Key result
Machine learning model integrating hemodynamics and cardiac energetics predicts 1-year advanced HF events with ~0.81 AUC.
Why the study?
Clinicians face challenges predicting which end-stage heart failure patients will require left ventricular assist devices or transplantation or are at risk of death. Traditional risk models often fail to accurately stratify patients due to complex interactions among guideline-directed medical therapy, individual responses, and haemodynamic or shock phenotypes.
Does a machine learning model integrating invasive haemodynamic and cardiac energetics data accurately predict the 1-year composite outcome of death, left ventricular assist device implantation, or heart transplantation in patients with heart failure?
Cohort (n=2,904)
Yes
Does a machine learning model integrating invasive haemodynamic and cardiac energetics data accurately predict the 1-year composite outcome of death, left ventricular assist device implantation, or heart transplantation in patients with heart failure?
Integrating invasive haemodynamics and cardiac energetics into an XGBoost machine learning model provides high negative predictive value (92%) and good discrimination (AUC 0.81) for 1-year advanced heart failure therapies or death, offering a quantitative tool to identify high-risk patients whose risk may be underestimated by standard cardiac output assessments alone.
Answers are grounded in this paper and cite their sources.
May aid 1-year prognostication in heart failure; extends traditional models but requires prospective validation before adoption.
Aims: In the evolving landscape of heart failure treatment, clinicians face challenges in predicting which end-stage heart failure patients will require heart replacement therapies, such as left ventricular assist devices or transplantation, or are at risk of death. Traditional risk models often fail to accurately stratify patients due to complex interactions of guideline-directed medical therapy, individual responses, and haemodynamic or shock phenotypes. Methods and results: This study introduces a novel approach using machine learning to predict patient outcomes by analysing a comprehensive set of haemodynamic variables, including metrics tied to cardiac energetics-an emerging field focused on heart energy dynamics during failure. A dataset of 2904 patients across the spectrum of heart failure was used to train and test the models, which underwent rigorous model selection, cross-validation, and hyperparameter tuning on 70% of the data, using forward feature selection. Metrics for the remaining 30% used for testing resulted in an area under the curve of 0.81, a Matthews correlation coefficient of 0.35, and an accuracy of 0.74 to predict the composite endpoint of survival free from death, left ventricular assist devices, or transplant at 1 year. We also demonstrate consistency across the different patient databases used for training, validation, and testing, supported by an F1 score of 0.65 overall and 0.82 for patients who received medical management. The model shows a promising negative predictive value of 0.92, but external validation with a prospective database is needed before clinical application. Conclusion: This approach is novel in integrating haemodynamics and cardiac energetics with machine learning towards more individualized and effective treatment for heart failure patients.
No takes yet. Share an insight, caveat, or question.
Blanco et al. (2026) conducted a cohort in Heart failure (n=2,904). XGBoost machine learning model integrating haemodynamic and cardiac energetics variables (LVMPS, SPAP, DAP, LVCPO, RVMPS, RVCPO) vs. Standard and advanced haemodynamic inputs using logistic regression models was evaluated on Composite endpoint of left ventricular assist device implantation, heart transplantation, or death at 1 year (AUC 0.81, MCC 0.35, Accuracy 0.74, NPV 0.92). An XGBoost machine learning model integrating haemodynamics and cardiac energetics predicted 1-year death, LVAD, or heart transplant with an AUC of 0.81, accuracy of 0.74, and NPV of 0.92.
Synapse has enriched 10 related papers. Consider them for comparative context: