<p>Pediatric acute kidney injury (AKI) often presents insidiously and progresses rapidly. Traditional diagnostic criteria based on serum creatinine and urine output are markedly delayed and insufficient to capture injury patterns across different etiologies. This paper aims to summarize recent advances in pediatric AKI biomarker research since the release of the ADQI 23 (2020) consensus. Focusing on three major clinical scenarios—cardiac surgery, sepsis, and nephrotoxic drugs—it reviews early biomarker evidence and explores their potential applications in risk stratification. At the mechanistic level, this paper outlines key pathological pathways in pediatric AKI progression: oxygenation-perfusion imbalance after cardiac surgery, endothelium-immune dysregulation driven by sepsis, and tubular-mitochondrial injury associated with nephrotoxic exposure. In CS-AKI, uNGAL shows the earliest elevation within hours after cardiopulmonary bypass, followed by sequential changes in IL-18, L-FABP, and KIM-1. [TIMP-2] × [IGFBP7] and exosomal miRNA aid in identifying high-risk or severe AKI. In SA-AKI, suPAR and glycocalyx/endothelial injury markers (e.g., syndecan-1, Angpt-2/sTM/Tie-2), combined with urinary DKK3 and complement Ba, can be used for early risk stratification and predicting poor outcomes. In NT-AKI, uNGAL has high negative predictive value for excluding severe AKI, while uKIM-1, uCysC, uOPN, and multi-biomarker combinations can indicate subclinical tubular injury earlier after drug exposure. Overall, single biomarkers struggle to cover AKI heterogeneity. Future efforts should integrate functional dynamic assessments (e.g., FST, RRI), scenario-based multi-biomarker combinations, and AI dynamic models to propose evidence-based, scenario-stratified identification pathways. These will serve as structured references for prospective studies and clinical workflow optimization.</p> Graphical abstract <p></p>