Type I disease is relatively infrequent (5–10 per cent) in patients with symptoms severe enough to warrant consideration for surgical revascularization. It is more common in younger patients in whom there is a lower incidence of associated coronary disease, hypertension, and diabetes. However, these patients often have abnormal lipid profiles (such as Frederickson Type IV hyperlipidemia). Type I disease is also unusual in that there is a high incidence of women with this disease pattern (nearly 50 per cent of patients).

The majority of patients with symptomatic aortoiliac occlusive disease will have a more widespread distribution of disease. Approximately 25 per cent of patients have extension of the disease process into the external iliac arteries (type II) and 65 per cent have the more widespread type III pattern of disease involving the aortoiliac segments and the infrainguinal arteries. Patients with Type III disease and the so-called ‘tandem lesions’ or combined segment disease present with more severe incapacitating claudication and may also develop limb-threatening ischaemia with rest pain and/or ischaemic ulcers and gangrene. The indication for revascularization in these patients is often limb salvage rather than claudication alone.

The clinical diagnosis of symptomatic aortoiliac disease is usually accurately established on the basis of a complete history and physical examination. Claudication occurring in the proximal muscle groups of the lower extremity, occurrence of the pain after walking a predictable distance, relief of the pain with only several minutes rest, and impotence in male patients is the classic description. Physical examination will often reveal auscultable bruits over the lower abdomen and the femoral regions. Femoral and distal pulses are usually diminished or absent and, in patients with more severe or type III disease, dependent rubor, elevation pallor and trophic skin changes may be seen in the feet and lower legs. However, claudication limited to the calf, the phenomenon of claudication despite normal distal pulses at rest, and pseudoclaudication may need to be considered in the differential diagnosis.

Some patients will present with symptoms of claudication limited to the calf despite the presence of haemodynamically significant inflow disease. This occurs most often in patients with multilevel (type III) disease, and it is important that they are recognized since inflow must be corrected to improve the symptoms. Physical examination will usually reveal diminished femoral pulses or bruits. Laboratory assessment of vascular patency, particularly good quality preoperative arteriography, sometimes including pressure measurements (see below), allows accurate assessment of the adequacy of aortoiliac inflow.

The presence of significant aortoiliac occlusive disease despite the normal pulses at rest is an important source of diagnostic confusion. This phenomenon is the physiological correlate of Ohm's Law: the stenosis of the aortoiliac segment may not be haemodynamically significant at the blood flow rates seen in the resting patient; with exercise, however, the rate of flow across the stenotic segment is increased, causing a greater drop in pressure across the stenosis. Pedal pulses that were present at rest may therefore vanish after 2 min on the treadmill, confirming the vascular aetiology of the symptoms. It is, therefore, important to include some form of post-exercise ankle pressure measurement in the evaluation of these patients.

Claudicatory symptoms must also be differentiated from non-vascular causes of lower extremity pain such as radicular pain caused by nerve root irritation from spinal stenosis (pseudoclaudication) or intervertebral disc herniation. These patients will often describe pain induced by simply standing as well as on walking, and this history will help distinguish these patients from claudicants. Patients with pseudoclaudication often need to sit or lie down in order to relieve the pain as opposed to simply stopping walking. A careful history may reveal the sciatic distribution of the pain, suggesting its true aetiology.

It is generally agreed that limb-threatening ischaemia, clinically defined by the presence of untreated rest pain, ischaemic ulceration, or frank gangrene, will usually require major amputation; these signs and symptoms are, therefore, clear-cut indications for arterial reconstruction. In this patient population there are few contraindications to surgery, since revascularization by some means can usually be accomplished with morbidity and mortality rates equivalent to or lower than those associated with major amputations. Age is rarely, if ever, a contraindication. If direct aortoiliac reconstruction is deemed too great a risk, high-risk patients with multiple associated medical problems may be candidates for alternative techniques for lower extremity revascularization, such as extra-anatomical bypass, percutaneous transluminal angioplasty, atherectomy, or a combination of such procedures.

The need for surgical intervention must be dictated by the circumstances of each individual patient. Incapacitating claudication that prevents the patient from earning a living or that has a significant negative impact on the patient's desired lifestyle is generally considered an indication for surgery, provided that the patient is not at high risk for surgical complications, does not have a limited life expectancy secondary to associated medical problems, and has a generally favourable distribution of disease for correction. Patients with stable claudication will often experience significant improvement in their symptoms following conservative measures such as abstinence from smoking and pursuit of a daily exercise protocol, with weight reduction if appropriate. Surgical intervention should only be considered if these conservative treatments fail to improve the claudication.

An occasional but well recognized indication for aortoiliac arterial reconstruction is the phenomenon of atheroembolism (blue toe syndrome). The aortoiliac arteries are well-known as a source of emboli which may arise from the degeneration of unstable atherosclerotic plaque within this segment of the arterial tree. A clinical history of lower extremity arterial insufficiency is often lacking in these patients since the source of the atheroembolism may be a haemodynamically insignificant ulcerated or unstable plaque. Nevertheless, if the presentation is consistent with atheroembolism and if an aortogram reveals irregular, shaggy, or ulcerated atheromatous changes in the aortoiliac system (so-called ‘degenerative aorta’), aortobifemoral bypass and exclusion of the native aortoiliac system may well be indicated to avoid further episodes of distal embolic events since, if left untreated, repeated episodes of microembolization may result in extensive tissue loss.

Preoperative evaluation of the patient with aortoiliac disease routinely includes assessment of the patient's cardiac, pulmonary, and renal function. Systemic atherosclerosis accounts for the fact that at least 40 per cent of patients requiring peripheral vascular reconstructive procedures have significant coronary artery disease. Myocardial infarction is the cause of more than 50 per cent of the perioperative deaths in patients undergoing peripheral vascular surgery, and the detection and management of coronary disease is, therefore, important. Traditional clinical cardiac risk assessment may be difficult in the patient awaiting peripheral vascular surgery who, due to claudication, leads a sedentary lifestyle: the absence of a cardiac history cannot safely be assumed to imply the absence of severe coronary disease. Unfortunately, most screening tests lack sensitivity and specificity for predicting postoperative cardiac complications, and the vascular surgeon often has to decide whether or not a patient requires preoperative cardiac catherization. Even coronary angiography may not allow anatomical findings to be related to the perioperative risk of a cardiac event.

Tests aimed at identifying the subset of patients truly at high risk of suffering perioperative cardiac ischaemic events due to coronary artery disease are being developed: dipyridamole thallium myocardial scintigraphy is sensitive in this respect. The dipyridamole thallium study identifies those patients who might require myocardial revascularization prior to aortic surgery or in whom an alternative to direct aortoiliac reconstruction might be advisable.

The high incidence of cigarette smoking and chronic obstructive airways disease in the atherosclerotic population places these patients at increased risk for pulmonary complications. Smoking, even in patients with no detectable lung disease, is associated with increased postoperative atelectasis and pneumonia: abstinence from smoking for 2 weeks prior to surgery is critical. In patients with clinical evidence of chronic obstructive airways disease, pulmonary function studies are often useful in directing and assessing preoperative preparation, which often includes chest physiotherapy, bronchodilators, and antibiotics. An FEV&sub1; of less than 800 ml and arterial blood gas tests revealing evidence of CO&sub2; retention identifies patients at high risk for pulmonary complications. Such complications may be avoided by combined thoracic epidural and general anaesthesia with immediate postoperative extubation and postoperative pain management using the epidural route.

If clinical evaluation indicates the need for revascularization and if the patient is a reasonable surgical candidate, angiography is the next stop in the evaluation. The angiogram should not be used as a diagnostic tool: clinical evaluation, history, and physical examination, often aided by non-invasive vascular laboratory testing, is usually sufficient to diagnose the problem and to determine the need for intervention. The angiogram is used to provide the anatomical details the vascular surgeon needs to select the appropriate operative approach. It may also be used to determine whether the atherosclerotic lesions may be amenable to some form of endovascular procedure such as percutaneous transluminal balloon angioplasty or atherectomy.

The arteriographic study should include the entire abdominal aorta as well as the infrainguinal runoff. Biplane aortograms will demonstrate any significant orificial lesions of the coeliac or superior mesenteric arteries which might alter the surgical approach. Oblique views of the pelvis are useful in identifying hypogastric arterial lesions as well as disease involving the origins of the profunda femoral arteries which might modify the distal anastomotic technique. Anatomical variations and/or occlusive lesions of the renal arteries should also be noted.

In patients with multilevel (type III) aortoiliac disease it may be difficult to ascertain the haemodynamic significance of the lesions at various levels, but this information is crucial since it determines whether an inflow, outflow, or combination procedure will be needed. Non-invasive laboratory testing such as Doppler analogue waveform analysis, segmental pulse volume recording, and pulsatility indices are often misleading: the presence of combined superficial femoral and profunda femoral disease may produce results that suggest the presence of proximal disease. Therefore, measurement of pull-back femoral pressures during the arteriogram can be extremely valuable in assessing the haemodynamic significance of iliac stenoses. A transfemoral retrograde catheter approach, usually undertaken at the time of diagnostic arteriography, allows measurement of intra-arterial pressures proximal to the level of disease and in the iliac or femoral vessels distal to the lesion (Fig. 2) 241. A significant pressure gradient across the lesion is represented by a resting systolic pressure drop of more than 5 mmHg across the lesion or a fall in the femoral arterial pressure of more than 15 per cent in response to reactive hyperaemia, induced either pharmacologically (i.e. 30 mg papaverine) or with an occluding thigh cuff left in place for 3 to 5 min.

Direct aortoiliac reconstruction with an aortobifemoral bypass using a prosthetic graft represents the most definitive and durable means of revascularization. In a small selected group of patients, direct aortoiliac endarterectomy may be appropriate. Extra-anatomical procedures or combination procedures using both endovascular techniques and extra-anatomical bypass, are applicable for patients considered to be too high a risk for direct repair. Proper selection of the relevant procedure depends upon critical analysis of three factors: the patient's general condition, the extent and the distribution of the atherosclerotic disease seen on the arteriogram, and the vascular surgeon's own experience and preference.

Aortoiliac endarterectomy may be used in those patients with truly localized (type I) disease. The advantages of this technique include the reduced risk of infection of the arterial reconstruction since no prosthetic material is used, a reduced incidence of wound complications since no groin incisions are needed, and establishment of antegrade inflow into the hypogastric arteries, potentially improving vasculogenic impotence in men more reliably than is the case with aortofemoral bypass. At present, these advantages over prosthetic grafting are rather minimal and the issue of improved potency is unproven. Careful patient selection is required for this approach. The atherosclerotic plaque should terminate at the common iliac bifurcation, allowing a satisfactory endpoint to the endarterectomy to be achieved without extending the arteriotomy more than 1 to 2 cm into the external iliac artery. The surgeon must verify that a secure endpoint has been achieved either with or without the use of tacking sutures.

Aortofemoral bypass grafting is the preferred method of treatment of most patients with symptomatic aortoiliac disease. Initial graft patency rates approach 100 per cent, the 5-year patency rate exceeds 80 per cent, and the 10-year patency rate approaches 75 per cent. These excellent long-term results are due to several factors. First, use of aortofemoral bypass in preference to aortoiliac bypass or extended aortoiliofemoral endarterectomies has eliminated the problem of graft failure due to the progression of disease or recurrent obliterative lesions in the external iliac artery segment. Secondly, prosthetic graft design and quality have steadily improved. Currently, a variety of reliable prosthetic grafts are available which have minimal tendencies to dilate or degenerate over time, and which have high long-term patencies in large vessel, high-flow situations. Such improved prosthetic grafts minimize the theoretical advantages of all autogenous reconstruction by means of aortobi-iliac endarterectomy. Avoidance of graft limb redundancy and matching the diameter of the graft limb to that of the femoral vessels are important factors in achieving a good result. Other important technical considerations involve the choice of methods for performing the proximal and distal vascular anastomoses.

Some controversy continues to exist regarding the advantages of end-to-end and end-to-side anastomosis of the graft to the proximal aorta, although most vascular surgeons prefer the end-to-end technique for several reasons. Theoretically, the end-to-end anastomosis is haemodynamically more sound: there is less turbulence, better flow characteristics and, unlike the end-to-side configuration, there is no chance of competitive flow in the native iliac arteries. The end-to-end technique also carries a lower risk of intraoperative distal embolic events, since the restoration of flow through the distal native aorta, where application of the distal clamp, in the end-to-side technique, may have loosened atherosclerotic debris or thrombus, is not required. The end-to-end technique also provides better visualization of the lumen of the proximal aorta, thus allowing any aortic thromboendarterectomy that might be needed in conjunction with the anastomosis to be performed. Finally, when using the end-to-end anastomosis, a short segment of the native aorta can be resected and the prosthetic graft thus allowed to lie in the aortic bed, where tissue coverage and reperitonealization can more adequately be obtained, thus potentially reducing the risk of a late aortoenteric fistula. (Fig. 3) 242.

There are, however, specific situations in which the end-to-side anastomotic configuration is preferred. These situations include patients in whom the distribution of atherosclerotic occlusive lesions would prevent retrograde perfusion of the hypogastric arteries and/or a patent inferior mesenteric artery, those with superior mesenteric and coeliac artery occlusive disease in whom prograde perfusion of the inferior mesenteric artery may be vital to intestinal viability, and those in whom aberrant or accessory renal arteries arise from the distal abdominal aorta (Fig. 4) 243.

In patients in whom the atherosclerotic lesions have occluded the external iliac arteries bilaterally, transection of the proximal aorta, for purposes of performing end-to-end anastomosis, may devascularize the pelvis since no retrograde flow up the external iliac arteries from the level of the femoral anastomoses can occur. This potentially increases the risk of impotence, ischaemic colitis, buttock ischaemia, persistent hip claudication, and even paraplegia secondary to spinal cord ischaemia. Use of the end-to-side anastomosis in such circumstances allows continued native perfusion of the specific vessels in question while the graft bypasses the aortoiliac occlusive disease. The technical alternative to end- to-side anastomosis in these cases would be reimplantation of, or side limb bypass grafts to, the inferior mesenteric, hypogastric, or renal arteries.

When an aortic graft is inserted for the treatment of occlusive disease, the distal anastomosis should almost always be to the femoral artery. Extensive experience has revealed that creation of the distal anastomosis in the groin rather than to the external iliac artery avoids the increased graft failure rate seen due to progression of external iliac disease distal to the aortoiliac anastomosis. The use of prophylactic antibiotics, proper intraoperative skin preparation and draping, and meticulous surgical technique has avoided the increased graft infection rates that were feared with use of femoral level anastomoses.

Following completion of the proximal aortic anastomosis, the graft limbs are tunnelled to the groins retroperitoneally. In male patients surgical dissection in the region of the aortic bifurcation and the left common iliac artery is kept to a minimum to avoid injury to the autonomic nerve plexuses important for erectile and ejaculatory function. On the right side, the retroperitoneal tunnel is created directly anterior to the native iliac artery, placing the graft posterior to the ureter to avoid entrapment of the ureter by the limb of the graft. On the left, the tunnel is placed beneath the sigmoid mesocolon in a slightly lateral position to avoid nerve plexus injury. Again, care is taken to prevent ureteral entrapment. The distal anastomosis is then performed in an end-to-side fashion to the distal common femoral artery.

It is absolutely critical for graft limb patency for the distal anastomosis to allow adequate outflow, especially when the graft is placed in a patient with multilevel disease. Reconstruction for type III aortoiliac disease often involves placement of the distal anastomosis in a diseased common femoral artery in the setting of chronic occlusion of the superficial femoral artery and orificial stenosis of the profunda femoral artery. In such patients, ensuring adequate profunda outflow is the key to maintaining graft limb patency, and correction of any significant profunda stenosis is of paramount importance. If the profunda is free of orificial stenosis, measures 4 to 5 mm in diameter as determined by gentle graduated probe insertion, and has a length of 20 to 25 cm, outflow via the profunda alone is usually adequate to maintain graft patency.

The aortogram and run-off studies in patients with multilevel disease should include oblique views of the pelvis and groin to visualize the profunda orifice. One study reported a 59 per cent incidence of profunda stenosis in patients with lower extremity ischaemia in whom oblique views were obtained. If profunda stenosis is discovered during aortography or at surgery, there are several options which can be used to correct the stenosis. One standard technique involves extension of the femoral arteriotomy down the profunda, thus crossing the orificial stenosis (Fig. 5) 244. The end of the graft limb is then fashioned with a long bevelled tip allowing the heel of the anastomosis to lie on the common femoral artery and the toe to lie down on the profunda. The anastomosis thus creates a graft patch profundaplasty. Some surgeons prefer to use autogenous tissue (endarterectomized superficial femoral artery or saphenous vein) to perform a separate profundaplasty. In this case, the prosthetic graft is anastomosed to the common femoral artery proximal to the patch profundaplasty.

The patient with type III aortoiliac (multilevel) disease often presents the therapeutic dilemma of whether an outflow procedure (e.g. femoropopliteal bypass) must be performed simultaneously with the aortobifemoral procedure. Eighty per cent of patients presenting with ischaemic rest pain will obtain adequate symptom relief following the inflow procedure alone if adequate profunda outflow is obtained. In patients who present with significant tissue loss in the foot, restoration of pulsatile flow in the foot may be required for successful healing, thus necessitating simultaneous distal grafting at the time of the inflow procedure. When claudication is the presenting symptom, reports suggest that 80 per cent of patients will experience improvement with the inflow procedure alone, although, in the author's experience, only 35 per cent of patients experience total relief of claudication.

Identification of patients likely to require synchronous distal bypass to obtain sufficient relief of ischaemic symptoms remains difficult. Preoperative non-invasive haemodynamic testing may provide useful data, although some investigators have found such studies unreliable. If unequivocally severe inflow disease in the aortoiliac segments is identified by absent or weak femoral pulses, severe lesions seen on angiography, or marked pressure gradients confirmed by intra-arterial femoral pressure measurements at the time of angiography, significant improvement may be expected with the inflow procedure alone. If clinical evaluation reveals only mild to moderate aortoiliac inflow, a small, diffusely diseased profunda femoris, poor profunda collateral development, or significant infrapopliteal disease, unfavourable results may be predicted from an inflow procedure alone.

In these situations intraoperative monitoring of calf or ankle peripheral vascular resistance (PVR) or ABI's may be used to assess the haemodynamic improvement following completion of the aortofemoral bypass, although improvement in the ankle PVR/ABI ratio may be delayed in the cold, vasoconstricted lower extremity. Good clinical acumen and surgical experience remain essential in deciding when to proceed with distal reconstruction. The use of intraoperative, adjunctive endovascular techniques such as transluminal balloon angioplasty, atherectomy, and laser/thermal angioplasty, to improve distal occlusive lesions following proximal graft insertion may result in more liberal use of synchronous distal revascularization if current studies reveal these procedures to be safe, reliable, and effective. At present, they are not commonly employed in these circumstances by the authors.

Excellent early and late results of aortoiliofemoral reconstruction may be anticipated and may be achieved with highly acceptable patient morbidity and mortality rates. Numerous studies have reported 5-year patency rates of 85 to 90 per cent and 10-year rates of 75 per cent. Perioperative mortality rates of less than 3 per cent are commonplace in many centres. Patients with type III aortoiliac disease are most likely to have associated carotid and coronary arterial disease. Improved screening techniques and intensive perioperative monitoring associated with advances in anaesthetic management of these patients will result in still further improvements in perioperative complication rates.

While direct reconstruction for aortoiliac occlusive disease using aortobifemoral bypass is the most definitive treatment, selected high-risk patients may not be suitable candidates for major surgery performed under general anaesthesia. A variety of ‘extra-anatomical’ bypass grafts, or a combination of endovascular techniques and extra-anatomical bypass grafts enable the vascular surgeon to achieve needed revascularization in such patients with reduced risk and acceptable patency rates.

The definition of the high-risk patient is a subjective process. Patients with a history of recent myocardial infarction, significant angina, a positive dipyridamole thallium study, and/or a left ventricular ejection fraction less than 35 per cent (per radionuclide ventriculography) are at high risk of suffering a perioperative myocardial event. Severe pulmonary disease also significantly increases the risks with associated general anaesthesia and major abdominal surgery: patients with severe dyspnoea, CO&sub2; retention, or FEV&sub1; of less than 800 ml/s fall into this category. Intensive intraoperative and postoperative monitoring techniques as well as improved anaesthetic techniques using combined general and thoracic epidural anaesthesia will often allow standard direct aortoiliac reconstruction to be performed with acceptable risk, even in this high-risk category of patient. If the risk is deemed unacceptable, however, alternative bypass techniques can be offered.

Extra-anatomical bypass involves placement of a graft in an anatomical location remote from that of the artery which the graft bypasses. The most commonly employed extra-anatomical procedures include axillobifemoral, iliofemoral, and femorofemoral bypass. Such procedures may also be combined with various endovascular techniques in order to improve inflow to or outflow from the bypass graft. The procedure used depends upon the exact distribution of occlusive disease found on arteriographic studies.

When the iliac occlusive disease is limited to the iliac system on one side, use of the contralateral patent iliac system as the source of inflow will allow use of a shorter prosthetic conduit with resultant improved long-term patency rates. The femorofemoral bypass graft may be used in this situation (Fig. 6) 245. Failure of the graft due to progression of disease in the ‘donor’ iliac system is unusual: several authors have suggested that the increased flow through the donor iliac artery that occurs as a result of the femorofemoral bypass, may actually retard progression of atherosclerotic disease in the donor iliac system. Past concerns that the graft may create a ‘steal’, significantly reducing perfusion of the donor limb, are unfounded in the absence of any haemodynamically significant stenosis in the donor limb inflow. Severe outflow lesions on the donor side with occlusion or stenosis of the superficial femoral artery may result in a steal phenomenon, but this is not usually clinically significant.

The donor iliac artery must be evaluated with careful biplanar arteriography to establish clearly the absence of any haemodynamically significant inflow lesions. Physical examination and physiological testing are also important in assessing adequacy of the donor iliac system: non-invasive arterial testing may be very helpful in this assessment. The segmental pressures should disclose normal thigh pressures; pulse volume recordings at the thigh level should have an excellent contour, with brisk upstroke on the waveform, and femoral artery Doppler analogue waveforms should be triphasic. These non-invasive tests results suggest a patent iliofemoral system. When the superficial and profunda femoral arteries are diseased, these studies may falsely suggest disease at the iliac level: pull-back intra-arterial pressures obtained at the iliac and femoral levels will disclose the presence of any significant lesions. If arteriograms show an iliac stenosis that is apparently not haemodynamically significant, hyperaemic testing or pressure measurement obtained following administration of papaverine is necessary to be certain that the lesion will not become haemodynamically significant at the increased rates of flow that will occur after placement of the femorofemoral graft.

The status of outflow in the recipient limb is also a critical factor in determining the long-term patency of such grafts. One study showed that the patency rate of femorofemoral grafts was reduced from 92 per cent to 52 per cent if the superficial femoral artery in the recipient limb was occluded. The probability of salvaging the limb or of adequately relieving ischaemic symptoms by femorofemoral grafting is also reduced in the presence of significant outflow disease, even if the graft remains patent.

If the criteria above are applied when selecting patients for femorofemoral bypass the results are excellent and morbidity and mortality rates are low, even in high-risk patients. The procedure can be performed under epidural, spinal, or even local anaesthesia, completely avoiding any myocardial depressant effects of general anaesthesia and avoiding the need for mechanical ventilation in patients with severe pulmonary disease. The common femoral arteries are exposed through bilateral vertical groin incisions and the graft is placed in a subcutaneous, suprapubic tunnel from the donor to the recipient groin. A prosthetic graft of 8 mm in size is generally chosen. Many currently believe that externally reinforced polytetrafluoroethylene is the best available prosthesis for such application, since this is less likely to be occluded by external compression than are Dacron grafts. The conduit size should closely match that of the femoral arteries in order to prevent the deposition of laminated thrombus within the body of the graft that can occur when the graft is larger in diameter than the host vessels. The graft configuration which appears to provide the most optimal haemodynamics is a gentle ‘C’ configuration, in which the graft rises up through the suprapubic tunnel and then down again to the opposite femoral artery (Fig. 6) 245.

Five-year cumulative femorofemoral graft patency rates of 75 to 80 per cent are common; these figures, combined with the simplicity of the procedure, and its low morbidity and mortality rates have led numerous authors to suggest the use of these grafts in low-risk as well as high-risk patients with unilateral iliac disease. In the sexually potent male patient, this approach has the additional advantage of avoiding the postoperative sexual dysfunction which may occur with direct reconstructions of the aorta. In general, however, good-risk patients are still best served by use of direct aortoiliac reconstruction.

High-risk patients with aortic or bilateral iliac artery occlusive disease who present with limb-threatening ischaemia are candidates for axillofemoral bypass as an alternative to either direct aortoiliac reconstruction or primary amputation (Fig. 7) 246. In such patients the axillobifemoral bypass should be used in preference to the axillounifemoral graft, since the cumulative 5-year patency of the former is clearly superior. This improved patency rate has been attributed to the increased rates of flow in the axillary limb of the graft in the bilateral bypass.

The axillary portion of the bypass graft is performed by exposing the first portion of the axillary artery using a skin incision placed beneath the clavicle. The pectoralis major muscle is divided in the tendinous portion near its insertion: this provides excellent exposure of the artery and also allows the bypass graft to lie at the appropriate angle as it emerges from the subclavicular fossa and enters the subcutaneous tunnel on the chest wall. Bilateral vertical groin incisions are used to expose the femoral arteries. The graft is then passed in subcutaneous tunnels from the subclavicular dissection to the ipsilateral groin and from that groin to the contralateral femoral artery (Fig. 7) 246. This procedure usually requires general anaesthesia, at least during the blunt dissection of the subcutaneous tunnels: the subclavicular and femoral dissections may be performed under local anaesthesia. The prosthetic graft of choice is generally externally reinforced polytetrafluoroethylene, although some surgeons prefer Dacron grafts. The former will usually allow better results of graft thrombectomy if occlusion occurs. Prompt thrombectomy produces secondary patency rates of 60 to 70 per cent for occluded grafts. Although inferior to the long-term patency performance of aortobifemoral or femorofemoral grafts, axillobifemoral grafts offer a reasonable compromise in the high-risk individual. Paradoxically, the mortality rate of 5 to 10 per cent exceeds that of direct aortic grafting, although it must be remembered that such grafts are undertaken in patients in whom the risks associated with direct operation are almost certainly increased.

Recent technological advances have produced a variety of endovascular techniques which, because of their decreased invasiveness, may be used alone or as an adjunct to surgical bypass techniques for treating aortoiliac occlusive disease in high-risk patients. These techniques include intra-arterial thrombolytic therapy for native occlusive lesions, transluminal balloon angioplasty, laser or thermal assisted balloon angioplasty, and atherectomy. Long-term follow-up of results of many of these techniques should be restricted to the high-risk patient felt to be unsuitable for standard aortobifemoral bypass. This does not apply, however, to balloon angioplasty for localized iliac disease, which has been evaluated for nearly a decade, and has been verified to be the procedure of choice for truly localized iliac stenotic lesions. Long-term patency and relief of ischaemic symptoms following percutaneous transluminal angioplasty for local iliac disease produces results equivalent to those of surgical treatment in patients with limited disease, and its advantages in terms of cost and morbidity are obvious. Iliac percutaneous transluminal angioplasty may also be employed as an adjunct to surgical revascularization, providing iliac inflow when distal surgical procedures are required. Transluminal balloon angioplasty may be used in these situations to provide inflow for femorofemoral bypass of a more extensive chronic atherosclerotic iliac occlusion on the contralateral side (Fig. 6) 245. Adjunctive use of iliac percutaneous transluminal angioplasty in patients with bilateral iliac disease may also allow the use of shorter extra-anatomical bypass grafts, such as femorofemoral rather than axillofemoral grafts, and this may improve long-term patency rates. Long-term follow-up of patients treated using these combined techniques suggests that with appropriate patient selection (most importantly, those with fairly focal iliac disease), results equivalent or even superior to those achieved with standard arterial reconstruction can be achieved.

Bernhard VM, Ray LI, Militello JP. The role of angioplasty of the profunda femoris artery in revascularization of the ischemic limb. Surg Gynecol Obstet 1976; 142: 840.

Boucher CA, Brewster DC, Darling RC, Okada RD, Strauss HW, Pohost GM. Determination of cardiac risk of dipyridamole-thallium imaging before peripheral vascular surgery. N Engl J Med 1985; 312: 389

Brewster DC. Direct reconstruction for aortoiliac occlusive disease. In Rutherford RB, ed., Vascular Surgery. Philadelphia: WB Saunders Co., 1989; 667–691

Brewster DC, Darling RC. Optimal methods of aortoiliac reconstruction. Surgery 1978; 84: 739.

Brewster DC, et al. Femoral artery pressure measurement during aortography. Circulation 1979; 60 (Suppl. 1): 120.

Brewster DC, et al. Aortofemoral graft for multilevel occlusive disease. Predictors of success and need for distal bypass. Arch Surg 1982; 111: 1593.

Brewster DC, et al. Selection of patients for preoperative coronary angiography. Use of dipyridamole-stress thallium myocardial imaging. J Vasc Surg 1985; 2: 504.

Brewster DC, Meier GH, Darling RC, Moncure AC, LaMuraglia GM, Abbott WM. Reoperation for aortofemoral graft limb occlusion. Optimal methods and long-term results. J Vasc Surg 1987; 5: 305.

Brewster DC, et al. Long-term results of combined iliac balloon angioplasty and distal surgical revascularization. Ann Surg 1989; 210: 324.

Brief DK, Brener FJ, Alpert J, Parsonnet V. Cross-over femoro-femoral grafts followed up five years or more. Arch Surg 1975; 110: 1294.

Crawford ES, et al. Aortoiliac occlusive disease: Factors influencing survival and function following reconstructive operation over a twenty-five-year period. Surgery 1981; 90: 1055.

Dalman RL, Taylor LM, Moneta GL, Yeager RA, Porter JM. Simultaneous arterial inflow procedures and femoral popliteal/tibial bypass procedures. J Vasc Surg 1991; 13: 211.

Darling RC, et al. Aortoiliac reconstruction. Surg Clin N Am 1979; 59: 565.

DeBakey ME, Lawrie GM, Glaeser DH. Patterns of atherosclerosis and their surgical significance. Ann Surg 1985; 210: 115.

Flanigan DP, et al. Hemodynamic evaluation of the aortoiliac system based on pharmacologic vasodilatation. Surgery 1983; 93.

Harris PL, Cave Bigley DJ, McSweeney L. Aortofemoral bypass and the role of concomitant femorodistal reconstruction. Br J Surg 1985; 72: 317.

Hertzer NR, et al. Coronary artery disease in peripheral vascular patients. A classification of 1000 coronary angiograms and results of surgical management. Ann Surg 1984; 199: 223.

Hertzer NR, et al. The risk of vascular surgery in a metropolitan community, with observations on surgeon experience and hospital size. J Vasc Surg 1984; 1: 13.

Kalman PG, Hosang M, Johnston KW, Walker PM. The current role for femorofemoral bypass. J Vasc Surg 1987; 6: 71.

Karayannacos PE, Yashon D, Vasko JS. Narrow lumbar spinal canal with ‘vascular’ syndromes. Arch Surg 1976; 111: 803.

Lerich R, Moral A. The syndrome of thrombotic obliteration of the aortic bifurcation. Ann Surg 1948; 127: 193.

LoGerfo FW, et al. A comparison of the late patency rates of axillobilateral femoral and axillounilateral femoral grafts. Surgery 1971; 81: 33.

Malone JM, Moore WS, Golstone J. The natural history of bilateral aortofemoral bypass grafts for ischemia of the lower extremities. Arch Surg 1975; 110: 1300.

Mannick JA, Williams LE, Nasbeth DC. The late results of axillofemoral grafts. Surgery 1970; 68: 1038.

Peters RM. Identification, assessment and management of surgical patients with chronic respiratory disease. Prob Gen Surg 1984; 1: 432.

Pierce GE, et al. Evaluation of end-to-side v. end-to-end proximal anastomosis in aortobifemoral bypass. Arch Surg 1983; 117: 1580.

Picone AL, et al. Spinal cord ischemia following operations on the abdominal aorta. J Vasc Surg 1986; 3: 94.

Piotrowski J, Pearce WH, Jones DN, Whitehouse T, Bell R, Pratt A, Rutherford RD. Aortobifemoral bypass. The operation of choice for unilateral iliac occlusion? J Vasc Surg 1988; 8: 211.

Queral LA, et al. Pelvic hemodynamics after aortoiliac reconstruction. Surgery 1979; 86: 799.

Royster RS, Lynn R, Mulcare RJ. Combined aortoiliac and femoropopliteal occlusive disease. Surg Gynecol Obstet 1976; 143: 949.

Rutherford RB, et al. Serial hemodynamic assessment of aortobifemoral bypass. J Vasc Surg 1986; 4: 428.

Rutherford RB, Patt A, Pearce WH. Extra-anatomic bypass: A closer view. J Vasc Surg 1987; 5: 437.

Strandness DF. Collateral Circulation in Clinical Surgery. Philadelphia: WB Saunders Co., 1969.

Szilagyi DE, et al. A thirty-year survey of the reconstructive surgical treatment of aortoiliac occlusive disease. J Vasc Surg 1986; 3: 421.