S2767 Impact of Hospital Size on Outcomes and Resource Utilization Following Transjugular Intrahepatic Portosystemic Shunt (TIPS): A 7-Year Retrospective Analysis in the United States

Introduction: Hospital size may influence outcomes and resource use in complex procedures like transjugular intrahepatic portosystemic shunt (TIPS), yet its role remains unclear. Methods: Adult TIPS procedures (2016–2022) were identified in the National Inpatient Sample. Hospitals were stratified by size (small, medium, large) per AHRQ definitions. Patient and hospital characteristics were compared. Multivariable regression assessed associations between hospital size and in-hospital mortality, complications, length of stay (LOS), and total hospital charges (TOTCHG), adjusting for demographics, comorbidities, and hospital factors. Results: Among 163,135 TIPS patients, 21.3% were treated at small, 32.3% at medium, and 46.4% at large hospitals. Compared to medium and large hospitals, small hospital TAVR admissions were more likely to have female sex, White race, and Medicare/Medicaid payer. Crude mortality increased with hospital size (small: 2.6%, medium: 3.2%, large: 3.5%; P < 0.001). On adjusted analysis, large hospitals had significantly higher odds of mortality (aOR 1.13; 95% CI: 1.09–1.17; P < 0.001). Large hospital size was also associated with increased odds of bleeding (aOR 1.40), AKI (aOR 1.19), and DVT (aOR 1.46) (all P < 0.001). LOS increased with hospital size (small: 5.0 ± 0.06 days; large: 6.26 ± 0.05; P < 0.001), as did TOTCHG (small: $62,997; large: $95,587; P < 0.001). Conclusion: These results suggest that TIPS procedures performed at large hospitals are associated with higher in-hospital mortality, complication rates, and healthcare costs. This may reflect greater case complexity at tertiary centers and underscores the importance of improved risk adjustment and resource allocation in TIPS care. Importantly, the findings also highlight that small hospitals are performing well in delivering safe and effective TIPS outcomes. Further studies are warranted to understand the drivers of these disparities and inform system-level improvements.