Supplementary Material for: Comparative Performance of Novel and Standard Urinary Kidney Biomarkers for Prediction of Persistent Severe AKI and Long-Term Major Adverse Kidney Events

Background and hypothesis: Acute kidney injury (AKI) is common complication in the critically ill. Standard functional biomarkers are limited at predicting persistent severe AKI (PS-AKI) and long-term outcomes. This study evaluated the diagnostic and prognostic performance of a panel of urinary biomarkers (novel and standard) for predicting PS-AKI and major adverse kidney events (MAKE). Methods: This was an exploratory post-hoc analysis of the prospective Dublin Acute Biomarker Group Evaluation (DAMAGE) multicentred prospective observational cohort study. ICU AKI (KDIGO Stage 1–3) patients were included. Sixteen urinary biomarkers were measured on the day of AKI diagnosis. The primary endpoint was PS-AKI (Stage 2/3 AKI ≥48 hours). Discrimination was assessed using AUC and logistic regression models, and reclassification metrics (IDI, cfNRI). Secondary endpoints included MAKE90 and MAKE365. Tertile trends for CCL14 also evaluated. Results: Among 186 patients with AKI, 80 (43.0%) developed PS-AKI. Albumin (uAlb; AUC 0.82; 95% CI: 0.76–0.88), albumin/creatinine ratio (uAlb/Cr; AUC 0.79; 95% CI: 0.72–0.85), urine output (AUC 0.81; 95% CI: 0.74–0.87), and serum creatinine (AUC 0.77; 95% CI: 0.70–0.84) demonstrated the highest discrimination. In logistic regression analysis adjusted for a clinical model, IL-18 showed the strongest association with PS-AKI (aOR 3.09; 95% CI: 1.92–5.30), while uAlb, uAlb/Cr, cystatin C, CCL14, and MCP-1 were also significantly associated. CCL14 and uAlb tertiles showed a significant stepwise increase in PS-AKI. uAlb was the strongest discriminator for MAKE90 (AUC 0.70; 95% CI: 0.57–0.83). PiGST was negatively associated with MAKE365 (aOR 0.44; 95% CI 0.21–0.81). Conclusion: Urine output, uAlb, and uAlb/Cr outperformed several novel biomarkers and demonstrated strong discrimination for PS-AKI. CCL14 showed moderate discrimination and was associated with early adverse outcomes. These findings support integrating standard and novel biomarkers to personalise AKI management.