Glucose control and psychosocial outcomes with use of automated insulin delivery for 12 to 96 weeks in type 1 diabetes: a meta-analysis of randomised controlled trials

Diabetology & Metabolic Syndrome

Review

Amanda Godoi¹  Joao Roberto de Sa²  Larissa C. Hespanhol³  Isabele A. Miyawaki⁴  Caroliny H. Silva⁵  Cintia Gomes⁶  José Eduardo Riceto Loyola Júnior⁷  Eduardo M. H. Padrão⁸  Isabela A. F. de Souza⁹  Ariadne C. Silva^1,10  Isabela Reis Marques^1,11  Vittoria C. S. Moreira^1,12  Kavita Advani^1,13  Christi Oommen^1,13  Ashwin Mahesh^1,13 

[1] Cardiff University School of Medicine, Neuadd Meirionnydd, CF144YS, Cardiff, UK;Endocrinology Division, ABC School of Medicine and Federal University of Sao Paulo, Paulista School of Medicine, São Paulo, Brazil;Federal University of Campina Grande, Cajazeiras, Brazil;Federal University of Paraná, Curitiba, Brazil;Federal University of Rio Grande do Norte, Natal, Brazil;Federal University of Santa Maria, Santa Maria, Brazil;Hospital Alemão Oswaldo Cruz, São Paulo, Brazil;Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA;Petrópolis Medical School, Petrópolis, Brazil;UniEvangelica University Centre of Anapolis, Anapolis, Brazil;Universitat Internacional de Catalunya, Barcelona, Spain;University Israelita de Ciências da Saúde Albert Einstein, São Paulo, Brazil;University of Connecticut Health, Farmington, USA;

关键词: Closed-loop;    Automated insulin delivery;    HbA1c;    TIR;    Time in range;    Hypoglycaemia;    Glucose control;    Diabetes technology;    Type 1 diabetes;    T1DM;   

DOI  :  10.1186/s13098-023-01144-4

 received in 2023-05-26, accepted in 2023-07-31,  发布年份 2023

来源: Springer

【 摘 要 】

BackgroundGlycaemic control of Type 1 Diabetes Mellitus (T1DM) remains a challenge due to hypoglycaemic episodes and the burden of insulin self-management. Advancements have been made with the development of automated insulin delivery (AID) devices, yet, previous reviews have only assessed the use of AID over days or weeks, and potential benefits with longer time of AID use in this population remain unclear.Methods We performed a systematic review and meta-analysis of randomised controlled trials comparing AID (hybrid and fully closed-loop systems) to usual care (sensor augmented pumps, multiple daily insulin injections, continuous glucose monitoring and predictive low-glucose suspend) for adults and children with T1DM with a minimum duration of 3 months. We searched PubMed, Embase, Cochrane Central, and Clinicaltrials.gov for studies published up until April 4, 2023. Main outcomes included time in range 70–180 mg/dL as the primary outcome, and change in HbA1c (%, mmol/mol), glucose variability, and psychosocial impact (diabetes distress, treatment satisfaction and fear of hypoglycaemia) as secondary outcomes. Adverse events included diabetic ketoacidosis (DKA) and severe hypoglycaemia. Statistical analyses were conducted using mean differences and odds ratios. Sensitivity analyses were performed according to age, study duration and type of AID device. The protocol was registered in PROSPERO, CRD42022366710.ResultsWe identified 25 comparisons from 22 studies (six crossover and 16 parallel designs) including a total of 2376 participants (721 in adult studies, 621 in paediatric studies, and 1034 in combined studies) which were eligible for analysis. Use of AID devices ranged from 12 to 96 weeks. Patients using AID had 10.87% higher time in range [95% CI 9.38 to 12.37; p < 0.0001, I2 = 87%) and 0.37% (4.77 mmol/mol) lower HbA1c (95% CI − 0.49% (− 6.39 mmol/mol) to – 0.26 (− 3.14 mmol/mol); p < 0·0001, I2 = 77%]. AID systems decreased night hypoglycaemia, time in hypoglycaemia and hyperglycaemia and improved patient distress, with no increase in the risk of DKA or severe hypoglycaemia. No difference was found regarding treatment satisfaction or fear of hypoglycaemia. Among children, there was no difference in glucose variability or time spent in hypoglycaemia between the use of AID systems or usual care. In sensitivity analyses, results remained consistent with the overall analysis favouring AID.ConclusionThe use of AID systems over 12 weeks, regardless of technical or clinical differences, improved glycaemic outcomes and diabetes distress without increasing the risk of adverse events in adults and children with T1DM.

【 授权许可】

CC BY   
© BioMed Central Ltd., part of Springer Nature 2023

【 图 表 】

Fig. 2

Fig. 3

12888_2023_5168_Article_IEq2.gif

Fig. 6

Fig. 7

Fig. 3