Automated insulin delivery systems help control type 1 diabetes better than traditional therapies, according to a review of 46 randomized clinical trials involving over 4,000 participants.
Automated insulin delivery (AID) systems were compared to standard insulin delivery methods from outpatient clinical trials lasting at least 3 weeks. The systems evaluated included hybrid closed-loop (HCL), advanced hybrid closed-loop (AHCL), and full closed-loop (FCL) technologies.
A key outcome was time-in-range—the percentage of time blood glucose remained between 70 and 180 mg/dL. Compared with insulin pump therapy, HCL systems increased time-in-range by 19.7%, AHCL by 24.1% , and FCL by 25.5%. The findings were supported by moderate to high certainty of evidence, depending on the system.
“Compared with pump therapy, the percentage of time in the range 70–180 mg/dl was greater with AID use,” the authors stated. [This fact was already introduced above. It's redundant, please delete]
AID systems also reduced time spent in hyperglycemia. Compared with insulin pumps, AHCL and FCL lowered time above 180 mg/dL by 19.6% and 23.3%, respectively. Reductions in time above 250 mg/dL were 14.8% for AHCL and 17.7% for FCL, both with moderate to high certainty, according to study investigators led by Anna Stahl-Pehe, of the Leibniz Center for Diabetes Research at Heinrich Heine University Düsseldorf, Germany.
Time spent in hypoglycemia—defined as below 70 mg/dL or 54 mg/dL—was slightly reduced with HCL and AHCL systems, though evidence was weaker. HCL reduced time below 70 mg/dL by 3.6%, and AHCL by 3.4%, both relative to insulin pumps. The certainty of this evidence was rated low. Reductions in time below 54 mg/dL were not statistically significant.
Changes in hemoglobin A1c (HbA1c) over 3 months were modest. A clinically meaningful reduction of 1% was observed only with AHCL compared to insulin pump therapy.
Treatments were grouped into seven categories: multiple daily injections, continuous subcutaneous insulin infusion, sensor-augmented pumps, predictive low glucose management systems, and the three types of AID systems. Most trials were small in size and short in duration; several outcomes showed signs of publication bias.
"The results provide information for clinical decisions on the actual achievable benefits of HCL, AHCL and FCL therapy compared with other insulin treatment modalities without neglecting heterogeneity and uncertainties in this context," noted Stahl-Pehe, and colleagues. "The scope and depth of our results may be particularly useful for advocacy and evidence-based recommendations."
The investigators used a standard grading system to assess the certainty of evidence, which ranged from very low to high. While AID systems consistently improved glycemic control, the authors noted that additional high-quality studies are needed, particularly for newer systems like FCL.
No conflicts of interest were reported.
Source: eClinicalMedicine
