Measuring Quality in Pediatric Heart Transplantation-An Important but Challenging Goal.

Pediatric heart transplantation (HT), first reported in 19681 and adopted on a broad scale during the 1980s, remains essentially the only therapy that can reverse symptoms of end-stage heart failure (HF) and restore a high quality of life. In the current era, early survival after pediatric HT now exceeds 90% in the US2; this remarkable statistic reflects a sustained cooperative effort from diverse sources, including the individual contributions of dedicated pioneers; small groups of practitioners who recognized at an early stage the value of prospective multicenter clinical and outcomes data collection through registries, such as the Pediatric Heart Transplant Society; and regulatory agencies that have provided oversight and maintained public trust in this scarce, valuable resource. Nonetheless, many challenges remain in ensuring that HT is available to the greatest number of children while assuring the best possible outcomes. One such challenge derives from the fact that pediatric HT is a lowvolume endeavor, with only 400 to 500 HTs performed in pediatric patients annually in the US.2 This combination of relatively low individual center volume and low early mortality rate leads to difficulty in discriminating between programs on the basis of early mortality, which is the most commonly used measure of a program’s quality. In other words, given that most programs do well: (1) is it possible to elucidate the characteristics of programs that enjoy the best post-HT outcomes? (2) do these characteristics differ from those whose outcomes are inferior? and (3) can this be accomplished in a constructive manner that programs can use to drive improvement initiatives? The study by Singh et al3 provides important insights into these questions through an analysis of the US Organ Procurement and Transplantation Network database, a publicly available data set that collects information on all solid organ transplants performed in the US since 1987. Building on prior work from their group, which established that most of the differences in post-HT survival among centers could be detected by evaluating survival 90 days post-HT,4 Singh et al3 divided HT programs into 3 groups (high-performing, medium-performing, and low-performing), according to standardized mortality ratios for observed 90-day post-HT survival, and compared these groups with expected 90-day post-HT survival for each recipient (according to recipient risk), expected 90-day survival for each center (according to case-mix), and the incidence of major post-HT complications, such as rejection and acute kidney failure, in a population sampled during a 10-year period from 2006 through 2015. Patients older than 18 years of age were excluded, as were patients undergoing retransplant or multiple organ transplant. Importantly, centers not performing HT in children younger than 10 years of age, who were inactive for 5 or greater consecutive years, or who performed less than 10 HT during the study period were excluded from analysis. Several noteworthy findings emerge from the analysis by Singh and colleagues.3 First is the lowvolume nature of pediatric HT and the wide variation in annual volume among centers; the median (range) annual transplant volume was 5.9 (1.0-19.2) HTs per year. The model created by the authors to estimate 90-day post-HT survival, which included the variables of age, diagnosis, mechanical ventilation, mechanical circulatory support, baseline kidney dysfunction, and serum bilirubin, suggested good discriminative ability, with a C statistic of 0.8, between those who were expected to survive HT and those who were not. Second, among the centers sampled, 90-day survival was excellent at 94.8%, which is based on the observed median (range) 90-day mortality of 5.2% (0% to 23.1%); therefore, the lowest performing center achieved a 90-day survival rate of nearly 80%. When centers were stratified into high-, mediumand low-performance groups according to standardized mortality ratios, no significant + Related article