Risk factors for 1-year allograft loss in pediatric heart transplant patients using machine learning: An analysis of the pediatric heart transplant society database

BACKGROUND

Pediatric heart transplant patients are at greatest risk of allograft loss in the first year. We assessed whether machine learning could improve 1-year risk assessment using the Pediatric Heart Transplant Society database.

METHODS

Patients transplanted from 2010 to 2019 were included. The primary outcome was 1-year graft loss free survival. We developed a prediction model using cross-validation, by comparing Cox regression, gradient boosting, and random forests. The modeling strategy with the best discrimination and calibration was applied to fit a final prediction model. We used Shapley additive explanation (SHAP) values to perform variable selection and to estimate effect sizes and importance of individual variables when interpreting the final prediction model.

RESULTS

Cumulative incidence of graft loss or mortality was 7.6%. Random forests had favorable discrimination and calibration compared to Cox proportional hazards with a C-statistic (95% confidence interval [CI]) of 0.74 (0.72, 0.76) versus 0.71 (0.69, 0.73), and closer alignment between predicted and observed risk. SHAP values computed using the final prediction model indicated that the diagnosis of congenital heart disease (CHD) increased 1 year predicted risk of graft loss by 1.7 (i.e., from 7.6% to 9.3%), need for mechanical circulatory support increased predicted risk by 2, and single ventricle CHD increased predicted risk by 1.9. These three predictors, respectively, were also estimated to be the most important among the 15 predictors in the final model.

CONCLUSIONS

Risk prediction models used to facilitate patient selection for pediatric heart transplant can be improved without loss of interpretability using machine learning.

Contemporary outcomes of pediatric cardiac transplantation with a positive retrospective crossmatch

BACKGROUND

A positive crossmatch (+ XM) has traditionally been associated with adverse outcomes following pediatric heart transplantation. However, more recent studies suggest that favorable intermediate-term outcomes may be achieved despite a + XM. This study's hypothesis is that children with a + XM have similar long-term survival, but higher rate of complications such as rejection, coronary allograft vasculopathy (CAV), and infection, compared to patients with a negative (-) XM.

METHODS

The Pediatric Heart Transplant Society Registry (PHTS) database was queried from 2010-2021 for all patients <18 years of age with a known XM. Baseline demographics were compared between + XM and - XM groups using appropriate parametric and non-parametric group comparisons. Cox Proportional Hazards Modeling was used to identify risk factors for post-transplant graft loss, rejection, and CAV.

RESULTS

Of 4599 pediatric heart transplants during the study period, XM results were available for 3914 (85%), of which 373 (9.5%) had a + XM. Univariate analysis showed lower 10-year survival for patients with + XM (HR = 1.3, p = .04). Multivariate analyses revealed no significant difference in 10-year survival in the 2 groups; however, time to first rejection (p = .0001) remained significantly shorter in the + XM group.

CONCLUSIONS

Pediatric patients transplanted across a + XM experience earlier rejection; however, after multivariate adjustment, + XM is not independently associated with intermediate-term graft loss. The risk of heart transplantation against a + XM must be balanced with the ongoing risk of waitlist mortality.

Perceptions of pediatric deceased donor consent: A survey of organ procurement organizations

BACKGROUND

Children awaiting transplantation face a high risk of waitlist mortality due to a shortage of pediatric organ donors. Pediatric donation consent rates vary across Organ Procurement Organizations (OPOs), suggesting that some OPOs might utilize more effective pediatric-focused donor recruitment techniques than others. An online survey of 193 donation requestor staff sheds light on the strategies that OPO staff utilize when approaching potential pediatric deceased organ donors.

METHODS

In collaboration with the Association of Organ Procurement Organizations, the research team contacted the executive directors and medical directors of all 57 of the OPOs in the US. Of these, 51 OPOs agreed to participate, and 47 provided contact information for donation requestor staff. Of the 379 staff invited to participate in the survey, 193 provided complete responses.

RESULTS

Respondents indicated more comfort approaching adult donors than pediatric donors, and they endorsed approach techniques that were interpersonal and emotional rather than professional and informative. Respondents were accurate in their perceptions about which donor characteristics are associated with consent. However, respondents from OPOs with high consent rates (according to data from the Scientific Registry of Transplant Recipients), and those from OPOs with low consent rates were very similar in terms of demographics, training, experience, and reported techniques.

CONCLUSIONS

Additional research is needed to better determine why some OPOs have higher consent rates than others and whether the factors that lead to high consent rates in high-performing OPOs can be successful when implemented by lower-performing OPOs.