Is it time to change the routine for surveillance endomyocardial biopsy after pediatric heart transplantation?

Graft rejection remains one of the most feared complications after pediatric heart transplantation (HT) due to its association with increased graft failure and mortality.1 Endomyocardial biopsy (EMB) monitoring originated in the early days of transplantation to monitor for rejection and initiate treatment prior to the clinical expression of graft dysfunction. Despite decades of experience, the pediatric HT community lacks an evidencebased standardized routine surveillance biopsy (RSB) protocol and significant practice variation exists between institutions.2 Duong et al3 is the latest group to evaluate the impact of RSB practice variation on rejection and graft survival. The authors linked the Pediatric Health Information System (PHIS) and Scientific Registry of Transplant Recipients (SRTR) databases to evaluate the impact of the firstyear RSB frequency on outcomes. This approach allowed for the inclusion of nearly 2,900 patients, easily the largest investigation to date. Centers performing 4 or more RSB in the first year were classified as high frequency, and those performing less than 4 were classified as low frequency. Graft survival and incidence of coronary allograft vasculopathy (CAV) were not significantly different between the groups at median followup of 3.9 years, while highfrequency surveillance was associated with an increased rate of diagnosed but not treated rejection and cost of care. The authors should be commended for their work on this challenging topic. The study adds important data on the additional costs of highfrequency surveillance in the first year after HT while showing that this additional screening had no significant clinical benefit. The inherent limitations to this type of study were appropriately discussed. Multicenter retrospective studies are limited by lack of knowing the true RSB frequency of each center, varying definitions of rejection, different thresholds for augmentation of immunosuppression, and lack of longterm followup. Singlecenter studies on RSB frequency and outcomes have the benefit of a more complete dataset and uniform protocol, but the results may not be reproducible in a larger population. These limitations make it difficult to enact practice changes, especially for highfrequency RSB centers leery of adjusting their protocols. The data supporting RSB reduction has been growing over the past decade. The 2010 International Society for Heart and Lung Transplantation guidelines for the care of heart transplant patients recommended RSB in the first year after HT for both pediatric and adult recipients.4 The group also agreed that echocardiography could replace RSB in infants to reduce RSB in the first year. RSB beyond the 5th year posttransplant was deemed optional in both pediatrics and adults. More recent publications have shown that lowfrequency RSB surveillance protocols are not associated with increased mortality,5 and RSB beyond the second year may have low diagnostic yield in younger patients.6 Taken together, these reports support using RSB early after transplant when the risk of rejection is the highest and gradually weaning or even cessation over the subsequent years in patients with a low risk for rejection. One of the biggest hurdles to acting on this growing data set is the comfort provided by pathology. Many transplant teams, patients, and families draw comfort from a negative biopsy even if the pretest probability for rejection is quite low. What if we could provide the same level of comfort through noninvasive means without sacrificing confidence in the assessment? Clinicians will always utilize a thorough history and physical, graft functional assessment via echocardiogram, and immunosuppression levels to regularly screen for rejection. While EMB remains the gold standard for the diagnosis of rejection, we now have several noninvasive means of assessing the likelihood of acute rejection. Duong et al. commented on brain natriuretic peptide, strain imaging, and magnetic resonance imaging as additional means of assessing for rejection. Donorderived cellfree DNA (ddcfDNA) assessment has been shown to be a sensitive assay for acute rejection in adult heart transplant recipients.7 Limited data exist in pediatrics.8 Our institution recently began using a ddcfDNA assay in place of RSB for lowrisk patients starting at nine months posttransplant. The results suggesting a low risk for acute rejection— along with a reassuring history, examination, echocardiogram, and immunosuppression levels— have allowed us to safely forego RSB without new graft dysfunction (manuscript under review). There is still much to learn about the utility and limitations of ddcfDNA in assessing for rejection after pediatric HT, but early investigations show promise that this and other noninvasive tools may one day significantly reduce the frequency of routine EMB. In conjunction with the current work by Duong et al., there appears to be enough supporting evidence to move forward with testing the hypothesis that lowfrequency RSB protocols in select lowrisk patients can be followed without an increased risk of clinically significant rejection, CAV, or graft loss. Clearly defined lowand highfrequency protocols, definitions for rejection, criteria for treatment of rejection, accompanying noninvasive screening data, and longterm outcomes assessment are key study points to provide the evidence necessary to update and standardize rejection surveillance after pediatric HT.