Effect of the UNOS policy change on rates of rejection, infection, and hospital readmission following heart transplantation

Mortality prediction of heart transplantation using machine learning models: a systematic review and meta-analysis

Introduction

Machine learning (ML) models have been increasingly applied to predict post-heart transplantation (HT) mortality, aiming to improve decision-making and optimize outcomes. This systematic review and meta-analysis evaluates the performance of ML algorithms in predicting mortality and explores factors contributing to model accuracy.

Method

A systematic search of PubMed, Scopus, Web of Science, and Embase identified relevant studies, with 17 studies included in the review and 12 in the meta-analysis. The algorithms assessed included random forests, CatBoost, neural networks, and others. Model performance was evaluated using pooled area under the curve (AUC) values, with subgroup analyses for algorithm type, validation methods, and prediction timeframes. The risk of bias was assessed using the QUADAS-2 tool.

Results

The pooled AUC of all ML algorithms was 0.65 (95% CI: 0.64, 0.67), with no significant difference between machine learning and deep learning models (p = 0.67). Among the algorithms, CatBoost demonstrated the highest accuracy (AUC 0.80, 95% CI: 0.74, 0.86), while K-nearest neighbor had the lowest accuracy (AUC 0.53, 95% CI: 0.50, 0.55). A meta-regression indicated improved model performance with longer post-transplant periods (p = 0.008). When pooling only the best-performing models, the AUC improved to 0.73 (95% CI: 0.68, 0.78). The risk of bias was high in eight studies, with the flow and timing domains most commonly contributing to bias.

Conclusion

ML models demonstrate moderate accuracy in predicting post-HT mortality, with CatBoost achieving the best performance. While ML shows potential for improving predictive precision, significant heterogeneity and biases highlight the need for standardized methods and further external validations to enhance clinical applicability.

Systematic review registration

https://www.crd.york.ac.uk/PROSPERO/view/CRD42024509630, CRD42024509630.

Use of mechanical circulatory support and survival for heart and heart-kidney transplant recipients in the new allocation system

Background

The 2018 United Network for Organ Sharing organ allocation change aimed to distribute donor hearts to the sickest patients on the waitlist. Whether this change differentially affected outcomes in heart-only vs heart-kidney transplant recipients is unknown.

Methods

This study used the Scientific Registry of Transplant Recipients to compare outcomes, including survival, of heart-only and heart-kidney transplant recipients from 2015 to 2021, from the old vs new allocation system, including use of mechanical circulatory support (MCS), prior to transplant.

Results

During the study period, 16,696 patients underwent heart transplant alone (9,320 in the old and 7,376 in the new system) and 1,156 patients underwent heart-kidney transplant (529 in the old and 627 in the new system). For both heart and heart-kidney transplant populations, there was a 3- to 5-fold increase in the use of temporary MCS. Heart-only recipients had worse survival when temporary MCS was used in the old allocation system. Heart-only recipients with durable MCS had worse survival both in the old and the new allocation system. There was no difference in survival in heart-kidney recipients in the old vs new allocation system, regardless of MCS use.

Conclusions

The new heart allocation system was associated with increased use of temporary MCS in both heart and heart-kidney recipients. However, this change only differentially affected survival in heart-only recipients with improved survival if on temporary MCS, but worse survival if on durable MCS. Unlike prior studies, heart-kidney recipients did not have different outcomes after the heart allocation change, which may reflect outcomes in more current times.

Rejection Requiring Treatment within the First Year following Heart Transplantation: The UNOS Insight

(1) Background: Heart failure is an extremely impactful health issue from both a social and quality-of-life point of view and the rate of patients with this condition is destined to rise in the next few years. Transplantation remains the mainstay of treatment for end-stage heart failure, but a shortage of organs represents a significant problem that prolongs time spent on the waiting list. In view of this, the selection of donor and recipient must be extremely meticulous, considering all factors that could predispose to organ failure. One of the main considerations regarding heart transplants is the risk of graft rejection and the need for immunosuppression therapy to mitigate that risk. In this study, we aimed to assess the characteristics of patients who need immunosuppression treatment for rejection within one year of heart transplantation and its impact on mid-term and long-term mortality. (2) Methods: The United Network for Organ Sharing (UNOS) Registry was queried to identify patients who solely underwent a heart transplant in the US between 2000 and 2021. Patients were divided into two groups according to the need for anti-rejection treatment within one year of heart transplantation. Patients' characteristics in the two groups were assessed, and 1 year and 10 year mortality rates were compared. (3) Results: A total of 43,763 patients underwent isolated heart transplantation in the study period, and 9946 (22.7%) needed anti-rejection treatment in the first year. Patients who required treatment for rejection within one year after transplant were more frequently younger (49 ± 14 vs. 52 ± 14 years, p < 0.001), women (31% vs. 23%, p < 0.001), and had a higher CPRA value (14 ± 26 vs. 11 ± 23, p < 0.001). Also, the rate of prior cardiac surgery was more than double in this group (27% vs. 12%, p < 0.001), while prior LVAD (12% vs. 11%, p < 0.001) and IABP (10% vs. 9%, p < 0.01) were more frequent in patients who did not receive anti-rejection treatment in the first year. Finally, pre-transplantation creatinine was significantly higher in patients who did not need treatment for rejection in the first year (1.4 vs. 1.3, p < 0.01). Most patients who did not require anti-rejection treatment underwent heart transplantation during the new allocation era, while less than half of the patients who required treatment underwent transplantation after the new allocation policy implementation (65% vs. 49%, p < 0.001). Patients who needed rejection treatment in the first year had a higher risk of unadjusted 1 year (HR: 2.25; 95% CI: 1.88-2.70; p < 0.001), 5 year (HR: 1.69; 95% CI: 1.60-1.79; p < 0.001), and 10 year (HR: 1.47; 95% CI: 1.41-1.54, p < 0.001) mortality, and this was confirmed at the adjusted analysis at all three time-points. (4) Conclusions: Medical treatment of acute rejection was associated with significantly increased 1 year mortality compared to patients who did not require anti-rejection therapy. The higher risk of mortality was confirmed at a 10 year follow-up. Further studies and newer follow-up data are required to investigate the role of anti-rejection therapy in the heart transplant population.