If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The use of mechanical circulatory support as a bridge to transplantation in pediatric patients: An analysis of the United Network for Organ Sharing database
Division of Cardiothoracic Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
International Center for Health Outcomes and Innovation Research, Department of Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
International Center for Health Outcomes and Innovation Research, Department of Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
Division of Cardiothoracic Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
Division of Cardiothoracic Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
International Center for Health Outcomes and Innovation Research, Department of Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
International Center for Health Outcomes and Innovation Research, Department of Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
Department of Pediatrics (Cardiology), Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
Division of Cardiothoracic Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
Address for reprints: Jonathan M. Chen, MD, Pediatric Cardiac Surgery, Children's Hospital of New York, 3959 Broadway, Suite 2-273, New York, NY 10032.
Division of Cardiothoracic Surgery, Children's Hospital of New York–Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York
The use of mechanical circulatory support to bridge pediatric patients to cardiac transplantation presents unique challenges because of the difficult anatomy and physiology in these patients.
Methods
The United Network for Organ Sharing provided deidentifed patient-level data. The study population included 2532 transplantations performed on patients less than 19 years old in status 1/1A/1B between 1995 and 2005. Mechanical circulatory support was used in 431 patients: 241 (9.5%) received ventricular assist devices, 171 (6.8%) underwent extracorporeal membrane oxygenation, and 19 (0.8%) received intra-aortic balloon pumps.
Results
Patients supported on ventricular assist devices had similar levels of hospitalization and intensive care use and less need for inotropic support (P < .0002) than had those not needing support. Five- and 10-year posttransplantation survival was better in patients receiving ventricular assist devices and patients not receiving mechanical circulatory support than in patients receiving extracorporeal membrane oxygenation or intra-aortic balloon pumping (P < .0001). Among mechanically supported patients, patients with a body surface area of less than 0.30 (odds ratio, 1.70; 95% confidence interval, 1.18–2.43) and those requiring extracorporeal membrane oxygenation (odds ratio, 1.65; 95% confidence interval, 1.15–2.35) or intra-aortic balloon pumping (odds ratio, 1.91; 95% confidence interval, 1.02–3.56) had higher long-term mortality. The use of a ventricular assist device at transplantation did not predict higher long-term, posttransplantation mortality.
Conclusions
Pediatric patients requiring a pretransplantation ventricular assist device have long-term survival similar to that of patients not receiving mechanical circulatory support. Early survival among patients undergoing extracorporeal membrane oxygenation and infants is poor, reinforcing the need for improvements in device design and physiologic management of infants and neonates.
The use of mechanical devices to bridge adults to heart transplantation has been well established. Despite higher immunologic sensitization in mechanically bridged patients,
Pediatric patients, because of both their smaller size and their often complex anatomy and physiology, present a unique set of challenges that has resulted in a slower adoption of MCS in this population.
Historically, extracorporeal membrane oxygenation (ECMO) has been used to support pediatric patients with end-stage heart failure. Children with heart failure have been supported for significantly longer on ECMO than have adults.
Heretofore, most reports of VAD use in the pediatric population have been anecdotal, and few reports have directly compared the variety of options for MCS in this population.
This report uses data from the United Network for Organ Sharing (UNOS) database to assess posttransplantation outcomes in patients requiring MCS at transplantation. Our goals were to (1) compare the clinical status at transplantation of patients with the various methods of MCS and (2) to identify risk factors for short- and long-term mortality in these groups.
Materials and Methods
Data Collection
UNOS provided deidentified patient-level data from the Thoracic Registry (data source no. 021606-4). Use of these data is consistent with the regulations of our university's institutional review board. Records with incomplete data were excluded from analyses requiring those data points.
Study Population
The study population consists of 2532 transplantations performed on patients less than 19 years of age in status 1/1A/1B between January 1, 1995, and December 31, 2005. Patients were stratified by the presence and type of MCS at the time of transplantation: none, VAD, ECMO, or intra-aortic balloon pump (IABP).
Data Analysis
Data were analyzed by using SAS 9.13 software for Windows (SAS Institute, Cary, NC). The primary outcome was survival; other outcomes were 30-day mortality and in-hospital complications. Continuous variables are reported as means ± standard deviation and were compared by using the Student t test (with the Bonferroni correction). Ordinal variables were compared by using the χ2 test. All P values are 2-sided. Multivariate regression (stepwise, P < .05) was also performed. Kaplan–Meier analysis and Cox proportional hazards regression (stepwise, P < .05) were used for time-to-event analysis; patients without accurate follow-up times were excluded from these analyses. Risk, odds, and hazard ratios are reported, with 95% confidence intervals (CIs) in parentheses. Survival function estimates for strata of the explanatory variables were calculated by using the BASELINE statement of PROC PHREG.
Results
Baseline Demographics and Incidence of MCS Use
There were 1101 female and 1431 male patients; median age was 5 years (range, 0–18 years). Of 2532 patients, 431 (17.0%) required MCS, either VAD (241 [9.5%] patients), ECMO (171 [6.8%] patients), or IABP (19 [0.8%] patients), at transplantation. Baseline demographics and clinical data are shown in Table 1. Over the study period, the total number of transplantations and the percentage of patients requiring MCS have increased (P = .0013, Figure 1).
Table 1Baseline pretransplantation variables by need for and type of mechanical circulatory support
Figure 1Histogram illustrating the number of transplantations (status 1/1A/1B, age ≤18 years) performed by year. Transplantations requiring mechanical support are shown in cross-hatching, and all others are shown in white. The line indicates the percentage of transplantations requiring mechanical support.
Patients receiving ECMO had a higher incidence of postoperative complications before discharge than had those with VADs, including cardiac reoperation (15.4% vs 12.7%; odds ratio [OR], 1.26; 95% CI, 0.71–2.24), need for dialysis (17.2% vs 9.1%; OR, 2.09; 95% CI, 1.16–3.80), drug-treated postoperative infection (48.2% vs 32.1%; OR, 1.97; 95% CI, 1.30–2.97), and noncardiac surgical procedures (30.7% vs 17.2%; OR, 2.13; 95% CI, 1.32–3.44). Among all patients, these complications resulted in high 30-day posttransplantation mortality (Table 2).
Table 2High risk of 30-day mortality with in-hospital complications for pediatric heart transplant recipients (<19 years old) between 1995 and 2005
Predictors of 30-day mortality included age less than 1 year (OR, 1.72; 95% CI, 1.29–2.30), body surface are (BSA) of less than 1.00 m2 (OR, 1.64; 95% CI, 1.18–2.27), a diagnosis of congenital heart disease (OR, 2.15; 95% CI, 1.60–2.88), and poor pretransplantation clinical status as reflected in the need for ECMO (OR, 3.75; 95% CI, 2.53–5.57), prostaglandin E infusion (OR, 1.73; 95% CI, 1.16–2.58), ventilator support (OR, 2.62; 95% CI, 1.94–3.52), intensive care (OR, 1.62; 95% CI, 1.15–2.27), and recent transfusions (OR, 2.05; 95% CI, 1.50–2.81). Donor variables did not contribute to early mortality (data not shown). Multivariate analysis confirmed the poor early survival among patients with poor clinical status (Table 3). Subset analysis was performed of patients without congenital heart disease. In this population the need for ECMO (OR, 2.88; 95% CI, 1.46–5.68) but not the need for a VAD (OR, 1.14; 95% CI, 0.62–2.10) remained strong predictors of early mortality.
Table 3Multivariate regression analysis of risk of 30-day posttransplantation mortality among pediatric heart transplant recipients (<19 years old) between 1995 and 2005
Ten-year survival for all patients was 56.8%. Patients with VADs and those in the group receiving no MCS had better long-term survival (Figure 2); analysis of noncongenital patients was limited by small sample size, but there was a similar trend (see Figure E1). In multivariate analysis both of these groups continued to have better long-term outcomes (Table 4).
Figure 2Kaplan–Meier curve illustrating 10-year survival stratified by the need for and type of mechanical circulatory support. Vertical bars indicate 95% confidence intervals of the survival function at selected time points. The number of patients at risk at each time point is given below the graph (P < .0001).
Table 4Cox proportional hazards regression of posttransplantation mortality among pediatric heart transplant recipients (<19 years old) between 1995 and 2005
Figure 2 demonstrates a late drop off in survival among the patients undergoing VAD between 5 and 10 years after transplantation. Among patients surviving at least 5 years, the primary predictor of death was age greater than 13 years (OR, 4.27; 95% CI, 2.47–7.38). Patients older than 13 years had poor long-term survival (88.0% at 1 year, 66.0% at 5 years, and 44.6% at 10 years) compared with other age groups (83.3%, 71.9%, and 62.1%, respectively; P = .0070)
Patients requiring MCS
Among patients requiring MCS, Cox hazards for poor long-term survival included a BSA of less than 0.30 m2 (hazard ratio [HR], 1.70; 95% CI, 1.18–2.43) and the need for ECMO (HR, 1.65; 95% CI, 1.15–2.35) or IABP (HR, 1.91; 95% CI, 1.02–3.56) at transplantation. Predicted survival curves based on patient size and need for mechanical support demonstrate the significant effect of these factors in determining long-term outcomes (Figure 3) Cox regression of patients without congenital heart disease revealed similar predictors of poor posttransplantation survival (see Table E1).
Figure 3Ten-year survival function estimates by patient size and need for mechanical support.
Primary causes of death among patients requiring VADs did not differ significantly from those among patients not requiring mechanical support (graft failure, 31.3%; infection, 6.0%; cardiac arrest, 7.6%; other cardiovascular causes, 10.5%; multiple organ failure, 6.1%; and noncompliance, 4.6%). In patients with ECMO, death more often resulted from multiple organ failure (16.9%; OR, 2.43; 95% CI, 1.22–4.86).
Discussion
This study demonstrates that the use of VADs in children results in excellent long-term posttransplantation survival. This corroborates the results from the recent study by Blume and colleagues,
with comparable 10-year survival rates of approximately 50% to 60% in all patients, including those with VADs. The survival curves clearly demonstrate that the largest risk of mortality occurs in the immediate postoperative period. In this range outcomes among patients requiring support with VADs were nearly identical to those of patients not requiring support.
The late drop off in survival apparent in the VAD group (Figure 1) appears to be attributable largely to the overrepresentation of older patients (>13 years) in that group. These older patients were at higher risk for transplant atherosclerosis and consequent graft failure over the long term. We speculate that this is largely due to poor compliance with the immunosuppressive regimen in the adolescent population, a phenomenon previously described both in heart transplantation and abdominal organ transplantation.
Although VADs were associated with excellent posttransplantation outcomes, similar outcomes were not seen in patients bridged with ECMO. Previous series have demonstrated the marginal survival associated with the use of ECMO, whether for temporary postcardiotomy support
only 34% of patients could be successfully bridged to transplantation with ECMO. Although slightly better results were reported by the group from Michigan, still only 57% of listed patients survived to transplantation.
Our results demonstrate that the high risk of death associated with the use of ECMO does not end when the device is explanted and a new heart is implanted. Especially in the first 30 days after transplantation, patients undergoing ECMO before the operation had higher rates of end-organ failure (especially renal failure) and incurred a resultant higher mortality rate. Notably, the effect of ECMO on early posttransplantation survival was independent of the cause of heart failure.
After this early period, the survival curves among all 3 groups (those without MCS, ECMO, and VAD) are parallel, and ECMO ceases to be a predictor of poor outcome among patients surviving the initial 30 days. Thus, those patients who reach transplantation with adequate end-organ function and survive the perioperative period appear to have equivalent long-term outcomes, regardless of the need for or type of mechanical support.
The poor clinical and functional status of patients bridged with ECMO can be clearly seen in the high rate of ventilator and inotropic support and in the nearly universal need for intensive care. The high requirement for inotropic support in patients using ECMO illustrates the poor ventricular off loading and marginal hemodynamics provided by venoarterial ECMO. In contrast, patients with VADs had a significantly lower need for inotropes and had hospitalization and intensive care rates nearly identical to those of patients not requiring mechanical support.
Unfortunately, although VAD implantation had better outcomes than those with ECMO independent of patient size, the smallest patients continued to do poorly with mechanical support, whether VAD or ECMO. It is likely that most of the patients with BSAs of less than 0.30 m2 included in the study population were supported by centrifugal pumps rather than pulsatile VADs. Such pumps do not convey the same advantages as the pulsatile VADs, including patient mobility and improved rates of extubation and hospital discharge, and therefore might contribute to the poorer outcomes in this population. These results can improve with the ongoing development of VADs specifically designed for implantation into the pediatric population.
Limitations
These data have several limitations. First, long-term follow-up (between 5 and 10 years) is only available in a small number of patients, and therefore outcomes in that time period are less amenable to detailed analysis. Second, the limited time points for collection of data in the UNOS registry (at listing, at transplantation, and at follow-up) preclude the analysis of clinical status at the time of device implantation or over the course of mechanical support.
As such, we could not analyze the length of time supported by a particular device, which might have been a particularly strong predictor of end-organ dysfunction and poor outcome. More importantly, without information about clinical status at device implantation, we are unable to answer the most important questions: “Which patients are likely to benefit from mechanical support, and when should it be initiated?” The nature of the UNOS dataset further limits our ability to answer these questions because it does not capture those patients who have a VAD but are not bridged to transplantation (either because they recover or die before listing).
Finally, although data submission at these time points is mandated, the completeness of submitted data is not; therefore several variables that might have been of interest were insufficiently populated and were eliminated from analysis.
Fortunately, several of these limitations, most importantly the question of clinical status at initiation of mechanical support, might soon be addressed by the Interagency Registry for Mechanically Assisted Circulatory Support database.
In conclusion, we have examined the UNOS thoracic organ transplantation registry and evaluated the outcomes of pediatric patients requiring MCS at the time of transplantation. This has shown that long-term posttransplantation survival in patients receiving VADs is similar to that seen in those not requiring mechanical support. In contrast, patients bridged to transplantation with ECMO often have evidence of end-organ dysfunction at transplantation and a correspondingly high mortality, especially in the immediate posttransplantation period. Despite the excellent outcomes overall with the use of VADs, the smallest patients continue to have poor early survival independent of the type of support used, and it remains to be seen whether the newer devices will mitigate this difference.
We thank UNOS for supplying these data and Katarina Anderson for her assistance with our analysis.
Figure E1.
Kaplan–Meier curve shows 10-year survival stratified by the need for and type of mechanical circulatory support among patients without congenital heart disease. Vertical bars indicate 95% confidence intervals of the survival function at selected time points. The number of patients at risk at each time point are given below the graph (P = .0289).
Performed by using forward stepwise selection. Because of the smaller sample size associated with excluding patients with congenital heart disease, selection criteria were broadened to a P value of less than .20.
∗ Performed by using forward stepwise selection. Because of the smaller sample size associated with excluding patients with congenital heart disease, selection criteria were broadened to a P value of less than .20.
01695-9&id=gr1.jpg){kind=link}
01695-9&id=gr2.jpg){kind=link}
01695-9&id=gr3.jpg){kind=link}
01695-9&id=fx1.jpg){kind=link}