Controlled versus uncontrolled diabetes and the risk of myocardial infarction

Abstract

OBJECTIVES: To investigate the risk of myocardial infarction (MI) attributed to diabetes.

METHODS: Questionnaire data and samples from 429 cases with MI and 434 controls from the Maltese Acute Myocardial Infarction (MAMI) Study were analysed. HbA1c estimation was performed on EDTA plasma using ion-exchange HPLC (Bio-Rad Variant II Analyzer). For cases this was assayed at time of MI. Uncontrolled diabetes was deemed present in self-reported diabetics who had HbA1c levels higher than 6.5% and controlled diabetes in self-reported diabetics who had normal HbA1c levels (≤6.5%). Undiagnosed diabetics were subjects with elevated HbA1c levels who did not report being diabetic. Odds ratios (AdjOR) were adjusted for age, gender, smoking, drinking alcohol, hypertension, hypercholesterolaemia and BMI.

RESULTS: Diabetes was reported in 13.7% of controls out of which 9.2% were uncontrolled. An additional 2.5% of controls have previously undiagnosed diabetes. The risk of MI was higher in undiagnosed [AdjOR 3.5(95%CI 1.6-7.8)] and in uncontrolled diabetics [AdjOR 2.7(95%CI 1.6-4.5)] when compared to non-diabetics. Controlled diabetics showed no increased risk of MI [AdjOR 0.9(95%CI 0.4-2.1)]. An elevated HbA1c level was associated with a 3-fold increased risk of MI [AdjOR 3.0(95%CI 1.9-4.7)]. Self-reported diabetes was associated with a 2-fold risk [AdjOR 2.1(95%CI 1.4-3.2)]. The median duration of diabetes in women with MI was longer than in controls (15 years versus 8 years, p<0.05). The risk associated with elevated HbA1c appeared to be higher in women [AdjOR 6.8(95%CI 2.4-19.1)] than men [AdjOR 2.5(95%CI 1.5-4.1)]. This risk difference persisted even after adjustment for duration of diabetes.

CONCLUSION: An elevated HbA1c level is a better indicator of the risk of MI than self-reported diabetes with the risk being particularly high in women. Undiagnosed and uncontrolled diabetes are strong risk factors for MI. Efforts at controlling diabetes are warranted since controlled diabetics showed no increased risk of MI.