A Randomized, Double-Blind, Pilot Study of Rifaximin 550 mg Versus Placebo in the Prevention of Travelers' Diarrhea in Mexico During the Dry Season


Jose Flores, MD, Paseo del Conquistador #611, Cuernavaca, Morelos, Mexico. E-mail: jose.flores.figueroa@uth.tmc.edu


Background. Rifaximin has been shown to be effective in treating and preventing travelers' diarrhea (TD) during the summer season.

Methods. The goal of this double-blinded multicenter trial was to assess the efficacy and safety of rifaximin 550 mg administered once daily for 14 days compared with placebo in the prevention of TD during the dry season in Mexico.

Results. There were 101 participants randomized. Overall, 25 participants developed TD during the 3 weeks of the study: 22% from the rifaximin group and 29% from the placebo group (p = 0.4). Mild diarrhea (defined as only one or two unformed stools during a 24-h period plus at least one abdominal symptoms) developed in only 3 (6%) participants taking rifaximin compared with 10 (21%) taking placebo during the first week of study (p = 0.03). No clinically significant or serious adverse events were reported.

Conclusions. Antibiotic prophylaxis of TD in Mexico during the dry season needs to be further studied and its benefits weighed against the benefits of self-treatment.

Travelers' diarrhea (TD), which occurs in approximately 40% of international travelers visiting high-risk areas,1 is caused by bacteria in approximately 80% of cases.2 A variety of drugs with antimicrobial effects have been used in the prevention of TD during periods of risk of no greater than 2 weeks, including doxycycline,3 bismuth subsalicylate,4 trimethoprim-sulfamethoxazole,5 and fluoroquinolones.6 Prophylaxis with antibacterial drugs is not generally recommended because of adverse effects of systemically absorbed drugs and risk of antimicrobial resistance for drugs that have important uses outside the gut.

Rifaximin is a nonsystemic, gut-selective antibiotic that has activity against enteric bacterial pathogens causing TD in multiple areas of the world,7 and has been shown to be effective in treating TD in studies carried out in Mexico.8 Previous clinical trials have been carried out during summer months in Mexico showing that a once daily dose of rifaximin (one, two, or three 200 mg tablets) was effective in preventing TD.9,10 The present study is a follow-up study of TD prevention using the rifaximin 550 mg tablet evaluated during the less rainy season in Mexico. The primary objective of this trial was to assess the safety and efficacy of rifaximin 550 mg compared with placebo in the prevention of TD during late summer, fall, and winter months in Mexico.

Patients and Methods

University of Texas physicians participated in the formal student orientations held on campuses in three Spanish schools in Cuernavaca, Mexico, and one Spanish school in Guadalajara, Mexico, from July 25, 2009 to January 16, 2010. Students were provided with health hints on staying well in Mexico, including describing the problems of accidents, altitude, constipation, and diarrhea, and offering strategies to prevention of TD. The prophylaxis clinical trial was then described. Eligible participants were ≥18 years of age traveling to Mexico for academic studies. In the week before traveling to Mexico, they could not have experienced diarrhea or received an antibacterial drug with expected activity against prevalent enteric pathogens (ie, fluoroquinolones, macrolides, azalides, or trimethoprim-sulfamethoxazole).

Treatments were randomly assigned 1 : 1 to receive one rifaximin 550 mg tablet (Xifaxan Tablets, Salix Pharmaceuticals, Inc, Morrisville, NC, USA) or one placebo tablet (identical in appearance to rifaximin tablet) administered orally once daily at the morning. The subjects were provided with their study medication at enrollment and were treated for 14 days on a double-blind basis. Each group was followed for a third week off medication as part of the study.

TD was defined as three or more unformed stools during a 24-hour period plus at least one of the following abdominal symptoms: nausea, vomiting, fecal urgency, or tenesmus. Mild diarrhea (MD) was defined as one or two unformed stools during a 24-hour period plus at least one of the described abdominal symptoms for TD. When a subject experienced TD, he or she was instructed to have a stool sample collected and submitted to our local laboratories, where it was shipped by overnight courier to Houston for determination of bacterial pathogens by previously described culture methods11 and presence of parasites in stool by enzyme immunoassay using commercially prepared kits for Giardia, Entamoeba, and Cryptosporidium (Alexon, Sunnyvale, CA, USA).

The study was approved by the committee for the protection of human subjects of the University of Texas Health Science Center at Houston. All participants provided written informed consent.

Statistical Analysis

The sample size selected (50 in each group) was based on comparable sample sizes in previous prophylactic studies that have been conducted9,10 and by a calculation of 95% power, 0.05 significance level, 80% protection rate for prophylaxis, and a 40% attack rate for the placebo with a 10% dropout rate. The primary end point was reduction in occurrence of diarrhea during each of the 2 weeks of study. Significance was calculated by two-sided Chi-squared test unless a comparison group had less than five elements, in which case Fisher's exact test was used. All calculations were carried out using Stata 10.0 (College Station, TX, USA).


A total of 101 subjects were enrolled into the study. Three participants dropped out from the study: one from the rifaximin group and two from the placebo group. Therefore, 98 subjects completed the study and were analyzed. Fifty-four participants were female (55%) and 44 (45%) were male, each treatment group had similar proportions of males and females (p = 0.2). The overall mean age at enrollment was 25 years (range, 18–67 y). Thirty-six (37%) participants were enrolled during the summer months, while 62 (63%) during the fall and winter months. Eight (8%) participants were enrolled in Guadalajara and 90 (92%) in Cuernavaca.

As noted in Table 1, 14 participants developed TD during the 2 weeks of study: 6 of 50 patients (12%) from the rifaximin group and 8 of 48 patients (17%) from the placebo group (p = 0.5). We also did not observe a difference in the occurrence of MD between the rifaximin and placebo groups (p = 0.3) during the 3-week study period. We only saw a statistical difference during the first week of study for the development of MD (p = 0.03). No difference in the occurrence of MD or TD between the two groups was seen during the second and third weeks of study.

Table 1.  Diarrhea attack rates among adults traveling to Mexico during late summer, fall, and winter months taking either 550 mg rifaximin or placebo
 Patients, n (%)p Value
  1. *One participant dropped out from study.

  2. Two participants dropped out from study.

  3. Fisher's exact test was used for comparisons when a group value was <5 elements.

Overalln = 50n = 48n = 98
 Travelers' diarrhea11 (22%)14 (29%)25 (26%)0.4
 Mild diarrhea11 (22%)15 (31%)26 (26%)0.3
 Total22 (44%)29 (60%)51 (52%)0.2
First wkn = 50n = 48n = 98
 Travelers' diarrhea3 (6%)5 (10%)8 (8%)0.5
 Mild diarrhea3 (6%)10 (21%)13 (13%)0.03
 Total6 (12%)15 (31%)21 (21%)0.02
Second wkn = 47n = 43n = 90
 Travelers' diarrhea3 (6%)3 (7%)6 (7%)1.0
 Mild diarrhea8 (17%)4 (9%)12 (13%)0.4
 Total11 (23%)7 (16%)18% (20%)0.4
Third wkn = 43n = 38n = 81
 Travelers' diarrhea5 (12%)6 (16%)11 (14%)0.6
 Mild diarrhea01 (3%)1 (1%)0.5
 Total5 (12%)7 (18%)12 (15%)0.4
Patients with TD who provided stool sample10 (91%)7 (50%)17 (68%)
Bacterial diarrhea6 (60%)2 (29%)8 (47%)0.3

Twelve of 36 (33%) participants enrolled during the summer developed TD: 6 of 19 (32%) from the rifaximin group and 6 of 17 (35%) from the placebo group (p = 0.9). Meanwhile, 12 of 62 (19%) participants enrolled during the fall and winter developed TD: 5 of 31 (16%) from the rifaximin group and 7 of 31 (23%) from the placebo group (p = 0.5). No difference in the incidence of MD or TD was observed during each of the 3 weeks in participants treated during the summer months. We observed that participants taking rifaximin during the fall and winter months were also less likely to develop MD during the first week of study compared with those taking placebo during the same timeframe (2 of 30 [6.7%] vs 8 of 29 [28%]; p = 0.04).

Seventeen of the 25 participants (68%) suffering from TD provided a stool sample for microbiological analysis before any antibiotic rescue therapy was administered: 10 (91%) from the rifaximin group and 7 (50%) from the placebo group. Bacterial diarrhea was detected in eight participants: six (60%) in the rifaximin group and two (29%) in the placebo group (p = 0.3; Table 1). Enteroaggregative Escherichia coli was the most common bacteria isolated (4 of 8), followed by enterotoxigenic E coli (3 of 8), diffusely adherent E coli (2 of 8), and Salmonella spp. (1 of 8). Two participants had mixed infections. No other bacterial or parasite pathogens were found. Three E coli isolates showed high minimum inhibitory concentration (MIC) for rifaximin (≥512 µg/mL): two of them isolated from participants in the rifaximin group and one taking placebo.

No clinical significant or serious adverse events were reported during the study. One participant in the placebo group developed mild transient lymphopenia, and another participant also in the placebo group developed asymptomatic mild indirect bilirubinemia (2.7 mg/dL) and mild aspartate transaminase elevation (46.0 IU/L). Investigators did not consider these adverse events to be drug related.


Rifaximin 550 mg was safely administered to international students during their time in Mexico the late summer and fall of 2009 and winter of 2009 to 2010. During the 2 weeks of study, 8 of 48 (17%) of placebo-treated subjects experienced TD. This is the lowest rate of diarrhea among students that we have reported in our trials to date. A lower rate would also be expected while studying subjects later in the year (September and later) when the rains have stopped. Also, significant decrease of fecal–orally transmitted diseases among travelers to Latin America and the Caribbean has been reported, probably due to improved hygienic standards.12 The proportion of diarrheal episodes caused by noroviruses increases during the winter months, whereas the rate of bacterial diarrhea decreases,13 although stool samples obtained from this study were not tested for norovirus.

In the current study, rifaximin failed to prevent TD compared with placebo, probably because of the low attack rate for illness. Rifaximin did provide protection against MD during week one of study among participants enrolled during late summer and nonsummer months. Similar to our study, another recent clinical trial using daily 1100 mg rifaximin conducted in Turkey between July 2007 and February 2008 also failed to show a statistically significant difference in the development of TD among participants taking rifaximin or placebo (p = 0.2).14

The prior clinical trials using rifaximin tablets that showed protection against TD9,10 were conducted in a different region of Mexico, and participants were enrolled only during the summer months.

This study has some limitations. The power was calculated taking in consideration a higher attack rate from prior similar studies. Also, not every participant suffering from diarrhea provided a stool sample for analysis. Only 50% of subjects taking placebo with TD provided a sample versus more than 90% of the subjects taking rifaximin.

The side effect profile of the rifaximin 550 mg appears to be comparable to results reported for the 200 mg. The one tablet, once daily administration of rifaximin, will likely be considered more convenient to take than multiple 200 mg tablets, and travelers may be more convenient with its use.


This study was supported by a grant through the University of Texas Health Science Center from Salix Pharmaceuticals, Inc.

Declaration of Interests

H. L. D. has received honorarium for speaking and consulting from Salix Pharmaceuticals, Inc. All the other authors state they have no conflicts of interest to declare.