Stability of hemoglobin AIc levels on repetitive determination in diabetic out-patients.

Hemoglobin (Hb) AIc levels were measured biweekly for 14 weeks in 49 diabetic out-patients and 20 nondiabetic subjects. As conventional indices of diabetes control, urine tests were performed four times per day, and plasma glucose concentrations were measured 2 h post breakfast. The diabetic group ranged in age from 15-73 yr and in duration of diabetes from 6 months to 45 yr. Eleven subjects were on diet therapy alone, three were on oral hypoglycemic agents, and 35 were on insulin therapy. The nondiabetic group ranged in age from 18-65 yr. On entry to the study, the mean (+/-SD) Hb AIc and plasma glucose levels of the diabetic group (8.60 +/- 2.11% and 142.5 +/- 99.9 mg/dl) were significantly higher than in the control group (4.68 +/- 0.60% and 102.4 +/- 21.7 mg/dl; P less than 0.001) and remained so throughout the study. The mean coefficient of variation for Hb AIc did not differ significantly between the control group and either the diet therapy or insulin therapy diabetic groups. Urine test values averaged over 2-week periods for each diabetic subject showed a high degree of stability. The mean Hb AIc levels for individual diabetic subjects correlated with the mean plasma glucose levels (r = 0.544; P less than 0.001), proportion of 2% urine tests (r = -0.798; P less than 0.001). These data provide further support for Hb AIc as a measure of diabetes control and, in addition, provide the first direct evidence that a single Hb AIc determination in a 3-month period is adequate for this purpose when the subjects are on a stable therapeutic regimen.