Systematic review of endoluminal vacuum-assisted therapy as salvage treatment for rectal anastomotic leakage.

Background: Endoluminal vacuum-assisted therapy (EVT) has been introduced recently to treat colorectal anastomotic leakage. The aim of this study was to evaluate the safety and efficacy of EVT in the treatment of anastomotic leakage and rectal stump insufficiency after Hartmann's procedure.

Methods: A systematic search of MEDLINE, Scopus and Cochrane databases was performed using search terms related to EVT and anastomotic leakage or rectal stump insufficiency in line with the PRISMA checklist. Observational studies, RCTs and case series studies published to July 2017 were included. Primary outcomes of the review were the success of EVT, defined as complete or partial healing of the anastomotic defect and associated cavity, and the rate of stoma reversal after EVT. Secondary outcomes included the duration of treatment to complete healing, complications of treatment and the need for further intervention. A meta-analysis was conducted. The potential effect of clinical confounders on the failure of EVT was investigated using the random-effects meta-regression model.

Results: Of 476 articles identified, 17 studies reporting on 276 patients were ultimately included. The weighted mean rate of success was 85·3 (95 per cent c.i. 80·1 to 90·5) per cent, with a median duration from inception of EVT to complete healing of 47 (range 40-105) days. The weighted mean rate of stoma reversal across the studies was 75·9 (64·6 to 87·2) per cent. Twenty-five patients (9·1 per cent) required additional interventions after EVT. Thirty-eight patients (13·8 per cent) developed complications. The weighted mean complication rate across the studies was 11·1 (6·0 to 16·2) per cent. Variables significantly associated with failure included preoperative radiotherapy, absence of diverting stoma, complications and male sex.

Conclusion: EVT is associated with a high rate of complete healing of anastomotic leakage and stoma reversal. It is an effective option in appropriately selected patients with anastomotic leakage.