Thiazolidinediones and the risk of incident strokes in patients with type 2 diabetes: a nested case-control study

Purpose To determine whether the use of thiazolidinediones (TZDs) decreases the risk of incident strokes in patients with type 2 diabetes. Methods We conducted a nested case-control study within a population-based cohort from the UK General Practice Research Database (GPRD). The cohort comprised patients over the age of 40 who were prescribed a first oral hypoglycemic agent between I January 1988 and 30 June 2008. Cases included all subjects who experienced a first stroke during follow-up. Up to 10 controls were matched to each case on age (+/- 2 years), sex, date of cohort entry (+/- 1 year), and duration of follow-up. Rate ratios (RRs) of stroke associated with TZD use, including rosiglitazone and pioglitazone, relative to combinations of other oral hypoglycemic agents, were estimated using conditional logistic regression. Results The cohort comprised 75 717 users of oral hypoglycemic agents, of whom 2417 had a stroke during follow-up. The rate of stroke in users of TZDs given as monotherapy (RR: 1.20, 95%CI: 0.77, 1.86) or in combination with other oral hypoglycemic agents (RR: 0.78, 95%CI: 0.58, 1.04) was not lower than combinations of other oral hypoglycemic agents. The RRs were similar for rosiglitazone and pioglitazone. Conclusions The results of this study indicate that TZDs do not appear to decrease the incidence of first strokes. Copyright (C) 2009 John Wiley & Sons, Ltd.