Frailty Predicts Mortality and Procedural Performance in Patients With Non-Variceal Upper Gastrointestinal Bleeding

Abstract

Introduction

Nonvariceal upper gastrointestinal bleeding (NVUGIB) is a common cause of hospitalization in the United States, with approximately 400 000 admissions annually and a 5%–10% mortality rate. This study aimed to evaluate frailty's impact on NVUGIB outcomes.

Methods

We utilized the 2019 National Readmission Database (NRD) to identify adult patients (≥ 18 years) admitted with a principal diagnosis of NVUGIB using ICD-10-CM codes. NVUGIB hospitalizations were stratified by frailty using the hospital frailty risk score (HFRS) of 5 or more as the cut-off for frailty. Multivariate regression analyses were conducted to analyze the outcomes. STATA 14.2 was used for statistical testing.

Results

Among 218 647 NVUGIB admissions, 99 892 (45.69%) were frail. Frail patients were older, more often female, and had higher comorbidity burdens. They showed significantly greater in-hospital mortality (adjusted odds ratio [aOR] 5.64, 95% CI 4.94–6.44; p < 0.001), acute kidney injury (5.85), respiratory failure (6.93), septic shock (40.94), hemorrhagic shock (2.64), vasopressor use (4.36), mechanical ventilation (6.04), and ICU admission (5.41). Although frail patients had higher odds of esophagogastroduodenoscopy (EGD) with intervention (1.04; p < 0.001), they were less likely to receive EGD within 24 h (0.75; p < 0.001). They also had higher odds of rebleeding (1.18; p < 0.001) and radioembolization (2.69; p < 0.001). Length of stay increased by 2.30 days, total charges rose by $28 518, discharge to rehabilitation was more frequent (3.12; p < 0.01), and 30-day readmission was higher (15.24% vs. 11.43%, HR 1.16; p < 0.001).

Conclusion

Frailty independently predicts worse clinical outcomes and increased resource use in NVUGIB. Recognizing frailty may improve risk stratification and guide more tailored management strategies for this high-risk population.