Cholecystectomy does not worsen progression or outcomes in non-alcoholic fatty liver disease

Cholecystectomy is a frequently performed surgical procedure for symptomatic cholelithiasis, which is reported to be more common in patients with non-alcoholic steatohepatitis (NASH), given the common risk factors. However, the data remains unclear on the association of cholecystectomy with NASH. We performed a retrospective study to examine the association of cholecystectomy and NASH.

Medical charts of patients with steatohepatitis related liver disease at a tertiary care center from 2004 to 2011 were stratified by cholecystectomy and defined by its history and/or absence of gallbladder on ultrasonography. Logistic regression model was built for predictors of cholecystectomy. Patients with NASH were stratified based on timing of cholecystectomy. The diagnosis of NASH and timing of cholecystectomy were compared based on baseline characteristics and outcomes (liver disease complications and survival) on follow up. Kaplan-Meier curves were generated for the two group comparisons. Chi-square and unpaired t-tests were used for comparing outcomes on follow up. P value <0.05 was considered significant.

Analysis of 584 patients [379 non-alcoholic fatty liver disease (NAFLD)] showed that patients with cholecystectomy (N=191) were more likely to be female (57% vs. 44%), diabetic (53% vs. 37%), have liver biopsy (43% vs. 25%) and diagnosis of NAFLD (80% vs. 58%) P<0.001 for all. NAFLD diagnosis was associated with 2.79 folds odds of cholecystectomy. Among 379 (192 cholecystectomy) NAFLD patients, cirrhosis and female gender were associated with over 2 and 1.5 folds of cholecystectomy. Of 141 patients with data on timing of cholecystectomy, 55 (39%) with cholecystectomy at or after NAFLD diagnosis vs. 86 with cholecystectomy within median of 6 years prior to NAFLD diagnosis were similar on all characteristics except on model for end-stage liver disease (MELD) score (9.2±8.4 vs. 6.4±7.1, P=0.045). Of 28 with available histology data, there were no differences on histology based on timing of cholecystectomy. On a median follow up of 5 years, timing of cholecystectomy did not impact on development of cirrhosis (74% vs. 67%, P=0.45), ascites (31% vs. 38%, P=0.76), variceal bleeding (11% vs. 16%, P=0.44), hepatic encephalopathy (22% vs. 29%, P=0.74), hepatocellular carcinoma (HCC) (15% vs. 9%, P=0.59), and patient survival (95% vs. 98%, P=0.3).

Cholecystectomy is associated with NAFLD diagnosis. We did not find cause and effect of cholecystectomy in the development of severity of NAFLD. Prospective studies are suggested to examine the role of cholecystectomy and bile acids in the pathogenesis of NAFLD.