In a Perspective article in this issue of Diabetes Care, Dr. Selvin (1) suggests that the glucose management indicator (GMI) is not a useful measure for glycemic management, since it often differs from laboratory-measured HbA1c. As noted by Dr. Selvin, “Nonglycemic factors can influence HbA1c, including red blood cell turnover, other red blood cell characteristics, and genetic variation in hemoglobin. These nonglycemic factors can affect the association of HbA1c with true average glucose exposure.” We fully agree with this statement. It explains the discordance between laboratory-measured HbA1c and a continuous glucose monitoring (CGM) estimate of HbA1c from mean glucose, which we termed the GMI (2). Just as Dr. Selvin makes the case that GMI does not “perform well as a substitute for HbA1c,” one could use the same data to make the case that HbA1c does not perform well as a substitute for the true glucose exposure as measured by CGM.
We proposed the term GMI (2) in response to a request from the U.S. Food and Drug Administration to use a term other than estimated HbA1c, as patients found the discordance between estimated HbA1c and measured HbA1c to be confusing. Thus, we proposed a distinctly different term (GMI) to refer to what the HbA1c level would be expected to be based solely on mean glucose.
We agree with Dr. Selvin that just referencing mean glucose potentially would be better than transforming mean glucose values into GMI. However, at this time, most patients and many clinicians do not have a perspective on how to interpret mean glucose and how much of a change in mean glucose is clinically meaningful, whereas with HbA1c, which has been used in patient management for decades, this is better understood. However, after decades of using HbA1c as our guide to clinical management, we are far from achieving our desired glycemic outcomes (3). While we believe that the GMI fills a need currently for the interpretation of CGM data, we will welcome the day when CGM-measured mean glucose is sufficiently well understood that there is no need to convert it to GMI for interpretation of the true glucose exposure, which Brownlee (4) outlined as the driving force of diabetes microvascular complications.
There is now a U.S. Food and Drug Administration quality designation to help ensure that the current generation of CGM sensors are sufficiently accurate and reliable to provide acceptable data for clinical decision-making (5). Use of CGM supports precision diabetes management by utilizing metrics such as time in target range, time below target range, mean glucose (or GMI for now), and coefficient of variation, along with a visualized glucose profile to personalize medication and lifestyle adjustments. We see the GMI as the bridge to move from the use of primarily HbA1c for diabetes management to the use of CGM metrics and profiles, along with the HbA1c when deemed helpful, for the personalized management of diabetes. Thus, it is not time to change course on the use of GMI but to use GMI to help cross over into more precise diabetes management.
Article Information
Duality of Interest. R.M.B.’s institution has received funding on his behalf for research study support, consulting, or serving on a scientific advisory board for Abbott Diabetes Care, Ascensia, Bigfoot Biomedical, Inc., CeQur, Dexcom, Embecta, Eli Lilly, Hygieia, Insulet, Medtronic, Novo Nordisk, Onduo, Roche Diabetes Care, Sanofi, United Healthcare, Vertex Pharmaceuticals, and Zealand Pharma. R.M.B.’s employer, nonprofit HealthPartners Institute, contracts for his services, for which he receives no personal income. R.W.B. reports that his institution has received funding on his behalf for grant funding, study supplies, and consulting fees from Insulet, Tandem Diabetes Care, and Beta Bionics, grant funding and study supplies from Dexcom, grant funding from Bigfoot Biomedical, study supplies from Medtronic, Ascencia, and Roche, consulting fees and study supplies from Eli Lilly and Novo Nordisk, and consulting fees from Embecta, Vertex Pharmaceuticals, Hagar, Ypsomed, Sanofi, and Zucara. No other potential conflicts of interest relevant to this article were reported.
Handling Editors. The journal editor responsible for overseeing the review of the manuscript was Matthew C. Riddle.
