Risk of diabetes in the new diagnostic category of impaired fasting glucose: a prospective analysis.

OBJECTIVE: To prospectively evaluate progression to diabetes in individuals with impaired glucose regulation as defined according to fasting glucose alone or an oral glucose tolerance test (OGTT) (i.e., both fasting and postload glucose) to compare the ability of these two screening methods to identify people at high risk of developing diabetes. RESEARCH DESIGN AND METHODS: A working population of 1,245 nondiabetic telephone company employees aged 40-59 years was studied by OGTT in 1980. Participants were classified according to baseline fasting glucose only (as encouraged by the American Diabetes Association [ADA]) or OGTT (as recommended by the 1998 World Health Organization [WHO] consultation). Progression to diabetes was evaluated 11.5 years later according to the 1997 ADA criteria of a fasting plasma glucose level > or =7.0 mmol/l. RESULTS: With the use of the OGTT, baseline prevalence of impaired glucose regulation was substantially higher than that with fasting glucose alone (7.2 vs. 3.2%); the two groups only overlap for 40.9% of the cases because a fairly large number of people with postload hyperglycemia (59.1%) have normal fasting glucose. Progression to diabetes in participants with normal fasting glucose and postload hyperglycemia is significantly more frequent than that of people with normoglycemia (32.5 vs. 7.2%; P < 0.001) and not significantly different from that of people with both fasting and postload hyperglycemia (i.e., 44.0%). However, the former are not identified as being at unusually high risk of diabetes unless an OGTT is performed. When the use of fasting glucose alone or OGTT was validated as a marker of progression to diabetes, sensitivity was substantially higher for the OGTT (33.3 vs. 9.0%) without major differences in specificity (92.6 vs. 97.0%). CONCLUSIONS: These data (the only data so far available in Caucasians) support the viewpoint that for the identification of people at high risk of diabetes, the use of the OGTT should be maintained.