# Comparative analysis of prognostic assessment in hospitalized heart failure patients: a comprehensive evaluation of KDIGO and WRF classifications

## TL;DR

This study compares two methods for identifying kidney dysfunction in heart failure patients and finds one method better predicts hospital mortality.

## Contribution

The study evaluates KDIGO and WRF criteria for acute kidney injury in heart failure patients and identifies KDIGO as more effective for predicting mortality.

## Key findings

- KDIGO criteria showed higher discriminatory power for predicting in-hospital mortality compared to WRF criteria.
- Modified KDIGO was more accurate than WRF for identifying mortality and early AKI in hospitalized patients.
- AKI severity was associated with increased mortality and adverse kidney effects across all definitions.

## Abstract

The definition of acute kidney dysfunction in patients with acute decompensated heart failure (ADHF) remains unclear. This study aimed to compare two sets of criteria for acute kidney injury (AKI), namely, the kidney disease: improving global outcomes (KDIGO) and worsening renal function (WRF) classification, in hospitalized patients with ADHF.

We utilized a multi-institutional database with 17,684 cases of hospitalizations for HF. AKI was defined using KDIGO, WRF-serum creatinine (Scr), and WRF-estimated glomerular filtration rate (eGFR) criteria. The study compared the performance of these criteria in predicting in-hospital mortality and employed logistic regression to assess associations with mortality, HF hospitalization, and major adverse kidney effects (MAKE). A sensitivity analysis was conducted to compare the modified KDIGO (mKDIGO) with the traditional AKI criteria.

The incidences of ADHF according to the KDIGO, WRF-Scr, and WRF-eGFR criteria were 28.6%, 29.9%, and 29.9%, respectively. KDIGO exhibited higher discriminatory power compared with WRF-Scr and WRF-eGFR for in-hospital mortality[area under the curve (AUC):73.6% vs. 71.6% vs. 71.2%]. On all definitions, ADHF was predicted to have an increase in mortality and MAKE, with mortality increasing stepwise with AKI severity. A sensitivity analysis revealed mKDIGO to be more accurate than WRF criteria for identifying in-hospital mortality and recognizing AKI early.

In hospitalized patients with ADHF, KDIGO is a more effective predictive tool for in-hospital mortality compared with WRF classification. Integrating a newer severity-staging classification into WRF criteria may enhance their predictive association with poor prognosis and enable early intervention.