Aims Spain has been one of the worst affected countries by the COVID-19 pandemic. A very strict lockdown at home was imposed with a tough restriction of mobility. We aimed to evaluate the impact of this exceptional scenario on glucose profile of patients with T1D prone to hypoglycemia using standalone continuous glucose monitoring. Methods Patients with T1D prone to hypoglycemia using multiple daily injections and either a Dexcom G5® or a Free Style Libre® CGM systems for at least 6 months under the funding of National Health Service were included in an observational, retrospective study. Data were collected in two periods: pre-lockdown (PL), February 23rd-March 7th and within lockdown (WL), April 1st–14th 2020. The primary outcome was the difference in the proportion of time in target glucose range of 70–180 mg/dL (TIR). Additional glucometric data were also analysed. Results 92 patients were included: 40 women, age 42.8 ± 3.9 years, disease duration of 23.1 ± 12.6 years. Seventeen patients used Dexcom G5® and 75 Free Style Libre®. TIR 70–180 mg/dL (59.3 ± 16.2 vs 62.6 ± 15.2%), time > 180 (34.4 ± 18.0 vs 30.7 ± 16.9%), >250 (11.1 ± 10.6 vs 9.2 ± 9.7%) and Glucose Management Indicator (7.2 ± 0.8 vs 7.0 ± 0.8%) significantly improved (PL vs WL, respectively, p < 0.05). Time in hypoglycemia remained unchanged. Conclusions Lockdown conditions imposed by the COVID-19 pandemic may be managed successfully in terms of glycemic control by population with T1D prone to hypoglycemia using CGM. The strict daily routine at home could probably explain the improvement in the time in glycemic target without increasing the time in hypoglycemia.

(1) Background: nocturnal hypoglycemia (NH) is one of the most challenging side effects of multiple doses of insulin (MDI) therapy in type 1 diabetes (T1D). This work aimed to investigate the feasibility of a machine-learning-based prediction model to anticipate NH in T1D patients on MDI. (2) Methods: ten T1D adults were studied during 12 weeks. Information regarding T1D management, continuous glucose monitoring (CGM), and from a physical activity tracker were obtained under free-living conditions at home. Supervised machine-learning algorithms were applied to the data, and prediction models were created to forecast the occurrence of NH. Individualized prediction models were generated using multilayer perceptron (MLP) and a support vector machine (SVM). (3) Results: population outcomes indicated that more than 70% of the NH may be avoided with the proposed methodology. The predictions performed by the SVM achieved the best population outcomes, with a sensitivity and specificity of 78.75% and 82.15%, respectively. (4) Conclusions: our study supports the feasibility of using ML techniques to address the prediction of nocturnal hypoglycemia in the daily life of patients with T1D on MDI, using CGM and a physical activity tracker.