New Score Predicts Risk for Death on Heart Transplant List

Core Concepts

A new US candidate risk score outperforms existing systems in ranking heart transplant candidates by medical urgency.

Abstract

Standalone Note here
TOPLINE:

US candidate risk score (US-CRS) surpasses the 6-status system in ranking heart transplant candidates.
METHODOLOGY:

US-CRS developed by adding predictors to the French-CRS.
Study evaluated US adult heart transplant candidates from 2019 to 2022.
Performance assessed by AUC for death without transplant within 6 weeks and overall survival concordance.
TAKEAWAY:

16,905 heart transplant candidates listed, 4.7% died without a transplant.
US-CRS model includes various factors like mechanical circulatory support, bilirubin, and B-type natriuretic peptide.
US-CRS model outperformed French-CRS and 6-status model in predicting death within 6 weeks.
IN PRACTICE:

US-CRS offers better discrimination and may improve ranking by medical urgency.
Uncertainty remains about its performance in disadvantaged groups and the need for exception requests.
SOURCE:

Study led by William F. Parker, MD, MS, PhD, published in JAMA on February 13, 2024.
LIMITATIONS:

Potential recall or misclassification bias due to reported variables.
Underreporting possible due to death verification methods.
Lack of external dataset validation for US-CRS.
DISCLOSURES:

Parker funded by NIH, reported grants from NIH and Greenwell Foundation.
Kittleson reported no conflicts of interest.

Stats

A total of 16,905 heart transplant candidates were listed during the study period.
796 patients (4.7%) died without a transplant.
The AUC for death within 6 weeks was 0.79 for the US-CRS model, 0.72 for the French-CRS model, and 0.68 for the 6-status model.
The overall concordance index was 0.76 for the US-CRS model, 0.69 for the French-CRS model, and 0.67 for the 6-status model.

Quotes

"The US-CRS has better discrimination than the current 6-status ranking system [and] may be useful for ranking candidates by medical urgency."

Key Insights Distilled From

New Score Predicts Risk for Death on Heart Transplant List

by Marilynn Lar... at www.medscape.com 02-14-2024

https://www.medscape.com/viewarticle/new-score-predicts-risk-death-heart-transplant-list-2024a1000359

Deeper Inquiries

How might the US-CRS impact the allocation of heart transplants?

The US-CRS, a new continuous multivariable allocation score, has the potential to significantly impact the allocation of heart transplants by providing a more objective and data-driven approach to ranking candidates based on medical urgency. Unlike the current therapy-based 6-status system, the US-CRS incorporates a wider range of clinical, laboratory, and hemodynamic data, allowing for a more comprehensive assessment of each candidate's risk for death without a transplant. By improving the discrimination and rank ordering ability compared to existing systems, the US-CRS may lead to a fairer and more efficient allocation of heart transplants, ensuring that those in most urgent need receive priority.

What are the potential ethical implications of relying on a scoring system for transplant decisions?

Relying on a scoring system like the US-CRS for transplant decisions raises several potential ethical implications. One concern is the risk of algorithmic bias, where certain groups of patients, such as women and minorities, may be disadvantaged if the scoring system does not adequately account for their specific risk factors. This could lead to disparities in access to heart transplants and perpetuate existing healthcare inequalities. Additionally, there is a question of transparency and accountability in how the scoring system is developed and implemented. It is crucial to ensure that the scoring system is fair, unbiased, and regularly updated based on the latest evidence to prevent any unintended consequences or discrimination in the allocation of heart transplants.

How can advancements in predictive models like the US-CRS influence healthcare disparities?

Advancements in predictive models like the US-CRS have the potential to both exacerbate and mitigate healthcare disparities. On one hand, if not carefully designed and validated, these models could inadvertently perpetuate disparities by failing to accurately capture the risk factors of marginalized populations, leading to unequal access to life-saving treatments like heart transplants. However, if predictive models are developed with a focus on equity and inclusivity, they can help identify and address disparities in healthcare by providing a more objective and standardized approach to decision-making. By incorporating a diverse range of variables and continuously evaluating and adjusting the models, healthcare providers can better understand and mitigate disparities in access to heart transplants and other critical medical interventions.