Reproducibility of the diagnosis of diabetes over a 30-month follow-up: the Paris Prospective Study.

OBJECTIVE To describe the change in diabetic status over 30 months. RESEARCH DESIGN AND METHODS Cohort study of 5,400 Caucasian men from the Paris Prospective Study, aged 44-55 years, who were not known as having diabetes at baseline. Oral glucose tolerance tests were performed at baseline and after 30 months. RESULTS At baseline, diabetes was diagnosed in 2.9% of the men by fasting plasma glucose (FPG) > or =7.0 mmol/l and in 0.9% by isolated postchallenge hyperglycemia (IPH) (FPG <7.0 mmol/l and 2-h plasma glucose concentration > or =11.1 mmol/l), i.e., one in four of all men with newly diagnosed diabetes. Thirty months later, 42% of the men with diabetes diagnosed by FPG reverted to nondiabetic status, compared with 72% of those with diabetes diagnosed by IPH (P < 0.0001). For the men with diabetes diagnosed by FPG at baseline, diabetes had been diagnosed by a physician at 30 months in 11.5%, in contrast to only 3.9% of those with diabetes diagnosed by IPH (P < 0.05). For the 51 men with diabetes diagnosed by IPH at baseline, those who reverted to nondiabetic status had a lower frequency of family history of diabetes (P < 0.1), a higher mean corpuscular volume (P < 0.08), and a significantly higher total cholesterol concentration (P < 0.006) at baseline; in contrast, for the 156 men with diabetes diagnosed by FPG at baseline, the men who reverted to nondiabetic status and those who remained diabetic had similar characteristics. CONCLUSIONS In this epidemiological study, diabetes diagnosed by one FPG concentration was more stable than diabetes diagnosed by one IPH; in clinical practice, the diagnosis of diabetes requires confirmation of the hyperglycemia.