Effectiveness of rifaximin in prevention of diarrhoea in individuals travelling to south and southeast Asia: a randomised, double-blind, placebo-controlled, phase 3 trial

Summary Background Travellers' diarrhoea causes substantial acute and long-term morbidity. Chemoprophylaxis with fluoroquinolones or rifaximin is effective in prevention of diarrhoea in individuals travelling to Latin America and Africa. Little evidence is available to support the protective effect of antimicrobial drugs in south and southeast Asia, where enteroinvasive and antibiotic-resistant bacteria cause a substantial proportion of diarrhoeal episodes. We aimed to assess the effectiveness of rifaximin in prevention of diarrhoea in individuals travelling to south and southeast Asia. Methods We did this double-blind, placebo-controlled, single-centre, parallel-group, clinical trial in Tubingen, Germany, between Nov 12, 2009, and Sept 3, 2012. Individuals aged 18–64 years who were planning a 6–28 day journey to south and southeast Asia were randomly assigned (1:1), according to a randomisation list (permuted block size of eight) generated by an independent statistician, to receive placebo or rifaximin 200 mg tablets twice daily. All members of the study team, including investigators, those assessing outcomes, and data analysts, were masked to treatment allocation. The primary endpoint was time to the first episode of classic travellers' diarrhoea, defined as three or more loose stools in 24 h, accompanied by one or more enteric symptoms. Analyses were by intention to treat and per protocol. This study is registered with ClinicalTrials.gov, number NCT00979056. Findings We randomly assigned 258 participants to rifaximin (n=129) or placebo (n=129), of whom 239 (93%) returned a completed diary and were included in the primary effectiveness analysis. 48 (41%) of 117 participants in the placebo group and 30 (25%) of 122 in the rifaximin group reported classic episodes of travellers' diarrhoea. From departure to 7 days after return, rifaximin provided 48% protection (95% CI 16–68) by lowering the incidence of travellers' diarrhoea from 1·99 (1·50–2·64) per 100 person-days in the placebo group to 1·04 (0·72–1·48) in the intervention group (incidence rate ratio 0·52, 95% CI 0·32–0·84; p=0·005). The number needed to treat was 5·70 (95% CI 3·44–16·69) to prevent one case of classic travellers' diarrhoea during the first 3 weeks of follow-up. The per-protocol analysis essentially corroborated the findings from the intention-to-treat analysis. We recorded one serious adverse event in a participant in the rifaximin group who had grade 3 right lower quadrant abdominal pain 72 h after the last intake of study drug. The complaints were considered unlikely to be related to use of the drug. Interpretation Rifaximin is moderately effective in prevention of diarrhoea in individuals travelling to south and southeast Asia. Similar studies are needed to inform travellers and practitioners about the effectiveness of this drug at other popular destinations. Funding University of Tubingen, Program for Applied Clinical Research