S1052 Hydrogen Sulfide Levels on Breath Testing Are Associated With GERD and Esophageal Symptoms, Independent of PPI Use

Introduction: Breath testing (BT) is used in the workup of patients with chronic gastrointestinal symptoms and measures hydrogen (H2), methane (CH4), and now, hydrogen sulfide (H2S). The relationship between breath test results and gastroesophageal reflux disease (GERD) is underexplored. Recently, we showed that GERD subjects have an increased levels of the H2S producer, Bilophila species, in the foregut. H2S can have detrimental effects on lower esophageal sphincter tone. In this study, we examine the relationship between breath H2S among GERD and non-GERD in a nationwide study. Methods: Consecutive subjects undergoing an in-home 3-gas BT were asked to complete a symptom questionnaire, medication intake (e.g., proton pump inhibitors [PPI]), and if they ever were diagnosed with reflux. Subjects prepared for BT per the North American Consensus. After fasting, subjects provided an end-expiratory breath sample and then ingested lactulose (10 g) or glucose (75 g) at the discretion of the physician. Breath samples were collected in consecutive bags every 15 minutes for 2 hours. Bags were then compiled and processed using a specialized gas chromatography instrument that measured H2S at baseline and at each time point. H2S >2.0 ppm was considered positive. Results: Of 6,000 subjects, 3,029 (50.4%) completed a breath test and questionnaire (F = 74.5%, age = 50.3 ± 18.1 years). Of these, a total of 1,064 subjects had GERD with an average age higher than subjects without reflux (54.6 ± 17.5 years vs 49.0 ± 46.8 years, respectively) (P < 0.0001). As expected, more subjects with reflux were on PPI (P < 0.0001), and had more severe “burning esophagus” sensation than those without reflux (P < 0.0001). Subjects with reflux had significantly higher H2S area under the curve (AUC) when compared to those without reflux (P = 0.039), and this signal was still observed when reflux subjects not taking PPI were compared to subjects without reflux also not taking PPI (P = 0.042). In addition, the symptom severity of the “burning esophagus” sensation correlated positively with H2S AUC (P = 0.002). Subjects without reflux but taking PPI had higher H2S AUC when compared to subjects without reflux and not taking PPI (129.2 ± 31.6 vs 122.7 ± 29.7, respectively) (P = 0.001). As previously described, H2S correlated with diarrhea severity as well (P = 0.003). Conclusion: This study demonstrates that elevated H2S levels on breath testing are associated with GERD with or without PPI use. In addition, H2S levels correlate with the severity of GERD symptoms as well as diarrhea.