Impact of insulin initiation on glycaemic variability and glucose profiles in a primary healthcare Type 2 diabetes cohort: analysis of continuous glucose monitoring data from the INITIATION study

Aim: To use continuous glucose monitoring to examine the effects of insulin initiation with glargine, with or without glulisine, on glycaemic variability and glycaemia in a cohort of people with Type 2 diabetes receiving maximum oral hypoglycaemic agents in primary healthcare. Methods: We conducted a post hoc analysis of continuous glucose monitoring data from 89 participants at baseline and at 24 weeks after insulin commencement. Indicators of glycaemic variability (standard deviation, J-index and mean amplitude of glycaemic excursion) and glycaemia (HbA, mean glucose, area under the glucose-time curve) were assessed. Multi-level regression analysis was used to identify the predictors of change. Results: Complete glycaemic variability data were available for 78 participants. Of these participants, 41% were women, their mean (sd) age was 59.2 (10.4) years, the median (interquartile range) diabetes duration was 10.4 (6.5, 13.3) years and the median (interquartile range) baseline HbA was 82.5 (71.6, 96.7) mmol/mol [9.7 (8.7, 11.0)%]. At baseline, BMI correlated negatively with standard deviation (r = -0.30) and mean amplitude of glycaemic excursion (r = -0.26), but not with J-index; HbA correlated with J-index (r = 0.61) but not with mean amplitude of glycaemic excursion and standard deviation. After insulin initiation the mean (sd) glucose level decreased [from 12.0 (3.0) to 8.5 (1.6) mmol/l; P < 0.001], as did the median (interquartile range) J-index [from 66.9 (47.7, 95.1) to 36.9 (27.6, 49.8) mmol/l; P < 0.001]. Baseline HbA correlated with a greater J-index reduction (r = -0.45; P < 0.001). The mean amplitude of glycaemic excursion and standard deviation values were unchanged. The baseline temporal profile, showing elevated postprandial morning glucose levels, was unchanged after insulin initiation, despite an overall reduction in glycaemia. Conclusion: Insulin initiation reduced hyperglycaemia but did not alter glycaemic variability in adults with Type 2 diabetes receiving maximum oral hypoglycaemic agents. The most significant postprandial excursions were seen in the morning, which identifies prebreakfast as the most effective target for short-acting insulin therapy. What's new?: We report the largest study to date of glycaemic variability using continuous glucose monitoring in people with Type 2 diabetes of long duration, before and after commencement of insulin therapy. Data were obtained in the primary care setting, which is where the majority of patients with Type 2 diabetes are treated in countries such as the UK and Australia. Insulin initiation reduced hyperglycaemia but did not alter glycaemic variability. The largest postprandial excursions were related to breakfast, suggesting that this time point should be targeted when intensifying treatment with rapid-acting insulin or other agents.