Jacinta I. Reddigan, Chris I. Ardern, Michael C. Riddell and Jennifer L. Kuk
Aims: To examine all-cause and cardiovascular disease (CVD) mortality risk in individuals categorized as normal, impaired or type 2 diabetic using clinical cutoffs for fasting plasma glucose (FPG), 2-hour plasma glucose (2hPG) and glycated hemoglobin (HbA1c).
Methods: The sample included 5,424 adults with undiagnosed diabetes from the Third National Health and Nutrition Examination Survey with public-access mortality data linkage (follow-up=8.5 ±2.3 years; 685 deaths). The association between the glycemic measures and all-cause and CVD mortality were analyzed with the measures as continuous and categorical variables. FPG, 2hPG and HbA1c were categorized using the American Diabetes Association criteria for normal, impaired and type 2 diabetes.
Results: When analyzed as a continuous variable, 2hPG was most strongly associated with both all-cause and CVD mortality. However, after categorizing each measure using clinical cutoffs, impaired and type 2 diabetic levels of FPG and HbA1c, but only type 2 diabetic levels of 2hPG, were signifi cantly associated with all-cause mortality. For CVD mortality, impaired (HR=1.68 [1.16-2.44]) and type 2 diabetic (HR=1.88 [1.11-3.18)] levels of HbA1c were found to be a signifi cant predictor of mortality risk. However, only type 2 diabetic levels of 2hPG and not impaired levels were signifi cantly associated with CVD mortality. FPG was not a signifi cant predictor of CVD mortality.
Conclusions: Clinically relevant categories of HbA1c provide more prognostic information for all-cause and CVD mortality risk than 2hPG. Therefore, given the ease and lower cost of measurement, HbA1c should be considered a bene fi cial diagnostic and prognostic alternative screening tool in the clinical setting.