Endovascular aortic aneurysm repair (EVAR) was first performed in 1987 ; however, a more popularized adaptation was introduced in the early 1990s. Since then, EVAR’s popularity and methods have expanded with a constantly evolving and growing list of techniques, devices, and indications. Several large-scale trials have proven the early morbidity and mortality benefit of EVAR over open infrarenal aortic aneurysm repair (OAR) in a variety of patient populations; however, trials with earlier device models have demonstrated a possible expiration date to this benefit and a higher rate of graft complications and need for reinterventions with EVAR compared with OAR, with an estimated complication rate in most series of approximately 30% and a 5-year reintervention rate of 20% to 30%. While the Dutch Randomized Endovascular Aneurysm Management (DREAM) trial 2017 update noted a 38.8% reintervention rate for EVAR compared with 21.1% for OAR at 12 years, it found no significant survival difference (38.5% EVAR versus 42.2% OAR). This provides further corroboration of EVAR’s safety but critiques its durability and reemphasizes the potential for improvement.
As with any other procedure, a critical factor to minimize complications, both peri- and postoperatively, is the preoperative evaluation and preparation. The Society for Vascular Surgery (SVS) published reporting standards for EVAR, outlining definitions of success when performing an EVAR ( Table 4.1 ). Preoperative risk stratification and medical optimization prior to implantation are as critical as technical preparations with imaging studies and device selection for the overall success of the procedure with minimal complications. A thorough and accurate understanding of a patient’s preoperative status and mitigation of physiological risk is of critical importance in safe peri- and postoperative care. However, all untoward outcomes cannot be avoided. This chapter aims to describe the general complications of EVAR and practices to prevent, recognize, and manage them as they arise.
Cardiovascular
Pulmonary
Renal
Thromboembolic
Ischemia
Sedation
Radiation
Postimplantation syndrome
Infection
| Successful device introduction and deployment |
| Adequate proximal and distal fixation | |
| Access from remote site without artificial conduit | |
| Absence of rupture, death, or conversion to open surgery | |
| Absence of Type I or III endoleak | |
| Graft patency without kinks, twists, or occlusions | |
| Technical success |
| Absence of aneurysm expansion (≥5 mm diameter or ≥5% volume) | |
| Absence of graft migration (≥10 mm proximal edge) | |
| Absence of graft dilation (≥20% diameter) | |
| Absence of graft infection | |
| Absence of failure of graft integrity | |
| Primary success | Success by above terms without need for additional intervention |
| Primary assisted success | Success with additional unplanned endovascular intervention |
| Secondary success | Success with additional unplanned surgical intervention |
Predicting Complication
Most reinterventions after EVAR are performed with the overall goal of prevention of aneurysm expansion and rupture. Given that there is approximately a 1% per year risk of rupture across series after EVAR with most of these being fatal, lifelong surveillance is recommended and remains a major downside of EVAR compared with open repair. A systematic review identified predictors of reintervention after EVAR to attempt to tailor surveillance schedules to risk-stratified individuals. Although no conclusive amendments to surveillance were devised, common high-risk characteristics were identified. Across studies, the most common and the strongest predictors of reintervention remain morphologic characteristics, such as preoperative abdominal aortic aneurysm (AAA) size greater than 65 mm, common iliac artery size greater than 20 mm, tortuous diseased iliac arteries, complex aneurysm neck anatomy (increased diameter, decreased length, increased angulation), and non-adherence to the instructions for use.
The SVS published a complex Anatomic Severity Grading score for predicting various aortic complications, factoring in a multitude of morphologic characteristics such as aortic neck diameter, length, and angulation; iliac vessel tortuosity, calcification, and diameter; presence and number of branch vessels; as well as several other factors. To date, there are several such risk prediction tools of reintervention risk after EVAR. The St George’s Vascular Institute (SGVI) risk score ( Fig. 4.1 ), which has performed well in external validation studies, dichotomizes patients into high- and low-risk groups for reintervention risk based on two parameters: maximum aortic diameter and common iliac artery diameter. The DREAM Study Group performed a retrospective validation study of the SGVI score and found that it reliably predicts a 5-year reintervention risk between high- and low-risk groups (hazard ratio=4.06). The implications of such scoring tools lie in the potential to customize postoperative patient care and graft surveillance; however, there remains a need for further prospective studies and external multi-center validation to validate their widespread introduction and to affect changes to management guidelines.
Systemic Complications
EVAR patients can be elderly and frail with multiple comorbidities and have a high incidence of cardiac and cerebrovascular disease, both known and occult. As a result, preoperative risk stratification is paramount to ensure safe care. Chaikof et al. published a medical comorbidity grading system ( Table 4.2 ) based on the systemic complications of EVAR published by the SVS in 2002, dividing and weighing risk based on cardiac, pulmonary, and renal factors, which generates a global assessment of a patient’s clinical risk. Morbidity and mortality are also predicted by frailty, defined loosely as a diminution of physiologic reserve. The Modified Frailty Index (mFI), for example, has been used to demonstrate this association between frailty and morbidity and mortality after EVAR. A study of 17,668 patients in the National Surgical Quality Improvement Program (NSQIP) database who underwent elective EVAR demonstrated a reliable and direct association, with increasing morbidity and mortality with each point increase in the mFI.
| Components | Pts | Characteristics | |
|---|---|---|---|
| Major |
| 0 |
|
| 1 | Asymptomatic or Miocardial Infarction (MI) MI>6 months ago or occult Ml on EKG or fixed defect on reperfusion scan | ||
| 2 | Stable angina or significant reversible defect on reperfusion scan or significant silent ischemia on Holter monitor or EF 25% to 45% or controlled ectopy or asymptomatic dysrhythmia or compensated CHF | ||
| 3 | Unstable angina or symptomatic or poorly controlled ectopy or dysrhythmia or poorly compensated CHF or EF <25% or MI <6 months ago | ||
| 0 | Asymptomatic and normal chest x-ray and PFT within 20% of predicted | |
| 1 | Asymptomatic or mild exertional dyspnea and mild chronic parenchymal changes on chest x-ray and PFT 65% to 80% of predicted | ||
| 2 | Between 1 and 3 | ||
| 3 | Vital capacity <1.85 L or FEV 1 <1.2 L or <35% of predicted or maximal voluntary ventilation <50% of predicted or PCO 2 >45 mmHg or supplemental oxygen use or pulmonary hypertension | ||
| 0 | No known renal disease and normal serum creatinine | |
| 1 | Moderately elevated serum creatinine but <2.2 mg/dL | ||
| 2 | Serum creatinine 2.5 mg/dL to 5.9 mg/dL | ||
| 3 | Serum creatinine >6 mg/dL or hemodialysis or history of kidney transplant | ||
| Minor |
| 0 | None |
| 1 | Controlled with single drug | ||
| 2 | Controlled with two drugs | ||
| 3 | Uncontrolled or requires >2 drugs | ||
| 0 | <55 years | |
| 1 | 55–69 years | ||
| 2 | 70–79 years | ||
| 3 | >80 years | ||
Cardiovascular Complications
Cardiac complications remain the most common cause of post-EVAR mortality. According to a 2017 study of 24,813 patients in the NSQIP Participant Use Data Files who underwent EVAR between 2005 and 2013, the overall 30-day mortality for EVAR was 1.4% and the incidence of postoperative myocardial infarction (MI) or cardiac arrest was 1.4%. These events increased to 8.5% and 4.3%, respectively, among patients with three comorbidities from the surgical risk factor inclusion criteria list ( Box 4.1 ).
MI within 6 months
Any history of cardiac intervention
Angina within 30 days
CHF within 30 days
History of COPD within 30 days
Cr >2.26 mg/dL
Hemodialysis within 2 weeks
Preoperative cardiopulmonary evaluation ( Fig. 4.2 ) begins with a detailed history of prior cardiac events and interventions as well as the existence of coronary artery disease equivalents and active disease. All patients should have routine blood work, a chest radiograph, and an electrocardiogram as indicated. Further cardiac workup is dictated by their risk stratification. Patients with acceptable exercise tolerance and those classified as low risk by the Vascular Study Group of New England Cardiac Risk Index do not require further cardiac testing, whereas patients with low or unknown exercise tolerance and a “high risk” classification merit further workup. Patients with new cardiac murmurs and/or signs or symptoms concerning unrecognized heart failure should have a preoperative echocardiogram and a stress test prior to elective EVAR to address potential valvular disease and/or heart failure and reduce the risk of perioperative adverse cardiac events. Medical management to optimize the patient is recommended prior to intervention. Patients with stable cardiac disease have not been shown to receive any mortality benefit from coronary revascularization unless it was planned prior to the decision to pursue aneurysm repair.
Cardioprudent medications such as beta blockers, statins, and antiplatelet agents should be evaluated and optimized appropriately, preferably one month prior to surgery. The beneficial effects of statins, which are derived from both serum cholesterol-lowering and pleiotropic effects, have been shown in multiple prospective randomized studies to reduce perioperative cardiac mortality significantly. Beta blockade should not be discontinued or initiated abruptly prior to surgery. However, in high-cardiac-risk individuals, initiation of beta blockade a week to a month prior to surgery can confer a survival benefit.
Concern regarding bleeding risk and other complications with the continued use of antiplatelet agents or other anticoagulants has been evaluated in several studies. One such study of 407 patients on chronic antiplatelet or anticoagulation therapy who underwent EVAR found no significant difference in the rate of endoleaks or aneurysmal sac expansion compared with patients not on therapy over a median follow-up period of 18 months. Antiplatelet agents can be safely continued through the perioperative period for EVAR. Chronic anticoagulation should be evaluated on a case-by-case basis to determine whether cessation or a heparin bridge is appropriate depending on indication for therapy and individual bleeding risk.
Pulmonary Complications
Tobacco use and variants of chronic obstructive pulmonary disease (COPD) can be intimately related to aneurysm development. Many EVAR patients may have underlying pulmonary disease, which can be diagnosed or undiagnosed at the time of presentation. According to an analysis of the NSQIP database, 30% of EVAR patients are active smokers at the time of intervention. With the decreased use of general anesthesia with EVAR compared with OAR, the incidence of respiratory complications is lower, but continues to be a significant source of morbidity and the second most common cause of mortality after EVAR. The incidence of post-EVAR pneumonia is 1.3%, unplanned reintubation 1.5%, and prolonged ventilator dependence 1.2%.
Severe COPD has been spirometrically defined as FEV 1 to FVC ratio of less than 0.7 or FEV 1 less than 30% of predicted. Severe COPD, decreased FEV 1 (1.9 L among nonsurvivors versus 2.3 L among survivors), and decreased FVC (3.0 L among nonsurvivors versus 3.4 L among survivors) have been independently associated with significantly lower 5-year survival after EVAR. Therefore, it is crucial to characterize and to optimize respiratory status preoperatively prior to intervention. Despite this, pulmonary function assessment prior to EVAR is often overlooked. A prospective observational study of 200 EVAR patients found when all patients underwent preoperative pulmonary function tests (PFTs), 57% of patients were found to have previously undiagnosed and untreated respiratory disease.
Active smokers should quit smoking at least 2 to 4 weeks prior to intervention to exceed the postcessation window of hypersecretion and impaired ciliary clearance ( Fig. 4.3 ). A risk index for pulmonary complications, such as the ARISCAT risk index, can stratify patients by risk to determine who may require further preoperative testing. Given the association between poor pulmonary function and post-EVAR mortality, any patient with history or physical examination for respiratory disease should undergo preoperative PFTs and medical optimization of pulmonary function.
Renal Complications
Renal dysfunction following EVAR has a wide range of reported incidences depending on the definition of postoperative renal dysfunction used. A recent systematic review of post-EVAR renal dysfunction estimated a risk of clinically relevant renal dysfunction at one year to be 18%. This injury is attributed to a combination of contrast-induced nephropathy from intraoperative contrast usage, surveillance imaging, renal infarction secondary to suprarenal renal fixation or microembolic disease from juxtarenal device manipulation, and ischemia-reperfusion injury. Moreover, this injury has potential life-long implications. A study of 212 consecutive EVAR patients identified a significantly increased rate of chronic kidney disease in EVAR patients compared with risk-controlled nonoperated aneurysm patients (23.5% versus 6.7%), and postprocedural acute kidney injury (AKI) was a significant risk factor for CKD.
Furthermore, postoperative AKI has been directly associated with increased incidence of cardiovascular events and all-cause mortality. A study of 947 patients undergoing EVAR followed over a period of 33 months postprocedurally found that those who developed AKI had a higher risk of mortality (32.1% versus 1.7%) than those who did not develop AKI. Pre-existing renal disease was the most significant risk factor for post-EVAR AKI. These studies identified other variables such as longer operative time, larger volume of contrast used, preoperative ACE inhibitor use, decreased preoperative cardiovascular reserve, and advanced age as prominent risk factors for perioperative kidney injury.
Pre- and intraoperative intravenous hydration may mitigate contrast-induced nephropathy (CIN), although results of studies regarding its efficacy are conflicting. Risk of CIN can also be decreased through the utilization of carbon dioxide as the contrast medium whenever feasible. One series achieved an 85% technical success rate for EVAR solely using carbon dioxide angiography, with only one case requiring reversion to iodinated contrast because of inadequate image quality. The patients with endoleaks were able to be diagnosed using carbon dioxide angiography and treated using very small volumes of iodinated contrast material. The risk of CIN can also be avoided through the increasing use of contrast-enhanced ultrasound (CEUS) for graft surveillance, which carries no additional risk of renal dysfunction as opposed to serial Computer Tomographic Angiography (CTA) examinations.
The association between suprarenal fixation and postoperative renal dysfunction has not yet been clearly elucidated in well-powered, long-term, controlled studies. Suprarenal fixation, while shown to decrease the incidence of EVAR graft migration, may be a nidus for thrombus with partial or complete occlusion of a renal artery over time. A study of 2574 EVAR patients found that suprarenal fixation was associated with a significantly higher incidence of serum creatinine increase of at least 0.5 mg/dL (3.7% versus 2%) and twice the likelihood of clinical renal deterioration. However, this study, as with the majority of other similar reports, failed to account completely for patient and graft selection bias, as patients for whom suprarenal fixation was utilized tended to have more severe juxtarenal disease at the time of intervention.
Another important consideration is the presence of an accessory renal artery (ARA), which can have a prevalence of up to one-third of the population. A clinically relevant ARA originates inferior to the main renal artery, supplies one-third of the renal parenchyma, and has a diameter greater than 4 mm. Exclusion of a clinically relevant ARA can lead to renal infarction with or without a decrease in glomerular filtration rate (GFR). A study of 34 patients with ARA who underwent EVAR compared with 102 matched patients without ARA found a significantly higher incidence of depressed GFR in the ARA group (decrease of 10.7 versus 1.2 mL/min/1.73 m 2 ). This study demonstrated that the presence of ARA is a clinically significant anatomic variant that merits attention in the preoperative workup with potentially serious implications if overlooked. Furthermore, there are techniques described to preserve clinically relevant ARA such as parallel graft placement during the EVAR implantation. In a small series of nine infrarenal EVARs that had chimney grafts placed in clinically relevant ARAs, over a mean follow-up of 13 months, eight of the nine patients had either no change or an improvement in their CKD stage. Given that short- and long-term renal decline is associated with increased morbidity and mortality, preservation of renal parenchyma and function should be considered in regard to clinically relevant ARA.
Deep vein thrombosis and pulmonary embolism
The incidence of acute deep vein thrombosis (DVT) after EVAR according to the NSQIP database is 0.5%, with similar rates across multiple large series. There exists some concern that percutaneous EVAR (PEVAR) with suture-mediated “pre-close” technique may precipitate development of a DVT. A series of 52 PEVAR patients retrospectively analyzed demonstrated a 4% incidence of proximal lower extremity DVT, which is comparable to that of other vascular procedures including open aneurysm repair. Pulmonary embolism after EVAR is an even less common complication, reported at 0.2% in the NSQIP database; however, any provider caring for a patient with an AAA must be aware of the heightened risk of venous thromboembolic disease among aortic aneurysm patients. A cohort study of the National Health Research Institutes database in Taiwan noted that, after matching age, prothrombotic comorbidity, and length of hospitalization, untreated patients with AAA had a 1.89-fold higher risk of DVT and/or PE compared with non-AAA patients. Therefore, while postoperative EVAR patients do not warrant heightened DVT or PE testing unless otherwise clinically indicated, clinical suspicion should be maintained given their higher baseline risk.
Ischemic complications
The main mechanisms of ischemia are cholesterol or thrombus embolization, thrombosis, and inadvertent vessel coverage, most commonly affecting the lower extremities, pelvis, and kidneys. However, atheromatous debris can embolize to every conceivable vascular bed. In a study of a series of 311 infrarenal EVAR patients, the total ischemic complication rate was 9%, with 6.8% lower extremity ischemia, 2.3% pelvis ischemia, and 0.6% spinal cord ischemia. The presentation naturally depends on the location of ischemia. Skin manifestations may occur in the form of petechiae, cyanosis, livedo reticularis, or frank necrosis.
Spinal Cord
Spinal cord ischemia resulting in paraparesis or paraplegia after infrarenal EVAR is an extremely rare complication and one that is not commonly reported in studies. In an analysis of the EUROSTAR database, the incidence of spinal cord ischemia was 0.2%. While a rare complication, this highlights the importance of understanding the patient’s anatomy and vascular supply to the spinal cord prior to intervention, which should be reviewed prior to graft deployment.
EVAR patients presenting with paraparesis or paralysis require expeditious workup with an immediate CT of the abdomen and pelvis to evaluate for hematoma compressing the cord followed by an MRI to confirm spinal cord infarction if the graft is MR-compatible. Neurosurgical consultation for possible lumbar drain placement should be made early.
Lower Extremity
Ischemia of the lower extremities can be a result of embolism as well as thrombotic occlusion. The incidence of leg ischemia during EVAR is approximately 7%. Delineation of the vascular status of the lower extremity preoperatively and careful perioperative monitoring of the vascular exam are paramount to awareness of ischemic time and the possibility of compartment syndrome and consequent need for decompressive fasciotomy. The use of large-caliber sheaths can cause occlusive ischemic insult. Therefore, the largest sheath should be removed first in order to evaluate for ischemia while maintaining access in the event that a diagnostic angiogram is indicated. In other cases, patients may present days to weeks postoperatively with claudication, rest pain, or even frank gangrene. Management of lower extremity ischemia takes many different forms depending on the etiology and the presentation: catheter-directed thrombolysis, angioplasty and stenting, embolectomy, endarterectomy with or without patch angioplasty, bypass grafting, and major amputation.
Colon
Intestinal ischemia has a reported incidence of 1.5% to 3% after EVAR. While uncommon, vigilant perioperative monitoring for this potentially devastating complication is of utmost importance. The prevalence of mesenteric atherosclerosis, often occult, has been found on autopsy studies to be in the range of 6% to 10%. While intestinal perfusion can be maintained through various anastomoses between the celiac (CA), superior mesenteric (SMA), and inferior mesenteric arteries (IMA), if one vessel is chronically occluded and a widely patent IMA is excluded by stent graft deployment, the single remaining visceral branch may provide inadequate perfusion. Alternatively, atheroembolism to any of the visceral branch vessels can cause similar hypoperfusion injuries.
The effect of hypogastric artery interruption has been shown in some series to be associated with a significantly increased number of pelvic complications, but in others bilateral embolization has not been shown to be associated with any serious ischemic complications. A small retrospective study of 16 patients who underwent preoperative hypogastric artery occlusion found no association with serious pelvic complications, including bowel ischemia. The only complications noted were buttock claudication and sexual dysfunction. Ultee et al., however, noted in a series of 4675 EVAR patients that unilateral hypogastric artery coverage yielded a 1:7 odds ratio of bowel ischemia.
Risk factors for bowel ischemia are difficult to define cleanly given that it shares many factors with the frequent comorbidities of AAA patients, such as smoking, dysrhythmia, and heart failure. Moghadamyeghaneh et al. conducted a retrospective study of 3082 EVAR patients who developed postoperative mesenteric ischemia and determined a number of risk-adjusted factors associated with postoperative mortality ( Table 4.3 ). Ruptured aneurysm and transfusion requirement were the strongest predictors of post-EVAR bowel ischemia, theorized to be caused by the interruption of perfusion through the IMA in these scenarios. Patients who developed bowel ischemia had a fourfold increased risk of postoperative mortality, further underscoring the gravity of this complication.
