Predictors of Longitudinal Cognitive and Motor Delay in Pediatric Heart Transplant Recipients and Associated Survival Outcomes

A. Bilgili¹, L. M. Brinkley¹, O. M. Sharaf¹, Z. Brennan², G. J. Peek³, M. S. Bleiweis³, J. P. Jacobs^3
1University of Florida College of Medicine, Gainesville, Florida ²Cedars-Sinai, Royal Oak, Michigan ³University of Florida, Gainesville, Florida

Purpose: Children undergoing cardiac surgery are at a higher risk of delayed functional development and ensuing diminished quality of life. This study aims to comprehensively investigate the predictive factors associated with the longitudinal development of (1) cognitive and (2) motor delays in pediatric heart transplant recipients and subsequent patient survival.

Methods: The United Network for Organ Sharing (UNOS) Registry was queried for pediatric patients ( < 18 years) who received a first-time isolated heart transplant between March 2008 and December 2022. Patients with incomplete follow-up data were excluded. Two patient cohorts were developed: 1) those with motor development data and 2) those with cognitive development data. Cohorts were subsequently stratified based on the presence or absence of motor delay (MD) or cognitive delay (CD). Univariate analysis was performed to assess baseline, perioperative, and postoperative characteristics. Post-transplant survival was compared in each cohort using univariate Cox regression and subsequent survival analysis. Multivariable logistic regression was used to adjust for 29 baseline and operative variables to create a predictive model for post-transplant MD and CD. Receiver operating characteristic (ROC) curves were generated and subsequent area under the curves (AUC) were calculated to assess individual model performance.

Results: A total of 3847 (n=3267 [no MD], n=580 [definitive MD]) and 3446 (n =2689 [no CD], n=757 [definitive CD]) patients were included across MD and CD cohorts, respectively. Cohorts had an overlap of 3189 patients. Compared to patients without MD, patients with definitive MD were younger (age [years]: 4.27±5.59 vs 6.98±6.14, p< 0.001) and smaller (weight [kg]: 18.7±19.7 vs 29.2±25.3, p< 0.001; height[cm] 91.3±38.5 vs 113±42.9, p< 0.001], with similar rates of male patients (male, 56.2% [n=326] vs 55% [n=1797], p=.645). Similarly, patients with definitive CD were younger (p < 0.001) and smaller (p < 0.001) but were mostly male (60% [n=454] vs 53.4% [n=1435], p= 0.002). Patients with definitive MD were also likely to have definitive CD and vice versa (p < 0.001). Additionally, patients with a definitive delay at follow-up had worse longitudinal survival, with hazard ratios of 2.82 (95% CI: 2.32-3.44, p< 0.001) and 1.67 (95% CI: 1.32-2.05, p< 0.001) for MD and CD cohorts, respectively (Figure 1A). Multivariate analysis revealed considerable overlap but also differences in independent predictors of developing delay in both cohorts (Table 1). Upon model predictive validation, ROC curves signified AUCs of 0.756 and 0.707 for MD and CD cohorts, respectively (Figures 1B and C).

Conclusion: Our findings demonstrate the presence of multiple predictive factors associated with the development of longitudinal cognitive and motor delays following pediatric heart transplantation. Patients with a delay at follow-up have varying demographic characteristics, differences in operative complications, and worse long-term survival.

Identify the source of the funding for this research project: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.