Abstract
Background
Children listed for heart transplantation face the highest waiting list mortality in solid-organ transplantation medicine. We examined waiting list mortality since the pediatric heart allocation system was revised in 1999 to determine whether the revised allocation system is prioritizing patients optimally and to identify specific high-risk populations that may benefit from emerging pediatric cardiac assist devices.
Methods and Results
We conducted a multicenter cohort study using the US Scientific Registry of Transplant Recipients. All children <18 years of age who were listed for a heart transplant between 1999 and 2006 were included. Among 3098 children, the median age was 2 years (interquartile range 0.3 to 12 years), and median weight was 12.3 kg (interquartile range 5 to 38 kg); 1294 (42%) were nonwhite; and 1874 (60%) were listed as status 1A (of whom 30% were ventilated and 18% were on extracorporeal membrane oxygenation). Overall, 533 (17%) died, 1943 (63%) received transplants, and 252 (8%) recovered; 370 (12%) remained listed. Multivariate predictors of waiting list mortality include extracorporeal membrane oxygenation support (hazard ratio [HR] 3.1, 95% confidence interval [CI] 2.4 to 3.9), ventilator support (HR 1.9, 95% CI 1.6 to 2.4), listing status 1A (HR 2.2, 95% CI 1.7 to 2.7), congenital heart disease (HR 2.2, 95% CI 1.8 to 2.6), dialysis support (HR 1.9, 95% CI 1.2 to 3.0), and nonwhite race/ethnicity (HR 1.7, 95% CI 1.4 to 2.0).
Conclusions
US waiting list mortality for pediatric heart transplantation remains unacceptably high in the current era. Specific high-risk subgroups can be identified that may benefit from emerging pediatric cardiac assist technologies. The current pediatric heart-allocation system captures medical urgency poorly. Further research is needed to define the optimal organ-allocation system for pediatric heart transplantation.
Keywords: pediatrics, transplantation, heart, survival, heart-assist devices
Introduction
Of all patients wait-listed for solid-organ transplantation in the United Sates, children listed for heart transplantation face the highest waiting list mortality regardless of age.1 To address this problem, in 1999, the United Network for Organ Sharing (UNOS) implemented a major change in the way donor hearts were allocated by assigning higher priority to sicker status 1 patients2 (ie, status 1A patients as determined by circulatory support requirements) who were less likely to survive a prolonged wait period. Over the same timeframe, after the landmark study by West and colleagues in 2001,3 the practice of listing infants across all blood types has increased steadily, a development that has the potential to shorten wait times for infant candidates considerably.3,4
The collective impact of these changes on present-day waiting list mortality is unknown, in large part because earlier studies were conducted primarily in the 1990s, before changes in organ-allocation practice occurred.5–12 In addition, earlier studies were limited by smaller sample sizes or single-institution experiences,9,10,12,13 the findings of which may not be generalizable owing to regional differences in practice or may be underpowered to detect important national trends. A contemporary analysis of the primary risk factors associated with waiting list mortality that included all US patients would be useful for 3 specific reasons: (1) To help policy makers determine whether the current organ-allocation system is serving children with end-stage heart disease optimally, (2) to better define specific high-risk populations that may benefit from emerging mechanical circulatory support technologies, and (3) to determine more precisely where the national organ shortage for pediatric donor hearts is most critical (especially with respect to age and size) as part of a nationwide effort to establish pediatric-specific organ-donation goals.
Methods
Study Population and Data Source
All pediatric subjects less than 18 years of age who were listed for first orthotopic heart transplantation in the United States between January 20, 1999, and July 12, 2006, were identified retrospectively through the US Scientific Registry of Transplant Recipients. The Scientific Registry of Transplant Recipients is an internally audited, mandatory, government-sponsored, solid-organ transplant registry that collects information on all solid-organ transplants in the United States. Demographic and clinical information is reported by transplanting centers to the Organ Procurement and Transplantation Network, supplemented by data from the Social Security Administration and the Center for Medicare and Medicaid Services. January 20, 1999, marks the point in time at which status 1 patients were subdivided into status 1A and status 1B patients. July 12, 2006, marks the point in time at which older status 1A children within 500 miles were given priority over status 1B children within the region. Patients listed for heart retransplantation or multivisceral transplants were excluded. All patients were followed up from the time of listing for heart transplantation until death or the day of last observation on August 3, 2007.
Study Definitions and Outcome Measures
The primary study hypothesis was that among children listed for orthotopic heart transplantation, mechanical ventilation is associated with reduced waiting list survival after adjustment for other patient factors. Time on the waiting list was defined as time from initial listing for heart transplantation to the time of waiting list removal due to transplant, death, or recovery. Subjects who died were considered to have reached the primary end point (ie, had an event). Subjects were censored at the time of transplantation or recovery. All other subjects who remained on the waiting list were censored on August 3, 2007. All clinical and demographic variables were defined at the time of listing for heart transplant unless otherwise specified. Race/ethnicity data (categories included black, white, Hispanic, Asian, American Indian/Alaska Native, Native Hawaiian/Pacific Islander, multiracial, and other) were analyzed as reported by the transplanting center. Glomerular filtration rate was estimated with the Schwartz formula.14
Statistical Analysis
Summary statistics are presented as median (interquartile range [IQR]) or number (percent). Patient characteristics were compared across cardiac diagnostic subgroups with the χ2 test for categorical variables and the Kruskal-Wallis test for continuous variables. Survival time on the waiting list was estimated by the Kaplan-Meier method. Univariate relationships between patient characteristics and waiting list mortality were evaluated with the log-rank test. Multivariable analysis was performed with the Cox proportional hazards model and a stepwise selection technique. Only risk factors that were statistically significant at the 0.05 level were retained in the final multivariable models. These models were then reevaluated with control for UNOS region. Analyses were performed with SAS version 9.1 and Stata version 10.0.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Results
Study Cohort
Of 3416 pediatric patients listed for heart transplant during the study period, 3098 met the inclusion criteria (308 were excluded owing to heart retransplantation and 10 owing to multivisceral transplantation). The baseline characteristics of the study cohort are summarized in Table 1. Of 3098 children listed for first orthotopic heart transplant, the median age was 2 years (IQR 0.3 to 12 years), and the median weight was 12.3 kg (IQR 5 to 38 kg); 1359 (44%) were female, and 1294 (42%) were nonwhite. The primary cardiac diagnosis that led to heart transplant listing was congenital heart disease (CHD) in 1494 (48%), cardiomyopathy in 1186 (38%), and myocarditis in 178 (6%).
Table 1.
Characteristics of Patients by Cardiac Diagnosis
| Variables | CHD (n=1494) | Myocarditis (n=178) | Cardiomyopathy (n=1186) | Other (n=240) | Total (n=3098) | p * |
|---|---|---|---|---|---|---|
| Age, y | 1 (0-7) | 4 (1-12) | 6 (0.8-13) | 8 (0.9-14) | 2 (0.3-12) | <0.001 |
| Weight, kg | 8.2 (3.5-22.3) | 16 (9.1-44.9) | 18.8 (7.6-49.5) | 28.2 (8.2-50.4) | 12.3 (5.1-37.6) | <0.001 |
| Body surface area, m2 | 0.42 (0.23-0.87) | 0.72 (0.44-1.4) | 0.78 (0.39-1.5) | 1.0 (0.41-1.5) | 0.57 (0.3-1.26) | <0.001 |
| Female | 39 | 49 | 49 | 42 | 44 | <0.001 |
| Nonwhite race | 37 | 50 | 47 | 38 | 42 | <0.001 |
| UNOS listing status | <0.001 | |||||
| 1A | 63 | 77 | 56 | 59 | 60 | |
| 1B | 12 | 11 | 15 | 15 | 13 | |
| 2 | 25 | 12 | 29 | 26 | 26 | |
| Blood type, % | 0.80 | |||||
| A | 37 | 38 | 33 | 34 | 35 | |
| AB | 4 | 5 | 4 | 4 | 4 | |
| B | 11 | 10 | 12 | 11 | 11 | |
| O | 49 | 47 | 50 | 51 | 49 | |
| Prostaglandin support, % | 12 | 0 | <1 | 2 | 6 | <0.001 |
| Invasive hemodynamic support, % | <0.001 | |||||
| ECMO support | 23 | 28 | 11 | 16 | 18 | |
| Ventilator support | 31 | 31 | 29 | 33 | 30 | |
| Other support (1A) | 46 | 60 | 51 | 51 | ||
| Inotropic support | 46 | 69 | 54 | 53 | 51 | <0.001 |
| Dialysis | 2 | 2 | 1 | 2 | 2 | 0.46 |
| Creatinine, mg/dL | 0.6 (0.4 to 0.8) | 0.6 (0.4 to 0.9) | 0.6 (0.4 to 0.8) | 0.7 (0.4 to 1.0) | 0.6 (0.4 to 0.8) | <0.001 |
Values represent median (IQR) or percentage
χ2 Test or Kruskal-Wallis test.
Overall, 1874 children (60%) were listed as status 1A, 418 (13%) as status 1B, and 806 (26%) as status 2. Among children listed as status 1A, 570 (30%) were supported with mechanical ventilation, 346 (18%) were supported by extracorporeal membrane oxygenation (ECMO), and 958 (51%) did not receive either type of support. Overall, children listed because of CHD were younger (P<0.001) and smaller (P<0.001 for both weight and body surface area) than children listed because of cardiomyopathy or myocarditis. Children with cardiomyopathy were less likely to be supported with ECMO or mechanical ventilation than children with CHD or myocarditis (P<0.001).
Survival
Among 3098 children listed for heart transplants, 533 (17%) died, 1943 (63%) received a transplant, 252 (8%) were removed from the waiting list because of recovery, and 370 (12%) remained alive on the waiting list on August 3, 2007. Table 2 summarizes the univariate predictors of waiting list mortality.
Table 2.
Univariate Predictors of Waiting List Mortality
| Univariate Predictors | |||
|---|---|---|---|
| Variable | Survived (n=2565) | Died (n=533) | p * |
| Age, y | 3 (0.3-12) | 1 (0.1-8) | |
| Weight, kg | 13.4 (5.5-39.5) | 8.8 (3.7-24.8) | |
| Weight categories | <0.001 | ||
| <10 kg | 42 | 54 | |
| 10-19 kg | 17 | 17 | |
| 20-39 kg | 16 | 12 | |
| 40-59 kg | 14 | 8 | |
| >60 kg | 11 | 9 | |
| Female | 44 | 44 | 0.74 |
| Nonwhite race | 40 | 51 | <0.001 |
| UNOS listing status | <0.001 | ||
| 1A | 58 | 74 | |
| 1B | 15 | 8 | |
| 2 | 28 | 18 | |
| Cardiac diagnosis, % | <0.001 | ||
| CHD | 45 | 64 | |
| Cardiomyopathy | 41 | 23 | |
| Myocarditis | 6 | 5 | |
| Other | 8 | 8 | |
| Blood type | 0.10 | ||
| A | 37 | 27 | |
| O | 48 | 57 | |
| B | 11 | 11 | |
| AB | 4 | 4 | |
| Prostaglandin support | 6 | 9 | <0.001 |
| Invasive hemodynamic support | <0.001 | ||
| ECMO support | 16 | 28 | |
| Ventilator support | 29 | 37 | |
| Other support (1A) | 55 | 36 | |
| Inotropic support | 50 | 57 | <0.001 |
| Dialysis | 1 | 4 | <0.001 |
| Creatinine, mg/dL | 0.6 (0.4-0.8) | 0.6 [0.4-0.9] | |
| GFR <50 mL · min−11.73 m−2 | 17 | 33 | <0.001 |
| Year of listing | 0.24 | ||
| 1999-2002 | 52 | 58 | |
| 2003-2006 | 48 | 42 | |
Values represent median (IQR) or percentage. GFR indicates glomerular filtration rate.
Log-rank test.
Table 3 summarizes the multivariable predictors of waiting list mortality. Among all children listed, independent predictors of waiting list mortality included ECMO support, ventilator support, CHD, listing status 1A, dialysis support, and nonwhite race. A glomerular filtration rate <50 mL • min−1 ∙ 1.73 m−2 was also found to be an independent predictor of mortality but was collinear with dialysis and thus was not included in the final model. Age, weight, and body surface area were not statistically significant predictors of waiting list mortality after adjustment for other covariates in the model. All of the variables in the final model remained statistically significant after adjustment for region and year of transplantation.
Table 3.
Multivariate Predictors of Waiting List Mortality*
| Variable | Adjusted HRs | |||
|---|---|---|---|---|
| All Patients | Status 1A Only | |||
| HR (95% CI) | P | HR (95% CI) | P | |
| ECMO | 3.1 (2.4–3.9) | <0.001 | 3.0 (2.3–3.8) | <0.001 |
| Ventilator support | 1.9 (1.6–2.4) | <0.001 | 1.9 (1.5–2.4) | <0.001 |
| Cardiac diagnosis of CHD | 2.2 (1.8–2.6) | <0.001 | 2.1 (1.7–2.6) | <0.001 |
| Dialysis | 1.9 (1.2–3.0) | 0.006 | 2.0 (1.3–3.2) | 0.004 |
| UNOS listing status 1A | 2.2 (1.7–2.7) | <0.001 | ... | |
| Nonwhite race, % | 1.7 (1.4–2.0) | <0.001 | 1.7 (1.4–2.0) | <0.001 |
| Year of listing 1999–2002 | 1.2 (1.0–1.5) | 0.040 | ||
Cox proportional hazards model.
Because a large majority of the 533 deaths occurred among children listed as status 1A, we performed a secondary analysis to determine the risk factors associated with mortality among children listed as status 1A (Table 3). Except for listing year (1999 to 2002 versus 2003 to 2006), which became significant in the 1A subgroup analysis, the multivariate predictors, hazard ratios, and 95% confidence intervals were essentially unchanged compared with the overall analysis.
Figure 1 shows the estimated survival for all children listed for heart transplant according to UNOS listing status (Figure 1A) and for all children listed as status 1A according to the level of invasive hemodynamic support (Figure 1B). No appreciable difference was found in overall waiting list mortality for patients listed as status 2 versus status 1B (11.7% versus 10.5%). By contrast, among patients listed as status 1A, a substantial difference was found in risk of waiting list mortality based on the level of invasive hemodynamic support (ie, required ECMO, mechanical ventilation, or neither).
Figure 1.
Kaplan-Meier survival for all children listed for heart transplant according to listing status (A) and for those children listed as status 1A according to invasive hemodynamic support(B).
Status 1A Risk Stratification
Table 4 summarizes the observed waiting list mortality of status 1A patients with risk stratification by subgroup. Among children listed as UNOS status 1A, a 7-fold difference was found in the 90-day risk of death on the waiting list on the basis of patient characteristics, with a range from 5% to 39%. The 14-and 30-day waiting list mortality variation for patients listed as status 1A was even more striking. For example, a child weighing <10 kg who was supported on ECMO for CHD (n=155) had a 12-fold higher risk of death by 14 days without transplantation (20.7% versus 1.5%) and an 8-fold higher risk of death by 30 days without transplantation (32% versus 4%) than a child weighing >10 kg with cardiomyopathy who was supported with inotropes alone (n=263). Figure 2 ⇓⇓ shows the competing outcomes for wait-listed children based on UNOS status at the time of listing and competing outcomes for children listed as status 1A according to their level of invasive hemodynamic support.
Table 4.
Risk Stratification of Status 1A Candidates Based on Observed Waiting List Mortality for Patient Subgroups
| 7 Days | 14 Days* | 30 Days | 60 Days | 90 Days | 6 Months | Overall | |
|---|---|---|---|---|---|---|---|
| All status 1A candidates (n=1874) | 4.9 | 8.5 | 13.7 | 18.0 | 18.9 | 20.3 | 21.1 |
| ECMO (n=346) | 10.1 | 17.9 | 25.7 | 29.8 | 30.4 | 30.6 | 31.5 |
| Cardiomyopathy | 5.7 | 11.4 | 15.7 | 21.4 | 21.4 | 21.4 | 24.3 |
| Weight >10 kg (n=34) | 0 | 8.8 | 14.7 | 23.5 | 23.5 | 23.5 | 26.5 |
| Weight <10 kg (n=36) | 11.1 | 13.9 | 16.7 | 19.4 | 19.4 | 19.4 | 22.2 |
| Myocarditis | 7.7 | 12.8 | 20.5 | 23.1 | 23.1 | 25.6 | 25.6 |
| Weight >10 kg (n=26) | 3.9 | 7.7 | 15.4 | 15.4 | 15.4 | 19.2 | 19.2 |
| Weight <10 kg (n=13) | 15.4 | 23.1 | 30.8 | 38.5 | 38.5 | 38.5 | 38.5 |
| CHD | 11.2 | 19.6 | 28.5 | 32.7 | 33.6 | 33.6 | 33.6 |
| Weight ≥10 kg (n=56) | 5.4 | 14.3 | 17.9 | 21.4 | 25.0 | 25.0 | 25.0 |
| Weight <10 kg (n=155) | 13.6 | 20.7 | 31.6 | 36.1 | 36.1 | 36.1 | 36.1 |
| Mechanical ventilation (n=570) | 6.0 | 10.4 | 17.0 | 22.1 | 23.5 | 25.1 | 25.4 |
| Cardiomyopathy | 4.6 | 8.3 | 16.0 | 17.0 | 18.0 | 19.6 | 19.6 |
| Weight ≥10 kg (n=69) | 10.1 | 17.4 | 27.5 | 29.0 | 29.0 | 29.0 | 29.0 |
| <10 kg (n=122) | 1.6 | 3.3 | 9.0 | 9.8 | 11.5 | 13.9 | 13.9 |
| Myocarditis | 2.4 | 7.1 | 7.1 | 9.5 | 9.5 | 9.5 | 11.5 |
| Weight ≥10 kg (n=20) | 5.0 | 15.0 | 15.0 | 15.0 | 15.0 | 15.0 | 20.0 |
| <10 kg (n=22) | 0 | 0 | 0 | 4.6 | 4.6 | 4.6 | 4.6 |
| CHD | 7.3 | 12.5 | 19.8 | 27.4 | 29.2 | 30.9 | 31.3 |
| Weight ≥10 kg (n=55) | 9.1 | 14.6 | 16.4 | 25.5 | 27.3 | 27.3 | 27.3 |
| Weight <10 kg (n=231) | 6.9 | 12.1 | 20.4 | 27.7 | 29.4 | 31.6 | 32.0 |
| No ventilation or ECMO (n=958) | 2.3 | 4.1 | 7.4 | 11.3 | 12.1 | 13.8 | 14.7 |
| Cardiomyopathy | 1.3 | 2.3 | 4.3 | 6.3 | 6.8 | 7.1 | 7.6 |
| Weight ≥10 kg (n=263) | 0.8 | 1.5 | 3.8 | 4.9 | 5.7 | 5.7 | 6.1 |
| Weight <10 kg (n=127) | 2.4 | 3.9 | 5.5 | 9.5 | 9.5 | 10.2 | 11.0 |
| Myocarditis | 1.8 | 3.6 | 7.1 | 10.7 | 10.7 | 12.5 | 14.3 |
| Weight ≥10 kg (n=45) | 2.2 | 4.4 | 6.7 | 11.1 | 11.1 | 13.3 | 15.6 |
| Weight <10 kg (n=10) | 0 | 0 | 10.0 | 10.0 | 10.0 | 10.0 | 10.0 |
| CHD | 3.5 | 5.8 | 10.8 | 16.6 | 18.0 | 21.0 | 22.1 |
| Weight <10 kg, no PGE (n=159) | 1.9 | 4.4 | 8.2 | 12.6 | 14.5 | 16.4 | 17.0 |
| Weight ≥10 kg (n=161) | 4.4 | 6.8 | 13.0 | 18.6 | 19.9 | 24.2 | 24.8 |
| Weight <10 kg, PGE (n=107) | 3.7 | 5.6 | 11.2 | 19.6 | 20.6 | 23.4 | 26.2 |
Values represent percentage of eligible patients who were removed from the waiting list due to death during the specified time frame. PGE indicates prostaglandin E infusion.
14 Days is the standardized time interval for status 1A justification by UNOS.
Figure 2.
Competing outcomes for wait-listed children according to UNOS status at the time of listing (A, B, and C), and for children listed as status 1A according to their level of invasive hemodynamic support: ECMO (D), mechanical ventilation (E), and neither ECMO nor mechanical ventilation (F).
Among those listed as status 1A, the following subgroups of children were found to be at 30% or greater risk of waiting list mortality based on observed mortality (Table 4): (1) Children weighing <10 kg who were listed because of CHD and who required mechanical ventilation (mortality 32%, n=231), and (2) children weighing <10 kg who were listed because of CHD and who required ECMO (mortality 36.1%, n=155). Children with a predicted risk of waiting list mortality of ≥20% included most children listed as status 1A for CHD and most children who required either mechanical ventilation (waiting list mortality 25%, n=570) or ECMO (waiting list mortality 31.5%, n=346).
Figure 3 summarizes the weight distribution of children who died while on the waiting list in the current era. Overall, 10% of patients weighed <3 kg, 34% weighed <5 kg, 54% weighed <10 kg, 64%
Figure 3.
Number of children who died while on the waiting list according to weight at listing (n=533). The numbers above each bar denote the total number of children who were listed for heart transplant within each weight category.
Discussion
In this study, we found that over a 6-year period, 533 US children with severe heart failure died while on the heart transplant waiting list before a suitable donor heart could be identified. Expressed as a rate, children awaiting heart transplantation experience the single highest waiting list mortality compared with all other age groups and all other solid organs in transplant medicine.1 Although the average status 1A pediatric patient is at higher risk of waiting list mortality statistically, status 1A patients as a group represent a large and heterogeneous population whose risk of waiting list mortality varies by as much as 10-fold or more based on patient-specific factors. The single most important patient factor predictive of waiting list mortality is the level of invasive hemodynamic support, as defined by ECMO versus mechanical ventilation versus inotropic support alone. Other patient factors associated with waiting list mortality include cardiac diagnosis, dialysis, and nonwhite race/ethnicity.
These findings are consistent with earlier reports from the 1990s that found that ECMO, former listing as status 1 (predecessor of the 1A/1B classification system), and CHD were associated with waiting list mortality in children5–7,9,10,13; however, no studies have analyzed waiting list outcomes since the pediatric heart-allocation system was revised in 1999. Consequently, the present report has 2 advantages over earlier reports in that (1) it analyzes outcomes since 1999, which permits a focused look at waiting list mortality under the present allocation system and practice conditions, and (2) it captures all children officially listed for a heart transplant in the United States, which provides the necessary statistical power to identify several important national trends for the first time. Specifically, this is the first published report (1) to identify nonwhite race and mechanical ventilation as powerful independent risk factors for waiting list mortality across the pediatric age spectrum, (2) to describe the striking variability in waiting list mortality observed among children listed as status 1A, and (3) to exclude blood type as an independent factor associated with waiting list mortality in the current era.
Our finding that the level of invasive hemodynamic support (ie, ECMO support versus mechanical ventilation versus inotropic support alone) is associated with waiting list survival is reasonably intuitive; however, we were surprised at the magnitude of effect, specifically, that the level of invasive hemodynamic support appears to be a much stronger predictor of waiting list mortality and therefore a more accurate reflection of medical urgency than UNOS listing status itself, the current system used to categorize children according to medical urgency.2,15 We believe the relatively poor correlation between UNOS listing status and medical urgency in pediatrics stems largely from the heterogeneity of the status 1A patient cohort resulting in greater waiting list mortality variability within UNOS listing groups than between listing groups. This heterogeneity is likely driven by 2 factors: (1) Greater numbers of high-risk children are listed as status 1A because of more widespread use of technologies like ECMO16 in pediatrics, combined with fewer absolute contraindications to transplant,17 and (2) greater numbers of low-risk patients are listed as status 1A because of less stringent status 1A criteria for children. (NB: Adult status 1A criteria generally necessitate pulmonary artery catheter placement, whereas pediatric status 1A criteria can be met with as little as minor dosing adjustments in intravenous medications.) It is likely that the less stringent status 1A criteria in pediatrics are largely responsible for the disproportionately large number of children who qualify for the highest tier of medical urgency (>60% of children at the time of listing and nearly three quarters [72%] by the time of transplantation).
The findings of the present study have several implications. First, our findings raise questions about whether the current allocation system is structured optimally to reduce pediatric transplant mortality. Because the current system captures medical urgency poorly, children facing markedly different short-term mortality risks are forced to compete directly for the same scarce donor organs. For example, under the current system, a child actively listed for heart transplant who is on ECMO support with days to live must compete directly with a child being supported by inotropes alone, who, according to our data, has a relatively low imminent risk of death. Consequently, an available heart is offered first to the child who has accumulated more status 1A wait time, rather than to the child who is likely to die without transplant. Ultimately, because the majority of pediatric patients and virtually all at-risk pediatric patients are listed as status 1A, “first come, first served” has functionally supplanted medical urgency as the primary determinant of pediatric donor heart allocation for the majority of children awaiting heart transplantation.
The discrepancy between medical urgency and waiting list seniority, a major problem in solid-organ transplantation historically, raises the possibility that some pediatric deaths could be prevented by moving away from an allocation system that relies heavily on waiting time toward an allocation system that better reflects medical urgency while also incorporating potential transplant benefit. Such a shift would be consistent with the Final Rule on Organ Allocation published by the US Department of Health and Human Services,15 which discourages the use of waiting time as a criterion for organ allocation and encourages organ sharing based on medical urgency (while avoiding futile transplantations) with the use of objective medical criteria that cannot be manipulated by patients or physicians. Accordingly, the US allocation systems for liver and lung allografts18–23 have undergone major revisions recently to bring them into compliance with contemporary standards. We believe a similar reappraisal is warranted for pediatric hearts, not only because of the excessively high waiting list mortality1 but also because of better data on the factors that drive pretransplantation and posttransplantation attrition.24,25 Because some factors such as ECMO may be associated with both pretransplantation5–7,9,10 and posttransplantation24,25 mortality, revising the current system will require complex simulation analyses to balance competing risks, as has been done successfully with the lung-allocation score recently.26
Second, the present findings suggest that specific high-risk subgroups of patients can be identified who may be suitable candidates for emerging pediatric mechanical support devices. By the same token, our findings suggest that low-risk subgroups can also be identified for whom investigational devices should generally not be used at the present time. Because of this heterogeneity, clinicians and investigators will need to use caution in selecting patients for evolving mechanical circulatory support devices26,27 and in developing selection criteria for clinical trials to obtain regulatory approval.28,29 For example, patients with an estimated waiting list mortality of less than 10% (eg, stable children with cardiomyopathy who are on inotropes) are unlikely to benefit from device therapy if the device itself carries a risk that could be higher. The use of such devices in such patients could not only expose children to unnecessary risks but could also undermine the interpretability of data in support of a regulatory claim of efficacy or probable benefit, the legal threshold for Food and Drug Administration approval in the United States.
Third, although age and size were not independently associated with waiting list mortality, the present findings indicate that the vast majority of children dying while on the waiting list weigh <10 to 15 kg (Figure 3). The skewed weight distribution draws needed attention to precisely where the national organ shortage for pediatric donor hearts is most critical: among infants and toddlers. Creating greater public awareness is a key first step for organ-donation advocates who are looking for high-impact strategies to reduce pediatric waiting list mortality.30 Although it would be beneficial to improve organ donation among children of all ages and sizes, the present findings suggest that the greatest benefit would come from a successful campaign to increase organ donation among infants and toddlers. The disproportionate number of deaths among infants and toddlers further underscores the need to develop reliable miniaturized mechanical circulatory support devices for infants and smaller children,26,27 similar to the approved ventricular assist devices that are widely available for larger children and adults.31–34
We were surprised to find that nonwhite race/ethnicity was associated with waiting list mortality, particularly among children listed as status 1A, a group that is uniformly hospitalized and usually under the watchful eye of intensivists. Contributing factors may include differences in timing of presentation, access to care, delivery of medical treatment, disease progression, regional heterogeneity, misclassification of race/ethnicity by centers, or some combination thereof. Adult studies of waiting list mortality have reported mixed findings on the relationship between race/ethnicity and waiting list mortality.13,35 Further research is needed to explore the effect of race/ethnicity on waiting list mortality in transplant candidates across all ages.
The findings of the present study should be interpreted within the context of the study design. First, the analysis did not account for changes in listing status while patients were on the waiting list; however, changes in status would be expected to result in misclassification of risk factor assignment, which would lead to an underestimate of the true effects of a given risk factor, which in this analysis were all highly significant. Second, the primary analysis did not account for patients who were removed from the waiting list because of clinical deterioration rather than death; however, secondary analyses using the combined outcome of death or delisting due to clinical deterioration yielded similar results. Lastly, all retrospective studies are inherently susceptible to selection bias that could skew findings if a nonrandom population of patients were selected for analysis; however, because the Scientific Registry of Transplant Recipients captures all patients officially listed for transplant in the United States, it is unlikely that patient selection bias would play a major role in the findings of this retrospective analysis.
In summary, despite improvements in pediatric heart allocation over the past decade, pediatric heart transplant waiting list mortality remains unacceptably high in the current era and is an outlier in transplant medicine. The current pediatric heart-allocation system captures medical urgency poorly, which raises the possibility that the current allocation system may not be prioritizing scarce donor hearts optimally. Although status 1A patients are at higher risk of waiting list mortality statistically, status 1A patients as a group represent a large and heterogeneous population. Independent risk factors for waiting list mortality can be used to risk-stratify children, which can help facilitate patient selection of emerging pediatric cardiac assist devices and guide pediatric donor allocation in a manner that is consistent with contemporary organ-allocation standards. Lastly, most children who die while on the waiting list are those who weigh <10 kg, which underscores the tremendous need for reliable pediatric mechanical support devices for the smallest children. Targeted efforts to expand infant donation through expanded neonatal intensive care unit donation or more widespread acceptance of donation after cardiac death36 are urgently needed.
CLINICAL PERSPECTIVE.
Despite improvements in pediatric heart allocation over the past decade, children listed for heart transplantation face the highest waiting list mortality in solid-organ transplant medicine. Data on waiting list outcomes since the pediatric heart-allocation system was revised in 1999 are limited. This study examines waiting list outcomes from all 3098 children <18 years of age listed in the United States for primary heart transplant during the period from 1999 to 2006. Overall, 533 children (17%) died, whereas 63% received transplants and 8% recovered. Although status 1A patients were at higher risk of waiting list mortality than status 1B or status 2 patients, waiting list mortality varied by a greater degree within status 1A and was best predicted by the level of invasive hemodynamic support (defined as extracorporeal membrane oxygenation versus ventilator versus neither). The study thus demonstrates that the current pediatric heart-allocation system captures medical urgency among those waiting for a heart poorly. Because patients on a higher level of invasive support at listing may also be at higher risk of posttransplantation death, further research is needed to determine what changes in the current pediatric allocation system will reduce overall (pretransplantation and posttransplantation) mortality in children listed for a heart transplant. Lastly, the study demonstrates that the vast majority of children who die on the waiting list weigh <10 to 15 kg, which underscores the need to develop and refine new technologies to support the smallest children with advanced heart failure and to expand opportunities for infant organ donation.
Acknowledgments
The authors wish to thank Drs John E. Mayer, Heung B. Kim, and Leslie B. Smoot for thoughtful review of the manuscript. The data reported here are based on Organ Procurement Transplant Network data as of August 3, 2007.
Sources of Funding
This study was supported in part by the Alexia Clinton Family Fund and by the National Institutes of Health under award number: T32HL007572. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Disclosures
None.
References
- 1.McDiarmid S. Death on the pediatric waiting list: scope of the problem.. Paper presented at: Summit on Organ Donation and Transplantation; San Antonio, Tex.. March 2007. [Google Scholar]
- 2.Renlund DG, Taylor DO, Kfoury AG, Shaddy RS. New UNOS rules: historical background and implications for transplantation management: United Network for Organ Sharing. J Heart Lung Transplant. 1999;18:1065–1070. doi: 10.1016/s1053-2498(99)00075-3. [DOI] [PubMed] [Google Scholar]
- 3.West LJ, Pollock-Barziv SM, Dipchand AI, Lee KJ, Cardella CJ, Benson LN, Rebeyka IM, Coles JG. ABO-incompatible heart transplantation in infants. N Engl J Med. 2001;344:793–800. doi: 10.1056/NEJM200103153441102. [DOI] [PubMed] [Google Scholar]
- 4.West LJ, Karamlou T, Dipchand AI, Pollock-BarZiv SM, Coles JG, McCrindle BW. Impact on outcomes after listing and transplantation, of a strategy to accept ABO blood group-incompatible donor hearts for neonates and infants. J Thorac Cardiovasc Surg. 2006;131:455–461. doi: 10.1016/j.jtcvs.2005.09.048. [DOI] [PubMed] [Google Scholar]
- 5.Addonizio LJ, Naftel D, Fricker J, Morrow WR, Kirklin JK, McGiffin DC, Dodd D, Bernstein D. Risk factors for pretransplant outcome in children listed for cardiac transplantation: a multi-institutional study. J Heart Lung Transplant. 1995;14:S48. Abstract. [Google Scholar]
- 6.McGiffin DC, Naftel DC, Kirklin JK, Morrow WR, Towbin J, Shaddy R, Alejos J, Rossi A. Pediatric Heart Transplant Study Group. Predicting outcome after listing for heart transplantation in children: comparison of Kaplan-Meier and parametric competing risk analysis. J Heart Lung Transplant. 1997;16:713–722. [PubMed] [Google Scholar]
- 7.Morrow WR, Naftel D, Chinnock R, Canter C, Boucek M, Zales V, McGiffin DC, Kirklin JK. Pediatric Heart Transplantation Study Group. Outcome of listing for heart transplantation in infants younger than six months: predictors of death and interval to transplantation. J Heart Lung Transplant. 1997;16:1255–1266. [PubMed] [Google Scholar]
- 8.Morrow WR, Frazier E, Naftel DC. Survival after listing for cardiac transplantation in children. Prog Pediatr Cardiol. 2000;11:99–105. doi: 10.1016/s1058-9813(00)00041-2. [DOI] [PubMed] [Google Scholar]
- 9.Nield LE, McCrindle BW, Bohn DJ, West LJ, Coles JG, Freedom RM, Benson LN. Outcomes for children with cardiomyopathy awaiting transplantation. Cardiol Young. 2000;10:358–366. doi: 10.1017/s1047951100009665. [DOI] [PubMed] [Google Scholar]
- 10.Mital S, Addonizio LJ, Lamour JM, Hsu DT. Outcome of children with end-stage congenital heart disease waiting for cardiac transplantation. J Heart Lung Transplant. 2003;22:147–153. doi: 10.1016/s1053-2498(02)00670-8. [DOI] [PubMed] [Google Scholar]
- 11.Feingold B, Bowman P, Zeevi A, Girnita AL, Quivers ES, Miller SA, Webber SA. Survival in allosensitized children after listing for cardiac transplantation. J Heart Lung Transplant. 2007;26:565–571. doi: 10.1016/j.healun.2007.03.015. [DOI] [PubMed] [Google Scholar]
- 12.Pollock-BarZiv SM, McCrindle BW, West LJ, Manlhiot C, VanderVliet M, Dipchand AI. Competing outcomes after neonatal and infant wait-listing for heart transplantation. J Heart Lung Transplant. 2007;26:980–985. doi: 10.1016/j.healun.2007.07.024. [DOI] [PubMed] [Google Scholar]
- 13.Chen JM, Weinberg AD, Rose EA, Thompson SM, Mancini DM, Ellison JP, Reemtsma K, Michler RE. Multivariate analysis of factors affecting waiting time to heart transplantation. Ann Thorac Surg. 1996;61:570–575. doi: 10.1016/0003-4975(95)01031-9. [DOI] [PubMed] [Google Scholar]
- 14.Schwartz GJ, Haycock GB, Edelmann CM, Jr, Spitzer A. A simple estimate of glomerular filtration rate in children derived from body length and plasma creatinine. Pediatrics. 1976;58:259–263. [PubMed] [Google Scholar]
- 15.63 Federal Register 16295, at 16332 1998 (codified at 42 CFR part 121) [Google Scholar]
- 16.Conrad SA, Rycus PT, Dalton H. Extracorporeal Life Support Registry Report 2004. ASAIO J. 2005;51:4–10. doi: 10.1097/01.mat.0000151922.67540.e9. [DOI] [PubMed] [Google Scholar]
- 17.Canter CE, Shaddy RE, Bernstein D, Hsu DT, Chrisant MR, Kirklin JK, Kanter KR, Higgins RS, Blume ED, Rosenthal DN, Boucek MM, Uzark KC, Friedman AH, Young JK. Indications for heart transplantation in pediatric heart disease: a scientific statement from the American Heart Association Council on Cardiovascular Disease in the Young; the Councils on Clinical Cardiology, Cardiovascular Nursing, and Cardiovascular Surgery and Anesthesia; and the Quality of Care and Outcomes Research Interdisciplinary Working Group. Circulation. 2007;115:658–676. doi: 10.1161/CIRCULATIONAHA.106.180449. [DOI] [PubMed] [Google Scholar]
- 18.Austin MT, Poulose BK, Ray WA, Arbogast PG, Feurer ID, Pinson CW. Model for end-stage liver disease: did the new liver allocation policy affect waiting list mortality? Arch Surg. 2007;142:1079–1085. doi: 10.1001/archsurg.142.11.1079. [DOI] [PubMed] [Google Scholar]
- 19.Egan TM, Kotloff RM. Pro/con debate: lung allocation should be based on medical urgency and transplant survival and not on waiting time. Chest. 2005;128:407–415. doi: 10.1378/chest.128.1.407. [DOI] [PubMed] [Google Scholar]
- 20.Freeman RB, Jr, Edwards EB. Liver transplant waiting time does not correlate with waiting list mortality: implications for liver allocation policy. Liver Transpl. 2000;6:543–552. doi: 10.1053/jlts.2000.9744. [DOI] [PubMed] [Google Scholar]
- 21.Levine GN, McCullough KP, Rodgers AM, Dickinson DM, Ashby VB, Schaubel DE. Analytical methods and database design: implications for transplant researchers, 2005. Am J Transplant. 2006;6:1228–1242. doi: 10.1111/j.1600-6143.2006.01277.x. [DOI] [PubMed] [Google Scholar]
- 22.Rosen HR, Prieto M, Casanovas-Taltavull T, Cuervas-Mons V, Guckelberger O, Muiesan P, Strong RW, Bechstein WO, O'Grady J, Zaman A, Chan B, Berenguer J, Williams R, Heaton N, Neuhaus P. Validation and refinement of survival models for liver retransplantation. Hepatology. 2003;38:460–469. doi: 10.1053/jhep.2003.50328. [DOI] [PubMed] [Google Scholar]
- 23.Travaline JM, Cordova FC, Furukawa S, Criner GJ. Discrepancy between severity of lung impairment and seniority on the lung transplantation list. Transplant Proc. 2004;36:3156–3160. doi: 10.1016/j.transproceed.2004.11.009. [DOI] [PubMed] [Google Scholar]
- 24.Boucek MM, Aurora P, Edwards LB, Taylor DO, Trulock EP, Christie J, Dobbels F, Rahmel AO, Keck BM, Hertz MI. Registry of the International Society for Heart and Lung Transplantation: tenth official pediatric heart transplantation report: 2007. J Heart Lung Transplant. 2007;26:796–807. doi: 10.1016/j.healun.2007.06.006. [DOI] [PubMed] [Google Scholar]
- 25.Davies RR, Russo MJ, Mital S, Martens TM, Sorabella RS, Hong KN, Gelijns AC, Moskowitz AJ, Quaegebeur JM, Mosca RS, Chen JM. Predicting survival among high-risk pediatric cardiac transplant recipients: an analysis of the United Network for Organ Sharing database. J Thorac Cardiovasc Surg. 2008;135:147–155. doi: 10.1016/j.jtcvs.2007.09.019. [DOI] [PubMed] [Google Scholar]
- 26.Duncan BW. Pediatric mechanical circulatory support in the United States: past, present, and future. ASAIO J. 2006;52:525–529. doi: 10.1097/01.mat.0000237588.44529.d1. [DOI] [PubMed] [Google Scholar]
- 27.Baldwin JT, Borovetz HS, Duncan BW, Gartner MJ, Jarvik RK, Weiss WJ, Hoke TR. The National Heart, Lung, and Blood Institute Pediatric Circulatory Support Program. Circulation. 2006;113:147–155. doi: 10.1161/CIRCULATIONAHA.105.571422. [DOI] [PubMed] [Google Scholar]
- 28.Almond CS, Chen EA, Berman MR, Less JR, Baldwin JT, Linde-Feucht SR, Hoke TR, Pearson GD, Jenkins K, Duncan BW, Zuckerman BD. High-risk medical devices, children and the FDA: regulatory challenges facing pediatric mechanical circulatory support devices. ASAIO J. 2007;53:4–7. doi: 10.1097/01.mat.0000247958.84788.3a. [DOI] [PubMed] [Google Scholar]
- 29.Rinaldi JE, Chen EA, Berman MR. Pediatric circulatory support: an FDA perspective. ASAIO J. 2005;51:533–535. doi: 10.1097/01.mat.0000181508.26853.dc. [DOI] [PubMed] [Google Scholar]
- 30.Organ Donation Breakthrough Collaborative [November 1, 2008];About the collaborative. Available at: www.organdonationnow.org.
- 31.Blume ED, Naftel DC, Bastardi HJ, Duncan BW, Kirklin JK, Webber SA. Outcomes of children bridged to heart transplantation with ventricular assist devices: a multi-institutional study. Circulation. 2006;113:2313–2319. doi: 10.1161/CIRCULATIONAHA.105.577601. [DOI] [PubMed] [Google Scholar]
- 32.Frazier OH, Rose EA, Oz MC, Dembitsky W, McCarthy P, Radovancevic B, Poirier VL, Dasse KA. Multicenter clinical evaluation of the HeartMate vented electric left ventricular assist system in patients awaiting heart transplantation. J Thorac Cardiovasc Surg. 2001;122:1186–1195. doi: 10.1067/mtc.2001.118274. [DOI] [PubMed] [Google Scholar]
- 33.Hill JD, Reinhartz O. Clinical outcomes in pediatric patients implanted with Thoratec ventricular assist device. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu. 2006:115–122. doi: 10.1053/j.pcsu.2006.02.019. [DOI] [PubMed] [Google Scholar]
- 34.Miller LW, Pagani FD, Russell SD, John R, Boyle AJ, Aaronson KD, Conte JV, Naka Y, Mancini D, Delgado RM, MacGillivray TE, Farrar DJ, Frazier OH. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med. 2007;357:885–896. doi: 10.1056/NEJMoa067758. [DOI] [PubMed] [Google Scholar]
- 35.Lietz K, Miller LW. Improved survival of patients with end-stage heart failure listed for heart transplantation: analysis of organ procurement and transplantation network/U.S. United Network of Organ Sharing data, 1990 to 2005. J Am Coll Cardiol. 2007;50:1282–1290. doi: 10.1016/j.jacc.2007.04.099. [DOI] [PubMed] [Google Scholar]
- 36.Boucek MM, Mashburn C, Dunn SM, Frizell R, Edwards L, Pietra B, Campbell D. Pediatric heart transplantation after declaration of cardiocirculatory death. N Engl J Med. 2008;359:709–714. doi: 10.1056/NEJMoa0800660. [DOI] [PubMed] [Google Scholar]