Heart failure & cardiomyopathies
Prediction of Donor Heart Acceptance for Transplant and Its Clinical Implications: Results From The Donor Heart Study
Authors' conclusion
Predictors of donor heart acceptance for transplantation have changed significantly over the last 2 decades, likely reflecting evolving evidence regarding their impact on posttransplant outcomes.
Study context
- Population
- Potential heart donors in the US
- Participants
- 73,948 vs 4,130
- Exposure
- Prediction algorithms for donor heart acceptance
- Nationwiden = 73,948
- DHSn = 4,130
- Outcome
- Donor heart acceptance for transplant
Methodological detailsnationwide registry, machine learning prediction model, 50% training / 50% test split
- Nationwide registry
- Machine learning prediction model
- 50% training / 50% test split
Main finding
As reported in the abstract
Predictors of donor heart acceptance
Older age increasingly predictive of discard over time (time interaction)
0.0001
Drug use, infection and mild cardiac diagnostic abnormalities less influential over time (time interaction)
<0.05 for all
Observational data: associations, not effects.
Further findings
| Outcome | Values |
|---|---|
| Prediction performance of random forest model (test subset) | Random forest AUC 0.908 · Accuracy 0.831 |
Prediction performance of random forest model (test subset)
- Random forest outperformed other prediction algorithms; basis of web-based prediction tool
Original publication
Prediction of Donor Heart Acceptance for Transplant and Its Clinical Implications: Results From The Donor Heart Study. Circ Heart Fail. Published 23 September 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →