Efficacy of Rifaximin in Prevention of Travelers' Diarrhea: A Meta-Analysis of Randomized, Double-Blind, Placebo-Controlled Trials


Corresponding Author: Haibin Dai, PhD, Department of Pharmacy, Second Affiliated Hospital, Zhejiang University School of Medicine, 88 Jiefang Road, Hangzhou 310009, China. E-mail: haibindai@zju.edu.cn



Rifaximin has been used successfully for the prevention of travelers' diarrhea (TD), the most general cause of disability among international travelers to developing tropical and semitropical regions.


We sought to better evaluate the efficacy of rifaximin in the prevention of TD. Randomized controlled trials (RCTs) of rifaximin for the prevention of TD published in Pubmed, the Cochrane Central Register of Controlled Trials, Embase, and the Science Citation Index were searched. [Correction added on 3 October 2012, after first online publication: the phrase “protection of TD” was replaced with “prevention of TD”.] The primary efficacy outcome was occurrence of TD over a 2-week treatment period. Secondary outcomes were requirement for antibiotic treatment, occurrence of mild diarrhea (MD), occurrence of TD in the third week after drug withdrawal, incidence of TD associated with isolation of diarrheagenic Escherichia coli (ie, ETEC, EAEC), and adverse events.


Four RCTs with 502 participants were included in the systematic review. Rifaximin treatment showed a significant protection against TD (risk ratios, RR: 0.41, 95% CI: 0.30–0.56, p < 0.00001) and needed antibiotic-treated TD (relative risk [RR]: 0.30, 95% confidence interval [CI]: 0.18–0.49, p < 0.00001). There was no significant difference between rifaximin and placebo in the occurrence of MD (RR: 1.11, 95% CI: 0.78–1.59, p = 0.55) and the occurrence of TD in the third week after drug withdrawal (RR: 0.73, 95% CI: 0.30–1.73, p = 0.47). Enterotoxigenic E. coli was the major cause of TD, and all trials reported no differences in adverse events between rifaximin and placebo.


Rifaximin can prevent TD caused by non-invasive enteric pathogens. Further research is needed for the treatment of invasive enteric pathogens. [Correction added on 3 October 2012, after first online publication: the phrase “Rifaximin can protect TD” was replaced with “Rifaximin can prevent TD”.]

The most general cause of disability among international travelers to developing tropical and semitropical regions is diarrhea. Travelers' diarrhea (TD) occurs in 15%–50% of individuals traveling to high-risk regions of southern Asia, Africa, Latin America, and the Caribbean (Haiti and the Dominican Republic).[1] Although TD is a non-fatal illness, it causes serious morbidity and is disruptive to any travel plan. Individuals with TD experience an average of 24 hours of total disability.[2] Affected individuals may experience persistent diarrhea lasting for weeks, months, or years.[3] More than 10% of people experiencing TD go on to develop post-infectious irritable bowel syndrome (PI-IBS).[4]

Enterotoxigenic E. coli (ETEC and EAEC) cause approximately half of TD in Latin America, Africa, South Asia, and the Middle East.[5, 6] It was first shown by Kean[7] that antibiotics can prevent a large proportion of TD. In the 1970s and 1980s, doxycycline and fluoroquinolones were successfully used to prevent TD.[8, 9] A National Institutes of Health (NIH) Consensus Development Conference in 1985, however, discouraged using prophylactic antibiotic treatment because of concern about absorbable antibiotics contributing to the development of resistance strains.[10] Rifaximin is a non-systemic, gut-selective antibiotic that has activity against enteric bacterial pathogens causing TD in multiple areas of the world.[11, 12] The small study size of previous studies has yielded inconsistent findings. The purpose of this meta-analysis was to integrate all available data to provide a clearer understanding of rifaximin's efficacy.


Data Sources

A systematic search of the literature in PubMed (up to November 2011), the Cochrane Central Register of Controlled Trials (Cochrane Library Issue 4, October 2011), Embase (up to November 2011), and the Science Citation Index (up to November 2011) was conducted to identify relevant randomized controlled trials (RCTs) for our meta-analysis. In addition, references from the trials were further searched manually to identify potentially relevant studies.

Study Selection

The following selection criteria were applied: (1) study design: randomized, controlled trial; (2) study population: healthy, adult civilian travelers or military members aged ≥18 years; (3) intervention: prophylactic administration of rifaximin; (4) comparison intervention: placebo; (5) outcome measures: the primary efficacy end point was occurrence of diarrhea during 14 days of treatment with rifaximin or placebo. TD was defined as passage of at least three unformed stools within a 24-hour period plus one or more of the following signs or symptoms of enteric infection: abdominal pain or cramps, nausea, vomiting, fever (≥37.8°C), fecal urgency, passage of gross blood or mucus in stool, tenesmus, or moderate to severe increase in intestinal gas.[13] Secondary end points included: incidence of the required antibiotic treatment, occurrence of mild diarrhea (MD; defined as a passage of one to two unformed stools during a 24-hour period plus at least one of the described abdominal symptoms for TD), incidence of TD occurring in the 7-day follow-up period, incidence of TD associated with isolation of diarrheagenic E. coli (ie, ETEC, EAEC), TD associated with unidentified pathogens, and any adverse events.

Data Extraction and Management

Two review authors independently extracted details of randomization methods, blinding of treatments, and outcome assessments. Standardized, detailed forms for extraction of data from the selected trials (Table 1) were developed. These included year of publication, author's name, RCT study design, setting, participant population, age, gender, intention to treat, dosage of rifaximin, duration, and method of follow-up. Any disagreement among authors in scoring or data abstraction was resolved by discussion.

Table 1. Characteristics of the trials included in the meta-analysis
StudyRCT study design, settingStudy populationITT (rifaximin vs placebo)Intervention (dosage, duration)Follow-up
  1. RCT = randomized controlled trial; ITT = intention to treat.

Armstrong and colleagues[15]Double-blind, Turkey≥18 y of age (median age 36), 88% were males, 73% were Whites95 (48 : 47)Rifaximin (1,100 mg qd, for 14 d) 
DuPont and colleagues[16]Double-blind, MexicoAt least 18 y of age108 (54 : 54)Rifaximin (200 mg tid, for 14 d)3 wk
Martinez-Sandoval and colleagues[18]Double-blind, multicenter, MexicoMedian age 21 y (range: 18–75 y), 65% were female201 (99 : 102)Rifaximin (600 mg once daily, for 14 d)1 wk
Flores and colleagues[17]Double-blind, multicenter, MexicoMean age 25 y (range: 18–67 y), 55% were female98 (50 : 48)Rifaximin (550 mg, for 14 d)1 wk

Quality Assessment

The methodological quality of the trials was assessed using the evaluation criterion of the Cochrane collaboration's tool for assessing risk of bias.[14] A quality review of each RCT was done to include random sequence generation, allocation concealment, blinding to participants and personnel, blinding to outcome assessment, incomplete outcome data, selective outcome reporting, and other potential sources of bias. Risk of bias for each domain was rated as high (seriously weakens confidence in the results), low (unlikely to seriously alter the results), or unclear (Cochrane hand book 5.1).

Data Analysis and Statistical Methods

Statistical analysis was performed using Review Manager Version 5.1.6 (Cochrane Collaboration, Oxford, UK). Heterogeneity of trial results was assessed by using chi-square test of heterogeneity and the I2 measure of inconsistency. Fixed effect risk ratios (RRs) for dichotomous variables and weighted mean differences for continuous variables with 95% confidence intervals (CIs) were calculated throughout the meta-analysis unless statistically significant heterogeneity was found (p < 0.10, I2 > 50%), in which case a random-effects model was used. To further test the robustness of the results, sensitivity analyses were also performed a priori. Evaluation of whether the model used (random-effects model vs fixed-effects model) would change the results was also done. Absolute risk reductions and numbers needed to treat (NNT) were also calculated, which represents the number of patients who must receive rifaximin instead of placebo to avoid one outcome event. Because of the limited numbers of studies available, the publication bias was not evaluated by examining funnel plots.


Randomized Controlled Trial Selection

A total of 245 citations were retrieved of which 241 were excluded due to various reasons (Figure 1). Four RCTs were included in this meta-analysis,[15-18] all of which were available as full articles.

Figure 1.

Flow diagram of the randomized controlled trials reviewed.

Study Characteristics

RCTs included in this meta-analysis are shown in Table 1. Five hundred and two participants from four trials were administered prophylactic treatment with rifaximin or placebo to prevent TD. One trial[16] included treatment doses of rifaximin at 200 mg qd, bid, and tid. Only the 200 mg tid data were selected for analysis as this closely approximated the dosing in the other studies. All studies were randomized, double-blind, placebo-controlled trials. The participants were US student travelers in Mexico in three studies, and US military personnel in Turkey in one study.

Risk of Bias

One study was at low risk of bias for random sequence generation and adequate allocation concealment[16] (not stated in three studies). All of the studies were blinded. Only one study was blinded to outcome assessment[15] (not stated in three studies).[16-18] All studies were at low risk of bias for incomplete outcome data. Three studies were at low risk of bias for selective reporting[16-18] (not stated in one studies).[15]

Quantitative Data Synthesis

Primary Efficacy Outcome (Occurrence of TD Over the 2 wk Treatment Period)

A total of 502 patients were included in the four trials comparing rifaximin with placebo for prevention of TD (Figure 2).[15-18] One-hundred forty-two patients developed TD of which 41 were in the rifaximin group and 101 were in placebo group. The included trials were homogeneous (test for heterogeneity: p = 0.16, I2 = 42%), and the incidence of TD was significantly different between the rifaximin group and the placebo group (RR: 0.41, 95% CI: 0.30–0.56, p < 0.00001). NNT was four, which implied that four patients must receive rifaximin to avoid one case of TD.

Figure 2.

Efficacy of rifaximin in the prevention of TD. The CI for each study is represented by a horizontal line, and the point estimate is represented by a square. The size of the square corresponds to the weight of the study. The diamond is centered on the summary RR of the observational studies, and the width indicates the corresponding 95% CI. TD = travelers' diarrhea; CI = confidence interval; RR = relative risk.

Secondary Efficacy Outcomes

Antibiotic Treatment of TD

Seventy two of 404 patients in three trials required antibiotic treatment for TD, 16 in the rifaximin group and 56 in the placebo group.[15, 18, 19] The included trials were not homogeneous (heterogeneity test: p = 0.11, I2 = 55%) so a fixed model was applied. The incidence of antibiotic treatment was significantly different between the rifaximin group and the placebo group (RR: 0.30, 95% CI: 0.18–0.49, p < 0.00001). NNT was five, which implied that one patient in every five would avoid antibiotic treatment for TD.

Occurrence of MD

There were 197 patients involved in three trials comparing rifaximin with placebo in whom the incidence of MD could be evaluated.[15, 17, 18] The included trials were homogeneous (heterogeneity test: p = 0.25, I2 = 28%). Rifaximin was not associated with significantly reduced incidence of MD (RR: 1.11, 95% CI: 0.78–1.59, p = 0.55).

Occurrence of TD in the Third Week after Drug Withdrawal

There were 153 participants involved in two trials comparing rifaximin with placebo, reporting the incidence of TD in the third week after drug withdrawal.[16, 17] After eliminating the first 2 weeks of data regarding diarrhea, the data were not homogeneous (heterogeneity test: p = 0.97, I2 = 0%). There was no significant difference (p = 0.47) in the incidence of TD in the third week after drug withdrawal between the two groups.

Microbiological Findings

Enterotoxigenic E. coli was the major cause of diarrhea and MD during the 2 weeks of drug administration.[16, 18] There was no significant difference between the rifaximin group and the placebo group in TD associated with diarrheagenic E. coli (ETEC or EAEC) (RR: 0.52, 95% CI: 0.24–1.09, p = 0.08). There was significant difference between the two groups in the incidence of unidentified pathogens associated with TD (RR: 0.37, 95% CI: 0.19–0.69, p = 0.002).[16, 17]

Adverse Events

All trials reported that there were no observed differences in adverse events between the rifaximin group and the placebo group. There was no clinically significant or serious adverse event in any of these studies.[15-18] There were no clinically relevant laboratory abnormalities reported.[16, 18]


This meta-analysis shows an advantage of rifaximin over placebo in preventing TD. [Correction added on 3 October 2012, after first online publication: the phrase “protecting TD” was replaced with “preventing TD”.] Prophylactic treatment would result in one patient in every four avoiding TD, and one in every five avoiding the need for antibiotic treatment. There were no significant differences between rifaximin and placebo in the incidence of diarrhea or MD after treatment was stopped. Enterotoxigenic E. coli was the major cause of diarrhea and MD in this study. All the trials reported no differences in the rate of adverse events between the two groups. Statistical analysis using fixed-effects model and random-effects model demonstrated similarly significant results.

There are some limitations in the present meta-analysis. Owing to limited numbers of studies available, use of funnel plots to evaluate publication bias was not possible. The research data were obtained from participants' diaries, so the outcome measurement has a degree of subjectivity. Owing to the lack of relevant information on the original works, such as microbiological findings, an adequate statistical analysis could not be performed. Finally, identifying the most effective dose or frequency of rifaximin was also not possible in this review.

Nearly all studies of TD were carried out in healthy adult subjects. The application of these findings to less healthy populations or different travel environments requires further validation.

Up to 40% of TD cases are of unknown etiology, even after the comprehensive microbiological evaluation.[19-21] Rifaximin can prevent illness caused by diarrheagenic E. coli including ETEC and enteroaggregative E. coli, but not against invasive bacterial strains. The use of rifaximin in geographic areas with different pathogenic bacteria requires further evaluation. In a volunteer study, it was found that shigellosis was prevented by prophylactic oral rifaximin.[22] Its efficacy in preventing diarrhea caused by other invasive organisms found in Asia, including Salmonella and Campylobacter, is not known.[21, 23]

The risk of acquiring TD in any geographic region is influenced by the season. Rainy seasons are associated with a higher risk than dry seasons. Local weather conditions and type of travel (ie, in camping and backpacking) can also affect the risk of acquiring TD.[24] Also, the incidence of diarrheal episodes caused by noroviruses increases during the winter months.[25]

The most common organisms developing resistance to rifaximin are aerobic Gram-positive cocci. Gram-negative organisms, such as E. coli, do not develop resistance to rifaximin after 3 to 5 days of therapy.[26-28] In spite of these advantages, owing to rifaximin's structural relationship to other rifamycins, the resistance rates to rifaximin in Enterococcus, Bacteroides, Clostridium, and Enterobacteriaceae range from 30% to 90% after 5 days of treatment. When rifaximin treatment is stopped, the resistant strains tend to disappear within 1 to 12 weeks.[29] Current recommendations advise treating diarrhea with azithromycin during rifaximin prophylaxis,[30] because of the increased risk of an invasive enteropathogen.

Two prospective, randomized, double-blind, placebo-controlled trials of 1,260 patients found that rifaximin gave better relief from IBS symptoms compared with placebo (40.7% compared with 31.7%, p < 0.001).[31, 32]

In summary, rifaximin can prevent TD caused by non-invasive enteric pathogens. Further research is needed regarding the treatment of invasive enteric pathogens. The risk of diarrhea should be weighed against the risk of adverse events and bacterial resistance when prescribing prophylactic antibiotics for TD.


This project was supported by the grant from the National Natural Science Foundation of China (81173040), and the Foundation from the Health Bureau of Zhejiang Province (2011KYA065, 2012RCA027). The authors wish to thank the Chinese Evidence-Based Medicine Center/The Chinese Cochrane Center and also Mr. Liming Wu for assistance in data collection and editorial assistance.

Declaration of Interests

The authors state they have no conflicts of interest to declare.