Glycemic variability(GV) is a better predictor of diabetes complications not detected by A1c.We evaluated extent of GV in South Indian T2D patients on insulin & OHA with continuous glucose monitoring (CGM) system (iPro® 2), for 3-5 days. CGM of 47 patients with A1c <11% were analyzed (82% male, age 53 ± 11 yrs, A1c 7.9 ± 1.4, duration of T2D 13 ± 8 yrs, BMI 26 ± 3.5). Standard Deviation (SD), Mean Absolute Difference (MAD), & descriptive measures of glycemic patterns in CGM reports were compared among A1c categories using t-test and Mann-Whitney U Test & among treatment groups using ANOVA.

Predictors of number of low excursions were identified using negative binomial regression models. Daily CGM values showed a mean highest value of 231 ± 60 mg/dl and mean lowest value of 84 ± 23 mg/dL. SD was 36.3 ± 15.9 and MAD% was 8.6 ± 3.8. Duration with glucose values above 140mg/dl was 45 ± 27% and duration with glucose values less than 70 mg/dl was 4 ± 7%. Multiple regression showed higher A1c statistically significantly associated with higher SD (b=4.1, t=2.2,p=0.02). Patients with A1c ≥7% had significantly higher duration of time more than 140 mg/dL but no significant difference in time below 70 mg/dL, when compared with patients with A1c <7%.

Patients had average 0.47±0.7 low excursions/day. Number of low excursions was statistically significantly associated with SD and T2D duration but not with A1c, gender, treatment groups, or BMI. This study contributes to normative data for GV & shows link of SD to higher A1c values, and suggesting possible utility of SD in predicting risk of hypoglycaemia.

Read More