Postprandial Breath H2 Using a Portable Handheld App-Connected Device to Predict the Presence of Small Intestinal Bacterial Overgrowth as Well as the Response to Antibiotic Treatment

Introduction: The lactulose hydrogen breath test (LHBT) is widely used to diagnose SIBO but its accuracy is variable. It is desirable to establish a biomarker that reliably predicts antibiotic response. We tested an app-connected personal breath analyzer (AIRE®, FoodMarble) for real-time measurement of postprandial (PP) H2 and symptoms. The aims of this study were: (1) to compare the performance of AIRE vs. a mail-in LHBT kit; (2) to evaluate daily PP breath H2 in suspected SIBO patients treated with antibiotics. Methods: This is a prospective study in which patients suspected to have SIBO in a tertiary level clinic are provided the AIRE device to monitor breath H2 concentrations before (baseline) and after (every 30 minutes) the morning and evening meals for a week. Thereafter a lactulose challenge is administered, and breath H2 concentrations measured by the mail-in kit and the AIRE device. A decision to treat with antibiotics for SIBO was made based on the clinical judgement of the treating physician (not solely on the H2 breath concentrations). After two weeks of treatment, the AIRE device was used to monitor PP H2 concentrations. Responders to antibiotics were defined as having an improvement of ≥ 3 on a 15 point PAGI-SYM questionnaire administered after treatment. Results: 35 patients suspected of SIBO recorded baseline and PP H2 and symptoms in response to their morning and evening meals for 1 week using the AIRE device; LHBTs were performed using AIRE and mail-in kits. 19 patients were prescribed antibiotics. There was near complete congruence between the AIRE and the mail-in kit (r = 0.99) for the LHBT. Of the 19 patients treated with antibiotics (26%) qualified as responders; 60% (3/5) of them had a positive LHBT on both AIRE and mail-in kit; 36% (5/14) of non-responders had a positive AIRE LHBT. Analysis of the PP H2 responses prior to treatment showed that the average (AM&PM) PP H2 was higher in responders compared to non-responders as well as healthy controls. (Fig. 1). Following treatment responder H2 normalized and was comparable to healthy subjects but there was no change in H2 in the non-responder group. Conclusion: AIRE performs as well as the standard LHBT for diagnosis of SIBO by established criteria. Daily PP H2 data shows promise in identifying those that are likely to respond to treatment and has the potential for providing a more reliable biomarker for SIBO. Using remote monitoring technologies to gather longitudinal breath data may enable focused and personalized SIBO treatment.Figure 1.: Post-prandial averages of AM and PM breath H2 in responders, non-responders and healthy controls, analyzed by 2-way repeated measures ANOVA.