Glycosylated hemoglobin in diabetes mellitus: correlations with fasting plasma glucose, serum lipids, and glycosuria.

The levels of glycosylated hemoglobin (GlHb), fasting plasma glucose (FPG), urine glucose, serum triglycerides, cholesterol, and HDL-cholesterol were monitored in several hundred patients attending an adult diabetic clinic. Thirty-five percent of the patients were being treated with insulin, 30% with diet alone, and 20% with sulfonylurea. Therapy was changed during the study in the remaining 15%. The levels of GlHb and FPG were significantly lower in the patients treated with diet alone than in the other groups. The overall coefficient of correlation (r) between FPG and GlHb was 0.64, with the highest r (0.70) for the diet group and the lowest r (0.55) for the insulin group. A follow-up study showed that in about 50% of the patients GlHb and FPG levels changed in different directions from one visit to the next. The absence of glycosuria was noted in two-thirds of the patients, while in 78% of the patients GlHb values were above the normal range. A weak but positive correlation was found between GlHb and serum triglycerides and cholesterol. There was no significant correlation between GlHb and HDL-cholesterol. The HDL-cholesterol levels were significantly lower in male than in female diabetic subjects, but no significant difference was found between the diabetic and nondiabetic subjects of the same sex. The results of this study suggest that (1) although there is a positive correlation between GlHb and FPG, the two tests cannot be used interchangeably in the evaluation of diabetic control; and (2) in the majority of the patients the absence of glycosuria tends to create the false assurance of satisfactory blood glucose control.