What question did this study set out to answer?

The study aims to develop an interpretable machine learning model that utilizes preoperative NT-proBNP and creatinine levels to predict AKI risk after cardiac valve surgery.

January 10, 2026The Heart Surgery Forum0 citationsOpen Access

Interpretable Machine Learning Analysis of Preoperative NT-proBNP and Creatinine for Predicting Acute Kidney Injury After Cardiac Valve Surgery

HFHuan null FuNanchang University YTYing null TianNanchang University SSShuigen null SongJinggangshan University

Key Result

A Random Forest model using preoperative NT-proBNP and serum creatinine achieved an AUC of 0.872 for predicting AKI after cardiac valve surgery in 333 patients.

Key Points

The study aims to develop an interpretable machine learning model that utilizes preoperative NT-proBNP and creatinine levels to predict AKI risk after cardiac valve surgery.
Conducted a retrospective cohort study of adults undergoing valve surgery with cardiopulmonary bypass.
Excluded patients with preoperative dialysis or emergency surgery, and those with missing biomarker data.
Developed five machine learning models, including logistic regression and Random Forest, using preoperative variables.
Employed synthetic minority oversampling technique to address class imbalance.
Evaluated model performance using area under the curve, sensitivity, specificity, and decision curve analysis.
Out of 333 patients, 106 experienced AKI (31.8%).
Random Forest model achieved an AUC of 0.872 and accuracy of 0.835.
Identified seven key predictors of AKI risk: age, NT-proBNP, creatinine, CPB duration, aortic cross-clamp duration, hemoglobin, and albumin.
SHAP analysis revealed NT-proBNP, creatinine, and aortic cross-clamp duration as leading risk factors.

Structured PICO

Does an interpretable machine learning model incorporating preoperative NT-proBNP and serum creatinine accurately predict acute kidney injury in adults undergoing isolated cardiac valve surgery?

Population

333 consecutive adults who underwent isolated valve surgery with cardiopulmonary bypass (CPB) at a single center in China. Key exclusions: preoperative dialysis, CKD Stage 4-5 (eGFR <30 mL/min/1.73 m2), emergent/salvage surgery, concomitant non-valve cardiac procedures, re-do sternotomy within 30 days, or missing NT-proBNP/SCr data.

Intervention

Random Forest machine learning model incorporating preoperative NT-proBNP, serum creatinine (SCr), and operative factors to predict acute kidney injury

Comparator

Other machine learning models (Logistic regression, support vector machine [SVM], extreme gradient boosting [XGBoost], and K-nearest neighbors [KNN])

Outcome

Any stage acute kidney injury (AKI) within 7 days after surgery according to Kidney Disease: Improving Global Outcomes (KDIGO) criteriahard clinical

An interpretable Random Forest machine learning model incorporating preoperative NT-proBNP and serum creatinine accurately predicts acute kidney injury following cardiac valve surgery, offering personalized risk evaluation.

Main Result

Absolute Event Rate: 0% vs 0%

Abstract

Background: Acute kidney injury (AKI) is a common and serious complication of cardiac valve surgery, and is associated with high mortality and healthcare costs. Existing prediction models for AKI are often incapable of capturing complex biomarker interactions. This study aimed to build an interpretable machine learning (ML) model that incorpates preoperative N-terminal pro-B-type natriuretic peptide (NT-proBNP) and serum creatinine (SCr) levels to predict of AKI risk in valve surgery patients. Methods: Consecutive adults who underwent isolated valve surgery with cardiopulmonary bypass (CPB) in the first affiliated hospital of Nanchang University from October 2016 to October 2021 were included in this retrospective cohort study. Patients who had preoperative dialysis or were having an emergency surgery were excluded as well as those with missing NT-proBNP/SCr data. The main outcome was any stage AKI within 7 days after surgery (Kidney Disease: Improving Global Outcomes, KDIGO criteria). Utilizing preoperative variables, five ML models Logistic regression, support vector machine (SVM), Random Forest (RF), extreme gradient boosting (XGBoost), and K-nearest neighbors (KNN) were developed after handling class imbalance synthetic minority oversampling technique (SMOTE). Key predictors were identified through feature selection techniques. Evaluation of model performance was done at area under the curve (AUC), sensitivity, specificity and decision curve analysis (DCA). SHapley Additive exPlanations (SHAP) values provided interpretability. Results: Among 333 patients eligible for inclusion, 106 experienced AKI (31.8%). Seven predictors were consistently selected: age, NT-proBNP, SCr, CPB duration, aortic cross-clamp (ACC) duration, hemoglobin and albumin. Overall, the RandomForest model outperformed the other models, with AUC of 0.872, accuracy of 0.835, sensitivity of 0.718, specificity of 0.923 and F1-score of 0.789 in the testing cohort (n = 91). DCA demonstrated excellent calibration and the highest net benefit with this model. SHAP analysis identified NT-proBNP, SCr, and duration of ACC as the three leading risk factors with clear, personalized risk evaluation. Conclusions: This novel, interpretable ML model leverages preoperative NT-proBNP and SCr to accurately predict AKI after cardiac valve surgery. It demonstrated promising predictive performance in internal validation, with the potential to surpass traditional models and have future potential for clinical application. Prospective trials are needed to assess whether model-guided interventions can truly reduce AKI incidence.

Ask AI

Helpful

Bookmark

View Full Paper