P3073 - Early Endoscopic Interventions in Left Ventricular Assist Device (LVAD) Patients Hospitalized for Non-Variceal Upper GI Bleeds Can Reduce Hospital Stays and Health Care Costs: A Nationwide Analysis

James Lee, MD¹, Rahul Tripathi, MD¹, Shivani Gupta, MD¹, Jasmine Lee, MD², Lisa Fisher, MD³, Daniel Jamorabo, MD⁴
1Stony Brook Medicine, Stony Brook, NY; 2Jefferson Health, Philadelphia, PA; 3Stony Brook University Hospital, Northport, NY; 4Northwell Health, Forest Hills, NY

Introduction: Left Ventricular Assist Device (LVAD) recipients are at increased risk of non-variceal upper gastrointestinal bleeding (NVUGIB). Esophagogastroduodenoscopy (EGD) remains the cornerstone of diagnostic and therapeutic management, but the optimal timing in this high-risk population remains unclear. Determining procedure timing may help inform clinical decision-making and improve resource allocation.

Methods: We conducted a retrospective cohort study using the National Inpatient Sample (NIS) from 2012 to 2022. Adult (≥18 years), non-elective hospitalizations with NVUGIB with LVAD were identified using ICD-9/10 codes. Early EGD was defined as occurring within 1 day (<24 hours) of admission. Survey-weighted multivariable linear and logistic regression models were used to evaluate associations between early EGD and inpatient outcomes, including in-hospital mortality, length of stay (LOS), and total hospital charges.

Results: There were 7,010 hospitalizations for NVUGIB in patients with LVAD, of whom 5,735 (81.8%) underwent EGD, and 2,610/5,735 (46.9%) received it early. The mean age was 64.0 years in the early EGD group compared to 64.5 years in the delayed group. Unadjusted outcomes showed that inpatient mortality was low and comparable between groups (0.57% early vs. 0.51% delayed). However, early EGD was associated with a significantly shorter length of stay (6.9 vs. 9.8 days, p < 0.001) and lower total hospital charges ($69,000 vs. $99,000, p = 0.014). In adjusted analyses, early EGD was not significantly associated with inpatient mortality (adjusted odds ratio [aOR] 1.65; 95% CI 0.29–9.52; p = 0.575), but it was associated with a significantly shorter length of stay (β = −2.90 days; 95% CI −4.31 to −1.80; p < 0.001) and lower total hospital charges (β = −$22,541; 95% CI −$39,369 to −$5,714; p = 0.009).

Discussion: Among LVAD patients hospitalized with NVUGIB, early EGD was associated with shorter hospital stays and reduced healthcare costs, suggesting a potential benefit in resource utilization. While no significant mortality difference was observed, this may reflect the low overall event rate in our study cohort. Future studies evaluating additional clinical endpoints such as transfusion requirements, ICU admissions, rebleeding rates, and 30-day readmissions are warranted to further define the benefits of early intervention on patient-centered and safety outcomes.

Disclosures:
James Lee indicated no relevant financial relationships.
Rahul Tripathi indicated no relevant financial relationships.
Shivani Gupta indicated no relevant financial relationships.
Jasmine Lee indicated no relevant financial relationships.
Lisa Fisher indicated no relevant financial relationships.
Daniel Jamorabo indicated no relevant financial relationships.

James Lee, MD¹, Rahul Tripathi, MD¹, Shivani Gupta, MD¹, Jasmine Lee, MD², Lisa Fisher, MD³, Daniel Jamorabo, MD⁴. P3073 - Early Endoscopic Interventions in Left Ventricular Assist Device (LVAD) Patients Hospitalized for Non-Variceal Upper GI Bleeds Can Reduce Hospital Stays and Health Care Costs: A Nationwide Analysis, ACG 2025 Annual Scientific Meeting Abstracts. Phoenix, AZ: American College of Gastroenterology.