Diabetes, Glycemic Control, and New-Onset Heart Failure

OBJECTIVE

Diabetes is a predictor of both coronary artery disease (CAD) and heart failure. It is unknown to what extent the association between diabetes and heart failure is influenced by other risk factors for heart failure.

RESEARCH DESIGN AND METHODS

We evaluated the association of diabetes and A1C with incident heart failure in outpatients with stable CAD and no history of heart failure (average follow-up 4.1 years).

RESULTS

Of 839 participants, 200 had diabetes (23.8%). Compared with patients who did not have diabetes, those with diabetes had an increased risk of heart failure (hazard ratio [HR] 2.17 [95% CI 1.37–3.44]). Adjustment for risk factors for CAD (age, sex, race, smoking, physical inactivity, obesity, blood pressure, and LDL cholesterol), interim myocardial infarction, and myocardial ischemia did not alter the strength of the association between diabetes and heart failure. After inclusion also of other risk factors for heart failure (left ventricular ejection fraction, diastolic dysfunction, and C-reactive protein) and medication use, diabetes remained an independent predictor of heart failure (HR 3.34 [95% CI 1.65–6.76]; P = 0.001). Each 1% increase in A1C concentration was associated with a 36% increased HR of heart failure hospitalization (HR 1.36 [95% CI 1.17–1.58]).

CONCLUSIONS

In patients with stable CAD who are free from heart failure at baseline, diabetes and glycemic control are independent risk factors for new-onset heart failure. The mechanisms by which diabetes and hyperglycemia lead to heart failure deserve further study, as the association is independent of baseline functional assessment of ischemia, systolic and diastolic function, and interim myocardial infarction.

Heart failure is an enormous burden of disease, leading to substantial health care costs. Despite advances in treatment, the number of heart failure hospitalizations has increased steadily. The 2005 Heart Failure Guidelines of the American College of Cardiology/American Heart Association and European Society of Cardiology emphasized the importance of identification and treatment of risk factors. Among the patients classified in the highest risk group are patients with diabetes. Diabetes is associated with incident heart failure in the general population  and with adverse outcomes among patients with already existing heart failure. Diabetes also predicts heart failure in patients with acute coronary syndromes. Whether diabetes predicts heart failure in patients with stable coronary artery disease (CAD) has not been evaluated in detail.

The precise underlying mechanism by which diabetes portends heart failure is unclear. In fact, it remains to be elucidated whether in this context the diagnosis of diabetes per se is more important than just the presence of inadequate glycemic control. CAD is the number one risk factor for heart failure in the developed world. Because diabetes is strongly associated with CAD, it is plausible to attribute the risk of heart failure associated with diabetes to the effects of CAD. However, although it is known that hyperglycemia predicts heart failure among diabetic patients with CAD (7), it is not known whether this risk is independent of CAD severity, CAD progression, or the presence of myocardial ischemia. Even in the absence of CAD, patients with diabetes show changes in myocardial performance that put them at risk for heart failure (diabetic cardiomyopathy).

To determine to what extent the association between diabetes and heart failure is influenced by other risk factors for heart failure (including interim myocardial infarction and the presence of baseline myocardial ischemia), we evaluated the risk of heart failure associated with diabetes in a cohort of outpatients with stable CAD. The cohort is derived from the Heart and Soul Study, which allows thorough investigation of the strength of the association between diabetes (both the diagnosis per se and the level of glycemic control) and future heart failure episodes, while taking into account the above-mentioned established and presumed risk factors.