LAUSR

The objectives were to determine the surgical site infection incidence (including superficial/deep) fter arterial intervention through non‐infected groin incisions and identify variables associated with incidence. MEDLINE, EMBASE and CENTRAL databases were searched for randomised controlled trials and observational studies of adults undergoing arterial intervention through a groin incision and reported surgical site infection. Infection incidence was examined in subgroups, variables were subjected to meta‐regression. One hundred seventeen studies reporting 65 138 groin incisions in 42 347 patients were included. Overall surgical site infection incidence per incision was 8.1% (1730/21 431): 6.3% (804/12 786) were superficial and 1.9% (241/12 863) were deep. Superficial infection incidence was higher in randomised controlled trials (15.8% [278/1762]) compared with observational studies (4.8% [526/11 024]); deep infection incidence was similar (1.7% (30/1762) and 1.9% (211/11 101) respectively). Aneurysmal pathology (β = −10.229, P < .001) and retrospective observational design (β = −1.118, P = .002) were associated with lower infection incidence. Surgical site infection being a primary outcome was associated with a higher incidence of surgical site infections (β = 3.429, P = .017). The three‐fold higher incidence of superficial surgical site infection reported in randomised controlled trials may be because of a more robust clinical review of patients. These results should be considered when benchmarking practice and could inform future trial design.