Comparative effectiveness of medical treatment

Background: Non-alcoholic steatohepatitis (NASH) comprises a major healthcare problem affecting up to 30% of patients with obesity and the associated risk for cardiovascular and liver-related mortality. Several new drugs for NASH-treatment are currently investigated. No study thus far directly compared surgical and non-surgical therapies for NASH. This network meta-analysis compares for the first time the effectiveness of different therapies for NASH using a novel statistical approach. Methods: The study was conducted according to the PRISMA guidelines for network meta-analysis. PubMed, CENTRAL and Web of Science were searched without restriction of time or language using a validated search strategy. Studies investigating therapies for NASH in adults with liver biopsies at baseline and after at least 12 months were selected. Patients with liver cirrhosis were excluded. Risk of bias was assessed with ROB-2 and ROBINS-I-tools. A novel method for population-adjusted indirect comparison to include and compare single-arm trials was applied. Main outcomes were NASH-resolution and improvement of fibrosis. Results: Out of 7,913 studies, twelve randomized non-surgical studies and twelve non-randomized surgical trials were included. NASH-resolution after non-surgical intervention was 29% [95% confidence interval (CI): 23-40%] and 79% (95% CI: 72-88%) after surgery. The network meta-analysis showed that surgery had a higher chance of NASH-resolution than medication [odds ratio (OR) =2.68; 95% CI: 1.44-4.97] while drug treatment was superior to placebo (OR =2.24; 95% CI: 1.55-3.24). Surgery (OR =2.18; 95% CI: 1.34-3.56) and medication (OR =1.79; 95% CI: 1.39-2.31) were equally effective to treat fibrosis compared to placebo without difference between them. The results did not change when only new drugs specifically developed for the treatment of NASH were included. Conclusions: Metabolic surgery has a higher effectiveness for NASH-therapy than medical therapy while both were equally effective regarding improvement of fibrosis. Trials directly comparing surgery with medication must be urgently conducted. Patients with NASH should be informed about surgical treatment options.