To define the target value of coefficient of variation (CV) in Chinese diabetes patients
A total of 3007 diabetes patients underwent continuous glucose monitoring (CGM). Type 2 diabetes were divided into groups according to the treatment: group 1, non-insulinotropic agent alone (n=138); group 2, oral therapy with insulinotropic agent (n=761); group 3, basal insulin therapy (n=100); group 4, premix insulin (n=784) and group 5, intensive insulin therapy (n=612). Another 612 type 1 diabetes were included as group 6. %CV and time per day within, below and above target glucose range were computed. TBR3.9 ≥4% and TBR3.0 ≥1% were defined as excessive hypoglycemia.
Type 2 diabetes with premixed or intensive insulin regimens had increased %CV than those on oral therapy or basal insulin. The upper limit of %CV in group 1 was 33%, which was adopted as the threshold to define excess glycemic variability. In each group, the percentage of patients with TBR3.9≥4% was significantly greater in subgroup with %CV>33% than ≤33% (P<0.001). In well-controlled patients who had TBR3.9<4%, TIR3.9-10>70% and TAR10<25%, the 95th percentile of %CV was 32.70% (close to 33%). The optimal cutoff values for %CV as identifier of TBR3.9≥4% by ROC analysis were different in groups with different treatment or with different A1C ranges (27% in A1C 6-7% group; 30.1% in A1C 9-10% group and 35.1% in A1C≥10% group)
The target value of %CV seems to be 33% in Chinese diabetes. However, patients with different A1C ranges or different therapy should be considered to use a personized %CV target.