Risk of cardiovascular events and all-cause mortality in patients treated with thiazolidinediones in a managed-care population.

BACKGROUND This study directly compares risk of acute myocardial infarction (AMI), acute heart failure (AHF), or all-cause death among pioglitazone- and rosiglitazone-treated patients in a managed-care population. METHODS AND RESULTS Patients ≥18 years of age, newly initiated on rosiglitazone or pioglitazone between January 1, 2001, and December 12, 2005, were included. The date of the first pharmacy claim for rosiglitazone or pioglitazone was defined as index date. Patients were excluded if they had <1 year continuous eligibility preindex or a preindex insulin claim. Primary outcome measure was time to composite event of AMI, AHF or death among pioglitazone- and rosiglitazone-treated patients. The National Death Index database was accessed to obtain date of death for patients who died during the study period. Propensity score matching was used to control for potential confounders. The Cox proportional hazards model was used to evaluate effects of exposure to rosiglitazone and pioglitazone on time to event. A total of 36 628 patients (58% male; mean age, 54 years) were identified. Of the rosiglitazone-treated patients, 602 (4.16%) had an AMI, AHF, or death compared with 599 (4.14%) propensity score-matched pioglitazone-treated patients. No significant difference was observed between matched groups for risk of composite event (hazard ratio, 1.03; 95% confidence interval, 0.91 to 1.15; P=0.666) when patients were followed from index date until end of study period, termination of enrollment status, or diagnosis of AMI/AHF/death. CONCLUSIONS In this retrospective cohort study directly comparing rosiglitazone and pioglitazone with a propensity score-matched population that includes mortality data, no significant differences were found in the risk of AMI, AHF or death.