Proper timing or ERCP and cholecystectomy on acute cholecystitis: a systematic review and meta-analysis.

Abstract

Purpose: To determine if endoscopic retrograde cholangiopancreatography (ERCP) should be performed with surgery or as a different step, on acute cholecystitis, and which strategy has the least complications and morbimortality.

Methods: Various databases (PubMed, Embase, Scopus, Web of Science, Science Direct, Cochrane Central Register of Controlled Trials, CINAHL, Latin American and Caribbean Health Sciences Literature, clinical trials, Google Scholar) were searched for randomized trials comparing the different timings for ERCP and cholecystectomy. No language or time restrictions were applied. Risk of bias was assessed with RoB 2.0 (Cochrane's Risk of Bias 2), and evidence certainty evaluated using Grading of Recommendations Assessment, Development and Evaluation. Data synthesis used R-4.1.0 Project for Statistical Computing for Windows, with meta-analysis via fixed-effects model and I2 for heterogeneity.

Results: Eleven studies was used, and meta-analysis was performed independently for each outcome. Different outcomes were evaluated, with preoperative ERCP as an intervention and intraoperative ERCP as the control: length of stays (four trials with mean differences - MD = -1.44; 95% confidence interval - 95%CI -3,87-0,98); bile leak (odds ratio - OR = 0.67; 95%CI 0.11-4.09); cholangitis (OR = 1.32; 95%CI 0.29-5.98); bleeding from sphincterotomy (OR = 0.98; 95%CI 0.20-4.86); wound infection (OR = 0.33; 95%CI 0.04-3.14); incisional bleeding (OR = 0.5; 95%CI 0.04-5.70); elevated amylase activity (OR = 5.22; 95%CI 2.17-12.59); acute pancreatitis (OR = 4.61; 95%CI 1.72-12.38); operative time (MD = -6,26; 95%CI -37.24-24.73); failure rate (OR = 1.74; 95%CI 0.99-3.05); conversion (OR = 1.34; 95%CI 0.6-2.96); morbidity (OR = 2.75; 95%CI 1.7-4.47).

Conclusions: Risk of bias was significant due to lack of blindness. The morbidity, pancreatitis, and elevated amylase activity outcomes were the only ones to find statistical significance and favored the intraoperative approach.