We have recently shown that the majority of patients with irritable bowel syndrome (IBS) have an abnormal lactulose breath test to suggest the presence of bacterial overgrowth of the small intestine. In open label and double blind treatment of IBS subjects with antibiotics, a dramatic improvement in clinical symptoms are observed. In these studies, the antibiotic chosen was neomycin, which is noted to have an efficacy of 20-25% in normalizing the lactulose breath test. A more efficacious antibiotic is needed. Therefore the aim of this study is to determine the efficacy of rifaximin in normalizing the lactulose breath test in IBS subjects with concomitant improvement in clinical symptoms.

1. Rome I criteria positive IBS subjects will be enrolled in the study after inclusion and exclusion criteria (see below for description of inclusion and exclusion criteria) are applied. There will be no preference given to constipation or diarrhea predominant forms of the condition.
2. Subjects will have a 7-day run in phase during which bowel movement consistency and number will be recorded (see Appendix A for bowel movement record).
3. At the end of the 7-day run in, subjects will present after a 12 hour fast for a lactulose breath test.
4. Prior to breath testing, subjects will be asked to fill out a symptom questionnaire (Appendix B) based on the previous 7 days. Completion of the questionnaire is expected prior to starting the breath test to prevent symptoms from the lactulose interfering with the testing.
5. Subjects will then provide a baseline breath sample after which 10g of lactulose will be administered. Breath samples will then be obtained every 15 minutes for 180 minutes. The test will be plotted graphically and sealed by the technician in an envelope. The subject and the investigators will be blinded to the results of the test. See below for details of the breath testing.
6. Subjects will then be randomized to receive either placebo or rifaximin (400 mg tid) for 10 days. Compliance will be tested by pill count.
7. On completion of the treatment, bowel movement consistency and number will again be recorded for 7 days.
8. 7 days after completion of the antibiotic or placebo, subjects will return for a follow up questionnaire (Appendix C) and breath test.
9. Subjects will then be sent a weekly questionnaire for a period of 2 months. During the final 7 days of the 2 months, the bowel movement consistency and number will be tracked again.

SPECIFIC METHODS:

Subject selection and enrollment:

Subjects will be selected for all studies based on the Rome I criteria (36). This will be the preferred method of identifying IBS patients to avoid pre-selecting patients with C- IBS. The goal is to evaluate methane as the determinant of transit, not constipation or diarrhea symptoms as through Rome II criteria (37). All subjects who are receiving a prokinetic drug at the time of enrollment will need to have a washout period of 7 days before starting the study protocol. Subjects will be identified through advertising in printed media and through the clinical operations of the GI motility program at Cedars-Sinai Medical Center.

Exclusion Criteria:

Subjects with a history of inflammatory bowel disease (38), diabetes (39), previous intestinal surgery (40-42), cirrhosis (43-47), celiac disease, probiotic use, current proton pump inhibitor use (48-54), recent antibiotic use (past 3 months), history of bowel obstruction (55), narcotic use (56) or age greater than 65 years (57-63) will be excluded. Most of these conditions are known to influence enteric bacteria levels. Women of childbearing years will undergo pregnancy testing before participating in the study (See below for details on the pregnancy test used). Women with positive pregnancy tests will be excluded.

Lactulose Breath test (LBT):

Subjects will present to the GI Motility clinic after a 12 hour fast. After a baseline breath sample and ingestion of 10g-lactulose syrup (Inalco Spa, Milano, Italy, packaged by xactdose Inc., South Beloit, IL), end-alveolar breath samples will be taken every 15 minutes for a three hours. Breath samples will be analyzed using a model SC Quintron gas chromatograph (Quintron Instrument Company, Milwaukee, WI) which determines breath hydrogen and methane concentration in parts per million (ppm). Subjects will not be allowed to smoke or exercise within 2 hours prior to performing the test (64). Methane exchanges quickly across the alveolar membrane, so hyperventilation can falsely lower the concentration in breath. Subjects will also be required to sit for the duration of the breath test such that exertion will not increase alveolar clearance of this gas. If modest hyperventilation occurs, the concentration of carbon dioxide in the breath sample will be used to correct for quality of alveolar sampling. Note that in the clinical study breath test results will be blinded to the investigator and patient to avoid influencing the interpretation of transit and serotonin levels.

Questionnaires:

Three questionnaires will be used in this study. The first is a Stool Form and Frequency Questionnaire (SFFQ). This questionnaire will be designed to determine the consistency of the stool based on standard consistency plots. The second questionnaire is designed to determine the duration and degree of IBS symptoms on visual analogue scores. This questionnaire is specifically administered before the first breath test and is designed to determine baseline symptoms and severity. The final questionnaire is a follow up questionnaire. This questionnaire also plots the degree of symptoms based on visual analogue scores. However, subjects are further asked to rank their degree of improvement based on percent improvement. In these last two questionnaires, subjects are asked to consider their answers as a summary of the preceding 7 days.

Pregnancy Testing:

In all human studies, pregnancy testing will be performed 1 week prior to initiating study. The test will specifically be performed in women of childbearing years. This will involve a blood draw for the quantitative determination of Beta HCG. The testing is an immunoassay for determination of total Beta HCG (Abbott Axsym Test) and performed by Cedars-Sinai Clinical Laboratory (Los Angeles, California).

Inclusion Criteria:

• Subjects will be selected for all studies based on the Rome I criteria. This will be the preferred method of identifying IBS patients to avoid pre-selecting patients with C- IBS. The goal is to evaluate methane as the determinant of transit, not constipation or diarrhea symptoms as through Rome II criteria. All subjects who are receiving a prokinetic drug at the time of enrollment will need to have a washout period of 7 days before starting the study protocol. Subjects will be identified through advertising in printed media and through the clinical operations of the GI motility program at Cedars-Sinai Medical Center.

Exclusion Criteria:

• Subjects with a history of inflammatory bowel disease, diabetes, previous intestinal surgery, cirrhosis, celiac disease, probiotic use, current proton pump inhibitor use, recent antibiotic use (past 3 months), history of bowel obstruction, narcotic use or age greater than 65 years will be excluded. Most of these conditions are known to influence enteric bacteria levels. Women of childbearing years will undergo pregnancy testing before participating in the study (See below for details on the pregnancy test used). Women with positive pregnancy tests will be excluded.