An observational study of cholecystectomy in patients receiving tegaserod.

BACKGROUND: Registrational studies of patients treated with tegaserod for irritable bowel syndrome (IBS) suggest an increased risk for cholecystectomy versus treatment with placebo. OBJECTIVE: To study cholecystectomy rates in association with tegaserod within a large administrative medical claims database. METHODS: Patients were drawn from a large population within the US with commercial medical insurance. The primary analysis consisted of a comparison of the observed incidence rate for cholecystectomy claims among a large cohort of new-to-therapy tegaserod users with an incidence rate published for tegaserod-naive patients classified with IBS within the same insured population. RESULTS: An inception cohort of 7475 individuals with up to 103 weeks of claims history following initiation of therapy with tegaserod was identified. After a follow-up of 3 months (and thus similar to the longest registrational trials), the observed cholecystectomy incidence rate was 340 per 10,000 person-years (95% CI 258, 442). The rate of cholecystectomy was highest in the earliest months of observation following initiation of tegaserod. The observed cholecystecomy incidence rate is 2.9 times higher than an IBS-specific rate of 119 per 10,000 person-years as published for patients so classified within the same insured population. CONCLUSION: Based on a large, inception cohort, we report a strong temporal association between the initiation of tegaserod therapy and an increased rate for cholecystectomy. The effect size at 3 months was similar to the relative risk for cholecystectomy reported in registrational studies comparing tegaserod with placebo. As misclassification of initial diagnosis for patients presenting with biliary colic-like symptoms may occur, precise measurements of tegaserod-related relative risk for cholecystectomy from observational studies are problematic and will require prospective studies.