The Effect of Transplant Center Characteristics On Survival After Pediatric Hematopoietic Cell Transplantation

To date, neither the effect of transplant center volume nor other transplant center-specific characteristics have been described for pediatric hematopoietic cell transplantations (HCT) in the U.S. Therefore, using data reported to the Center for International Blood and Marrow Transplant Research (CIBMTR), we explored the effect of transplant center characteristics on day-100 and 1-year survival for pediatric HCT in the U.S. for the years 2008 and 2009.

Center Directors of all pediatric and combined adult and pediatric transplant centers that performed HCT in patients younger than 18 years were surveyed. Eighty-four transplant centers were queried and 68 centers responded (response rate 80%) representing patients from 42 pediatric and 26 combined adult and pediatric centers in the U.S. The survey instrument collected relevant transplant center characteristics, including transplant unit activities and resources, medical care team structure, and physician and healthcare provider characteristics. Patient, disease and HCT characteristics on 2228 consecutive pediatric allogeneic transplants performed at these centers between 2008–2009 were obtained from the CIBMTR. All centers performed transplantations during both study years. Prior to analysis of potential center effects, stepwise multivariate logistic regression was built to determine patient-related (age, performance score, sex, race), disease-related (disease and disease status) and HCT-related (donor [matched related, mismatched related, matched unrelated, mismatched unrelated], HCT-conditioning regimen [myeloablative vs. reduced intensity conditioning regimens], and graft-versus-host disease prophylaxis) factors associated with mortality. Thereafter, marginal logistic regression models were built using the generalized estimating equation approach to determine the effect of individual center characteristics on mortality, adjusting for significant patient, disease and HCT characteristics.

The median pediatric and pediatric and adult combined transplant center volume during the study period were 55 and 144, respectively. The results of multivariate analysis are shown in Table 1. After adjusting for performance score, donor type, and disease/disease status at transplant, day-100 mortality risks were higher at transplant centers that performed 16 or fewer allogeneic transplants over the two-year study period and at transplant centers where the number of inpatients per on-service transplant physician was five or fewer (median number of transplants was 24 per year). After adjusting for age, donor type, disease and disease status at transplant and CMV serostatus, 1-year mortality risks were also higher at transplant centers where the number of inpatients per on-service transplant physician was five or fewer, and in the absence of a dedicated transplant outpatient clinic during the weekend.

These data suggest that center volume is important to maximize survival for pediatric transplantation. Centers where there were fewer than five inpatients per on-service transplant physician were associated with higher mortality risks at both early and later post-transplant periods. Furthermore, a potentially modifiable center characteristic that may improve survival for pediatric HCT patients is access to specialized care (outpatient transplant clinic) on the weekend.

No relevant conflicts of interest to declare.