Abstract
Aims: To evaluate the efficacy and safety of closed‐loop automated insulin delivery (AID) systems with the conventional insulin delivery arm in adults with type 2 diabetes mellitus (T2DM). Methods: A systematic review and meta‐analysis of 10 randomised controlled trials ( n = 786; AID: 436, control: 350) was conducted per PRISMA guidelines. Data from studies were pooled using a random‐effects model to derive weighted mean differences (WMDs) and 95% confidence intervals (CIs). Results: AID significantly increased time in normoglycemia (WMD: +22.14%; 95% CI: 17.05, 27.23; 319 min/day; P < 0.00001) and reduced hyperglycemia (WMD: −21.04%; 95% CI: −6.23, −15.84; −303 min/day; P < 0.00001), mean glucose (WMD: −1.54 mmol/L; 95% CI: −2.37, −0.70; P = 0.0003), and HbA1c (WMD: −0.80%; 95% CI: −1.48, −0.12; P < 0.00001). Glucose SD improved significantly (WMD: –0.46; 95% CI: −0.74, −0.18; P = 0.002). No significant differences were found in hypoglycemia (WMD: −0.11%; 95% CI: −0.32, 0.09; −1.6 min/day; P = 0.28), total daily insulin (WMD: +4.80 IU; 95% CI: −1.87, 11.47; P = 0.16), or CV (WMD: −0.52%; 95% CI: −2.08, 1.04; P = 0.52). Total adverse events were higher with AID (RR: 1.61; 95% CI: 1.20, 2.17; P = 0.002), but serious adverse events (RR: 1.61; 95% CI: 0.83, 3.13; P = 0.16) and severe hypoglycemia (OR: 2.13; 95% CI: 0.22,20.83; P = 0.52) were not significantly different. Conclusions: Closed‐loop AID significantly improves glycemic control in T2DM without increasing serious adverse events.

This publication has 30 references indexed in Scilit: