Outcomes In Pediatric Heart Transplantation Research Articles

The Reality of Limping to Heart Transplant

Factors associated with poor outcomes in pediatric heart transplantation are well defined. It is often felt that transplantation is the ultimate goal, regardless of how ill a patient may be. We hypothesize that transplantation for certain palliated congenital patients may have high morbidity and require careful consideration and counseling.

Early Implantation of Ventricular Assisted Devices (VADs) in Failing Single Ventricle Physiology Patients Can Lead to Better Outcomes in Pediatric Heart Transplantation

The most common cause for infant heart transplantation is congenital heart disease (CHD), with the most complex being single ventricle (SV) patients. Due to their singular anatomy and prior surgeries, support of SV patients with a VAD remains challenging as one pump supports both circulations. We sought to describe our single center success in transplantation of SV patients post-VAD implantation.

Supplemental Use of Blood Cardioplegia Before Graft Implantation in Pediatric Heart Transplantation

Optimal preservation of donor myocardium in heart transplantation is unknown. Despite many novel approaches, the best option currently available is meticulous usage of cardioplegia (crystalloid or blood-based). Adult studies have shown improved outcomes with supplemental administration of cardioplegia before graft implantation. Similar studies in children are lacking. The aim of this study is to analyze if supplemental use of blood cardioplegia before graft implantation can improve outcomes in pediatric heart transplantation.

A New Combined Model for End-Stage Liver Disease Excluding INR (CMELD-XI) Score Better Predicts Pediatric Heart Transplant Outcomes

The Model for end-stage liver disease (MELD) score, derived based on serum creatinine, bilirubin, and INR, was developed to accurately predict the probability of waitlist mortality in patients awaiting a liver transplant. Subsequently, a MELD excluding INR (MELD-XI) score was adapted to predict mortality after heart transplant (HT). Our objective was to devise a new combined scoring model (defined as CMELD-XI, derived by adding the MELD-XI score with independent predictors of 1-year mortality) to accurately predict mortality after pediatric HT (PHT).

Geographic Distance From Transplant Center Does Not Impact Pediatric Heart Transplant Outcomes.

Many pediatric heart transplant recipients live a significant distance from their transplant center. This results in families either traveling long distances or relying on outside physicians to assume aspects of their care. Distance has been implicated to play a role in congenital heart disease outcomes, but its impact on heart transplantation has not been reported. The aim of this study was to assess the impact of distance on pediatric heart transplant outcomes. The Scientific Registry of Transplant Recipients database was queried for all pediatric heart transplant recipients from large US children's hospitals (1987-2014). Patients were stratified into 4 groups (<20, 20-50, 50-100, and >100 miles) based on distance. Survival curves were generated and compared using the log-rank test. Cox proportional hazards regression was performed to adjust for differences between groups. A total of 4768 patients were included in the analysis, of which 1435 (30.1%) were <20 miles, 940 (19.7%) were 20 to 50 miles, 806 (16.9%) were 50 to 100 miles, and 1587 (33.3%) were >100 miles from their transplant center. There was no difference in posttransplant survival based on distance after adjusting for patient age, gender, ethnicity, blood type, diagnosis, listing status, and the need for pretransplant ventricular assist device, extracorporeal membrane oxygenation, or ventilator support. There is no significant difference in graft survival after pediatric heart transplantation based on patient distance from their transplant center. Our data suggest the current strategy of transitioning some aspects of transplant care to local physicians or management from a distance does not increase posttransplant mortality risk.

Invited Commentary

Invited Commentary

Obesity class does not further stratify outcome in overweight and obese pediatric patients after heart transplantation.

The effect of obesity stratification on pediatric heart transplant outcomes is unknown. The UNOS database was queried for patients ≥2-<18years listed for heart transplant and stratified by BMI: normal (BMI>5%-≤85 percentile), overweight (BMI=86%-95 percentile), class 1 (BMI=100%-120% of 95 percentile), class 2 (BMI=121%-140% of 95 percentile), and class 3 obesity (BMI>140% of 95 percentile). A total of 5056 individuals were listed for transplant, with 71% normal, 13% overweight, 10% class 1, 4% class 2, and 2% class 3 obesity. Waitlist survival was not different between groups. Post-transplant survival was decreased in overweight and combined obese groups vs normal, with no further difference between overweight and obese classes. Overweight and obese patients had higher listing status and were more likely to have ventilator, inotrope, and mechanical circulatory support at listing. After transplant, there was an association of overweight-obese patients with diabetes and rejection requiring hospitalization. Stricter definition of normal weight reveals overweight-obese status was an independent risk factor for poorer post-transplant survival, without further effect by stratification of weight class. However, because there is no difference in waitlist survival, this study does not allow the selection of absolute weight-based criteria regarding transplant listing and suggests the need to look further for modifiable risk factors post-transplant.

The Number of Refusals for Donor Organ Quality Does Not Impact Heart Transplant Outcomes in Children

Waitlist mortality is more than 12% for pediatric heart transplantation, with strikingly high rates of organ refusal, many of which are due to donor quality. However, some centers use these organs despite refusals by other centers for donor quality. We hypothesize that the number of refusals for donor quality (RDQ) does not affect pediatric heart transplantation outcomes. Pediatric heart transplants from 2000 to 2015 were identified using the United Network for Organ Sharing database and were matched against the potential transplant recipients dataset with donor refusal codes. Refusals for donor quality were counted for each organ. The population was divided into two groups: RDQ-low (0 to 3 RDQs, n= 3,404) and RDQ-high (>3 RDQs, n= 1,585). Posttransplant outcomes of both cohorts were analyzed. Of 4,989 pediatric heart transplants, 75% (n= 3,770) had 1 or more RDQ (median 3 RDQ; interquartile range, 1 to 7). The RDQ-lower group (0 to 3 RDQs) and the RDQ-higher group (>3 RDQ) had similar posttransplant survival (p= 0.41) and freedom from retransplantation (p= 0.37). Both groups had similar posttransplant survival even for high-risk recipient cohorts: adolescents (p= 0.06), congenital heart disease (p= 0.87), retransplantation (p= 0.47), extracorporeal membrane oxygenation (p= 0.61), mechanical ventilation (p= 0.24), and poor renal function at transplant (p= 0.46). In addition, recipient subgroups who had donors with increasing number of RDQ (>6, >9, >12, and >15 RDQ) also had similar posttransplant survival compared with the RDQ-low group (p= 0.63, p= 0.62, p= 0.92, and p= 0.50, respectively). The outcome of pediatric heart transplantation is not affected by the number of RDQ, even for high-risk recipients. The use and interpretation of donor quality refusal code should be considered carefully while selecting or refusing donors in this era of supply and demand mismatch.

(699) - Center Variability in Selecting High-Risk Donors: Does It Affect the Outcomes in Pediatric Heart Transplantation?

(699) - Center Variability in Selecting High-Risk Donors: Does It Affect the Outcomes in Pediatric Heart Transplantation?

The impact of cognitive delay on pediatric heart transplant outcomes.

The presence of CD may be viewed as a relative contraindication to transplantation; however, its impact on pediatric HTx outcomes is poorly characterized. The aim of this study was to assess the impact of CD on pediatric HTx outcomes using academic progress as a surrogate measure of cognitive performance. The OPTN database was queried for all pediatric HTx recipients (2004-2014) with reported academic progress. Multivariable analysis assessed the impact of DGL and the need for SE on post-HTx graft survival. A total of 2245 children were included: 1707 (76%) within grade level, 269 (12%) with DGL, and 269 (12%) who required SE. The need for SE was not a risk factor for post-HTx mortality; however, DGL was an independent risk factor for worse post-HTx outcomes (AHR 1.4, 95% CI 1.02, 1.79, P=.03). Patients who require SE have similar outcomes compared to those without CD, likely secondary to significant parental involvement. Children with DGL demonstrate inferior post-HTx survival, which could result from less parental oversight in children perceived to maintain compliance. Ensuring adequate social support for patients with evidence of CD may help to improve outcomes.

Urgent listing exceptions and outcomes in pediatric heart transplantation: Comparison to standard criteria patients

United Network for Organ Sharing (UNOS) policy enables listing exceptions to avoid penalizing patients with waitlist mortality not captured by standard criteria. Outcomes among patients listed by exception have not been analyzed. We performed a retrospective analysis of pediatric (≤17 years of age, n = 4,706) listings (2006 to 2015) for primary, isolated heart transplantation within the UNOS data set, assessing Status 1A exception (n = 211, 4.5%) use across regions and patient characteristics and evaluating waitlist outcomes compared with candidates listed using standard criteria. Death or removal for reason other than transplant did not differ between exception and standard criteria patients at 1 month (11.7% vs 16.2%, p = not statistically significant [NS]), 2 months (18.2% vs 29.0%, p = 0.11) or overall (16.1% vs 22.0%, p = NS) on the waitlist. Rates were higher than among Status 1B patients (1 month: 2.8%; 2 months: 5.6%; overall: 14.9%; p < 0.0001). The cumulative incidence of competing risks (transplantation, death/removal for reasons other than transplant and removal) did not differ when comparing Status 1A exception patients with Status 1A standard criteria patients. Use of 1A exceptions varied across UNOS regions (1.9% to 22.3%, p < 0.0001). Risk-adjusted modeling identified patients more (hypertrophic cardiomyopathy: odds ratio [OR] = 2.8, 95% confidence interval [CI] 1.5 to 5.0; restrictive cardiomyopathy: OR = 2.7, 95% CI 1.7 to 4.3) and less (low socioeconomic status: OR = 0.7, 95% CI 0.5 to 1.0) likely to use an exception. Use of exceptions was uncorrelated with regional outcomes. Waitlist mortality among Status 1A exception patients is similar to that among those listed by standard criteria. However, variation in exception use across geography and demography may contribute to inequities in access to transplantation, particularly for those with low socioeconomic status. Standardization of practices may decrease regional variation and minimize inequities.

Donor-recipient height ratio and outcomes in pediatric heart transplantation.

Height matching in pediatric HTx has been proposed as a superior method of evaluating graft size, but no studies have examined survival advantage for height-matched donor-recipient pairs. We hypothesized that in pediatric patients with DCM, an oversized donor improves survival and aimed to define the optimal height ratio in this patient group. Pediatric primary HTx recipients with DCM between 10/89 and 09/12 were identified in the OPTN database. Patients were stratified into three donor-recipient height and weight ratio categories. One- and five-yr survival was compared using Kaplan-Meier analysis and HRs were computed. A total of 2133 children with DCM who underwent HTx during the study period were included. Unadjusted one-yr survival was worse for DRHR <0.87 (HR, 2.15 [95% CL, 1.30, 3.53]; p < 0.01). This difference was not present at five yr post-HTx or when stratified by weight. After adjustment for other risk factors affecting transplant survival, height matching was no longer significant. Although height matching appears to predict short-term survival better than weight in pediatric HTx recipients with DCM, other factors play a more important role as height matching loses significance in multivariate analysis.

(36) - Number of Refusals for Donor Quality Does Not Impact Post-Transplant Outcomes in Pediatric Heart Transplantation

(36) - Number of Refusals for Donor Quality Does Not Impact Post-Transplant Outcomes in Pediatric Heart Transplantation

(1165) - Impact of Geographic Distance from Transplant Center on Pediatric Heart Transplant Outcomes

Many families live a significant distance from their transplant center, prompting outside physicians without formal transplant training to assume aspects of their routine care. Distance from surgical center has proven important in congenital heart disease outcomes; however, the impact of this distance after heart transplantation is unknown. We aimed to determine the impact of geographic distance from transplant center on mortality after pediatric heart transplantation. The SRTR database was queried for all pediatric (<18 years) heart transplants performed at PHTS centers from 1987 to 2014. Recipients were stratified by distance from transplant center (DistTxC) using their home zip code with three cutoffs (≤20, 50, 100 miles). Survival curves were generated and compared using the log rank test. Cox proportional hazards regression was performed to adjust for differences between groups. A total of 4825 patients were included in the analysis. No difference was detected between groups by gender, age at listing, diagnosis, or the need for preoperative ECMO. DistTxC groups varied by blood type, need for pre-operative ventilator support, race, and status at listing. There was no significant difference in post-transplant mortality based on DistTxC (p=0.20, p=0.37, and p=0.43 for 20, 50, and 100 mile comparison groups) (figure), and this remained true after adjusting for differences in a multivariable model. There is no significant difference in mortality after pediatric heart transplantation based on geographic distance from the transplant center. Our data suggests the current strategy of transitioning some aspects of transplant care to local physicians does not increase post-operative mortality risk.

(175) - Increased Frequency of Surveillance Biopsy Does Not Improve Pediatric Heart Transplant Outcomes: Time for a Change in Practice?

(175) - Increased Frequency of Surveillance Biopsy Does Not Improve Pediatric Heart Transplant Outcomes: Time for a Change in Practice?

(32) - Recipient-Donor Height Ratio and Outcomes in Pediatric Heart Transplantation

(32) - Recipient-Donor Height Ratio and Outcomes in Pediatric Heart Transplantation

(35) - High BMI Predicts Poor Outcomes in DCM But Not CHD Patients: The Differential Impact of Obesity on Outcomes in Pediatric Heart Transplantation

Body mass index (BMI) is thought to have an important impact on waitlist and post-transplant mortality in children. Our aim was to investigate whether the impact of BMI varies by the etiology of the heart failure. UNOS has 3,254 listings for primary, isolated heart transplant in patients 3-18 yrs old (1995-2012) with either congenital heart disease (CHD) or dilated cardiomyopathy (DCM). Patients were stratified into 4 groups based on BMI%ile-for-age (BMI%): underweight (BMI% < 5), normal weight (5-84), overweight (85-94), or obese (≥ 95). Waitlist and post-transplant outcomes were assessed. Obesity was more common among DCM patients (18.7% vs. 9.91%) while underweight was more common in CHD patients (20.0% vs. 15.7%, p<0.0001). Waitlist survival was unaffected by BMI% category among CHD patients; underweight (HR 1.3, 1.0-1.6) and obese (1.2,1.0-1.5) DCM patients HR 1had worse survival and lower transplantation rates (OR0.6, 0.5-0.8). Overweight DCM (1.8, 0.8-4.2) patients and obese CHD (1.9, 0.9-4.0) patients had higherrisk-adjusted mortality prior to discharge. Overweight (HR 1.6, 1.2-2.2) and obese (1.2, 0.9-1.7) patients had higher 1-yr conditional mortality, whereas obese CHD patients had lower mortality (0.5, 0.3-1.0). Higher BMI% was a significant predictor of coronary allograft vasculopathy (p=0.02), and diabetes (p=0.0006) in DCM (but not CHD) patients (Figure). The impact of BMI% on waitlist and post-transplant mortality varies by diagnosis. Analyses of weight and nutritional status in heart failure and transplantation need to examine CHD and DCM patients separately. BMI alone is insufficient, and better measures of nutrition, causes of increased BMI (obesity vs fluid retention), and physical conditioning are required to better estimate risk.

HLA molecular epitope mismatching and long-term graft loss in pediatric heart transplant recipients

Although evidence links HLA allele mismatching to worse outcomes in pediatric heart transplantation, no studies to our knowledge have applied the quantification of structural HLA differences between donor and recipient to risk evaluation. We examine the association between molecular-level HLA mismatching and long-term graft loss in pediatric recipients of heart transplants. HLA Matchmaker was used to quantify the number of mismatched class-specific HLA eplets among 4,851 heart transplant recipients ≤18 years of age in the Scientific Registry of Transplant Recipients (1987-2012). Survival analysis was used to compare long-term probabilities of graft loss by number of eplet mismatches and allele mismatches stratified by eplet mismatches. Recipients with 10 to 20 or >20 class I (HLA-A and HLA-B) eplet mismatches experienced increased long-term graft loss compared with recipients with <10 class I eplet mismatches (adjusted hazard ratio = 1.23 [95% confidence interval = 1.06-1.42], adjusted hazard ratio = 1.27 [95% confidence interval = 1.08-1.50], respectively). Recipients with 2 to 4 class I allele mismatches had increased long-term graft loss compared with recipients with 0 to 1 class I allele mismatches. Neither class II (HLA-DR) eplet mismatching nor class II allele mismatching was associated with graft loss. On stratification by allele and structural eplet mismatching, only recipients with 2 to 4 class I allele mismatches and ≥10 class I eplet mismatches had an increased probability of graft loss compared with recipients with 0 to 1 class I allele mismatches (adjusted hazard ratio = 1.42 [95% confidence interval = 1.09-1.57]). Molecular-level HLA mismatching may aid in identifying recipients at increased risk of long-term graft loss who could benefit from intensified post-transplant surveillance and management.

Differential effect of body mass index on pediatric heart transplant outcomes based on diagnosis.

The impact of nutritional status on HTx waitlist mortality in children is unknown, and there are conflicting data regarding the role of nutrition in post-HTx survival. This study examined the influence of nutrition on waitlist and post-HTx outcomes in children. Children 2-18yr listed for HTx from 1997 to 2011 were identified from the OPTN database and stratified by BMI percentile. Multivariable logistic regression evaluated the influence of BMI on waitlist mortality. Cox proportional hazard regression assessed the impact of BMI on post-HTx mortality. When all 2712 patients were analyzed, BMI did not impact waitlist, one-, or five-yr mortality. However, when stratified by diagnosis, BMI>95% (AOR 1.96; 95% CI 1.24, 3.09) and BMI<1% (AOR 2.17; 95% CI 1.28, 3.68) were independent risk factors for waitlist mortality in patients with CM. BMI did not impact waitlist mortality in CHD and did not impact post-HTx outcomes, regardless of diagnosis. BMI>95% and BMI<1% are independent risk factors for waitlist mortality in patients with CM, but not CHD. This suggests differing risk factors based on disease etiology, and an individualized approach to risk assessment based on diagnosis may be warranted.

The influence of human leukocyte antigen matching on outcomes in pediatric heart transplantation.

Previous adult heart transplantation studies have demonstrated that donor-recipient human leukocyte antigen (HLA) matching results in reduced graft failure and improved patient survival. No study has examined these effects in children. This study investigated the effect of HLA matching on outcomes in pediatric heart transplantation. All pediatric heart transplantation data for patients 0-18 years of age available from the United Network for Organ Sharing Transplant Registry from 1987 to 2009 were analyzed retrospectively. Donor-recipient HLA matching at loci A, B, and DR (0-6) was compared with graft survival and recipient survival. For this study, 3,751 pediatric cardiac transplantation events with complete HLA matching data were identified and grouped as having 0 to 2 matches (3,416 events) or 3 to 6 matches (335 events). The 3- to 6-match group had less graft failure than the 0- to 2-match group (28.7% vs 34.4%; p = 0.035) and greater patient survival by 5 years (81% vs 72%; p = 0.045) and 10 years (66% vs 55%; p = 0.005) after transplantation. The HLA-DR matching alone resulted in less graft failure (p = 0.038) and improved patient survival (p = 0.017). A higher degree of HLA matching in pediatric heart transplantation is associated with decreased graft failure and improved patient survival. In this study, decreased graft failure rates and superior survival also were seen with DR matching alone.