Gang Wang, Yi Zhu, Xuefen Yan, Linjuan Xu, Jiaheng Wang, Jinle Liu, Li Yu
EASIX holds promise in predicting AKI in hospitalized MM patients, supporting its potential use in risk stratification and clinical decision-making.
BACKGROUND: To investigate the predictive value of Endothelial Activation and Stress Index (EASIX) for acute kidney injury (AKI) in hospitalized patients with multiple myeloma (MM).
METHODS: This retrospective study included 352 MM patients admitted to our hospital between January 2020 and December 2024. Patients were divided into a training set (n = 248) and a validation set (n = 104) using a time-based split. Patients were further classified into AKI (training n = 52, validation n = 22) and non-AKI groups (training n = 196, validation n = 82) according to occurrence of AKI during hospitalization. Factors associated with the development of AKI in MM patients were identified using logistic regression analysis. The predictive performance of EASIX for AKI was assessed by receiver operating characteristic (ROC) curve analysis.
RESULTS: In the training set, binary logistic regression analysis revealed that EASIX, International Staging System (ISS) stage, and bortezomib-lenalidomide-dexamethasone (VRD) chemotherapy regimen were independent factors influencing the development of AKI in MM patients during hospitalization (all P < 0.05). The area under the ROC curve (AUC) for EASIX predicting AKI was 0.897 (95% CI: 0.852-0.942). All 248 patients were followed up without loss to follow-up. The mortality rate was significantly different between the group with EASIX ≤ 3.16 (18.54%) and the group with EASIX > 3.16 (34.29%) (P < 0.05). Temporal validation in the time-split validation set confirmed higher EASIX in the AKI group (P < 0.001) and robust predictive performance (AUC = 0.893, 95% CI: 0.830-0.956), with consistent mortality differences between EASIX subgroups (P = 0.016).
CONCLUSION: EASIX holds promise in predicting AKI in hospitalized MM patients, supporting its potential use in risk stratification and clinical decision-making.