Details

Autor(en) / Beteiligte

• Karageorgiou, Vasilios
• Papaioannou, Theodoros G.
• Bellos, Ioannis
• Alexandraki, Krystallenia
• Tentolouris, Nikolaos
• Stefanadis, Christodoulos
• Chrousos, George P.
• Tousoulis, Dimitrios

Titel

Effectiveness of artificial pancreas in the non-adult population: A systematic review and network meta-analysis

Ist Teil von

• Metabolism, clinical and experimental, 2019-01, Vol.90, p.20-30

Ort / Verlag

United States: Elsevier Inc

Erscheinungsjahr

2019

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Artificial pancreas is a technology that minimizes user input by bridging continuous glucose monitoring and insulin pump treatment, and has proven safety in the adult population. The purpose of this systematic review and meta-analysis is to evaluate the efficacy of closed-loop (CL) systems in the glycemic control of non-adult type 1 diabetes patients in both a pairwise and network meta-analysis (NMA) context and investigate various parameters potentially affecting the outcome. Literature was systematically searched using the MEDLINE (1966–2018), Scopus (2004–2018), Cochrane Central Register of Controlled Trials (CENTRAL) (1999–2018), Clinicaltrials.gov (2008–2018) and Google Scholar (2004–2018) databases. Studies comparing the glycemic control in CL (either single- or dual-hormone) with continuous subcutaneous insulin infusion (CSII) in people with diabetes (PWD) aged <18 years old were deemed eligible. The primary outcome analysis was conducted with regard to time spent in the target glycemic range. All outcomes were evaluated in NMA in order to investigate potential between-algorithm differences. Pairwise meta-analysis and meta-regression were performed using the RevMan 5.3 and Open Meta-Analyst software. For NMA, the package pcnetmetain R 3.5.1 was used. The meta-analysis was based on 25 studies with a total of 504 PWD. The CL group was associated with significantly higher percentage of time spent in the target glycemic range (Mean (SD): 67.59% (SD: 8.07%) in the target range and OL PWD spending 55.77% (SD: 11.73%), MD: −11.97%, 95% CI [−18.40, −5.54%]) and with lower percentages of time in hyperglycemia (MD: 3.01%, 95% CI [1.68, 4.34%]) and hypoglycemia (MD: 0.67%, 95% CI [0.21, 1.13%]. Mean glucose was also decreased in the CL group (MD: 0.75 mmol/L, 95% CI [0.18–1.33]). The NMA arm of the study showed that the bihormonal modality was superior to other algorithms and standard treatment in lowering mean glucose and increasing time spent in the target range. The DiAs platform was superior to PID in controlling hypoglycemia and mean glucose. Time in target range and mean glucose were unaffected by the confounding factors tested. The findings of this meta-analysis suggest that artificial pancreas systems are superior to the standard sensor-augmented pump treatment of type 1 diabetes mellitus in non-adult PWD. Between-algorithm differences are also addressed, implying a superiority of the bihormonal treatment modality. Future large-scale studies are needed in the field to verify these outcomes and to determine the optimal algorithm to be used in the clinical setting. •In the non-adult population, AP increases time spent in target glycemic range•AP decreases time spent in hypoglycemia and hyperglycemia and the mean glucose•The most robust findings are the mean glucose reduction and the time in target range increase•In all analyses, the presence of an unannounced meal or exercise attenuated the AP superiority•In certain outcomes, network meta-analysis indirectly suggests a superiority of the bihormonal treatment modality