Effect of acute kidney injury on hospital-based outcomes in patients admitted for variceal hemorrhage. Analysis of national inpatient sample database

Introduction
Acute kidney injury (AKI) is known to be a marker of mortality in patients with cirrhosis and variceal hemorrhage.

Aim
To study the effect of AKI on hospital-based outcomes in patients with variceal hemorrhage.

Material and methods
We obtained data from the National Inpatient Sample for the years 2016–2018. Study inclusion criteria comprised adult variceal hemorrhage patients who also had AKI. The primary outcome of interest was in-hospital mortality. Secondary outcomes were length of stay, hospital charge, shock, blood transfusion, and ICU admission. We also determined the independent predictors of mortality in variceal hemorrhage patients using multivariate regression analysis. We used 2 different methods: multivariate logistic regression and propensity matching to adjust for confounders.

Results
The number of people included in this study was 124,430, of whom 32,315 (26%) had AKI. Mortality in variceal hemorrhage patients with AKI was 30.4% in comparison to 4.8% without AKI. The presence of AKI was associated with increased odds of mortality (AOR = 8.28, 95% CI: 7.45–9.20, p < 0.01), ICU admissions (AOR = 4.76, 95% CI: 4.42–5.13, p < 0.01), blood transfusion (AOR = 1.24, 95% CI: 1.15–1.32, p < 0.01), and shock (AOR = 3.41, 95% CI 3.07–3.79, p < 0.01). The patients with AKI also had increased length of stay and hospital charges. Higher Charlson co-morbidity index, African American race, and being admitted to large sized hospital were independently associated with increased mortality.

Conclusions
After analyzing the combined NIS dataset of 2016–2018, we concluded that patients admitted with variceal hemorrhage who has AKI are prone to adverse hospital outcomes.