Abstract

OBJECTIVE—Under the auspices of the National Institutes of Health, American Diabetes Association, and World Health Organization, expert committees lowered the fasting plasma glucose (FPG) concentration diagnostic for diabetes from 7.8 to 7.0 mmol/l and defined 6.1–6.9 mmol/l as impaired fasting glucose (IFG) and <6.1 mmol/l as normal fasting glucose. In 2003, IFG was lowered to 5.6–6.9 mmol/l and normal fasting glucose to <5.6 mmol/l. Reports of the relationship between glucose concentration and all-cause mortality have been inconsistent. It is not known if the 2-h plasma glucose (2hPG) concentration from an oral glucose tolerance test (OGTT) adds to the predictive power of FPG.

RESEARCH DESIGN AND METHODS—We followed 1,236 men for an average of 13.4 years to determine the relationship between both FPG and 2hPG and all-cause mortality.

RESULTS—Risk for mortality did not increase until the FPG exceeded 6.1 mmol/l. Risk increased by ∼40% in the 6.1–6.9 mmol/l range and doubled when FPG ranged from 7.0 to 7.7 mmol/l. A combination of the 2hPG and FPG allowed better estimation of risk than the FPG alone. Within any category of FPG, risk generally increased as the 2hPG increased, and within any category of 2hPG, risk generally increased as the FPG increased.

CONCLUSIONS—These data support the decision to lower the FPG diagnostic for diabetes from 7.8 to 7.0 mmol/l, show that both IFG and impaired glucose tolerance have risks between the normal and diabetic ranges, and show that the OGTT adds predictive power to that of FPG alone and should not be abandoned. The lowering of IFG to 5.6 mmol/l from 6.1 mmol/l, at least for mortality, is, however, not supported by our results.