Relationship between haemoglobin A1C values and recurrent cardiac events: A retrospective, longitudinal cohort study

OBJECTIVE: This study set out to analyse the impact of baseline glycosylated haemoglobin A1c (HbA1c) values on the incidence of recurrent cardiac events in patients prescribed optimal secondary prevention medications and receiving aggressive cardiac risk factor management. METHODS: This was a retrospective study conducted at Kaiser Permanente Colorado and included adults followed by a clinical pharmacy specialist-managed cardiac risk service (CPCRS) with an incident cardiac event and an HbA1c value measured within 1 year prior or 60 days after the incident cardiac event was identified. Cox proportional hazards models were constructed to assess the relationship between HbA1c levels and recurrent cardiac events (assessed as continuous and categorical measures) after adjustment for potential confounding variables. RESULTS: Of 5663 patients identified within an incident cardiac event between January 1999 and March 2005, 1270 (22.4%) patients had a baseline HbA1c value recorded. Of these 1270 patients, 215 (16.9%) had a recurrent cardiac event. Compared with the 'no recurrent event' cohort, the 'recurrent event' cohort were younger, less likely to have undergone an initial coronary artery bypass graft, and more likely to have undergone percutaneous coronary intervention with or without stent. The recurrent event cohort was also less likely to have purchased an HMG-CoA reductase inhibitor ('statin') [p = 0.043] at the time of the incident cardiac event. There was no significant difference in mean baseline HbA1c value between the cohorts. There were also no significant differences between the cohorts when categorized by baseline HbA1c <7% as referent compared with > or =7% to <8%, > or =8% to <9%, > or =9 to <10%, and > or =10%. Moreover, there was no significant difference between cohorts when HbA1c values <7% were compared with values >7% in the unadjusted analysis. Results remained non-significant after adjustment for sex, incident cardiac event type, baseline age, ss-blocker use, statin use and hyperlipidaemia. CONCLUSION: The results of this study suggest that an abnormal HbA1c is not predictive of recurrent cardiac events among patients with cardiovascular disease when other cardiovascular risk factors are being aggressively treated and appropriate secondary prevention medications are being taken. However, larger studies are warranted to validate these findings.