Aliment Pharmacol Ther 2012; 35: 66–75


Background  Thiazolidinediones (TZDs) have been used in the treatment of non-alcoholic steatohepatitis (NASH). However, the magnitude of treatment response associated with TZDs in improving liver histology in NASH has not been quantified systematically.

Aim  To conduct a meta-analysis of randomised, placebo-controlled clinical trials (RPCTs) using pioglitazone and rosiglitazone in the treatment of NASH.

Methods  Pubmed/MEDLINE and Cochrane Central Register of Controlled Trials 2010 were searched until September 2010 and four RPCTs were identified. Peto odds ratios (ORs) and their respective 95% confidence intervals (CIs) were used to assess the efficacy of TZDs in improving liver histological parameters.

Results  Four good quality RPCTs derived from three continents were included. The meta-analysis showed that TZDs (= 169) were significantly better than placebo (= 165) in improving ballooning degeneration, lobular inflammation and steatosis with combined ORs of 2.11 (95% CI, 1.33–3.36), 2.58 (95% CI, 1.68–3.97) and 3.39 (95% CI, 2.19–5.25) respectively. The improvement in combined necroinflammation with TZD (= 58) vs. placebo (= 52) was also statistically significant (combined OR 6.52[95% CI, 3.03–14.06]), but improvement in fibrosis was not. When pioglitazone (= 137) was analysed alone, the improvement in fibrosis with pioglitazone (= 137) vs. placebo (= 134) (combined OR 1.68 [95% CI, 1.02–2.77]) was statistically significant. The total body fat slightly decreased in the control, while it markedly and highly significantly increased with TZD treatment.

Conclusions  Thiazolidinediones significantly improve ballooning degeneration, lobular inflammation, steatosis and combined necroinflammation in patients with NASH. Pioglitazone may improve fibrosis. Larger randomised, placebo-controlled clinical trials are needed to examine the efficacy of thiazolidinediones in improving NASH fibrosis.