Non-invasive myocardial tissue deformation and discoordination indices predict cardiac allograft vasculopathy in pediatric heart transplantation patients.

ABSTRACT

There is an urgent need for non-invasive imaging-based biomarkers suitable for diagnostic surveillance of cardiac allograft vasculopathy (CAV) in pediatric heart transplant (PHT) patients. The purpose of this study was to comprehensively investigate left ventricular (LV) myocardial deformation in conjunction with electromechanical discoordination in PHT. PHT patients with and without CAV were evaluated for echocardiography derived global longitudinal strain (GLS) and electromechanical discoordination indices including systolic stretch fraction (SSF) and diastolic relaxation fraction (DRF). SSF was increased in CAV(+) patients at the time of CAV diagnosis (median CAV(+) 5.0 vs. median CAV(-) 0.0, P = 0.008) and in the echocardiogram preceding the CAV diagnosis (median CAV(+) 29.0 vs. median CAV(-) 0.0, P < 0.001). DRF was also increased in the echocardiogram that preceded CAV diagnosis in CAV(+) patients (0.31 ± 0.08 vs. 0.25 ± 0.05, P = 0.008). The final model using indices 6-12 months prior to CAV diagnosis included GLS, SSF, and DRF providing AUC of 0.94 with sensitivity 98.5%, specificity 80.0%, positive predictive value 85.0%, and negative predictive value 94.1%. Systolic and diastolic electro-mechanical discoordination indices are significantly worse in PHT patients experiencing CAV. Non-invasive imaging guided surveillance using echocardiographic myocardial deformation indices can be improved by adding SSF and DRF to standard GLS measurements.