Defining the target value of the coefficient of variation by continuous glucose monitoring in Chinese people with diabetes

AIMS/INTRODUCTION: To define the target value for the percentage coefficient of variation for glucose (%CV) as a measure of glycemic variability (GV) in Chinese diabetes patients. MATERIALS AND METHODS: This retrospective study included 3,007 diabetes patients who underwent continuous glucose monitoring for 3 days. Type 2 diabetes was divided into groups according to the received treatment: group 1, non‐insulinotropic agent (n = 138); group 2, insulinotropic agent (n = 761); group 3, basal insulin therapy (n = 100); group 4, premixed insulin (n = 784); and group 5, intensive insulin therapy (n = 612). Type 1 diabetes patients were included as group 6 (n = 612). %CV and percentage of time per day within, below (3.9mmol/L; TBR(3.9)) and above (10.0 mmol/L) the target glucose range (3.9–10.0 mmol/L) were computed. TBR(3.9) ≥4% was defined as excessive hypoglycemia. RESULTS: Type 2 diabetes with a premixed or intensive insulin regimen had an increased %CV compared with those receiving oral therapy or basal insulin. The upper limit of %CV in group 1 was 33%, which was adopted as the threshold to define excessive GV. For each treatment group, the percentage of people with TBR(3.9) ≥4% was significantly greater in the subgroup with %CV >33% than ≤33% (P < 0.001). In participants who achieved TBR(3.9) <4%, the time per day spent within the target glucose range of 3.9–10.0 mmol/L > 70% and time per day above 10.0 mmol/L <25%, the 95th percentile of %CV was 32.70%. Further receiver operating characteristic curve analysis showed that the cut‐off values of %CV for predicting TBR(3.9) ≥4% varied by the type of diabetes and glycated hemoglobin categories. CONCLUSIONS: A %CV of 33% was set as the threshold for excess glucose variability in Chinese diabetes patients. Meanwhile, glycated hemoglobin and the type of diabetes should be considered for the goal‐setting of %CV.