Hemoglobin A(1c)—Using Epidemiology to Guide Medical Practice: Kelly West Award Lecture 2020

The discovery that HbA(1c) was a valid and reliable measure of average glucose exposure was one of the most important advances in diabetes care. HbA(1c) was rapidly adopted for monitoring glucose control and is now recommended for the diagnosis of diabetes. HbA(1c) has several advantages over glucose. Glucose assessment requires fasting, has poor preanalytic stability, and is not standardized; concentrations are acutely altered by a number of factors; and measurement can vary depending on sample type (e.g., plasma or whole blood) and source (e.g., capillary, venous, interstitial). HbA(1c) does not require fasting, reflects chronic exposure to glucose over the past 2–3 months, and has low within-person variability, and assays are well standardized. One reason HbA(1c) is widely accepted as a prognostic and diagnostic biomarker is that epidemiologic studies have demonstrated robust links between HbA(1c) and complications, with stronger associations than those observed for usual measures of glucose. Clinical trials have also demonstrated that lowering HbA(1c) slows or prevents the development of microvascular disease. As with all laboratory tests, there are some clinical situations in which HbA(1c) is unreliable (e.g., certain hemoglobin variants, alterations in red blood cell turnover). Recent studies demonstrate that fructosamine and glycated albumin may be substituted as measures of hyperglycemia in these settings. Other approaches to monitoring glucose have recently been introduced, including continuous glucose monitoring, although this technology relies on interstitial glucose and epidemiologic evidence supporting its routine use has not yet been established for most clinical settings. In summary, a large body of epidemiologic evidence has convincingly established HbA(1c) as a cornerstone of modern diabetes care.