[Diagnosing type 2 diabetes mellitus: in primary care, fasting plasma glucose and glycosylated haemoglobin do the job].

OBJECTIVE To determine the validity of glycosylated hemoglobin (HbA1c) values as a method to diagnose type 2 diabetes mellitus (DM2) in a population at risk seen in primary care. DESIGN Cross-sectional analytical study. SETTING Data were obtained for the Raval Sud study population (epidemiologic study of alterations in glucose metabolism in a population at risk). PARTICIPANTS 454 subjects from this population (mean age, 65 +/- 3 years; 52% male) at high risk for DM2, seen at a primary care center, were included in the study. MAIN MEASURES We recorded demographic data and laboratory values for fasting plasma glucose (FPG), oral glucose tolerance test (OGTT), and HbA1c. The diagnostic criteria used for DM2 were those published by the WHO in 1999. Values for HbA1c were expressed as the number of standard deviations (SD) above the mean. RESULTS Levels of HbA1c correlated with FPG (r=0.72) and glucose levels 2 h after oral glucose overload (r=0.43). Thirty percent of the patients with FPG between 110 and 125 mg/dL had HbA1c values higher than the reference limits. A combined technique based on FPG>125 mg/dL or FPG 110-125 mg/dL with HbA1c > or = 3 SD (5.94%) showed a sensitivity of 92% and a specificity of 95%. CONCLUSIONS When FPG is inconclusive (110-125 mg/dL), an HbA1c value more than 3 standard deviations above the mean (>5.94%) is useful in suggesting a likely diagnosis of diabetes and identifying patients who require treatment.