Risk Factors for Recurrence after anal fistula surgery: A meta-analysis.

<bold>Background: </bold>Despite a burgeoning literature during the last two decades regarding perioperative risk management of anal fistula, little is known about its risk factors that influence postoperative recurrence. We performed a meta-analysis to summarize and assess the credibility of evidence of potential risk factors for anal fistula recurrence (AFR) after surgery.<bold>Methods: </bold>Pubmed and EMBASE without language restriction were searched from inception to April 2018 that reported risk factors which predisposed recurrence after anal fistula surgery. We excluded studies that involved patients with anal fistula associated with Crohn's disease. MOOSE guidelines were followed when this meta-analysis was performed. We used random-effects models to pool relative risks (RRs) with 95% confidence intervals (CIs). Evidence from observational studies was graded into high-quality (Class I), moderate-quality (Class II/III) and low-quality (Class IV) based on Egger's P value, total sample size and between-study heterogeneity.<bold>Results: </bold>Of 3514 citations screened, 20 unique observational studies comprising 6168 patients were involved in data synthesis. High-quality evidence showed that AFR was associated with high transsphincteric fistula (RR, 4.77; 95% CI, 3.83 to 5.95), internal opening unidentified (RR, 8.54; 95% CI, 5.29 to 13.80), and horseshoe extensions (RR, 1.92; 95% CI, 1.43 to 2.59). Moderate-quality evidence suggested an association with prior anal surgery (RR, 1.52; 95% CI, 1.04 to 2.23), seton placement surgery (RR, 2.97; 95% CI, 1.10 to 8.06), and multiple fistula tract (RR, 4.77; 95% CI, 1.46 to 15.51). High-quality evidence demonstrated no significant association with gender or smoking; moderate-quality evidence also suggested no association with age, tertiary referral, alcohol use, diabetes mellitus, obesity, preoperative seton drainage, high internal opening, postoperative drainage, mucosal advancement flap surgery, supralevator extensions, location or type of anal fistula.<bold>Conclusion: </bold>Several patient, surgery and fistula-related factors are significantly associated with postoperative AFR. These findings strengthen clinical awareness of early warning to identify patients with high-risk disease recurrence for AFR. [ABSTRACT FROM AUTHOR]