Intensive glucose control and recurrent cardiovascular events: 14-year follow-up investigation of the ACCORDION study.

Aims

While cardiovascular disease in patients with type 2 diabetes commonly progresses with the occurrence of repeated events, most trials consider the effect of glucose-lowering strategies only on the first event. We examined the Action to Control Cardiovascular Risk in Diabetes trial and its observational follow-up study (ACCORDION) to investigate the effect of intensive glucose control on multiple events and further identify any subgroup effects.

Materials and Methods

A recurrent events analysis, using a negative binomial regression model, was applied to estimate the treatment effect on different consecutive cardiovascular disease events, including non-fatal myocardial infarction, non-fatal stroke, hospitalisation from heart failure, and cardiovascular death. Interaction terms were used to identify potential effect modifiers. The robustness of the results was confirmed in sensitivity analyses using alternative models.

Results

The median duration of follow-up was 7.7 years. Of the 5128 participants in the intensive and 5123 in the standard glucose control arm, respectively, 822 (16.0%) and 840 (16.4%) participants experienced a single event; 189 (3.7%) and 214 (4.2%) participants experienced two events; 52 (1.0%) and 40 (0.8%) experienced three events; and 1 (0.02%) and 1 (0.02%) experienced four events. There was no evidence of a treatment effect, with a rate difference of 0.0 (−0.3, 0.3) per 100 person-years comparing intensive versus standard intervention, although with non-significantly lower event rates in younger patients with HbA1c < 7% and higher event rates in older patients with HbA1c ≥ 9%.

Discussion

Intensive glucose control may not affect cardiovascular disease progression except in select subgroups. Since time-to-first event analysis may miss beneficial or harmful effects of glucose control on the risk of cardiovascular disease, recurrent events analysis should be routinely analysed in cardiovascular outcome trials, particularly when investigating long-term treatment effects.