S669 The Effect of Outpatient Proton Pump Inhibitor Use on Peptic Ulcer Disease in Patients Presenting With Suspected Upper Gastrointestinal Bleeding

Abstract

Introduction: Proton pump inhibitors (PPIs) are prescribed to both prevent and treat peptic ulcer disease (PUD). There is a paucity of data regarding how outpatient PPI use affects the likelihood of finding PUD in patients with concern for an upper gastrointestinal bleed (UGIB). The aim of this study was to assess how active PPI use impacts PUD prevalence and severity in this population. Methods: An IRB-approved retrospective study was performed on all adult patients who underwent upper endoscopy between 1/2018 and 2/2021 at 2 affiliated hospitals within 5 days of admission for a suspected UGIB. Patients were excluded if endoscopy was performed up to 90 days prior to admission. Active PPI use was defined as once or twice daily dosing for at least 30 days. Demographic, procedural and clinical data were obtained via electronic records. Descriptive statistics included ANOVA and Chi-square tests. Multivariate analysis was conducted using polytomous logistic regression. Results: A total of 1122 unique charts were reviewed, with 696 patients included in the analysis (Table 1). Average age was 64 years; 405 (58%) were male. The most common symptoms were melena (67%), hematemesis (26%) and hematochezia (15%). Active PPI use was noted in 133 patients (19%). PUD was diagnosed in 350 patients (50.2%), with 74 patients (10.6%) classified as high risk Forrest class 1a/1b/2a/2b PUD where endoscopic intervention is indicated. Another 276 patients (39.7%) had low risk PUD with Forrest class 2c/3 lesions, and 346 patients (49.7%) had no PUD. A slim majority (51.2%) of PPI-naïve patients were found to have PUD on endoscopy, compared to 45.9% of active PPI users. PPI-naïve patients also were more likely to have high risk PUD than active PPI users (11.5% vs 8.3%), though neither reached statistical significance. A smaller percentage (20.4%) of active PPI users with PUD required blood transfusion compared to PPI-naïve patients (79.7%). Conclusion: It is surprising that there is no statistical difference in the percentage of patients with suspected UGIB found to have PUD based on active PPI use prior to admission. However, while PPIs may not decrease the likelihood of finding PUD, they do seem to decrease the severity of blood loss. The transfusion rate for active PPI users found to have PUD was a quarter of that found in PPI-naïve patients. Therefore, this study suggests PPI use at home may not prevent a PUD-related UGIB, but it may lessen the undesired effects of such an event.Table 1.: Mortality in MDS patients with GI bleed (OR).