Related Experiment Video
Updated: Jul 4, 2026
A Large Animal Model for Acute Kidney Injury by Temporary Bilateral Renal Artery Occlusion
Published on: February 2, 2021
Machine learning models for predicting postoperative acute kidney injury in pediatric cardiac surgery: a systematic
Noel Matthew Imaniku Sihombing1, Hashfi Fauzan Raz2, Sandra Rosa Uli Siahaan1
1Faculty of Medicine, Universitas Sumatera Utara, Medan, Indonesia.
Insights
Machine learning (ML) shows promise for predicting acute kidney injury (AKI) in pediatric cardiac surgery. However, limited external validation necessitates caution before widespread clinical adoption.
Area of Science:
- Pediatric critical care medicine
- Medical informatics
- Nephrology
Background:
- Acute kidney injury (AKI) affects up to 42% of pediatric cardiac surgery patients, increasing morbidity and mortality.
- Machine learning (ML) offers a potential solution for early risk stratification of pediatric cardiac surgery-associated AKI (CSA-AKI).
Purpose of the Study:
- To systematically review and meta-analyze the diagnostic accuracy of ML models for predicting pediatric CSA-AKI.
Main Methods:
- A systematic search of major databases (PubMed, ScienceDirect, Springer, DOAJ) was conducted.
- Study quality was assessed using QUADAS-2 and PROBAST+AI.
- A bivariate random-effects diagnostic meta-analysis was performed to estimate pooled performance metrics.
Main Results:
- The meta-analysis of seven studies showed a pooled SROC AUC of 0.91 for ML models predicting pediatric CSA-AKI.
- Internally validated models demonstrated high performance (AUC 0.93), while externally validated models showed lower, more clinically relevant performance (Sensitivity 0.70, Specificity 0.80).
- Substantial heterogeneity (I² = 81.48%) was observed across studies.
Conclusions:
- ML models demonstrate promising diagnostic accuracy for pediatric CSA-AKI prediction.
- Limited external validation and significant heterogeneity highlight the need for cautious interpretation.
- Further multicenter validation is recommended before clinical implementation of ML models for pediatric CSA-AKI.
Background:
Acute kidney injury (AKI) occurs in up to 42% of pediatric cardiac surgeries and is associated with prolonged intensive care, increased morbidity and in-hospital mortality. Machine learning (ML) has emerged as a promising approach for early AKI risk stratification by modeling complex, high-dimensional clinical data.
Objectives:
To systematically review and meta-analyze the diagnostic accuracy of ML models for predicting pediatric cardiac surgery associated AKI (CSA-AKI).
Methods:
We performed a systematic search of PubMed, ScienceDirect, Springer, and DOAJ. The review protocol was prospectively registered in PROSPERO (CRD420251145645). Study quality was assessed using QUADAS-2 tool and PROBAST + AI. A bivariate random-effects diagnostic meta-analysis was performed using Stata 17.0 to estimate the pooled area under the summary receiver operating characteristic curve (SROC AUC), sensitivity, specificity, likelihood ratios, and diagnostic odds ratio (DOR).
Results:
A meta-analysis of seven studies yielded a pooled SROC AUC of 0.91 (95% CI 0.88-0.93), driven predominantly by internally validated models (AUC 0.93, Sensitivity of 0.84, Specificity of 0.95). Externally validated models showed substantially lower performance (Sensitivity 0.70, Specificity 0.80), representing the more clinically relevant benchmark. A sensitivity analysis using median-performing models confirmed directional consistency (AUC 0.85, Sensitivity 0.75, Specificity 0.91). Substantial heterogeneity was observed (I 2 = 81.48%).
Conclusion:
ML models show promising accuracy for predicting pediatric CSA-AKI. Substantial heterogeneity and limited external validation warrant cautious interpretation and further multicenter validation before clinical use.
Systematic Review Registration:
https://www.crd.york.ac.uk/PROSPERO/view/CRD420251145645.
Related Concept Videos
01:22Acute Kidney Injury I: Introduction
01:30Acute Kidney Injury IV: Diagnostic Studies and Prevention
01:20Acute Kidney Injury V: Interprofessional Care
01:29Acute Kidney Injury II: Pathophysiology
01:22Acute Kidney Injury VI: Nursing Management
01:26Kidney Transplant II: Surgical Procedure