Endocrine Abstracts

Introduction: Continuous subcutaneous insulin infusion (CSII) therapy in type 1 diabetes (T1D) reduces the risk of hypoglycemia. Hypoglycemia remains a treatment-limiting factor. Impared hypoglycemia awareness (IHA) occurs in 25% of T1D cases and seems to be underestimated by continuous glucose monitoring (CGM). Glycemic variability (GV) is a increasingly valued parameter as a predictor of hypoglycemia and risk of chronic complications.

Objective: Analysis of the relationship between impared hypoglycemia awareness (IHA) and new metrics of glycemic control [estimated A1C (A1Ce), time in hyper- and hypoglycemia, GV] in T1D patients on CSII and CGM.

Methods: Cross-sectional, observational study of patients with T1D under CSII and intermittent-scanning GCM FreeStyle Libre® (active>70% of the time). Glycemic control assessed by the 30-day GCM (AGP report). IHA was defined by a score ≥4 on the Clarke Questionnaire (CQ).

Results: 43 cases were analyzed: 61.5% were female; mean age of 33.3±12.7 years. Mean onset of CSII at 26.9±12.2 years. The mean duration of T1D was 20.5±9.3 years. The mean A1Ce was 7.1±0.7% and glycemic variability was 39.3±8.0%. The time spent in hyperglycemia was 35.2±18.7% (8.5 hours/day): 24.0±11.5% between 180–250 mg/dl and 11.2±9.4% above 250 mg/dl. Time spent in hypoglycemia was 7.0±5.9% (1.7 hours/day): 4.6±3.3% between 54–70 mg/dl (level 1) and 2.4±3.2% <54 mg/dl (level 2). From the Clarke Questionnaire, we obtained a prevalence of 14.3% of IHA and 9.3% of ≥1 level 3 hypoglycemia per year. There was a moderate to strong correlation between glycemic variability and time in hypoglycemia (r=0.72; P<0.001), as opposed to time in hyperglycemia which showed no significant correlation (P=0.41). GV did not show correlation with T1D duration but approached statistical significance (P=0,06). In patients with A1Ce <7% medians of time in hypoglycemia were significantly higher – 7.0(4)% vs 3.0(3)%. Impared hypoglycemia awareness occurred in patients with lower A1Ce values (6.9±0.7% vs 7.2±0.7%) but without significant difference. There was no significant difference in time in hypoglycemia at 30 days in these cases.

Conclusion: Hypoglycemia occurred in a higher frequency than the goals especially for more demanding glycemic controls, correlating with the increase in glycemic variability (GV). GV was higher than desirable in most cases, underlining the difficulty of its optimization. Impared hypoglycemia awareness in this population had a lower prevalence than the overall estimated prevalence in T1D and did not show a significant correlation with the GCM data, highlighting the complexity of its pathophysiology and its possible undervaluation by glycemic control metrics.