Recurrent infection is more common after endovascular versus open repair of infected abdominal aortic aneurysm: Systematic review and meta-analysis.

OBJECTIVE Controversy has continued regarding the use of endovascular aneurysm repair (EVAR) vs open aneurysm repair (OAR) for infected abdominal aortic aneurysms (AAAs). In the present study, we investigated the comparative outcomes of EVAR and OAR for the treatment of infected AAAs. METHODS We conducted a systematic review and meta-analysis using the MEDLINE and EMBASE databases through May 2021. We included studies that had described both EVAR and OAR for the treatment of infected AAAs. The primary endpoints were the rates of recurrent infection and related rupture and/or death. Perioperative and 1-year mortality and readmissions and reinterventions were also analyzed. RESULTS Fourteen observational studies describing a total of 1203 patients (EVAR, 359 [29.8%]; OAR, 844 [70.2%]) were eligible for qualitative analysis. The baseline characteristics included diabetes mellitus (33.2%), fever at presentation (71.6%), rupture at diagnosis (26.1%), and positive blood cultures (52.5%). The mean follow-up period ranged from 12 to 40 months. The use of EVAR became more prevalent in recent years (2016-2020, 32.4%) compared with the former period (2010-2015, 13.8%; P < .0001). Fenestrated, branched, or concomitant visceral debranching EVAR was performed in 6.1% of cases. In OAR, surgical debridement was consistently performed, and in situ reconstruction was applied in 82.2% and an omental flap in 51.5%. In nine studies considered for quantitative analysis, the patients' background (EVAR, n = 264; OAR, n = 274) were statistically balanced. The crude rates of recurrent infection and related rupture or death were 13.6% (95% confidence interval [CI], 8.8%-18.5%) and 4.9% (95% CI 1.8%-8.0%), respectively. The pooled analyses depicted significantly higher rates of recurrent infection after EVAR than after OAR (relative risk [RR], 2.42; 95% CI, 1.80-3.27; P < .0001; I2 = 0%). Recurrent infection-related rupture or death (RR, 1.51; 95% CI, 0.70-3.23; P = .29; I2 = 0%), perioperative death (RR, 0.80; 95% CI, 0.39-1.65; P = .55; I2 = 35%), 1-year mortality (hazard ratio, 1.12; 95% CI, 0.97-1.28; P =.13; I2 = 0%), and readmission or reintervention (RR, 1.16; 95% CI, 0.74-1.82; P =.52; I2 = 0%) were not significantly different statistically between the two groups. Funnel plots showed no evidence of publication bias. Sensitivity analyses of leave-one-out meta-analysis confirmed higher rates of recurrent infection after EVAR. CONCLUSIONS EVAR has become more prevalent as the initial treatment of infected AAAs. Although operative and 1-year survival were similar between OAR and EVAR groups, recurrent infection was more frequent after EVAR. This limitation should be weighed in selecting patients for EVAR in infected AAAs. Postoperative graft and infection surveillance are critical, especially after EVAR.