Back

Hydrogen sulfide small intestinal bacterial overgrowth case registry

Goldenberg, J.; Nevitt, B.; Wentz, A.; Bradley, R.; Siebecker, A.

2023-03-07 gastroenterology
10.1101/2023.03.07.23286900 medRxiv
Show abstract

BackgroundThere is growing interest in hydrogen sulfide small intestinal bacterial overgrowth (H2S SIBO). However, basic questions including how clinicians are making the diagnosis, what symptoms are present, and what clinicians are using for treatment, remain unanswered. AimsTo address this, we created an online, survey-based, clinical registry of H2S SIBO cases. MethodsParticipants in this study were clinicians diagnosing and treating patients with H2S SIBO and input data on patient presentation, diagnosis, treatment, and treatment response. We describe the population and summarize our results using descriptive statistics. We use Pearsons chi-squared test and modified Poisson regression in exploratory analyses. Results131 total cases were submitted by 99 providers across a spectrum of health disciplines. The mean patient age was 45.6 (71.4% female). The most common symptoms were bloating (77.0%), constipation (50.8%) and abdominal pain (50.8%). Diagnosis was made based on flatline hydrogen in the 3rd hour of a lactulose breath test (42.5%), symptom presentation (empiric) (31.9%), or H2S levels (25.7%). The most common treatments used were a low sulfur diet (46.6%), oregano (44.0%), and bismuth (39.7%). Fifty-eight percent of cases were responders. Of the most common interventions used, only a low sulfur diet (73% responder; p=0.01) and bismuth (76% responder; p=0.01) were significantly associated with treatment response. Interestingly, response rates differed based on how H2S SIBO was diagnosed, with empiric underperforming flatline diagnoses (relative risk 0.60; p=0.04). ConclusionsThis case registry represents the largest collection of H2S SIBO cases to-date, providing important early descriptive information on this emerging diagnosis.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.