Comparative Effectiveness Of Long-Term Lifestyle And Pharmacological Interventions For Primary Prevention Of Type 2 Diabetes: A Network Meta-Analysis Of Randomized Controlled Trials

Background: Lifestyle modification and some pharmacological interventions have been proven effective in preventing type 2 diabetes (T2D) in previous primary prevention trials. However, there is controversy regarding their comparative effectiveness because of the lack of head-to-head trials for all possible comparisons among different interventions. Network meta-analysis enables us to simultaneously evaluate the relative effectiveness among all interventions by synthesizing both direct evidence from head-to-head trials and indirect evidence through more than one common comparator while maintaining randomization. Objective: We performed a network meta-analysis of randomized controlled trials (RCTs) to directly and indirectly compare lifestyle modification and five commonly prescribed medications (insulin sensitizers, insulin secretagogues, alpha-glucosidase inhibitors, weight-loss drugs, and renin-angiotensin system blockade), and to rank their relative effectiveness for primary prevention of T2D among adults with prediabetes at enrollment. Methods: We systematically searched PubMed, Embase, and CENTRAL from inception to October 2018 and identified long-term RCTs (follow-up ≥ 1year) that evaluated the interventions for T2D prevention in adults with prediabetes (e.g. impaired fasting glucose and impaired glucose tolerance). A network meta-analysis using the frequentist approaches was conducted to calculate the summary odds ratio (OR) and 95% confidence interval (CI) for each active intervention in comparison with controls (i.e. usual care alone or with placebo or control pills) and rank the effectiveness of these interventions by the surface under the cumulative ranking curve probabilities. Results: Thirty-three RCTs involving 54,910 individuals with prediabetes and 13,189 diabetic cases were included. Network meta-analysis showed that lifestyle modification (OR, 0.53; 95% CI, 0.42 to 0.66), insulin sensitizers (0.44; 0.31 to 0.62), alpha-glucosidase inhibitors (0.60; 0.42 to 0.87), and weight-loss drugs (0.39; 0.23 to 0.66) were significantly associated with decreased risk of incident T2D as compared with the controls. Neither insulin secretagogues (0.80; 0.51 to 1.26) nor renin-angiotensin system blockade (0.93; 0.62 to 1.38) were significantly associated with lower risk of developing T2D. Ranking of these interventions revealed weight-loss drugs being the optimal intervention for preventing T2D (89.5%), followed by insulin sensitizers (84.1%), lifestyle modification (66.7%), and alpha-glucosidase inhibitors (54.4%). Conclusions: Among the four interventions (lifestyle modification, insulin sensitizers, alpha-glucosidase inhibitors, and weight-loss drugs) which showed significant benefits on the primary prevention of T2D, weight-loss drugs seem to be the most effective.