Objectives
The aim was to assess peri-operative mortality of endovascular aneurysm repair (EVAR) vs. open repair for ruptured abdominal aortic aneurysm (AAA) and to investigate whether outcomes have improved over the years and whether there is an association between institutional caseload and peri-operative mortality.

Methods
Electronic information sources (MEDLINE, EMBASE, CINAHL and CENTRAL) were searched up to August 2019. A systematic review was carried out according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines using a registered protocol (CRD42018106084). Studies were selected that reported peri-operative mortality of EVAR for ruptured AAA. A proportion meta-analysis was conducted, and summary estimates of odds ratios (ORs) and 95% confidence intervals (CIs) for EVAR vs. open surgical repair were obtained using random effects models. Mixed effects regression models were developed to investigate outcome changes over time and with institutional caseload.

Results
One hundred and thirty-six studies were included in quantitative synthesis reporting a total of 267 259 patients (EVAR 58 273; open surgery 208 986). The pooled peri-operative mortality of EVAR and open surgical repair was 0.245 (95% CI 0.234–0.257) and 0.378 (95% CI 0.364–0.392), respectively. EVAR was associated with reduced peri-operative mortality (OR 0.54, 95% CI 0.51–0.57, p < .001). Meta-regression analysis found decreasing peri-operative mortality over the years following EVAR (p < .001) and open repair (p < .001), and a decreasing OR of peri-operative mortality in favour of EVAR (p = .053). Meta-regression found a significant positive association between peri-operative mortality and institutional case load for open repair (p = .004).

Conclusions
If EVAR can be done, it is a better treatment for ruptured AAA in view of the reduced peri-operative mortality compared with open surgery. The outcomes of both EVAR and open surgical repair have improved over the years, and the difference in peri-operative mortality in favour of EVAR has become more pronounced. There is a significant association between peri-operative mortality and institutional case load for open repair of ruptured AAA.