Association between allopurinol and cardiovascular outcomes and all-cause mortality in diabetes: A retrospective, population-based cohort study

Aim To assess the association between allopurinol and mortality and cardiovascular outcomes in an allopurinol-treated diabetes cohort. Materials and Methods We conducted a population-based retrospective cohort study in Ontario, Canada. Eligible subjects were >= 66 years old with diabetes and a first prescription for allopurinol between 1 April, 2002 and 31 March, 2012 and were followed until 31 March, 2016. The primary outcome was a composite: all-cause mortality, non-fatal cardiovascular event (myocardial infarction, revascularization procedure, or stroke) or congestive heart failure (CHF). Secondary outcomes were components of the primary outcome and pneumonia as a negative tracer. Allopurinol was modelled as time-varying exposed versus unexposed, daily dose category and cumulative dose using sex-specific multivariable Cox proportional hazards models. Results Over a median follow-up of 4.65 years (interquartile range 1.79-7.81), 16 266/23 103 males and 10 571/15 313 females experienced the primary outcome. Allopurinol was associated with a reduction in the primary outcome [adjusted hazard ratios (aHR) 0.77 (95% confidence interval 0.75-0.80) and 0.81 (0.78-0.84) for males and females, respectively], driven by marked reductions in all-cause mortality and modest reductions in cardiovascular events/CHF. There was no effect of cumulative allopurinol dose on any outcome, and allopurinol was also associated with reduced risk of pneumonia in males [aHR 0.88 (0.83, 0.93)]. Conclusions Allopurinol was associated with reduced mortality and cardiovascular outcomes. However, lack of cumulative dose effect and a positive tracer outcome in males suggests residual bias. Future research assessing whether allopurinol prevents vascular complications in diabetes requires a clinical trial.