Impact of left ventricular assist devices on 30-day readmission and outcomes in non-variceal upper gastrointestinal bleeding: a nationwide analysis

Introduction
Bleeding, especially non-variceal upper gastrointestinal bleeding (NVUGIB), remains the most common cause of readmission in left ventricular assist device (LVAD) patients. Any readmission after NVUGIB carries a worse prognosis.

Aim
To compare readmission outcomes in NVUGIB patients with and without LVAD.

Material and methods
We identified adult NVUGIB patients using the National Readmission Database 2018 employing International Classification of Diseases, Tenth Revision (ICD-10) codes. The patients were grouped based on LVAD history. Proportions were compared using the Fisher exact test, and multivariate Cox proportional regression analysis was used to compute adjusted p-values. We used Stata version 14.2 to perform analyses considering 2-sided p < 0.05 as statistically significant.

Results
The analysis included 322,342 NVUGIB patients, 1403 had a history of LVAD (mean age 64.25 years). The 30-day all-cause readmission rate in NVUGIB with LVAD was higher (24.31% vs. 13.92%, p < 0.001). Gastrointestinal bleeding as a readmission cause was more prevalent in the LVAD group. In patients with LVAD, NVUGIB readmissions required more complex endoscopic procedures, either requiring intervention during endoscopy or enteroscopy. There was no difference in mortality in NVUGIB readmissions (1.51% vs. 4.49%, p = 0.36); however, the length and cost of stay were higher in the LVAD group. Additionally, we identified novel independent predictors of readmission from NVUGIB in patients with LVADs.

Conclusions
Readmissions in NVUGIB patients after LVAD require complex haemostatic intervention and are associated with greater resource utilization. To reduce readmissions and associated healthcare costs, it is essential to identify high-risk patients.