Induction of clinical response and remission of inflammatory bowel disease by use of herbal medicines: A meta-analysis

Abstract

To evaluate the efficacy and tolerability of herbal medicines in inflammatory bowel disease (IBD) by conducting a meta-analysis. Electronic databases were searched for studies investigating efficacy and/or tolerability of herbal medicines in the management of different types of IBD. The search terms were: "herb" or "plant" or "herbal" and "inflammatory bowel disease". Data were collected from 1966 to 2013 (up to Feb). The "clinical response", "clinical remission", "endoscopic response", "endoscopic remission", "histological response", "histological remission", "relapse", "any adverse events", and "serious adverse events" were the key outcomes of interest. We used the Mantel-Haenszel, Rothman-Boice method for fixed effects and DerSimonian-Laird method for random-effects. For subgroup analyses, we separated the studies by type of IBD and type of herbal medicine to determine confounding factors and reliability. Seven placebo controlled clinical trials met our criteria and were included (474 patients). Comparison of herbal medicine with placebo yielded a significant RR of 2.07 (95%CI: 1.41-3.03, P = 0.0002) for clinical remission; a significant RR of 2.59 (95%CI: 1.24-5.42, P = 0.01) for clinical response; a non-significant RR of 1.33 (95%CI: 0.93-1.9, P = 0.12) for endoscopic remission; a non-significant RR of 1.69 (95%CI: 0.69-5.04) for endoscopic response; a non-significant RR of 0.64 (95%CI: 0.25-1.81) for histological remission; a non-significant RR of 0.86 (95%CI: 0.55-1.55) for histological response; a non-significant RR of 0.95 (95%CI: 0.52-1.73) for relapse; a non-significant RR of 0.89 (95%CI: 0.75-1.06, P = 0.2) for any adverse events; and a non-significant RR of 0.97 (95%CI: 0.37-2.56, P = 0.96) for serious adverse events. The results showed that herbal medicines may safely induce clinical response and remission in patients with IBD without significant effects on endoscopic and histological outcomes, but the number of studies is limited to make a strong conclusion.