Abstract
PURPOSE: To determine the frequencies of complications and risk factors for complications following endovascular abdominal aortic aneurysm (AAA) repair (EVAR). METHODS: Thirty-nine articles published between October 1995 and October 1999 in English, German, French, or Dutch were identified in electronic databases. All articles reported perioperative (30-day) complications on 2387 patients, while postoperative complications (>30 days) were reported on 1645 patients in 30 of 39 studies. Data were pooled and subjected to multivariable logistic regression analysis to identify risk factors for death, endoleak, rupture, conversion, and local and systemic complications. Effects of risk factors were expressed as odds ratios (OR) with 95% confidence interval (CI). RESULTS: Mean follow-up was 13.9 months (range 4-29). The perioperative death rate (3.7%) was lower than postoperative mortality (5.0% per annum). The risks for both endoleak (13.1%) and conversion (5.0%) were higher in the perioperative period than postoperatively (5.4% and 1.4% per annum, respectively). Independent risk factors for perioperative complications were: general anesthesia (death: OR = 5.1, 95% CI 1.9-13.3); EVT graft (endoleak: OR = 3.0, 95% CI 1.3-7.0); female sex (rupture: OR = 2.8, 95% CI 1.4-5.8); hypertension (conversion: OR = 0.03, 95% CI 0.0-0.3); age >70 years (conversion: OR = 3.5, 95% CI 1.3-9.2); and team experience >30 patients (conversion: OR = 3.0, 95% CI 1.2-1.6). Independent risk factors for postoperative complications were: predischarge examination (death: OR = 0.2, 95% CI 0.0-0.7); follow-up 30 days after operation (death: OR = 0.3, 95% CI 0.1-1.0); and female sex (rupture: OR = 1.4, 95% CI 0.5-4.4; conversion: OR 6.8, 95% CI 2.0-23.4; and systemic complications: OR = 2.9, 95% CI 1.1-7.5). CONCLUSIONS: The risk of complications after EVAR is high, supporting the cautious use of EVAR. Both patient characteristics and procedural variables were independent risk factors for complications. To avoid the limitations of this study, the results of randomized clinical trials must determine if EVAR offers a safe and durable alternative to open AAA repair.
