S695 Factors Affecting Likelihood of Peptic Ulcer Disease as the Etiology of Suspected Upper Gastrointestinal Bleeding

Abstract

Introduction: Patients presenting with suspected upper gastrointestinal bleeding (UGIB) have a broad differential diagnosis. Any process which affects mucosal integrity must be considered when developing a management plan. When treating UGIB a patient’s medical history may provide clues that prioritize consideration of certain diagnoses. This study aims to assess components of the medical history that affect the likelihood of finding a peptic ulcer during endoscopic evaluation. Methods: An IRB-approved retrospective study was performed to evaluate all adult patients who underwent endoscopy for suspected UGIB within five days of presentation, at two affiliated academic hospitals, between 2018 and 2022. Patients were excluded if they had prior endoscopy within 90 days of presentation. Demographic, procedural, and clinical data were obtained via electronic record. Descriptive statistics included t-tests and Chi-square tests. Multivariate analysis was conducted using logistic regression. (Table) Results: A total of 1392 charts were reviewed and 874 patients were included in the analysis. Average age was 64 years, and 525 (60%) were male. The most common symptoms were melena (68%), abdominal pain (32%) and hematemesis (28%). Peptic ulcer disease (PUD) was diagnosed in 397 patients (45%), with 97 patients (11%) classified as high-risk Forrest class 1a/1b/2a/2b lesions where endoscopic intervention is indicated. Another 300 patients (34%) had low risk PUD with Forrest class 2c/3 lesions, and 477 patients (55%) had no PUD. PPI-naïve patients were found to have PUD in 47% of endoscopies, compared to 36% in active PPI users. Multivariate analysis revealed PPI users were 37% less likely to have any ulcer (p< 0.0001). The risk of finding an ulcer increased by 2.0% for every year of age (p< 0.0001). High risk ulcers were more likely to be seen in males (OR 1.73, 95% CI 1.05-2.82), and a history of hepatic disease decreased the likelihood of finding these lesions (OR 0.56, 95% CI 0.37-0.85). Conclusion: Active outpatient PPI use leads to a significant decrease in the likelihood of finding PUD in patients with suspected UGIB. This result underscores the importance of considering other etiologies of bleeding in order to prevent a missed diagnosis and delay in care. The “protective” effect of hepatic disease likely stems from liver-related sources of UGIB present in this population. Even when PUD is suspected in these patients, the data suggest timely empiric treatment with antibiotics and vasoactive agents is warranted. Table 1. - Univariate & Multivariate Analysis of Factors Associated with Upper Gastrointestinal Bleed Presentations Any Ulcer High Risk Ulcer (Forrest 1a/1b/2a/2b) Low Risk Ulcer (Forrest 2c/3) No Ulcer Subjects (N=874) PPI Naïve (N=732) 346 (47%) 84 (11%) 262 (36%) 386 (53%) PPI Use (N=142) 51 (36%) 13 (9%) 38 (27%) 91 (64%) Active PPI Use OR 0.63 0.64 0.65 N/A (95% CI) (0.43-0.93) (0.33-1.23) (0.43-0.99) Age OR 1.02 1.04 1.01 N/A (95% CI) (1.01-1.03) (1.02-1.05) (1.01-1.02) Male Gender OR 1.29 1.73 1.21 N/A (95% CI) (0.97-1.72) (1.05-2.82) (0.88-1.64) Hepatic Disease History OR 0.56 0.86 0.46 N/A (95% CI) (0.37-0.85) (0.46-1.60) (0.28-0.73)