Impaired fasting glucose concentrations in nondiabetic patients with ischemic heart disease: A marker for a worse prognosis

Background The issue of whether glucose concentrations below the diabetic t hreshold may be predictive of increased cardiovascular risk has not yet bee n fully elucidated. The current study evaluates the prognosis of nondiabeti c patients with ischemic heart disease (IHD) and impaired fasting glucose ( IFG) over a 7.7-year Follow-vp period. Methods A total of 11,853 patients with documented coronary artery disease aged between 45 and 74 years were examined. Patients were divided into 3 gr oups on the basis of their fasting blood glucose levels at screening: nondi abetic individuals, patients with IFG, and undiagnosed diabetic patients. P atients who were on any type of pharmacologic antidiabetic treatment were e xcluded From the study. Mortality rates were assessed separately for each g roup. Results The population comprised 9773 nondiabetic patients (82.4%, glucose up to 109 mg/dL), 1258 patients with IFG levels (10.6%, glucose 110-125 mg/ dL), and 822 diabetic subjects (7%, glucose greater than or equal to 126 mg /dL). Patients were followed up from 6.2 to 9.0 years (mean follow-up perio d 7.7 +/- 1.5 years). Crude mortality was lower in the nondiabetic subjects than in the 2 other groups. All-cause mortality in the nondiabetic group w as 14.3% compared to 20.1% in patients with IFG and 24.3% in the undiagnose d (P<.001). Multivariate adjustment showed the lowest mortality in nondiabe tic subjects, who exhibited a survival rate of 0.86 at the end of the Follo w-up, wherease the lowest survival-0.75-was seen among undiagnosed diabetic patients (P=.0001). An intermediate value of 0.78 was documented for patie nts with IFG (P<.01). After multivariate analysis, with nondiabetic patient s as the reference group, IFG was identified as a consistent predictor of i ncreased all-cause and IHD mortality with hazard ratios of 1.39 (95% confid ence interval 1.21-1.59) and 1.29 (95% confidence interval 1.01-1.64), resp ectively. Conclusions The main finding of this study is the substantially increased m ortality rate among nondiabetic coronary patients with IFG, who had fasting glucose levels markedly lower than hitherto acknowledged as defining overt diabetes.