Pioglitazone and bariatric surgery are the most effective treatments for non-alcoholic steato-hepatitis: a hierarchical network meta-analysis

Aims: To compare different treatments for Non-Alcoholic Steato-Hepatitis (NASH) and to determine an effectiveness hierarchy. Materials-and-Methods: This is a systematic review and Bayesian Network Meta-Analysis including randomized-controlled trials or prospective trials with at least 6 months follow-up and histologically-proven NASH in adult participants. Monte Carlo simulations were performed, each generating 10,000 data points, and results are reported as medians and 95% Credibility Intervals (CrIs). A meta-regression was conducted to find the effects of BMI decrement or reduction of HOMA-IR on NAFLD Activity Score (NAS) change. Results: The review identified 48 eligible trials comprising 2356 adults (55.6% men). Data were pooled using a random-effects model. The most effective treatments in terms of NAS reduction per semester were Pioglitazone and Roux-en-Y Gastric-ByPass (RYGB) (-1.50; 95% CrI: -2.08, -1.00 for Pioglitazione and -1.00; 95% CrI: -1.70, -0.32 for RYGB). Pioglitazone was also the best therapy for steatosis and lobular inflammation reduction; RYGB was the best treatment for hepatocellular ballooning reduction, whereas antioxidants appeared to be best for fibrosis improvement. For each 1% decrement in BMI, NAS was reduced by 1.3% (?=1.28%, P=0.01). Conversely, 1% reduction of HOMA-IR index reduced NAS by 0.3% (?=0.31%,P<0.001). Treatments that were regarded as promising, such as Elafibranor, Simtuzumab, Selonsertib, Cenicriviroc, Obeticholic acid and Liraglutide did not reduce either NAS or liver fibrosis significantly. Conclusions: Pioglitazione and RYGB are the most effective therapies for NASH. Antioxidants may be effective in reducing liver fibrosis. Weight loss and improvement of hepatic insulin resistance are promising approaches in the treatment of NASH.