BACKGROUND: Pioglitazone, a drug for the treatment of type 2 diabetes mellitus has been associated with bladder cancer in observational studies. Diabetes mellitus itself has also been linked with bladder cancer. The objective was to estimate the risk of bladder cancer for diabetic patients using thialozidinediones (TZDs) compared with patients in other treatment stages of the disease.

METHODS: We performed a population-based cohort study (1996-2007) in the Danish National Health Registers. Oral antidiabetic drug users (n=179,056) were matched 1:3 by sex and year of birth to non-users. Hazard ratios (HRs) of bladder cancer were estimated using Cox proportional hazards models. Time-dependent adjustments were made for age, comorbidity, and drug use. Four different treatment stages were defined: current use of either a biguanide or a sulfonylureum (stage 1), current use of a biguanide and a sulfonylureum at the same time (stage 2), current use of TZDs (stage 3) and current use of insulin (stage 4).

RESULTS: Compared with non-diabetic controls, patients using antidiabetic medication experienced a 1.3-fold increased risk of bladder cancer (adjusted HR 1.3 [95%CI 1.2-1.4]). No major differences were observed between the different treatment stages. The risk of bladder cancer varied between 1.2 [95%CI 1.0-1.4] in stage 4 and 1.4 [95%CI 1.3-1.6] in stage 1. The risk of bladder cancer with TZD use (stage 3) was similar to the other groups (adjusted HR 1.3 [95%CI 0.6-2.7]).

INTERPRETATION: The association between TZD use and bladder cancer is probably confounded by the underlying disease.