Glycemic Efficacy and Safety by Using Insulin Degludec and Aspart Guided by a Clinical Decision Support System in Non-Critically Ill Inpatients with Type 2 Diabetes Mellitus

Background: Algorithm-based insulin dosing systems are increasingly used in hospitals and have shown the potential to efficiently and safely enable glycemic control. The goal of this study was to evaluate glycemic control using the ultralong-acting basal insulin degludec (IDeg) in combination with insulin aspart (IAsp) within an algorithm-driven electronic clinical decision support system (cDSS) in inpatients with type 2 diabetes (T2D). Methods: In this non-controlled single-arm pilot study, an electronic, algorithm-based cDSS was applied for the management of insulin treatment in an internal general ward. Thirty hospitalized patients with T2D (18 female, age 74.1 ± 10.9 years, HbA1c 72.4 ± 22.3 mmol/mol, BMI 28.6 ± 5.6 kg/m 2, diabetes duration 13.2 ± 11.6 years, creatinine 1.5 ± 1.2 mg/dL, length of hospital stay 9.1 ± 4.0 days) were included in the study. Capillary blood glucose (BG) was evaluated four times daily using a point-of-care device integrated into the hospital information system. In addition, all participants received a blinded continuous glucose monitoring (CGM; Abbott Freestyle Libre Pro) system. The primary endpoint was defined as the percentage of BG measurements within the target range of 3.9-7.8 mmol/L. Results: Overall, 722 BG values and 17,242 CGM data points were available. Of those, 52.2% and 55.0% were in the specified target area (3.9-7.8 mmol/L), respectively. Mean BG prior to study start was 11.9 ± 4.4 mmol/L and improved to 7.5 ± 1.9 mmol/L and 7.4 ± 1.4 mmol/L after 6 and 10 days of treatment. BG < 3.9, <3.0 and <2.2 mmol/L was 1.25%, 0.28% and 0%, respectively. Adherence to the total daily insulin dose suggested by the cDSS was 94.2%, and 99.5% of all basal and 85.3% of all bolus insulin suggestions were accepted by the nurses in charge. Basal-bolus therapy using the cDSS covered 85% of the participants' total hospital stay. Conclusions: Glycemic control using IDeg within an algorithm-driven cDSS could effectively and safely be achieved in the hospital and was highly accepted.