Temporal changes in left ventricular strain with the development of rejection in paediatric heart transplant recipients.

INTRODUCTION: Myocardial strain measurements are increasingly used to detect complications following heart transplantation. However, the temporal association of these changes with allograft rejection is not well defined. The aim of this study was to describe the evolution of strain measurements prior to the diagnosis of rejection in paediatric heart transplant recipients. METHODS: All paediatric heart transplant recipients (2004-2015) with at least one episode of acute rejection were identified. Longitudinal and circumferential strain measurements were assessed at the time of rejection and retrospectively on all echocardiograms until the most recent negative biopsy. Smoothing technique (LOESS) was used to visualise the changes of each variable over time and estimate the time preceding rejection at which alterations are first detectable. RESULTS: A total of 58 rejection episodes were included from 37 unique patients. In the presence of rejection, there were decrements from baseline in global longitudinal strain (-18.2 versus -14.1), global circumferential strain (-24.1 versus -19.6), longitudinal strain rate (-1 versus -0.8), circumferential strain rate (-1.3 versus -1.1), peak longitudinal early diastolic strain rate (1.3 versus 1), and peak circumferential early diastolic strain rate (1.5 versus 1.3) (p<0.01 for all). The earliest detectable changes occurred 45 days prior to rejection with simultaneous alterations in myocardial strain and ejection fraction. CONCLUSIONS: Changes in graft function can be detected non-invasively prior to the diagnosis of rejection. However, changes in strain occur concurrently with a decline in ejection fraction. Strain measurements aid in the non-invasive detection of rejection, but may not facilitate earlier diagnosis compared to more traditional measures of ventricular function.