Background: A higher Helicobacter pylori (H. pylori) prevalence was associated with increased rates of post-operative complications including increased marginal ulceration and leak rates. Limited studies described prevalence of H. pylori in patients undergoing bariatric surgery and H. pylori associated upper and lower gastrointestinal bleeding (GIB). Primary Aim of our study was to evaluate the prevalence of GIB in patients with H. pylori infection undergoing bariatric surgery. The secondary aim of the study was to identify odds of having GIB in patients with H. pylori infection in comparison without H. pylori infection.

Methods: A retro-spective cross-sectional observational study from nationwide inpatient sample (2007-2018) was planned amongst morbid obese adults undergoing bariatric surgery (Open roux-en-y gastric bypass, laparoscopic roux-en-y gastric bypass, laparoscopic adjustable gastric banding). Bariatric surgeries, H. pylori infection, GIB were identified using ICD 10 codes. Univariate analysis using chi-square test and unpaired t test and multivariate analysis using mix-effects survey logistic regression modeling were performed to find out prevalence and association of GIB among patients with H. pylori infection in comparison without H. pylori infection. Adjusted odds ratio (aOR) with 95% confidence interval (95%CI) were calculated using alpha criteria of 0.05 as statistically significant.

Results: Out of 167795 who underwent bariatric surgery, 1204 (0.72%) had GIB and 480 (0.29) had H. pylori infection. Patients with H. pylori were more older (51 vs 47-years old), 50-75 years age group (12.5% vs 2.5%), Black race (26.6% vs 15.4%), Hispanic race (26.6% vs 13.9%), lower median household income 0-25 percentile (44.6% vs 26.2%), having medicare (32.3% vs 19.2%), hypertensives (57% vs 56%), and CHF (4.2% vs 2.7%) in comparison to those without H. pylori. Prevalence of GIB was higher in H. pylori (9.37% vs 0.69%) in comparison without H. pylori . In demographics and comorbidity adjusted regression analysis, patients with H. pylori were associated with 143% higher (aOR 2.43, 95%CI 1.60-3.71, p< 0.0001) odds of GIB in comparison with patients without H. pylori. (c value/ ROC =0.94)

Conclusions: H. pylori infection had increased risk-adjusted occurrence of GIB amongst patients undergoing bariatric surgery. H. pylori testing may improve risk stratification for GIB and lower the health care cost burden. This cross-sectional study was confounded by disease severity, follow up, and causality between H. pylori and GIB.