Abstract P230: Measures of Glycaemia (Fasting and 2 hr Glucose, Glycosylated Haemoglobin) in Relation to Incident Vascular Disease in Non-diabetic Adults

Introduction: Glycosylated haemoglobin (HbA 1c ) has recently been accepted for diagnosing diabetes in New Zealand. A 2 hour 75g oral glucose test was used previously and remains an alternative for diagnosis. Hypothesis: We assessed the hypothesis that HbA 1c would do at least as well as fasting and 2 hour glucose levels for predicting vascular events. Methods: Baseline data from a primary care cardiovascular disease (CVD) risk register of approximately 220,000 patients were prospectively linked to hospitalization, mortality and lab test data. There were 21,720 patients without a history of diabetes or CVD who had an HbA 1c and oral glucose tolerance test. We compared the prognostic value of HbA 1c and fasting and CVD or all-cause mortality using Cox proportional hazards regression. Results: The median follow-up time was 5 years (range 0 to 11). For HbA 1c values of less than 5.0%, 5.0 to less than 5.5%, 5.5 to less than 6.0%, 6.0 to less than 6.5%, and 6.5% or greater, the adjusted (age, gender and ethnicity) hazard ratios (with 95% confidence intervals) for diabetic hospitalisations were respectively: 1.45 (0.81-2.45), 1.00 (reference), 1.27 (1.01-1.63), 2.34 (1.88 -2.95), and 6.31 (5.10-7.94), respectively. Baseline HbA 1c was also related to risk of renal complications and retinopathy. For CVD, the hazard ratios (HR’s) were 0.95 (0.51-1.62), 1.00 (reference), 1.16 (0.95-1.42), 1.19 (0.98-1.46), and 1.32 (1.09-1.61), respectively. All associations remained significant after adjustment for the baseline fasting and 2 hour glucose levels. No HR’s were significant for all-cause mortality after adjusting for ethnicity. The associations between fasting glucose levels and the risk of CVD or death from any cause were not significant in models with adjustment for age, gender and ethnicity as well as HbA 1c . The association between the 2 hour glucose levels and the risk of diabetes was significant, but attenuated, in the model with adjustment for age, gender and ethnicity as well as HbA 1c . Conclusions: In conclusion, in people with no diabetes and no CVD, baseline HbA 1c was associated with subsequent diabetes, CVD, diabetic renal complications and retinopathy. HbA 1c was more strongly associated with these vascular complications than were fasting glucose and 2-hour glucose.