Comparison Between Lactulose and Glucose Breath Tests as Predictors of Clinical Benefit From Rifaximin in Irritable Bowel Syndrome Patients

After signing the informed consent, all patients will be submitted to both glucose and lactulose breath tests within a 10 days period. The order of the test will be randomized using a random number list of allocation, which will not be known by the patient. For the lactulose test a 10 g and for the glucose test a 50 g dose will be used, as previously suggested. After the second test, patients will receive Rifaximin 400 mg TID (1200 mg/day) for 10 days. In the case of methane producers (defined as patients with basal methane levels > 3 ppm) Neomycin 500 mg BID (1000 mg/day) for ten days will be added, as suggested by Low et al. Two weeks after the antibiotic course, patients will be cited and the following data will be assessed: 1. Adequate relief for global IBS symptoms. 2. Adequate relief for bloating. 3. Evaluation of IBS severity using IBSSS. 4. Drug induced side effects. Only after that information is submitted by the patient, the results of the test will be revealed. Patients will be classified as responders or not according to their global symptoms adequate relief status. Using this as gold standard criterion, the test performance, in terms of sensitivity, specificity, positive and negative predictive value, will be calculated for the following tests: Glucose breath test. A ROC curve will be calculated to determine the best cutoff value Lactulose breath test using the rise occurring before 60 min, without the presence of a double peak. A ROC curve will be calculated to determine the best cutoff value Lactulose breath test using the rise occurring before 60 min, if there is a double peak. A ROC curve will be calculated to determine the best cutoff value Lactulose breath test using the rise occurring before 90 min, without the presence of a double peak. A ROC curve will be calculated to determine the best cutoff value Lactulose breath test using the rise occurring before 90 min, if there is a double peak. A ROC curve will be calculated to determine the best cutoff value The different tests will be compared in terms of their performance at the optimal cutoff value and calculating the area under the ROC curve. The sample size was calculated as previously suggested. Using Shah et al data, we assumed a sensitivity of 72% and a specificity of 66% for the lactulose breath test. Using an alpha and beta error of 5% and 20%, respectively, and a disease prevalence of 35% (SIBO in IBS), we estimate a sample size of 120 patients.