Metabolic surgery for the treatment of type 2 diabetes: A network meta-analysis of randomized controlled trials

Aim: To compare different types of metabolic surgery (MS) with medical therapy (MT) for the treatment of type 2 diabetes (T2D). Materials and methods: We conducted a network-meta-analysis (NMA) including randomized clinical trials comparing different MS techniques versus MT in people with T2D, with a duration of ≥24 weeks. Primary endpoints were glycated haemoglobin (HbA1c), fasting plasma glucose (FPG) and diabetes remission. Indirect comparisons of different types of surgery were performed by NMA. Mean and 95% confidence intervals for continuous variables, and Mantel−Haenzel odds ratios for categorial variables, were calculated using random effect models. Types of MS included: laparoscopic adjustable gastric banding (LAGB), Roux-en-Y gastric bypass, sleeve gastrectomy (SG), bilio-pancreatic diversion (BPD); greater curvature plication (GCP); one-anastomosis gastric bypass (OAGB); and duodenojejunal bypass. Results: The 24 retrieved trials included 1351 patients (1014 with MS and 337 with MT). The mean baseline BMI was 36.8 kg/m2. MS was associated with significantly greater reductions in HbA1c and FPG and greater diabetes remission when compared to MT. In the NMA, a significant reduction in HbA1c was observed with OAGB and SG. All surgical procedures were associated with a significant increase in diabetes remission, except GCP and LAGB. All procedures were associated with a reduction of body mass index (BMI). Conclusions: Metabolic surgery is an interesting option for the treatment of T2D, although further data are needed to demonstrate its long-term efficacy and safety. Present data are not sufficient to modify current recommendations, which consider MS a possible treatment for T2D in those with a BMI >35 kg/m2.