BIOLOGICAL VARIATION OF GLYCATED HEMOGLOBIN - IMPLICATIONS FOR DIABETES SCREENING AND MONITORING

OBJECTIVE - To assess the inherent potential of glycated hemoglobin as a screening test for type 2 diabetes by determining the biological va riation in nondiabetic subjects. RESEARCH DESIGN AND METHODS - HbA(1c) values were measured by high-performance liquid chromatography (HPLC) in 12 nondiabetic subjects (7 men and 5 women; median age, 40 years [ range, 21-55 years]) on 10 fortnightly occasions. The nondiabetic inde x of individuality (IOI) for HbA(1c) (i.e., the square root of the rat io of intra-to interindividual variance) was determined. Any lest with an IOI of 1.4 has the most potential in disease screening, while one of 0.6 will be of little value. RESULTS - The analytical variance cont ributed to 9% of the total test variance, intraindividual variance, 6% ; and interindividual variance, 85%. The IOI was, therefore, only 0.27 . Thus, nondiabetic HbA(1c) values vary markedly between subjects, whi le values in the same individual change little with time. As such, to lie outside the assay reference range, the HbA(1c) values of some nond iabetic subjects must exceed 12 SD from their usual mean value, while in others a change of only 2 SD would be sufficient. CONCLUSIONS - Thi s fundamental characteristic of HbA(1c) means that even if analytical methods improve, glycated hemoglobin measurements will always be of li mited value when screening for type 2 diabetes. If similar interindivi dual differences also exist in diabetic subjects, then patients with t he same glycemic control may vary by at least 1-2%, which has implicat ions in setting glycated hemoglobin targets.