A systematic review and meta-analysis of long-term reintervention after endovascular abdominal aortic aneurysm repair

Objective: Patients who undergo endovascular aneurysm repair (EVAR) often require reintervention after the index repair. The long-term rate of reintervention and how this has changed with newer device technology are poorly understood. Therefore, we performed a systematic review and meta-analysis of the available literature to determine long-term freedom from reintervention after EVAR and the change in reintervention rates over time. Methods: We performed a systematic review of MEDLINE, Embase, Cochrane Library, and ClinicalTrials.gov in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We included randomized controlled trials and observational studies that documented the rate of reintervention after EVAR. We performed a meta-analysis of Kaplan-Meier freedom from reintervention at each year after EVAR. We used linear regression to evaluate change in reintervention rate over time with newer device technology. Results: We included a total of 30 studies (randomized trials, n = 3; observational studies, n = 27) comprising 32,126 patients in this review and meta-analysis. Studies ranged in the implantation date of the EVAR device from 1996 to 2014. The probability of freedom from reintervention was 81% (95% confidence interval [CI], 77%-85%) at 5 years, 70% (95% CI, 65%-76%) at 10 years, and 64% (95% CI, 46%-79%) at 14 years. Linear regression demonstrated an improvement in freedom from reintervention when results were stratified by the year of device implantation. At 1 year, estimated freedom from reintervention improved from 90% in 1998 to 94% in 2008 (n = 26 studies; R² = 0.11; P =.10). At three years, estimated freedom from reintervention improved from 77% in 1998 to 90% in 2008 (n = 26 studies; R² = 0.27; P =.006). At 5 years, estimated freedom from reintervention improved from 68% in 1998 to 81% in 2008 (n = 22 studies; R² =0.12; P =.12). At 7 years, estimated freedom from reintervention improved from 51% in 1998 to 86% in 2011 (n = 22 studies; R² = 0.40; P =.015). Conclusions: EVAR patients remain at risk for reintervention indefinitely, and therefore lifelong surveillance is imperative. Encouragingly, reintervention rates have improved over time, with newer devices exhibiting lower rates. Reintervention rate remains an important metric for new devices and registries.