Probability of developing gallstones and related risk factors in 400 cirrhotics

Objective: To estimate the cumulative probability of developing gallstones and to evaluate possible risk factors in a large series of Italian cirrhotic patients, free of gallstones at enrolment. Setting: Gastroenterology unit at a Milan teaching hospital and four general hospitals in northern Italy. Patients: The study included 400 consecutive cirrhotic patients (257 men and 143 women, mean age 58 ± 10 SD and 61 ± 10 SD years, respectively). Cirrhosis was due to alcohol abuse (n = 169), chronic infection with hepatitis B virus (n = 31) or non-A, non-B hepatitis virus (NANBV; n = 107, including 75 positive for anti-hepatitis C virus), concomitant chronic hepatitis B virus or NANBV infection and alcohol abuse (n = 50), and miscellaneous causes (n = 43). At enrolment, 231 patients were classified as having Child's class A cirrhosis, 130 as B and 39 as C. Statistical analysis: The cumulative probability of developing gallstones was estimated by the Kaplan-Meier method and possible risk factors evaluated by logistic regression analysis. Results: During a median follow-up of 48 months (range 12 to 132 months), ultrasonography demonstrated the development of gallstones in 94 patients (23.5%, 53 men and 41 women), who were all informed of the finding. The proportion of symptomatic patients was 10% at diagnosis, whereas 16.2% manifested gallstone-related symptoms during follow-up. Cholecystectomy was carried out in three patients. The cumulative probability of gallstone development was 15 and 38% at 4 and 8 years, respectively. On logistic regression analysis, high body mass index, female gender and severity of liver disease were significantly associated with a higher risk of cholelithiasis. Conclusion: This longitudinal study provides further evidence of the clinical importance of gallstones in cirrhosis, already demonstrated in prevalence studies.