LETTER

Nowadays, around 40 to 60% of Spanish travelers to developing countries develop diarrhea (4). Different enteropathogens have been associated with the development of traveler's diarrhea. The levels of prevalence of these enteropathogens as a cause of traveler's diarrhea are 42% for diarrheagenic Escherichia coli, 19.4% for Shigella spp., 3% for Salmonella spp., 2% for Campylobacter spp., 2% for Yersinia spp., 2% for Aeromonas spp., and <2% for others (4).

Infectious diarrhea is usually a self-limited disease lasting a few days and does not require antibiotic therapy. In some cases, antimicrobial therapy is recommended (2); however, high levels of resistance to several antimicrobial agents have been described. To resolve the problem of this increase in resistance, the activities of new antimicrobial agents should be studied. Rifaximin is a nonabsorbable antibiotic (2, 5, 6) achieving concentrations of 4,000 to 8,000 µg/g in feces, with a common therapeutic dosage being 800 mg divided in two oral administrations (7). The main aim of this study was to evaluate the in vitro activity of rifaximin against enteropathogens isolated as a cause of traveler's diarrhea.

MICs of several antimicrobial agents for 177 enteropathogens (Table 1) were determined by the agar dilution method according to guidelines of the National Committee for Clinical Laboratory Standards (8). E. coli ATCC 29522 was used as a quality control strain.

MICs at which 50 and 90% of the isolates tested were inhibited (MIC₅₀s and MIC₉₀s, respectively) and the percentages of resistance were calculated for each antimicrobial agent used in this study and are shown in Table 1.

The conventional antimicrobial agents, such as ampicillin, cotrimoxazole, tetracycline, and chloramphenicol, showed no or very little activity against the enteropathogens producing traveler's diarrhea. A MIC₉₀ of ampicillin of greater than 128 µg/ml was observed against all of the microorganisms, whereas the MIC₉₀s of tetracycline and trimethoprim for all the microorganisms were 16 µg/ml. Only nalidixic acid and ciprofloxacin showed MIC₉₀s of <16 and 0.06 µg/ml, respectively, for all the enteropathogens tested, with the exception of Campylobacter jejuni, for which MIC₉₀s were >256 and 32 µg/ml, respectively. The MIC₉₀s of chloramphenicol for the different microorganisms were in a range from 8 to >128 µg/ml.

Cotrimoxazole and ampicillin have been widely used to treat traveler's diarrhea (3, 11, 12), and the long use and sometimes the misuse of these antibiotics have been associated with the increase of resistance levels (1, 10, 12).

MIC₅₀s and MIC₉₀s of rifaximin and rifampin were very similar. MICs of rifaximin ranged from 4 to 8 and from 4 to 16 µg/ml, and MICs of rifampin ranged from 4 to 16 and from 8 to 16 µg/ml, for all tested bacteria except Yersinia enterocolitica and C. jejuni.

In particular, the MIC₅₀ and MIC₉₀ of rifamixin of 64 and 128 µg/ml, respectively, were observed for Y. enterocolitica and a value of >128 µg/ml for both the MIC₅₀ and MIC₉₀ was achieved for C. jejuni. In this case and only for rifaximin, doses greater than 128 µg/ml were tested to determine the precise MIC₅₀s and MIC₉₀s, which were found to be 256 and 512 µg/ml, respectively.

The in vitro activities of rifaximin against strains from stock culture collections of four university-associated teaching hospitals had been previously reported by Hoover et al. (6). In that study, the activities of rifaximin against enteropathogens were found to be as follows: a MIC₅₀ of 4 to 8 µg/ml and a MIC₉₀ of >8 µg/ml. These results are in accordance with ours, even though our strains were isolated from patients who traveled to different geographical areas. In another study, Ripa et al. (9) tested rifaximin against Campylobacter and Yersinia strains collected from patients with diarrhea.

The main difference between these studies and ours is the MICs of rifaximin for Y. enterocolitica and C. jejuni, two microorganisms which were isolated from not more than 2% of patients with traveler's diarrhea in our laboratory (3).

In conclusion, rifaximin is a nonabsorbable antimicrobial agent, reaching high concentrations in the intestinal tract. The concentrations of rifaximin achieved in the intestinal tract are more than 10-fold higher than the MICs of this antimicrobial agent for the different enteropathogens used in our study. In particular, we observed a definitely good in vitro activity of rifaximin against several enteropathogens, such as E. coli, Shigella spp., and Salmonella spp., which is in accordance with clinical and microbiological outcomes of two recent studies of traveler's diarrhea (2; H. L. DuPont, Z. D. Jiang, C. D. Ericsson, J. J. Mathewson, J. Aldachi, E. Palazini, L. S. Riopel, D. Ashley, and F. Martinez-Sandoval, Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 2227, p. 698, 1999).