Endocrine Abstracts

Introduction: Glycemic variability (GV) is emerging as a measure of glycemic control in type 1 diabetic patients (T1DM) given that it provides an integrated picture of postprandial hyperglycemia and hypoglycemic episodes and is associated with diabetic complications and mortality. However, little is known about the best way to address this problem in clinical practice.

Aim: To characterize GV in T1DM patients followed in our center and to identify clinical predictors for higher GV.

Methodology: Retrospective cohort study, including adults with T1D diagnosed clinically for ≥ 12 monthsthat were using flash glucose monitoring (FGM) systems ≥70% of the time. Raw data fromFGMwas downloaded in each visit from July 1st 2018 to December 31st 2019, and two periods of 28 consecutive days were selected, for each patient, and analyzed the following parameters: % of time in range (TIR [70–180 mg/dl]), % of time below range (TBR [< 70 mg/dl]), % of time below 55 mg/dl (TBR55),% of time above range (TAR [> 180 mg/dl]), % of time above 250 mg/dl (TAR250) and coefficient of variation (CV). A cutoff threshold value of 36% for CV was assumed to separate stable from labile glycemic control.

Results: Inclusion of seventy-five T1DM patients (62.7% female), with mean age 46.7 ± 13.5 years and mean duration of T1DM 25.4 ± 13.3 years. Fifty (66,7%) patients were using CSII. Carb counting was used by 59 (78.7%) patients. Adjuvant therapy was used in 22 (29.3%) patients (21,3% metformin, 5.3% GLP-1 agonists, 4.0% DPP4- inhibitors and 4.0% SGLT2 inhibitors). Mean A1c was 7.6 ± 0.9% and 18 (25.7%) patients had HbA1c ≤ 7%. Microvascular complications were present in 38 (48.7%) patients (diabetic retinopathy in 45.3%, nephropathy in 24.2% and peripheral neuropathy in 17.3%) andmacrovascular complications in one (1.3%) patient. Median TIR/TBR/TAR were 49 (38–70)%; 5.0 (3–7.5)% and 44.5 (30.5–57.5)%, respectively. Nine (12%) patients with TIR > 70% and 29 (38.7%) with TBR < 4%. Mean CV was 41.2 ± 7.3 reflecting labile control in 80% (n = 60) patients. Vascular complications were more frequent in this group (58.3% vs 20%, P = 0.01). Disease duration was associated with higher CV [OR: 1,28 (CI 95%: 1.03–1.60)] but neither age, sex, body mass index, carbs counting, CSII or adjuvant therapy showed to be predictors of GV in our sample.

Conclusions: GV was high in this sample and associated with longer disease duration, translating the difficulties that both patients and health care professionals face every day and reinforcing the need for strategies to reduce GV effectively.