Six RCTs were included (2,899 participants, range 40 to 1,252). All studies reported adequate randomisation sequence generation methods but methods of allocation concealment were unclear for five of the trials. All studies were at low risk of bias for reporting outcome data and for selective outcome reporting. Only two studies reported adequate methods for outcome assessor blinding; reporting was unclear in four studies.
At 30 days the risk of all-cause mortality was lower in the endovascular group than in the open surgery group (RR 0.35, 95% CI 0.19 to 0.64; five trials). At two years the difference was not statistically significantly different (five trials) but abdominal aortic aneurysm-related mortality was significantly lower (RR 0.46, 95% CI 0.28 to 0.74; four trials) and re-intervention rates were higher (RR 1.48, 95% CI 1.06 to 2.08; five trials) in the endovascular group than in the open surgery group. There was no evidence of statistically significant heterogeneity and no evidence of publication bias.
At three years (or longer) there was no significant difference between groups in all-cause mortality and in abdominal aortic aneurysm-related mortality but the significant difference in rate of re-interventions was still evident (RR 2.53, 95% CI 1.58 to 4.05; four trials); statistically significant heterogeneity was evident at three years for the aneurysm-related mortality and re-intervention rates analyses. There was evidence of publication bias for the all-cause mortality analysis.
Sensitivity analysis results were reported.