Timing of cholecystectomy following endoscopic sphincterotomy: A population-based study

Brett D Mador, MD1, Avery B Nathens, MD, MPH, PhD1, Wei Xiong, MSc1, O N Panton, MB2, S M Hameed, MD, MPH2. 1Sunnybrook Health Sciences Centre, 2University of British Columbia

Introduction: Choledocholithiasis is commonly treated initially with endoscopic sphincterotomy, followed by cholecystectomy to definitively address the underlying problem of cholelithiasis. While the benefits of early cholecystectomy have been realized in other populations, the preferred timing for this subset of patients is less well established. We performed a large, population-based analysis to determine the frequency, benefits, and practice variance in regards to early cholecystectomy on a provincial level.

Methods and Procedures: Patients undergoing endoscopic sphincterotomy followed by cholecystectomy in British Columbia, Canada from January 2001 to December 2011 were identified using fee-code billing data. Multiple databases were linked to obtain information on demographics, admissions, procedures, mortality, and census geographic data. Student’s t-test and chi-square test were used to compare groups (early versus delayed cholecystectomy). Logistic regression analysis was performed for length of stay (LOS) and biliary re-admission data. Outcome data was risk-adjusted for age, gender, comorbidities, socio-economic status (SES), and year of procedure. Variability of early cholecystectomy rates across census areas was determined using funnel plots of adjusted odds ratios.

Results: 4287 patients met inclusion criteria. 1905 (44.4%) underwent early cholecystectomy, defined as surgery within 14 days of sphincterotomy. Mean interval to cholecystectomy was 2 days for the early cholecystectomy group and 61 days for delayed. There was a significant difference in hospital LOS favoring early cholecystectomy for patients with documented gallstone disease (p<0.05). Patients initially admitted to a surgical service were more likely to undergo early cholecystectomy (0.60 vs 0.36, p<0.001). There was no difference between groups in terms of mortality or SES. There was wide variability in rates of early cholecystectomy among census areas (range 0.20 – 0.66) and health regions (range 0 – 0.96) which was not explained by population density or geography.

Conclusion: A significant clinical benefit is associated with early cholecystectomy for gallstone disease post-sphincterotomy. Despite this, a large amount of clinical variance exists in regard to timing of cholecystectomy which seems to be primarily institution-dependent.

« Return to SAGES 2016 abstract archive