Abstract: SA-PO0293
External Validation of the Malhotra AKI Risk Score for Predicting the Development of AKI in Critically Ill Patients at St. Luke's Medical Center - Global City
Session Information
- AKI: Epidemiology and Risk Factors
October 24, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 101 AKI: Epidemiology, Risk Factors, and Prevention
Author
- Murakami, Kenneth George Conilas, Saint Luke's Medical Center, Taguig, NCR, Philippines
Background
Acute kidney injury (AKI) is a frequent and serious complication in critically ill patients, associated with increased mortality and healthcare costs. The Malhotra AKI risk score, a ten-variable bedside tool developed and validated in US-based ICU cohorts (AUROC 0.81), has not been externally validated in a Southeast Asian or Filipino population.
Methods
This retrospective, single-center, observational cohort study included adult patients (≥19 years) admitted to the Medical ICU of St. Luke’s Medical Center – Global City from January to December 2024. Patients with end-stage renal disease, prior kidney transplantation, AKI at admission, ICU stay <24 hours, or incomplete records were excluded. AKI was defined using KDIGO serum creatinine criteria. Discrimination was assessed by AUROC, calibration by the Hosmer–Lemeshow test, and the optimal cutoff by Youden’s J statistic.
Results
Of 434 patients included, 90 (20.7%) developed AKI. The Malhotra score demonstrated good discrimination (AUROC 0.753; 95% CI 0.702–0.803) and adequate calibration (Hosmer–Lemeshow p = 0.122). The optimal cutoff of ≥5, concordant with the original study, yielded a sensitivity of 97.7% (CI 91.3- 99.8%), specificity of 44.8% (CI 39.5-50%), positive predictive value (PPV) of 30.4% (CI 25.7-36.6%), and negative predictive value (NPV) of 98.7% (CI 95.0-99.9%). The score also showed poor discrimination for in-ICU mortality (AUROC 0.659) and fair discrimination for RRT need (AUROC 0.712). ICU length of stay did not differ significantly between risk groups (p = 0.682).
Conclusion
The Malhotra AKI risk score retains acceptable discriminative ability and adequate calibration in a Filipino ICU population, with an excellent NPV supporting its utility as a bedside rule-out screening tool. However, the low PPV warrants supplementation with clinical judgment when interpreting high-risk classifications.