Exploring Use of Closed-Loop Insulin Delivery System for Hospital-at Home: A Meta-analysis of Randomized Controlled Trials on Closed-Loop Insulin Delivery System in Type 2 Diabetes

Abstract

This meta-analysis aims to evaluate the use of closed-loop insulin delivery system (CLS), compared to regular insulin therapy, in controlling glucose levels in adults with type 2 diabetes mellitus (T2DM). This is to explore the use of CLS for optimal glucose management in Hospital-at-Home (HaH). HaH is a programme that brings monitoring and treatment to patients houses, instead of requiring them to stay in the hospital. Patients under HaH are commonly treated for infections, with a large percentage with T2DM. Fluctuations of glucose during acute illness call for frequent monitoring and titration of insulin. CLS has proven efficacy in glucose management for Type 1 diabetes. However, there were no articles reporting the use of CLS in HaH. This meta-analysis aims to evaluate Randomized Controlled Trials (RCTs) done for CLS in T2DM, to analyse its feasibility in HaH. RCTs that investigated adults >18 years old, diagnosed with T2DM, comparing CLS and regular insulin therapy. The search was performed on CINAHL, Embase and PubMed on 11 January 2023. A search was limited to 2012 to 2022, and language was not limited in the search. Two researchers independently screened for selection, extracted data, and appraised included articles. Of 398 articles identified from the 3 databases, 5 RCTs were selected and included in the review with a total sample size of 302 and an average age of 67.2 years. Studies were done in both hospitals and home. The quality of studies was high. Meta-analysis results show a significantly longer proportion of time within target glucose range in the intervention group, measuring a mean difference of 21.15 (95%CI, 17.33, 24.96). CLS showed efficacy in managing blood glucose in T2DM, regardless of their current medical conditions, pre-existing comorbidities, meals, or activity levels. There is potential CLS to be used for the remote monitoring and active titration of insulin for patients under HaH. However, as included studies had small sample sizes, more high quality RCTs, on larger sample sizes are warranted to ascertain its efficacy.