Figure 96-1. Classification of abdominal aortic aneurysms. More than 95% of all abdominal aortic aneurysms are infrarenal. Juxtarenal aneurysms extend to the level of the renal arteries, and suprarenal aneurysms to the level of the superior mesenteric. Aneurysms extending above the superior mesenteric artery and above are designated as thoracoabdominal aneurysms and are classified (Crawford extent I–IV) according to how far they extend cephalad and caudal.
PATHOGENESIS AND RISK FACTORS
The pathogenesis of AAAs remains unresolved, although it is an intense area of both experimental and clinical investigation. Multiple potential etiologic factors have been implicated including atherosclerosis,20 hemodynamics,21 collagen,22 collagenase,23 elastin,24 elastase,25 metalloproteinases,26 protease inhibitors,27 programmed cell death (apoptosis),28 neutrophils,29 and inflammatory mediators.30 The etiology is likely multifactorial with interaction between both environmental and genetic factors. Unfortunately, investigation into the potential mechanisms has not resulted in any effective biologic therapies. Elucidation of the pathogenesis has been complicated by the older age of patients at presentation and the absence of suitable animal models.
31,32 Identification of these risk factors is important to facilitate screening high-risk patient populations. AAAs are a disease process of aging and are rare among persons less than 50 years of age. Indeed, the mean age among patients undergoing repair across the country was 75.5 years in a recent Medicare population study.19 A meta-analysis of the population-based screening studies for AAAs reported that male sex had the strongest association (odds ratio, 5.69).33 The incidence of death resulting from AAAs for men of 60 to 64 years of age is 11-fold higher than that for women of the same age, but it is only 3-fold higher for men between 85 and 90. Furthermore, men account for approximately 80% of all AAA repairs performed nationally.19,34 The Aneurysm Detection and Management Veterans Affairs Cooperative Study Group (ADAM) reported that smoking was the strongest modifiable risk factor associated with AAAs >4 cm (odds ratio, 5.57) among the 73,451 veterans screened.35 Similarly, Wilmink et al.36 reported that abdominal aneurysms were 7.6 times more likely to develop in current smokers than in nonsmokers and that the duration of smoking rather than level of exposure appeared to correlate with their development. Decreasing prevalence of smoking has correlated with decreasing AAA deaths in the US and European population.17,37 Darling et al.38 prospectively analyzed patients undergoing repair of AAAs and reported that 15.1% had a first-degree relative with an aneurysm, in contrast to only 1.8% in the control group. Interestingly, the presence of a female family member with an aneurysm correlated with an increased risk for rupture. Larsson et al.39 reported from a population-based control study in Sweden that the relative risk of an AAA in first-degree relatives was 1.9 (95% CI, 1.6–2.2). However, the risk of an aneurysm was not affected by the gender of the index individual or relative.Multiple risk factors have been identified for the development of AAAs, including age, sex, race, smoking, hypertension, hyperlipidemia, peripheral vascular disease, myocardial infarction, and family history.
The presence of an aneurysm in the aorta, iliac, femoral, or popliteal arteries dramatically increases the risk for a new or additional AAA. Metachronous aneurysms may develop anywhere in the remaining native aorta, and are commonly seen during follow-up within the residual infrarenal aortic cuff. When this occurs, the etiology of the aneurysm is often unclear and may be related to degeneration of the native aorta, or a pseudoaneurysm that develops at the anastomosis. The incidence of aortoiliac aneurysms in patients with popliteal or femoral artery aneurysms is approximately 50%.13 Importantly, all patients found to have one of these peripheral artery aneurysms should undergo a computed tomography (CT) scan of the entire aorta from thorax to the iliac vessels to exclude a synchronous aneurysm in addition to being screened for other peripheral aneurysms. Interestingly, the reverse scenario is not true; patients with aortic or iliac artery aneurysms have a <5% chance of having a peripheral artery aneurysm, and evaluation beyond physical examination is likely not justified.40
The incidence of AAAs is increased among patients with an aortic dissection. Late or repeated operations are required in approximately 20% of patients by 10 years after an acute aortic dissection.41 The aneurysms may develop in either the thoracic or abdominal aorta, although the former site is more common. The term dissecting aneurysm is frequently used to describe fusiform degenerative aneurysms, although it is a misnomer; dissection and aneurysm degeneration are separate processes, but can occur in the setting of the other. Aneurysm false lumen degeneration can occur after a dissection, and dissection of the aorta involved in an existing aneurysm can also occur. Simply, a dissection is a tear within the aortic intima itself that leads to blood flow between the layers of the aorta (within the media) and extends for a variable length resulting in a “true” and “false” lumen.
Interestingly, the prevalence of AAAs and the rate of expansion are both increased among heart transplant patients.42–44 The responsible mechanisms remain unclear, but the obligatory chronic immunosuppression may contribute. Arguments have been made to initiate screening programs as part of pre- and posttransplant evaluations.
It is notable that the risk factors for AAA are similar to those associated with the development of atherosclerosis with the noted exception of diabetes mellitus.45,46 Furthermore, atherosclerotic changes are found almost universally within degenerative aneurysms at the time of repair, and the two disease processes are likely related, but it is unclear how. The pathogenesis may be distinct and not necessarily causative.
PRINCIPLES OF MANAGEMENT
The treatment goals for patients with AAAs are to prolong meaningful life, relieve symptoms, and prevent rupture. Because surgical treatment is the only effective means to achieve these goals, the crucial question that must be answered is whether the patient merits repair. The decision algorithm is relatively straightforward for patients with ruptured or symptomatic aneurysms, but more difficult for patients with asymptomatic, intact aneurysms in the elective setting. The decision to recommend prophylactic intervention in the elective setting is contingent on the balance between the risk of the procedure and the risk of expectant or nonoperative management within the context of the patient’s desires or wishes and their meaningful life expectancy. Appropriate assessment of these risks requires an understanding of the size-associated risk for rupture, the growth rate, and the morbidity and mortality associated with repair, as well as the natural history of a patient with the risk factors that led to aneurysm disease.
Understanding the natural history of untreated AAA requires knowledge of the physics associated with the vessel wall. The tangential stress (t) of a fluid-filled cylindrical tube is determined by the following equation:
t = Pr/d
where P is the pressure exerted by the blood (dyne/cm2), r is the internal radius (cm), and d is the thickness (cm) of the arterial wall.47 The tangential stress of a cylinder 0.2 cm thick with an internal radius of 0.8 cm and a fluid pressure of 150 mm Hg is 8 × 105 dyne/cm2 (Fig. 96-2). An increase in the internal radius (diameter) of the cylinder to 2.94 cm and a concomitant decrease in the wall thickness, as might occur with an aneurysm, would increase the tangential stress to 98 × 105 dyne/cm2. Thus, a 3-fold increase in diameter would result in a 12-fold increase in the tangential stress. Aneurysms rupture when the tangential stress exceeds the tensile strength of the vessel wall. It should be emphasized that the tangential stress varies directly with the radius of the cylinder (vessel), but is independent of its length.
Table 96-1).48 Although there is some variability in the data, it is generally appreciated that the rupture risk for aneurysms less than 5 cm in diameter is small, but increases considerably for those greater than 5.5 cm in men and 5 cm in women. These data may be simplified by using the rule of thumb that the annual rupture risk is ≤5% for a 5-cm aneurysm, ∼5% for a 5.5-cm aneurysm, 10% for a 6-cm aneurysm, and 20% for a 7-cm aneurysm. These numbers correspond to an estimated 5-year rupture risk of 50% for 6-cm aneurysms and 100% for those of 7 cm. Notably, both the ADAM35 and UK Small Aneurysm Trial49 that randomized patients with small aneurysms (4 to 5.5 cm) to open repair or surveillance reported that the rupture risk for surveillance was ≤1%/yr.The diameter of an AAA is the greatest predictor of rupture as would be predicted by the tangential stress of the vessel wall. The diameter of an aneurysm is determined by measuring its greatest diameter from outer wall to outer wall, preferably using an orthogonal image, throughout the extent of the aneurysm. Orthogonal refers to a cross-sectional measurement perpendicular to the long axis, along the centerline of flow to avoid overestimation due to tortuosity of the aorta. The collective annual rupture risks per aneurysm diameter are shown (
Figure 96-2. Cross-sectional view of a 2-cm-diameter cylinder that expands to a diameter of 6 cm while wall cross-sectional area remains constant. t, wall stress; d, wall thickness; ri, inside radius; r0, outside radius. Expansion of a 1-cm-diameter cylinder to a diameter of 3 cm with no change in wall cross-sectional area increases wall tensile stress 12-fold.
A variety of other factors have also been reported to increase the risk of aneurysm rupture including female gender,50–52 chronic obstructive pulmonary disease,50,53 smoking,50 hypertension,50,53 family history,54 and wall stress.55 The UK Small Aneurysm Trial reported that the rupture risk was increased for female gender (hazard ratio, 3), current smoking (hazard ratio, 1.5), severe COPD (hazard ratio, 0.6 per L FEV1), and higher mean arterial pressure (hazard ratio, 1.2/mm Hg). Fillinger et al.56 used finite element analysis to calculate the wall stress of AAAs and reported that the wall stress of symptomatic/ruptured aneurysms exceeds those for elective aneurysms. The impact of the aneurysm growth rate on rupture risk remains unclear and has been difficult to separate from aneurysm diameter alone.52 However, an aneurysm expansion of ≥1 cm/yr is generally considered worrisome and a potential risk factor for rupture.
Table 96-1 Estimated Annual Rupture Risk
The natural history of AAAs is to increase in size. The reported mean rate of growth has varied from 0.2 to 0.3 cm/yr in population studies57–59 to 0.4 cm/yr from referral practices60–62 with the latter figure (0.4 cm/yr) generally quoted as a reasonable estimate. Several factors including female gender, current smoking, and larger original diameter have been associated with an increased rate of growth as might be predicted from the risk factors for rupture.57,59,63 Interestingly, early studies showed that doxycycline and coenzyme A reductase inhibitors (i.e., statins) may inhibit aneurysm growth. More recent literature has shown promise for statins inhibiting growth and rupture of AAA; however, doxycycline has been proven ineffective in a randomized trial.63–67 It should be emphasized that these growth rates are mean values and that aneurysms do not always grow in a linear fashion, as might be predicted. The growth curve may be somewhat erratic or “staccato” with no growth detected during consecutive 6-month intervals followed by a growth of 0.6 cm during the next follow-up interval.68 Furthermore, it should be emphasized that the past rate of growth does not predict future growth; patients should not be lulled into a false sense of security if their aneurysm is relatively stable over time, and should be encouraged to continue monitoring with a vascular specialist.
The mortality rate associated with repair of an AAA depends on the status of the aneurysm (intact/asymptomatic, intact/symptomatic, ruptured), the physiologic state of the patient (age and medical comorbidities), and the method of repair (open vs. endovascular). Population studies from Medicare and the Nationwide Inpatient Sample (NIS) have reported that operative mortality rates for open repair of intact aneurysms in the United States range from 4.2% to 5.4% over the past decade.34 Predictably, the operative mortality rate increases with age, ranging from 2.2% among persons of 50 to 59 years of age to 9.2% among those >80. Interestingly, operative mortality rate has been shown to be significantly higher among women (6.1% vs. 3.7%). The mortality rate for open repair in the randomized trials comparing operative repair with surveillance for small aneurysms (UK Small Aneurysm Trial – 5.8%, ADAM – 2.7%) and open repair with endovascular repair (DREAM – 4.6%, EVAR Trial – 4.6%)69,70 are within the range of these nationwide series. Furthermore, a literature review examining the mortality rate of open AAA repair encompassing 64 individual studies reported a collective rate of 5.5%.71
The reported operative mortality rate for endovascular repair of intact AAAs has been consistently less than those for the open approach.72–78 Schermerhorn et al.74 reported that the operative mortality after endovascular aneurysm repair (EVAR) was 1.2% among Medicare beneficiaries during 2001 to 2004 while mortality within the NIS through 2005 was 1.3%. Results from the Dutch Randomized Endovascular Aneurysm Management (DREAM) Trial70 and the Endovascular Aneurysm (EVAR) Trial79 comparing open and endovascular repair have reported that the perioperative mortality rate is lower for endovascular repair and within the same range (DREAM – 1.2% vs. 4.6%; EVAR Trial – 1.7% vs. 4.7%).
The operative mortality rate for open repair of intact/symptomatic aneurysms among patients undergoing emergent repair exceeds that for elective repair and has ranged from 9% to 19%.13,80,81 Various explanations have been proposed for this increased mortality rate relative to that for intact/asymptomatic aneurysms including failure to maximize preoperative medical conditions, increased incidence of inadvertent venous injuries, and less experienced operative teams, although the true explanation remains unclear. However, in more recent data, De Martino et al.82 reported from the Vascular Study Group of Northern New England a perioperative mortality of 2.1% and 0% for open and endovascular repair for intact/symptomatic aneurysms, respectively. However, late mortality and in-hospital adverse events were still higher comparable to the asymptomatic cohort.
13 Indeed, a meta-analysis spanning 50 years and 77 studies reported that the operative mortality rate for the open repair of ruptured aneurysms was 48%.83 These figures correspond to an overall mortality rate of approximately 80% although this may be an underestimate. It is remarkable that the optimization of pre-hospital and emergency room care including a reduction in the mean transfer time from the emergency department to the operating room to 12 minutes did not appear to result in a decrease of the mortality rate of ruptured aneurysms.84 Notably, the operative mortality rate for ruptured aneurysms treated with the open approach has improved slightly over the past few decades with Bown et al.83 reporting a 3.5% reduction per decade.The actual mortality rate for ruptured AAAs is somewhat difficult to determine because a significant number of sudden deaths in elderly patients are likely secondary to ruptured aneurysms. It has been estimated that 50% of all patients with ruptured AAAs die outside the hospital, and that approximately 50% of those who actually undergo open repair do not survive.
The mortality rate for the endovascular repair of ruptured AAAs is lower than associated with the open approach, although data may be significantly skewed by selection bias.19,85–91 Rayt et al.88 reported a collective mortality rate of 24% for the endovascular approach from 31 studies encompassing 982 patients while other meta-analyses or systematic reviews have reported comparable rates.86,89 The potential to treat ruptured AAAs with the endovascular approach may represent the greatest contribution or benefit of the technology. However, further validation is necessary since the reports cited above likely reflect both patient selection and publication bias. In population data, Giles et al.91 and Schermerhorn et al.19 reported that the annual number of deaths across the country from both intact and ruptured aneurysms has decreased significantly since the introduction of EVAR.
CLINICAL PRESENTATION AND DIAGNOSIS
92 Given these fairly broad ranges, the authors concluded that physical examination could not be relied upon to exclude an AAA. Predictably, the accuracy of primary care physicians for detecting an AAA in patients with known aneurysms is only fair because of the limitations noted above and the failure to actually palpate the aorta during the physical examination. Intact abdominal aortic and iliac artery aneurysms may present with symptoms that lead to further investigation and the correct diagnosis although this is the exception rather than the rule. Rarely, enlargement of the aneurysm may cause vertebral erosion and chronic back pain. In addition, thrombosis of an AAA may cause acute ischemia in the lower torso, and aneurysms may be a source of arterial macroemboli or microemboli leading to acute ischemia of a lower extremity or digit, respectively.The overwhelming majority of AAAs are asymptomatic at the time of discovery. Most aneurysms are detected by abdominal or pelvic imaging studies, such as ultrasonography and CT, performed for other indications (e.g., chronic back pain, renal cysts) rather than on physical examination. Indeed, it is often difficult to palpate an AAA on physical examination because of its anatomic location in the posterior abdomen adjacent to the spine and these difficulties are exacerbated in the presence of truncal obesity. A literature review examining the accuracy of physical examination reported that the sensitivity ranged from 33% to 100%, the specificity ranged from 75% to 100%, and the positive predictive value ranged from 14% to 100%.
Patients with intact/symptomatic or ruptured aneurysms present with abdominal or back pain related to the aneurysm itself. The character of the pain is variable and ranges from dull to sharp. The pain is usually acute in onset and persistent. It may be superimposed on more chronic abdominal or back pain, but the presentation is usually not subtle, and the pain can be differentiated from more chronic complaints. In addition, the pain may radiate from the abdomen to the back, flank, inguinal region, or genitalia. Approximately 10% of patients with ruptured AAAs present with signs and symptoms similar to those of ureteral colic or other acute urologic problems.93 For example, the diagnosis of a ruptured aneurysm must be ruled out in a timely fashion in patients who might be at risk for AAA presenting with testicular pain who have a normal urinalysis and testicular examination.
Patients with a ruptured AAA may present anywhere along the spectrum from hemodynamically normal to profound shock. Their status depends on the ability of the tissues adjacent to the aorta to tamponade the bleeding. If the adjacent tissues effectively contain the bleeding, the patient may present in a hemodynamically stable state with essentially normal vital signs. However, it should be emphasized that this is usually a temporary situation, and health care providers should not be lulled into a false sense of security. A ruptured AAA is a true medical emergency that requires immediate operative repair regardless of the patient’s hemodynamic status. Tamponade can quickly degenerate into free intraperitoneal rupture and exsanguinating hemorrhage. Furthermore, vital signs may be misleading because patients can lose up to 15% of their blood volume (class 1 shock) without any appreciable change in their pulse rate or blood pressure. If the aneurysm initially ruptures freely into the peritoneal space, patients usually exsanguinate before they can seek medical attention.
Patients with a ruptured AAA may also present with either an aortoenteric or an aortocaval fistula, although both are relatively rare. Patients with an aortoenteric fistula may present with massive intestinal bleeding, but often present with a “sentinel” bleed that is small volume. The aorta may rupture through any portion of the bowel, although the duodenum and proximal small bowel are the most common sites. The overwhelming majority of aortoenteric fistulae result from the erosion of a prosthetic graft into the adjacent bowel (secondary aortoenteric fistula) rather than from an unrepaired aneurysm (primary aortoenteric fistula). The diagnosis of an aortoenteric fistula must be ruled out in all patients with gastrointestinal bleeding and either an AAA or a previous infrarenal aortic reconstruction. Patients with an aortocaval fistula generally present with high-output congestive heart failure, a continuous abdominal bruit, and edema of the lower extremities. The severity of the heart failure symptoms depends on the size of the fistula and the magnitude of the systemic shunt.
Several imaging studies are available to establish or confirm the diagnosis of an AAA. Indeed, the introduction of endovascular techniques for aneurysm repair has resulted in an evolution of these modalities. The generic imaging goals for patients with an AAA are to establish the diagnosis, determine the presence of rupture, determine the cephalad/caudal extent of the aneurysm, determine the feasibility of endovascular repair, appropriately size the aneurysm and access vessels for endovascular repair, screen for other visceral pathology, and screen for the presence of anatomic variants that would complicate operative repair, such as a left-sided vena cava or a horseshoe kidney. Although no imaging study satisfies every objective, ultrasound has emerged as the ideal screening study with CT arteriography being the definitive diagnostic test and the preferred modality for operative planning.
Abdominal ultrasound is a safe, simple, and inexpensive means of detecting AAAs (Fig. 96-3). It is relatively inexpensive and does not require the use of ionizing radiation or intravenous contrast. Furthermore, ultrasound units are portable and almost universally available in the hospital setting including the emergency room.94 The sensitivity of ultrasound for detecting AAAs is acceptable, and the technique is reproducible within 0.3 cm in experienced hands.95 However, the technique is quite operator-dependent and potentially confounded by the presence of bowel gas or extreme obesity. Ultrasound can accurately image the infrarenal aorta to its bifurcation but is less reliable for imaging the portions of the aorta proximal to the renal arteries and distal to the iliac vessels. Furthermore, it is not as reliable as CT for differentiating a ruptured from an intact aneurysm. However, ultrasound is an excellent tool for screening patients at high risk for an AAA and for confirming the presence of an aneurysm suspected by physical examination or clinical presentation. Further, it is a useful technique to follow patients with small aneurysms (<4 cm) when the exact measurement is not crucial. It should not be used to confirm the diagnosis of a ruptured aneurysm, nor should it be used as the sole imaging study before elective repair because it does not provide a complete image of the aorta and iliac arteries.
Figure 96-3. B-mode ultrasound image showing a transverse view of an infrarenal abdominal aortic aneurysm. Note the vessel wall and the large quantity of intraluminal clot surrounding the smaller, blood-filled center (dark circle).
CT arteriography overcomes many of the limitations of ultrasound and represents the current “gold standard” for imaging patients with AAAs (Fig. 96-4). The downside of CT is the greater expense and the potential harm that may be caused by the requisite ionizing radiation and intravenous contrast. Indeed, the evolution of EVAR and the required follow-up imaging studies have focused increased attention on the magnitude of long-term radiation injury. It is worth noting that the radiation dose associated with an abdominopelvic CT scan is 10 to 20 mGy while that for a routine chest radiograph is only 0.5 mGy. The incidence of allergic reactions to the contrast may be reduced by a steroid preparation while the potential nephrotoxicity can be reduced by acetylcysteine or sodium bicarbonate.96,97 Of course, the renal risks can be avoided altogether by not using contrast. However, the quality of the noncontrast images is less than optimal, and is generally not useful for determining eligibility for endovascular repair. CT is very sensitive for detecting both intact and ruptured aneurysms, and the images are reproducible within 0.2 cm.95 The quality of the CT images has continued to improve with each new generation of scanners, and the image acquisition times have decreased. Currently, CT imaging of the arterial tree from the ascending aorta to the femoral arteries can be obtained with an image acquisition time of less than 45 seconds. Given the quality and ease of imaging, CT arteriograms have essentially replaced traditional catheter-based diagnostic arteriograms for evaluation of aneurysms. Using axial CT images and 3D imaging software, a 3D image of the aorta can be obtained that allows measurements along the centerline of blood flow (Fig. 96-5), and dramatically improve the accuracy and consistency of sizing for endovascular repair.98 CT is also helpful for detecting other intra-abdominal pathology or anatomic variants that may impact the operative approach. Specific concerns include the location of the left renal vein and other associated venous anomalies, the location and size of the kidneys, and the characteristics of the aneurysm wall. CT is currently the sole diagnostic test performed before repair in the majority of cases and the imaging study of choice to confirm or refute the diagnosis of a ruptured AAA. In addition, CT is the serial imaging study of choice when aneurysms exceed 4 cm and approach the threshold for intervention.
Figure 96-4. CT scan demonstrates a large abdominal aortic aneurysm. Note that the majority of the lumen is filled with contrast.
Figure 96-5. 3D CT of the aorta and the iliac, femoral, and visceral arteries is shown. Note the infrarenal abdominal aortic aneurysm that extends to the aortic bifurcation.
Both magnetic resonance imaging (MRI) and catheter-based arteriography have been used as diagnostic imaging studies for patients with AAAs. The image quality and overall sensitivity of MRI is comparable to CT, but the technology is not as widely available and most surgeons are less familiar with interpreting the images. Furthermore, the technique is relatively contraindicated for patients with ferromagnetic devices (e.g., pacemakers, joints) or renal insufficiency, and imaging critically ill patients is cumbersome, if not prohibitive.
Importantly, aneurysms are frequently diagnosed during catheter-based arteriography performed for other purposes, but this should not be viewed as a diagnostic test of choice for AAAs. An arteriogram only delineates the lumen of the vessels (i.e., aorta, iliac and femoral arteries), and thus gives no indication of aneurysm extent or diameter. AAAs are often filled with laminated thrombus and may have a relatively normal appearing lumen. The “lumenogram” produced by the contrast reflects the patent lumen rather than the “true” lumen of the vessel (and the actual cross-sectional diameter of the aneurysm).
The diagnostic approach and initial treatment for patients with a potential ruptured aneurysm merit further comment. Because of the high attendant mortality rate, prompt diagnosis and repair are necessary. In a study from the Cleveland Clinic Vascular Registry, the operative mortality rate associated with ruptured AAAs increased from 35% when the initial diagnosis was correct to 75% when incorrect.99 Admittedly, the clinical presentation may be confusing, and delays in diagnosis are not uncommon. The classic triad of hypotension, abdominal pain, and a pulsatile abdominal mass was present in only 50% of patients with ruptured aneurysms in a single institutional series.100
Elderly patients who present to the emergency room in a hemodynamically unstable state with abdominal or back pain can often be evaluated quickly in the emergency department with ultrasound or CTA if they are stable enough for the examination. The potential causes of shock (i.e., hypovolemic, cardiogenic, septic, neurogenic) can usually be quickly differentiated by a brief history and physical examination. However, this clinical scenario is most suggestive of hypovolemic or hemorrhagic shock resulting from an intra-abdominal catastrophe. The differential diagnosis is extensive and includes pancreatitis, mesenteric infarction, acute Addisonian crisis, and rupture of a visceral artery aneurysm in addition to rupture of an AAA. A myocardial infarction can mimic a ruptured aneurysm in this patient population and potentially confounds the diagnosis, although it can usually be confirmed by the findings on electrocardiogram. Additional diagnostic imaging has not traditionally been considered necessary in this setting and, indeed, has been considered potentially harmful due to the obligatory delay in getting patients to the operating room. Due to the dramatic reduction in CT acquisition times and the potential feasibility of endovascular repair, abdominal/pelvic CT scans are allowable in this scenario to confirm the diagnosis and plan the operative procedure, provided it can be performed expeditiously. This approach has the added advantage of reducing the number of negative abdominal explorations in critically ill patients. A natural history study of patients with ruptured AAAs not offered operative repair reported that <15% of the patients died within 2 hours of hospital admission.101 Based upon these findings, the authors concluded that most patients with ruptured aneurysms are sufficiently stable to undergo a CT scan and this opinion has been supported by improved mortality results of EVAR for ruptured aneurysms.89
Several findings on CT are suggestive of a ruptured AAA including disruption of the calcium ring within the aortic wall, disruption of the aortic margins, retroperitoneal hematomas, mass lesions in the psoas region, displacement of the kidneys, abnormal soft tissues posterior to the aorta, effacement of the normal fat planes between the aorta and adjacent viscera, and abnormal retroperitoneal fluid collections (Fig. 96-6). Patients undergoing an emergent CT arteriogram to rule out a ruptured AAA should not receive oral contrast because of the delay associated with its administration and the confounding effects on the imaging of the vessels.
Figure 96-6. Contrast CT demonstrates a ruptured abdominal aortic aneurysm. Note the large retroperitoneal hematoma and the loss of the normal fat plane anterior and lateral to the left psoas muscle.
The diagnosis of a ruptured AAA should be considered in elderly patients that present to the emergency room hemodynamically stable with abdominal or back pain. Admittedly, the differential diagnosis for abdominal or back pain in this patient population is extensive, and the incidence of a ruptured AAA is small. An expeditious history and physical examination can usually determine the cause of the pain. A pulsatile abdominal aortic mass, an unexplained low hematocrit, or hemodynamic instability before presentation are particularly worrisome and increase the level of suspicion. The diagnosis of an AAA may be confirmed with a portable abdominal ultrasound in the emergency room.94 Indeed, the current trauma algorithms include abdominal ultrasound as a diagnostic technique for blunt trauma, and many centers have ultrasound units assigned to the emergency room and personnel who are appropriately trained. If ultrasound confirms the diagnosis of an aneurysm, further evaluation with CT should be obtained to rule out rupture. Alternatively, a CT may be obtained as the sole imaging study. Any findings consistent with a ruptured aneurysm on CT mandate direct transfer to the operating room and immediate repair. Aggressive fluid resuscitation should be avoided and mild hypotension (i.e., systolic pressure >80 mm Hg) tolerated in conscious patients due to the theoretical potential to cause the aneurysm to rupture into the peritoneal cavity and/or release the tamponade effect of the retroperitoneal tissue. If an intact aneurysm is found on CT without any suggestion of rupture, the next logical question is whether the aneurysm is the source of the pain. Symptomatic aneurysms are likely associated with an increased risk of rupture although the natural history remains poorly defined. Patients with symptomatic/intact aneurysms ≥5 cm in diameter should be admitted to a monitored setting and scheduled for urgent operative repair, usually the following day, provided no alternative causes for the pain are identified. Additional sources of the abdominal pain should be sought in patients with aneurysms smaller than 4 cm in light of the small rupture risk. The appropriate treatment for patients with aneurysms 4 to 5 cm is less clear. These aneurysms have the potential to rupture although the risk is small. It is recommended that these patients be admitted and the source of their pain further investigated. However, urgent operative repair is recommended if no additional causes are identified.
The role of screening for AAAs in asymptomatic patients has been partially clarified. The United States Preventative Task Force has issued a position statement advocating a single screening ultrasound in males 65 to 75 years of age who have a smoking history and selective screening for males without smoking history.102 These recommendations were based upon the results of a best-evidence systematic review that identified four population-based randomized controlled trials demonstrating that screening resulted in a reduction of aneurysm-related mortality.103 Notably, the Preventative Task Force stated that the literature did not substantiate screening for women even among those who smoke or have a family history and stated that the harms of screening outweighed the risks for screening women who have never smoked. Screening for AAAs in this subset of elderly men has been shown to be both cost effective104 and comparable to other screening programs in adult patients.105 Despite the Preventative Task Force’s recommendations, screening should likely be extended to other high-risk patient populations including patients with a first-degree relative with an AAA, evidence of a peripheral artery aneurysm, and those undergoing evaluation for heart transplantation. Medicare currently pays for a single screening ultrasound as part of the Welcome to Medicare physical examination for men who have smoked sometime during their life and for both men and women with a family history of AAAs.
All patients with symptomatic or ruptured AAAs should undergo operative repair unless they have an underlying medical condition, such as metastatic cancer, that precludes long-term survival or their quality of life is not sufficient to justify the intervention. The latter situation entails a difficult decision, but not offering operative repair should be considered in certain cases (e.g., a debilitated, demented patient in a nursing home) after discussion with the patient if he or she is alert and coherent and with the patient’s family.
The operative decision-making process for intact/asymptomatic AAAs is a complex one that needs to be tailored to the individual patient. Indeed, there is no single parameter that merits repair. The operative indications are contingent upon the size of the aneurysm, life expectancy, comorbidities, preference, and anatomic configuration. It is important to remember that the repair of an intact/asymptomatic AAA is a prophylactic operation that represents a balance between the operative risk and the future risk of rupture with the ultimate treatment goals to prolong life, relieve symptoms, and prevent rupture.
The diameter of the AAA is the best predictor of rupture as stated above and has been used as the most common indication for repair. There has been a change in the diameter-based operative criteria within the past few decades although this has been clarified more recently with level 1 evidence. It is interesting to note that the diameter threshold for good-risk patients has decreased from 6 cm to as low as 4 cm with latter recommendation from the guidelines of the national vascular surgical societies.95 Both the UK Small Aneurysm Trial49 and the ADAM Trial35 concluded that it was safe to follow patients with 4- to 5.5-cm aneurysms and that early operation did not confer any long-term survival benefit. As noted above, the rupture risk for patients in the surveillance group was <1%/yr. It is important to note that more than 60% of the patients in both studies ultimately underwent operative repair despite their initial randomization. A longer-term follow-up study from the UK Small Aneurysm Trial extending to 12 years did not demonstrate a survival benefit for the patients assigned to early surgery.106 However, it is important to note that almost all of the patients that survived ultimately required repair since their aneurysms continued to grow and exceed the 5.5-cm threshold. Indeed, the relevant question may not be whether patients with small aneurysm need to be repaired, but rather when they need to be repaired. In a separate publication from the UK Small Aneurysm Trial, Brown et al.50 reported that the rupture risk for women was over 4-fold higher, as noted above, suggesting that the 5.5-cm diameter threshold for repair may be too high for women. The proponents of smaller diameter-based thresholds for repair (i.e., <5.5 cm) have justified their approach stating that even small aneurysms rupture, aneurysms continue to increase in diameter and will likely need to be repaired, the patients’ medical conditions will likely deteriorate with age, and the operative mortality/morbidity rate for small aneurysms may be less. Although the level 1 evidence does not support this lower threshold, it is important to note that the size discrepancy between a 5.2-cm and a 5.5-cm aneurysm is very small and likely within the resolution of the imaging study.
The presence of medical comorbidities predictably impacts the perioperative mortality rate and threshold for repair. Steyerberg et al.107 identified several independent risk factors for operative mortality during open repair, including renal insufficiency (creatinine >1.8 mg/dL, congestive heart failure, ECG ischemia, pulmonary dysfunction, older age, and female gender) and these have remained consistent throughout the literature. Similarly, Beck et al.72 developed a predictive model for both open and endovascular repair using a prospective registry from the Vascular Study Group of New England. They reported that chronic obstructive pulmonary disease, suprarenal aortic clamp, renal insufficiency and advanced age (≥70) were predictive of mortality at 1 year after open repair with the mortality ranging from 1% to 67% depending upon the number of risk factors. Congestive heart failure and larger aneurysm diameter (≥6.5 cm) were the only predictive factors after endovascular repair with the mortality ranging from 4% to 23%. The consistent, dramatic impact of renal insufficiency was further emphasized by a national series that reported a 9-fold increase in mortality. Notably, the estimated glomerular filtration rate may be a better index of renal function than serum creatinine and, therefore, likely a better predictor of adverse outcome. Life expectancy is inseparable from comorbidities, but it should be emphasized that the average life expectancy for a 60-year-old and an 85-year-old man in the United States is 18 and 5 years, respectively.108
Patient preference should be factored into the operative decision process. Although the level 1 evidence suggests that it is safe to follow AAAs until they reach the 5.5-cm threshold, patients may not be willing to accept this small, finite risk and desire to have their aneurysm repaired a lower threshold. Indeed, patients often echo the justification for early repair proposed by surgeons.
The anatomic configuration of the aneurysm and/or the associated structures should factor into the operative decision process as well. Any technical factors that complicate the repair likely increase the perioperative mortality/morbidity including the need for suprarenal clamp application due to the obligatory renal/visceral ischemia, venous anomalies (e.g., left-sided vena cava), renal anomalies (e.g., horseshoe kidney), and inflammatory aneurysms. Admittedly, many of these technical concerns are relevant only to the open approach and can be overcome/avoided by endovascular repair provided that it is an option from an anatomic standpoint. Anatomic concerns relevant to endovascular repair include continued aneurysm degeneration of the aortic neck or iliac arteries that may make standard infrarenal endograft repair more difficult or impossible unless early repair is undertaken.
The introduction of the endovascular approach has challenged the operative indications for intact/asymptomatic aneurysms. Indeed, the significant decrease in the perioperative mortality rate reported in both the DREAM109 and EVAR Trials79 appear to justify lowering the threshold. However, it is important to note that the rupture rate for aneurysms less than 5.5 cm in the UK Small Aneurysm49 and ADAM110 Trials was <1%/yr which is still lower than the perioperative mortality rate for endovascular repair reported from DREAM,109 EVAR,79 and most of the national databases.73,74,111 Similarly, Finlayson et al.112 used a decision analysis model to determine the optimal diameter for open and EVAR and concluded that the endovascular approach lowers the operative threshold only for older patients in poor health.
The Joint Council of the American Association for Vascular Surgery and the Society for Vascular Surgery have released updated guidelines for the treatment of patients with AAAs that address the concerns highlighted above.48 They recommend that a diameter of 5.5 cm is an appropriate threshold for repair in the “average patient” with an intact infrarenal aneurysm, but emphasize the importance of individualizing each case. They state that rapid aneurysm expansion (>1 cm/yr), symptoms related to the aneurysm, and female gender merit repair at a smaller diameter while consideration should be given to earlier repair in young patients provided that the operative mortality rate is acceptable. Furthermore, they recommend that a larger diameter threshold is appropriate in higher risk patients and emphasize that there does not appear to be any justification to alter the operative threshold for the endovascular approach. A reasonable approach, and one that we have adopted in our own practice, is to use 5.5 cm as a threshold for repair in men and 5 cm for women.
Patients that do not meet the threshold for operative repair should be followed closely. It is important to educate the patients about their underlying disease process and emphasize the importance of long-term follow-up. Notably, Valentine et al.113 reported that 32% of patients with small aneurysms managed by “watchful waiting” were noncompliant with their follow-up plan. Furthermore, it is important to counsel patients about the presence of symptoms associated with rupture and the importance of seeking urgent medical attention. Guidelines have suggested that patients with aneurysms <3 cm in diameter should be re-imaged at 5 years while those with between 3 and 3.4 cm should be re-imaged at 3 years and those between 3.5 and 3.9 cm should be re-imaged at 1 year.114 Aneurysms >4 cm should likely be re-imaged every 6 months. As noted above, abdominal ultrasound is likely the most appropriate imaging study for aneurysms <4 cm with CT more appropriate above that threshold. It is important to note that up to 50% of the patients deemed a prohibitive operative risk and not offered elective repair will ultimately die from a ruptured aneurysm.115–119 These patients who are not candidates for elective repair should not be offered emergent repair in the event that their aneurysm ruptures or becomes symptomatic. Patients and families should be counseled on this matter preemptively to minimize difficulty and inform decisions when rupture occurs.
CHOICE OF OPEN OR ENDOVASCULAR REPAIR
After the decision to recommend operative repair has been made, the technique for repair needs to be determined. Admittedly, these decisions are somewhat interrelated since the decision to recommend operative repair in certain subsets of patients is oftentimes contingent upon whether they are candidates for the endovascular approach. The past two decades have witnessed a rapid evolution of the endovascular technique, and, indeed, this evolution has helped define our discipline. It has been clearly demonstrated that the technical success rates for endovascular graft repair are excellent and the need for intraoperative conversion to open repair negligible. The perioperative events and mid-term outcomes have been defined by level-1 evidence. Many of the technical limitations inherent to the earlier endovascular devices and anatomic limitations have been overcome. Despite the lack of definitive long-term outcome data, the endovascular approach for treatment of AAAs has been widely applied. Indeed, the endovascular approach eclipsed open repair in the United States in 2005 and recent data shows that 75% or more aneurysms are repaired in this manner.120 The choice of open or endovascular repair is complicated and contingent upon several factors including feasibility, outcome, comorbidities, compliance, cost, and preference. It is imperative that these issues, including their respective advantages/disadvantages or strengths/weakness, be addressed with the patients during the decision process to ensure proper informed consent.
The initial determinant of the approach is the anatomic configuration of the aorta, the aneurysm, and the access vessels. The commercially available endovascular devices come in a finite range of sizes and, thus, are suitable only if specific anatomic conditions are satisfied. Although the number of available devices and their specific characteristics are constantly evolving, the “generic” endograft consists of a fabric graft (i.e., polyester or ePTFE) and a metallic endo/exoskeleton (i.e., stainless steel or nitinol) that facilitates proximal/distal fixation by the radial force of the stent (Fig. 96-7). In some devices, proximal fixation is augmented by the presence of suprarenal hooks. The “generic” devices are modular and consist of either two (main body and contralateral iliac limb) or three (main body, contralateral iliac limb, ipsilateral iliac limb) components with a variety of additional ancillary pieces that allow proximal or distal extensions at the aortic and iliac ends, respectively. The initial experience with aortoaortic or “tube graft” configurations were unsuccessful due to problems with the distal landing site at the aortic bifurcation and have been abandoned. Indeed, the current strategy is to seat the proximal component of the bifurcated system as close to the lowest renal artery as possible and the distal components as close to the iliac bifurcation as possible to improve immediate seal and fixation and decrease the potential for later aneurysm degeneration at or adjacent to seal sites.
Figure 96-7. The current commercially available infrarenal endografts in the United States are shown: Cook Zenith (A); Medtronic Endurant (B); Trivascular Ovation (C); Gore Excluder (D); Endologix Powerlink (E); Lombard Aorfix (F). These are modular devices consisting of a main body, iliac limbs/extensions and main body extensions. Note the common theme of bilateral iliac limbs and a main body. There are many differences between these endografts, which make them each uniquely suited for some patients. On the top row (A,B,C), the devices all utilize a bare suprarenal stent construct to facilitate fixation. On the lower row (D,E,F), the devices utilize active fixation with hooks (D,F) or anatomic fixation at the aortic bifurcation (E).
Although there is some variability among the commercially available devices in terms of their specific sizes and anatomic constraints, the general anatomic requirements are somewhat similar. The infrarenal abdominal aorta must have a suitable, nonaneurysmal landing zone for the endograft that measures ≥10 to 15 mm in length and ≥17 to 19 mm, but ≤32 mm in diameter (ranges reflect differences between the commercial devices). Furthermore, the proximal neck angle should be ≤60 degrees for most devices (≤75 to 90 for some newer devices), as measured by the intersection of the centerline of the infrarenal aorta at the landing zone site and the centerline of the aneurysm through the aortic bifurcation. Newer devices do allow for more neck angulation, but the operating surgeon should keep in mind that more infrarenal neck may be required in more severely angulated necks to allow for appropriate seal and fixation of the device. The infrarenal neck should be relatively free of thrombus and calcification to facilitate a seal at the implantation site. It is notable that neck length requirements were initially ≥1 cm with the first generation of devices, which was increased to ≥1.5 cm with later generations, and now has again been decreased to >1 cm with some of the most recent devices. Indeed, the longer the infrarenal neck length and, therefore, the longer the device seal zone, the better. The iliac artery should have a suitable landing zone ≥20 mm with an associated diameter between 7 and 20 mm to facilitate both anchoring the graft and passing the main device into the aorta. The distal landing zone for the iliac limbs is usually in the common iliac artery although the anatomic constraints with regards to the size of the access vessels and the introduction of the devices are relevant for both the common and external iliac vessels.
The percentage of patients that are anatomically suitable for an endograft remains unresolved and is likely contingent upon the nature of the individual surgeon/institution practice and the available devices. This percentage of suitable patients has been highly variable with report of 30% in nationwide series,121 55% in a regional series,122 and 14% to 66% in institutional series.123–125 Notably, the reasons cited for exclusion include a short infrarenal neck (54%), inadequate iliac vessels (47%), and a wide infrarenal neck (40%).123
A variety of techniques have been described to overcome the anatomic limitations of the endovascular approach thereby extending its feasibility. Indeed, some modification has been described to overcome almost every anatomic contraindication. Stenosis within the access vessels can be overcome by dilation (i.e., balloon angioplasty, serial dilators) or by the use of an open prosthetic conduit or endovascular stent graft conduit.126 The open prosthetic conduit is usually anastomosed to the bifurcation of the common iliac artery and then tunneled through the retroperitoneal space below the inguinal ligament. Alternatively, an aorto-uni-iliac endograft can be configured and a femoral–femoral bypass performed. Notably, the patency rates of the femoral–femoral bypass graft in this setting have been reported to be excellent.127,128 Aneurysm degeneration of the common iliac artery can be overcome by a variety of techniques including simply using larger-diameter graft limbs, occluding/embolizing the internal iliac artery and seating the iliac limb in the external iliac artery or by bypassing the internal iliac artery (and seating the iliac limb in the external iliac artery). Although relatively simple to perform, internal iliac artery embolization has been associated with a moderate incidence of complications including buttock/thigh claudication (30% to 40%), sexual dysfunction, neurologic deficit/paraplegia, pelvic ischemia, and gluteal compartment syndrome.129,130 The claudication improves with time in the majority of patients, but can be quite debilitating, particularly in patients that did not claudicate preoperatively. Many of the other complications, although somewhat rare, are irreversible and can be catastrophic. Because of these concerns, concomitant bilateral internal iliac artery occlusion/embolization should be avoided when possible. Bypass of the internal iliac artery overcomes many of these limitations and has been associated with excellent results in terms of long-term graft patency although the procedure is somewhat challenging and adds significantly to the overall magnitude of the “less-invasive” procedure.126 Unfortunately, extending the indications for endovascular repair beyond those recommended by the manufacturers (i.e., instructions for use [IFU]) has been associated with an increase in the incidence of adverse events including decreased survival and a higher need for reintervention.131 Iliac bifurcation devices are currently commercially available outside the United States and will soon be available inside the United States, and will allow for preservation of internal iliac arteries in select patients with appropriate anatomy.132,133
Consideration of the perioperative and long-term outcomes after EVAR requires introduction of the concepts of endoleak and endotension. Simply, endoleak is the perfusion of the aneurysm sac outside the lumen of the endograft while endotension is the persistent pressurization within the excluded aneurysm sac. Endoleaks have been classified as types 1 through 4 based on the mechanism of the leak (Fig. 96-8). Type 1 leaks originate at either the proximal or distal attachment sites. Type 2 leaks come from branch vessels, such as the lumbar or inferior mesenteric arteries. Type 3 leaks are caused by fabric tears or problems at the graft interfaces of the modular devices, whereas type 4 leaks are usually transient (<24 hours) trans-graft extravasations that result from the porosity of the graft and needle holes. It should be emphasized that the entire concept of an endoleak is predicated on the ability to detect contrast or blood flow outside the lumen of the endograft and is, therefore, contingent on the sensitivity and specificity of the various imaging techniques. The major concern about both endoleaks and endotension is that the pressure transmitted to the aneurysm wall may cause the aneurysm to expand and/or rupture. The clinical significance of the various endoleak types is quite different. Both type 1 and 3 endoleaks are considered major adverse outcomes associated with an increased risk of rupture, and they merit urgent/emergent treatment.134 Type 2 endoleaks are generally considered less worrisome in terms of their rupture risk although they are associated with an increased risk for reintervention.134 They can generally be followed with serial imaging studies, but merit evaluation/intervention if the aneurysm sac continues to enlarge. Type 4 endoleaks are self-limited and benign. The clinical significance of endotension is likewise unresolved. Indeed, the concept itself is somewhat ambiguous given the limitations of actually measuring the pressure within the aneurysm sac. It should be noted that freedom from endoleak does not necessarily mean freedom from endotension given the observation that aneurysms can continue to enlarge in the absence of an identifiable endoleak.135
The perioperative complication rates appear to be lower after endovascular repair. As noted above, the randomized, controlled DREAM and EVAR Trials reported a significant decrease in the perioperative mortality rate (DREAM – 1.2% vs. 4.6%; EVAR Trial – 1.7% vs. 4.7%).70,79,109 The EVAR Trial also reported a trend toward a decrease in the combined operative mortality/severe complication rate (9.8% vs. 4.7%, p = 0.10). Multiple clinical trials have reported that the overall major complication rates after both endovascular and open repair are approximately 15% although the magnitude of the complications is less for the endovascular approach and primarily includes vascular access complications (e.g., hematoma, femoral artery injury).136 These clinical trials have likewise demonstrated that the total hospital length of stay, intensive care unit length of stay, operative blood loss, and time necessary to resume normal activities are all less for the endovascular approach.136 In addition, Hua et al.73 reported from the private sector NISQIP database that the perioperative complication rate after endovascular repair was 24%. Interestingly, the impact on sexual function remains unresolved. A survey of the DREAM participants demonstrated that sexual dysfunction was common after both endovascular and open repair, but returned to the baseline state at 3 months for both groups.137 In contrast, Xenos et al.138 reported significantly less orgasmic and erectile dysfunction after endovascular repair.
Figure 96-8. Type 1 to 4 Endoleaks. Type I and III represent the most dangerous forms of endoleaks, and represent what is essentially an unrepaired aneurysm. A: Type I leaks originate at either the proximal (Ia) or distal (Ib) attachement sites. Note the large blush of contrast outside of the graft lumen at the proximal fixation site, demonstrating a type Ia endoleak. B: Type 3 endoleaks are caused by fabric tears or problems at the graft interfaces of the modular devices. Note the contrast blush outside of the lumen of the graft at the modular interface. Type II endoleaks generally have a more benign natural history, but can cause aneurysmal growth and rupture. C: Demonstrates an inferior mesenteric arteriogram from a catheter (white arrow) in the meandering mesenteric artery. Filling of the aneurysm sac through a patent inferior mesenteric artery (red arrow) is evident. D: Demonstrates successful embolization of the IMA with coils demonstrated by black arrows. Note no further filling of the aneurysm sac.
The long-term results after the EVAR and DREAM Trials have failed to document any significant benefit for the endovascular approach in terms of almost every outcome measure analyzed.79,109 Somewhat surprisingly, there was no difference in all-cause mortality at 2 years in the DREAM Trial (survival, open – 89.6%, EVAR – 89.7%) or at 4 years in EVAR (mortality, open – 29%, EVAR – 26%). There were significant differences in terms of aneurysm-related mortality at these time points although the differences were fairly minimal and their relevance suspect. Both the complication rates (EVAR Trial; EVAR – 41%, open – 9%) and costs (EVAR Trial; EVAR – ₤13,257, open – ₤9,926) were significantly greater for the endovascular approach while there were negligible differences in the quality of life assessments.79,139,140
The mid-term results from the DREAM and EVAR Trials have been somewhat sobering and longer-term outcomes are necessary to further define the role of the endovascular approach. It is clear that the endovascular approach is not as secure a repair as the traditional, open alternative. The endovascular repair is associated with ongoing rupture risk that likely approaches 1%/yr.136,141 Notably, Schermerhorn et al.19 reported a 1.8% rupture risk among Medicare patients undergoing EVAR between 2001 and 2004. The rupture risk has been associated with poor patient selection, operator error, unrecognized/untreated endoleaks, large aneurysms, and device migration.136,142 Interestingly, the mortality rate associated with rupture after endovascular repair may be less than for de novo ruptures.143 Approximately 10% to 20% of patients develop an endoleak during the first year after endovascular repair.144–146 Admittedly, not all of these require remediation. The excluded aneurysms can continue to grow and, thereby, represent a risk for rupture. This has been correlated not only with the presence of endoleak as noted above, but also with the specific device and the baseline aneurysm size.147,148 Some type of structural failure including fabric tears, hook fractures, and suture breakage has been reported for almost every device. Design modifications have been implemented to overcome many of these deficiencies although they represent an ongoing concern that may not be manifest for years. As a consequence of all these potential limitations, reintervention is sometimes necessary either to prevent complications or to treat them. Aneurysm-related reinterventions occurred in 3.7% of patients per year after EVAR from 2001 to 2004 in the Medicare population.149 The majority of these remedial procedures are minor endovascular procedures, however with a risk of major open procedures of 0.4%/yr.141 In contrast, open aneurysm repair is associated with a less than 1%/yr incidence of aneurysm-related complications that require remedial procedures in long-term follow-up. Graft-related deaths have been reported in 2% of patients at 15 years.150
The presence of significant comorbidities and advanced age favors the endovascular approach. Although the operative threshold in regards to the aneurysm diameter measurement is the same, the endovascular approach may allow a subset of patients not considered suitable candidates for an open operation to have their aneurysms repaired. Indeed, EVAR has been shown to be feasible in patients with hepatic insufficiency151 and consistently safe in octogenarians.152,153 However, a modicum of clinical judgment is necessary, and the concept that AAA repair is a prophylactic operation and that not every patient merits treatment must be kept in mind. The EVAR Trial 2 randomized patients not fit for open repair to expectant management or endovascular repair and demonstrated no difference in aneurysm-related mortality, all cause mortality, or quality of life.140 Notably, the patients were truly “high risk” with a perioperative mortality rate of 9% and a 4-year mortality rate of almost 40%. Despite the potential nephrotoxicity associated with iodinated contrast, chronic renal insufficiency is not an absolute contraindication to endovascular repair. Strategies to reduce the associated risk can be employed including the administration of acetylcysteine and sodium bicarbonate.96,97 Alternatively, the procedure can be performed without contrast altogether by using intravascular ultrasound (IVUS) or CO2 as the contrast agent. Unfortunately, chronic renal insufficiency is a significant risk factor for adverse outcome after both open and endovascular repair. Furthermore, multiple studies have shown that both approaches are associated with a decrement of renal function.154–157 The endovascular approach may be associated with a greater decrement of function postoperatively, but the findings are somewhat equivocal; suprarenal fixation of the endovascular device does not seem to be associated with a greater decrement.157
The known device-related complications and the uncertainty about the long-term outcome after endovascular repair mandate indefinite surveillance, although the intervals and type of imaging for surveillance remain controversial. Whatever is chosen by the practitioner, patients must agree to comply with the prescribed protocol and their ability or desire to fulfill these expectations should be factored into the specific choice of procedure. It is important to note that although the incidence of complications declines with the number of negative postoperative CT scans; new endoleaks have been discovered many years after device implantation. Similarly, it is imperative that all surgeons that offer EVAR provide conscientious, long-term follow-up.
The cost comparisons between the open and endovascular approach have been somewhat inconclusive although the endovascular approach is likely more expensive. Clearly, the EVAR-1 and EVAR-2 trials demonstrated that the endovascular approach was more expensive.79,140 Although the shorter length of stay and lower incidence of major perioperative complications have resulted in a reduction in some of the hospital costs with EVAR, these benefits have been offset by increases in other hospital-related costs, namely the device cost. Notably, Sternbergh and Money158 reported that the cost of the device accounted for 52% of the total cost of the endovascular repair and estimated that the costs of open and endovascular repair were $12,546 and $19,985, respectively in the AneuRx Phase II clinical trial. Analysis of the hospital costs and reimbursement associated with endovascular repair among 7 medical centers (university hospital – 3, community hospital – 4) demonstrated a net loss of $2,162.158,159 However, others have reported that the contribution to the hospital margin on a daily basis may be superior for endovascular repair given the associated shorter duration of stay that permits higher throughput, fuller overhead amortization, and better use of inpatient beds.160 It is important to note that most of the analyses have focused on the hospital and device-related costs, but have failed to include the associated professional fees and the costs associated with long-term follow-up and reintervention, which all may be substantial. Importantly, Kim et al.161 reported that reimbursement for endovascular repair does not include long-term surveillance and secondary procedures. Indeed, it has been reported that the overall cost of endovascular repair may be twice that of the open approach.136
Despite the various advantages and disadvantages of the approaches outlined above, patients and providers seem to prefer the endovascular approach and these preferences appear to be one of the driving forces for the widespread application of the technique. Indeed, it is uncommon for patients that are candidates for either approach to elect open repair. Notably, Williamson et al.162 reported that 18% of all patients undergoing open repair would not undergo the procedure again. The potential to perform the endovascular repair completely percutaneously (i.e., no femoral artery exposure) clearly adds to the appeal of the approach.163,164 It is interesting to note that these patient preferences may not be sustained. A follow-up study from the DREAM trial reported that patients undergoing endovascular repair had a better quality of life initially, but those undergoing open repair had a better quality at 6 months and beyond.137
The proverbial “bottom line” for the open versus endovascular debate remains somewhat unresolved, but the discussion is merely academic at this point. EVAR, when anatomically possible, has essentially become the standard of care in most practices. Notably, the Agency for Health Care Research Quality conducted an evidence-based review published in 2006 and concluded that endovascular repair for aneurysms >5.5 cm did not improve patient survival or health status relative to open repair despite the fact that the perioperative outcomes were improved.165 Furthermore, they reported that the endovascular approach was associated with increased cost, complications, need for surveillance and need for remedial procedures. Finally, they concluded that it did not provide a benefit for patients unfit for open repair. Despite these recommendations, the trend toward dominance of EVAR over open repair has increased to a point where many vascular training programs are having difficulty with finding enough open aneurysm cases to sufficiently teach their trainees.166
The preoperative evaluation of patients undergoing elective AAA repair is similar to that of patients undergoing any major general or vascular surgical procedure. Patients undergoing endovascular repair should likely undergo the same preoperative work-up despite the perception that associated perioperative stresses are less. All patients should receive a complete history and a physical examination, electrocardiogram, and a chest radiograph. Routine laboratory studies, including a complete blood cell count with platelets, serum electrolytes/creatinine, and coagulation studies should be obtained. A specimen should be sent to the blood bank and the appropriate quantity of blood products cross-matched. This number can be determined from the historic operative transfusion requirements obtained from the blood bank but usually 2 to 4 units of packed red blood cells are sufficient. A thorough peripheral pulse examination should be included in the physical examination and validated with formal ankle–brachial indices. The anesthesiologist should see the patients preoperatively. In addition, patients should be started on an aspirin and a statin (HMG Co-A reductase inhibitors) and consideration made for starting a beta-blocker if they are not already on them. The ACC/AHA Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery recommend that the beta-blockers are “probably recommended” for vascular surgery patients who are at high cardiac risk or those with more than one clinical risk factor (risk factors – CAD, CHF, CVOD, DM, CRI).167 However, much of the data that this recommendation was based upon have been questioned due to reports of unscrupulous academic behavior by the investigators, and many now believe that initiation of beta-blockers before major vascular surgery is unwarranted and potentially harmful.168
Furthermore, the ACC/AHA Guidelines recommend that patients undergoing vascular surgery should be on a statin. Notably, a recent study demonstrated that statins were associated with a decreased incidence of perioperative mortality and nonfatal myocardial infarction after aneurysm repair.169 Indeed, all patients with atherosclerotic cardiovascular disease should likely be on aspirin, a statin medication, and an ACE inhibitor long-term as part of the AHA/ACC Guidelines for Preventing Heart Attack and Death in Patients with Atherosclerotic Cardiovascular Disease.170
All active medical problems, including abnormalities identified during the preoperative evaluation, should be controlled as well as possible before elective aneurysm repair. However, extensive diagnostic testing is probably unnecessary. Routine pulmonary function tests and measurement of arterial blood gases are not indicated, although they may be beneficial in selected patients with advanced chronic obstructive pulmonary disease.171 The presence of chronic obstructive pulmonary disease often complicates postoperative ventilator management, but it is unusual for a patient’s pulmonary disease to be sufficiently severe to preclude operation.172 Similarly, timed urine collections for creatinine clearance and other assessments of renal function have not proved beneficial despite the dramatic impact of preoperative renal insufficiency on perioperative outcome although it may be beneficial to calculate the estimated glomerular filtration rate.
The appropriate cardiac work-up before AAA repair is evolving and is somewhat institution-dependent. This controversy has been further complicated by the publication of the CARP Trial that examined the role of coronary artery revascularization before major vascular surgery among patients with significant coronary artery disease.173 Notably, the study reported that preoperative coronary artery revascularization did not reduce the incidence of either perioperative myocardial infarction or long-term mortality. It is important to emphasize that the overall objective of the preoperative cardiac work-up is to optimize the cardiovascular system and thereby reduce both the perioperative and long-term risk of myocardial infarction and death. Admittedly, the prevalence of coronary artery disease among patients undergoing AAA repair is quite high. Hertzer et al.,174 in a landmark publication, reported that 25% of 1,000 patients undergoing evaluation for peripheral vascular surgery (cerebral vascular occlusive disease, lower extremity arterial occlusive disease, AAA) had severe, surgically correctable lesions detected during cardiac catheterization; 6% had severe, uncorrectable disease, and only 8% had no evidence of disease. Interestingly, the incidence of surgically correctable disease was highest among patients undergoing evaluation for AAA. The most recent edition of the ACC/AHA Guidelines hase simplified the preoperative evaluation before elective vascular procedures.167,175 Briefly, patients with active cardiac conditions (unstable coronary syndromes, decompensated congestive heart failure, significant arrhythmias, and significant valvular disease) should be seen in consultation by a Cardiologist. Patients with good functional capacity as defined by the ability to generate at least four metabolic equivalents (METS, 4 METS – ability to walk up a flight of stairs) can undergo major vascular procedures without additional testing. Those patients that cannot generate 4 METS and have at least 3 clinical risk factors (see above) should be considered for further cardiac testing if the results will change the clinical management. Notably, there is insufficient evidence to support a reduced cardiac work-up for patients undergoing endovascular repair.176 It is important to emphasize that although the cardiac risk of endovascular repair may be less, the subset of patients undergoing the procedure are often older and sicker.
All patients should undergo some type of imaging modality as part of their preoperative evaluation to confirm the diagnosis and plan the procedure. Indeed, determining whether a patient is an endovascular candidate and appropriately sizing the device depend on the anatomic measurements obtained at the time of imaging. A CT arteriogram of the chest, abdomen, and pelvis is the optimal imaging study to visualize the aneurysm and is the only one required in most cases. Abdominal ultrasonography is insufficient as the sole imaging study before aneurysm repair in light of its inability to accurately define the cephalad extent of the aneurysm and the involvement of the iliac vessels.
Open Repair of Intact Abdominal Aortic Aneurysms
A significant amount of preparation is required in the operating room before making the incision and this preparation needs to be coordinated among the surgical and anesthetic teams for both the open and endovascular approaches. Although the decision about the choice of anesthesia is deferred to the anesthesiologists, inhalation agents and an endotracheal tube are used most frequently. Adjunctive epidural anesthesia may improve postoperative pain control177 and may be beneficial in patients with severe pulmonary disease.178 Adequate intravenous access should be established to facilitate resuscitation. Central venous access is usually obtained although not necessary. Electrocardiographic leads, an arterial catheter, and a Foley catheter should be placed for continuous monitoring of the electrocardiogram, arterial pressure, and urine output, respectively. In addition, a nasogastric tube should be inserted. A Swan–Ganz pulmonary artery catheter or a transesophageal echocardiogram probe should be inserted in patients with significant cardiac disease. However, routine use of pulmonary artery catheters in patients undergoing aortic surgery is not recommended and may be associated with a higher rate of intraoperative complications.179,180 Peripheral arterial pulses should be interrogated with either palpation or continuous-wave Doppler ultrasound and marked to facilitate confirmation after restoration of lower-extremity perfusion. Strategies to maintain core body temperature should be initiated.179,181 Specifically, the room temperature should be increased, warming devices should be attached to all intravenous infusion lines, and either a recirculating alcohol blanket or forced-air blanket should be applied. Bush et al.182 reported that hypothermia (<34.5°C) during AAA repair was associated with multiple physiologic derangements and adverse outcomes. Of note, this does not pertain to thoracoabdominal aortic surgery where hypothermia has shown some benefits in spinal cord ischemia reduction.183
Use of an intraoperative autologous transfusion device should be considered. However, a recent meta-analysis of 5 randomized, controlled trials reported that there is insufficient evidence to recommend its use during vascular surgery including aortic surgery.184 These devices should likely be used when a significant amount of blood loss is anticipated, such as during suprarenal or thoracoabdominal aortic aneurysm repairs. Furthermore, they can be helpful in patients who object to blood transfusions on religious principles. An extensive operative field from “nipples to toes” should be prepared with the use of topical antimicrobial agents. A first-generation cephalosporin or vancomycin should be administered prior to the incision.
AAAs may be repaired through several different incisions or approaches including midline, retroperitoneal, or transverse (supraumbilical straight, infraumbilical straight, infraumbilical curvilinear, bilateral subcostal). The incisions or approaches must be viewed as complementary since neither is perfect for every clinical scenario. Indeed, surgeons should be familiar with the various approaches and select the optimal one for the clinical setting. The determinants of the incision include the cephalad/caudal extent of the aneurysm, body habitus, presence of prior abdominal incisions, presence of abdominal wall stomas, comorbidities, additional intraoperative pathology, inflammatory aneurysms, renal anomalies, requirements for concomitant procedures, urgency of aortic control, and surgeon preference. The midline approach is preferable for patients with ruptured AAAs because aortic control at the level of the diaphragm can be obtained rapidly. The bilateral subcostal approach provides the best exposure and is the incision of choice for obese patients, those with extensive iliac artery aneurysms, those patients requiring concomitant renal artery revascularization, and those patients with juxtarenal aneurysms that require suprarenal aortic control. The retroperitoneal approach is optimal for patients with multiple previous abdominal incisions (“hostile abdomen”), abdominal wall stomas, suprarenal aneurysms, inflammatory aneurysms, and horseshoe kidneys. However, the retroperitoneal approach is limited by the inability to assess the intraperitoneal structures and the limited access to the right renal artery and right iliac vessels. It was previously contended that the retroperitoneal approach posed less of a physiologic insult than the transperitoneal approach and, therefore, was ideal for patients with advanced pulmonary or cardiac disease. However, this has not been supported by a prospective, randomized trial.185 A detailed description of the retroperitoneal approach is beyond the context of this chapter, but is available in most standard vascular surgical texts.
The sequence of steps used to repair an intact, infrarenal AAA after a bilateral subcostal incision can be summarized (Fig. 96-9). The abdomen is explored after the peritoneal cavity is entered, and both the gallbladder and colon carefully examined. The lower abdominal wall flap is immobilized to either the drapes or the pubic towel with the use of penetrating towel clips. The small bowel is manually retracted laterally to the right, and the duodenum is mobilized by incising the ligament of Treitz. The inferior mesenteric vein may be suture-ligated at this juncture to facilitate exposure. The tissue adjacent to the inferior mesenteric vein should be palpated to rule out a large, meandering mesenteric artery. This artery is an important visceral collateral and should be preserved. The retroperitoneum over the aorta is incised with the electrocautery, and the left renal vein is exposed. Self-retaining retractors (e.g., Bookwalter, Omni retractors) are then placed to facilitate further exposure. The small bowel is placed in a bowel bag, eviscerated, and retracted laterally to the right with the aid of malleable self-retaining retractors. The transverse colon and superior abdominal wall flap are retracted cephalad while the lower abdominal wall is further retracted caudal. The aorta immediately inferior to the renal arteries is exposed and both renal arteries visualized. The infrarenal aorta at this location is dissected circumferentially to facilitate placement of a transverse aortic clamp. However, this step may be omitted if a vertical clamp is used. It is important to identify the course of the renal vein and any venous anomalies on the preoperative CT scan to prevent inadvertently injuring these structures at this stage of the procedure. In the presence of a retroaortic renal vein or circumaortic collar, the aortic neck should not be dissected circumferentially and vascular control should be obtained with a vertical clamp. The retroperitoneum over the aorta is then incised further caudally and the incision extended along the course of the right common iliac artery. The extent of the caudal dissection depends on the anatomic configuration of the aneurysm. If the aneurysm extends to the aortic bifurcation, it is sufficient to dissect only the common iliac arteries provided a suitable site for clamp application is identified. If the aneurysm extends to the distal common iliac arteries, both the internal and external iliac vessels should be dissected free. This may be facilitated on the left side by mobilizing the sigmoid colon along its peritoneal reflection and reflecting it medially. The inferior mesenteric artery is then dissected free and vascular control obtained with a vessel loop. Patients are administered 100 units/kg of intravenous heparin, and the activated clotting time is confirmed to be twice the baseline value. Supplemental doses of heparin are administered throughout the procedure as dictated by the activated clotting time. Interestingly, a recent randomized, controlled trial reported that heparin does not reduce thrombotic events or increase bleeding during aneurysm repair, but is associated with a significant reduction in myocardial events.186
Figure 96-9. Steps involved in the standard repair of an infrarenal abdominal aortic aneurysm extending into the proximal common iliac arteries. A: The proximal duodenum is mobilized and the retroperitoneum overlying the aorta incised. The infrarenal aorta immediately below the renal vein is dissected. The iliac bifurcations are exposed, and vascular clamps are applied to the infrarenal aorta and distal common iliac arteries after adequate heparinization. A longitudinal arteriotomy is extended from the infrarenal aorta onto the right common iliac artery. B: Back bleeding from the lumbar arteries is controlled with “figure-of-eight” sutures. The proximal anastomosis is performed in an end-to-end configuration below the renal arteries. The distal anastomoses are performed at the common iliac bifurcation beyond the aneurysmal segments. The left limb of the graft is tunneled through the intact left common iliac aneurysm shell. C: The residual aneurysm shell is closed over the prosthetic graft, and the retroperitoneum is reapproximated to prevent erosion of the graft into the overlying bowel.