Type of preadmission glucose-lowering treatment and prognosis among patients hospitalised with myocardial infarction: a nationwide follow-up study

Aims/hypothesis We examined whether the type of preadmission glucose-lowering treatments explained differences in mortality rate and risk of readmission with myocardial infarction (MI) and heart failure following first-time hospitalisation for MI in patients with type 2 diabetes mellitus. Methods We conducted a nationwide population-based follow-up study among all Danish patients hospitalised with first-time MI from 1996 to 2004. Data on use of glucose-lowering drugs and other medications, comorbidities, socioeconomic status, laboratory findings, readmission with MI and heart failure, and death were obtained from medical databases. We computed mortality rates and rates of MI and heart failure readmission, according to type of glucose-lowering treatment and used Cox's proportional hazards regression analysis to compute hazard ratios (HRs) as estimates of relative risks. Results We identified 8,494 MI patients with type 2 diabetes mellitus. The overall cumulative 30 day and 1 year mortality rates were 22.2 and 36.6%, respectively. Patients not receiving any glucose-lowering drugs (adjusted 30 day HR: 0.79, 95% CI: 0.57-1.10) and users of any combination (adjusted 30 day HR: 1.43, 95% CI: 0.98-2.09) had the lowest and highest mortality rates, respectively, when compared with users of sulfonylureas. We found that glycaemic control had no impact on the risk estimates in a subanalysis including biochemical laboratory data. We found no differences in the risk of new MI and heart failure between the different glucose-lowering agents. Conclusions/interpretations Type of preadmission glucose-lowering treatment in monotherapy is not associated with substantial differences in prognosis following hospitalisation with MI. However, patients treated with any combination had increased mortality rates.