Post-transplant Outcomes of Children Bridged to Transplant With the Berlin Heart EXCOR Pediatric Ventricular Assist Device

Transcription

1 Post-transplant Outcomes of Children Bridged to Transplant With the Berlin Heart EXCOR Pediatric Ventricular Assist Device Pirooz Eghtesady, MD, PhD; Christopher S.D. Almond, MD, MPH; Christine Tjossem, BS; Deirdre Epstein, RN; Michiaki Imamura, MD; Mark Turrentine, MD; James Tweddell, MD; Robert D.B. Jaquiss, MD; Charles Canter MD; on behalf of the Berlin Heart Investigators Background Recent data suggest that Berlin Heart EXCOR Pediatric (EXCOR) ventricular assist device improves waiting list survival for pediatric heart transplant candidates. Little is known about their post-transplant outcomes. The aim of this analysis was to determine whether there was a difference in early survival for children bridged to transplant with EXCOR versus status 1A pediatric heart transplant patients not transplanted with ventricular assist device support. Methods and Results Pediatric heart transplant patients (n=106) bridged to transplantation with EXCOR were compared with a similarly aged cohort (n=1021) within the Organ Procurement and Transplant Network (OPTN) database (both cohorts from May 2007 to December 2010). In the EXCOR group, 12-month post-transplant survival (88.7%) was similar to OPTN patients listed status 1A who were not on ventricular assist device support at transplant (89.3%; P=0.85) and significantly better than 12-month survival in OPTN patients on extracorporeal membrane oxygenation at transplant (60.3%; P<0.001). Rejection (50%) was a significantly (P=0.005) higher cause of 12-month post-transplant mortality in the EXCOR compared with the OPTN group. Death after transplant was also higher in EXCOR patients with congenital heart disease compared with those with cardiomyopathy (26.1% versus 7.2%; P=0.02). Post-transplant survival was similar in EXCOR patients with 1 serious adverse event during ventricular assist device support as those without an event during support. Conclusions The 12-month post-transplant survival with EXCOR is comparable with overall pediatric heart transplant survival and superior to survival after extracorporeal membrane oxygenation. Neither adverse events during support nor factors associated with mortality during support influence post-transplant survival. Rejection was a significantly greater cause of post-transplant mortality in EXCOR than in OPTN patients. (Circulation. 2013;128[suppl 1]:S24-S31.) Key Words: heart-assist device heart failure pediatrics surgery transplantation Recently, the Food and Drug Administration granted approval under Humanitarian Device Exemption regulations for use of the Berlin Heart EXCOR Pediatric (EXCOR) ventricular assist device (VAD) 1 based on a prospective, nonrandomized, multicenter, single-arm investigational device exemption (IDE) study assessing the safety and probable benefit of the device. 2 In this trial, EXCOR led to a 90% survival to transplant or weaning off the VAD. Furthermore, the overall rate of serious adverse events was significantly less in the EXCOR group compared with the extracorporeal membrane oxygenation (ECMO) group. 3 What remains unknown is how outcomes in patients who have been bridged to transplantation supported with EXCOR compare with outcomes in children who were transplanted without VAD support. 4 The purpose of this study was to evaluate the outcomes at 1 year after transplant in children bridged with the EXCOR and to compare these outcomes with a contemporaneous group of children in the Organ Procurement and Transplant Network (OPTN) database, who were not on VAD support at the time of transplantation. Methods Study Populations and Data Sources Patients who were transplanted were identified from the EXCOR IDE study 5 database (May 9, 2007, up to the Food and Drug Administration approval of the device on December 16, 2011). The institutional review committee at each center approved this study, and written informed consent was obtained for all data collected by the sponsor, Berlin Heart Inc (The Woodlands, TX). The comparison group for this study (n=1021) was patients aged <16 years undergoing primary orthotopic heart transplantation and not on VAD support, identified from the OPTN database (May 2007 December 2010). All had 1-year follow-up data. A significant From the St. Louis Children s Hospital, St. Louis, MO (P.E., D.E., C.C.); Boston Children s Hospital, Boston, MA (C.S.D.A.); Berlin Heart, Inc, The Woodlands, TX (C.T.); Arkansas Children s Hospital, Little Rock, AR (M.I.); Riley Hospital for Children, Indianapolis, IN (M.T.); Children s Hospital of Wisconsin, Milwaukee, WI (J.T.); and Duke Children s Hospital, Durham, NC (R.D.B.J.). Presented at the 2012 American Heart Association meeting in Los Angeles, CA, November 3 7, Correspondence to Pirooz Eghtesady, MD, PhD, Washington University in St. Louis, Pediatric Cardiothoracic Surgery, Campus Box 8234, St Louis Children s Hospital, One Children s Place, Suite 5S50, St. Louis, MO eghtesadyp@wudosis.wustl.edu 2013 American Heart Association, Inc. Circulation is available at DOI: /CIRCULATIONAHA S24

2 Eghtesady et al Berlin Heart Outcomes After Transplant vs OPTN S25 number of patients in both EXCOR and OPTN groups were diagnosed with congenital heart disease (CHD). These subgroups, as well as those listed as status 1A or 1B, were evaluated. The status 1A designation includes patients who are ventilator dependent, VAD dependent, high-dose inotrope dependent, as well as patients supported with ECMO. Study Aim The aim of this analysis was to determine whether pediatric heart transplant patients bridged with EXCOR have different post-transplant outcomes compared with pediatric heart transplant patients not requiring VAD support as a bridge to transplant. Outcomes were measured 1 year after transplantation. End-organ dysfunction in the EXCOR group was defined as (1) renal dysfunction, estimated by the glomerular filtration rate adjusted for age using the Schwartz formula 6 and categorized as moderately impaired (glomerular filtration rate, percentile) or severely impaired (glomerular filtration rate <30% predicted); and (2) hepatic dysfunction, as reflected by the total serum bilirubin level, categorized as moderately increased ( mg/dl) and severely increased ( 1.2 mg/dl). All clinical and demographic variables for the EXCOR group were collected at the time of EXCOR implantation unless otherwise specified. Statistical Analysis Summary statistics are presented as means and SDs with medians and ranges or number (percent). Comparisons within the EXCOR data or comparisons of EXCOR with OPTN were made using the χ 2 or Fisher exact test (if appropriate) for discrete variables. Kaplan Meier survival data comparisons between EXCOR and OPTN were made, and testing was performed using the log-rank test. Analyses were performed using SAS version 9.2. Results Patient Survival and Mortality Of the 148 patients who received the EXCOR VAD, 113 patients were bridged to transplant, 23 patients died on support, 11 were weaned off the device, and 1 patient was still on VAD support. Of the EXCOR-transplanted patients, 1-year follow-up was available in 106 patients who are the subject of this report. A total of 94 patients from this group were alive, with 12 deaths (11.3%) occurring during this Table 1. Comparison Mortality After transplant With EXCOR Support and in OPTN Patients EXCOR same time interval. Four of the children who died were 1 year of age and 8 were >1-year old. Time to death in the EXCOR patients ranged from 1 to 343 days after transplant (median, days). A comparison of mortality after transplant in the EXCOR and OPTN groups is shown in Table 1. Overall, the outcomes were statistically similar when comparing the number of deaths in the EXCOR group (11.3%; 95% confidence interval [CI]=5.3% 17.4%) with the total number of deaths in the OPTN group (9.5%; 95% CI, 7.7% 11.3%), or the number of deaths in the OPTN patients with status 1A (10.7%; 95% CI, 8.6% 12.8%), or the number of deaths in the OPTN patients on ventilator support (15.9%; 95% CI, 9.5% 22.3%). Notably, however, patients supported with EXCOR did significantly (P<0.001) better than patients bridged to transplant with ECMO, wherein 40% (95% CI, 27.6% 51.8%) died after transplant. Only the OPTN status 1B patients (3.9% mortality; 95% CI, 0.2% 7.6%) did significantly (χ 2 P=0.04) better than those in the EXCOR group. Mortality in patients with CHD supported by EXCOR (26.1%; 95% CI, 8.1% 44.0%) was 10% higher than OPTN subjects with CHD (13.2%; 95% CI, 10.1% 16.3%), as well as OPTN subjects with CHD and status 1A (14.9%; 95% CI, 114.4% 18.5%). However, the outcomes of EXCOR CHD patients with the OPTN cohorts were not statistically significantly different. Cumulative mortality curves for the various groups are shown in Figures 1A, 1B, and 2. In nearly all instances, the most dramatic increase in mortality occurs early, with the steepest increase notable for the ECMO group (reaching a plateau by 75 days after transplant). Risk Factors for Post-transplant Mortality A univariate analysis of potential risk factors for post-transplant mortality in the EXCOR patients (n=12) who died is shown in Table 2. Death after transplant was significantly higher in patients with CHD compared with those with cardiomyopathy or myocarditis (26.1% versus 7.2%, respectively; Fisher exact test P=0.02; Table 2). In the survival analysis, those patients who OPTN Died % (95% CI) Died % (95% CI) EXCOR vs OPTN 12/ ( ) 97/ ( ) 0.55 EXCOR vs OPTN status 1A 12/ ( ) 90/ ( ) 0.85 EXCOR vs OPTN ventilator 12/ ( ) 20/ ( ) 0.32 EXCOR vs OPTN ECMO 12/ ( ) 25/ ( ) <0.001 EXCOR vs OPTN CHD status 1A 12/ ( ) 58/ ( ) 0.34 EXCOR CHD vs OPTN CHD/status 1A 6/ ( ) 58/ ( ) 0.15* EXCOR vs OPTN status 1B 12/ ( ) 4/ ( ) 0.04 EXCOR CHD vs OPTN CHD 6/ ( ) 62/ ( ) 0.11* EXCOR vs OPTN status 1A/no other support 12/ ( ) 46/ ( ) 0.12 Mortality after transplant in children supported with the Berlin Heart EXCOR Pediatric Ventricular Assist Device (n=106) as a bridge to transplant compared with data from patients in the OPTN database (n=1021). Twenty-three patients died while on EXCOR support and 12 in the period 1 y after transplant. Status 1A includes ventilator dependence as well. Status 1A includes ventilator dependence as well as ECMO. CHD indicates congenital heart disease; CI, confidence interval; ECMO, extracorporeal membrane oxygenation; EXCOR, Berlin Heart EXCOR Pediatric Ventricular Assist Device; and OPTN, Organ Procurement and Transplant Network. *Fisher exact test. OPTN data have ventilator, ECMO, or none; this includes only the none group. χ 2 P Value

3 S26 Circulation September 10, 2013 Figure 1. Cumulative mortality in EXCOR and Organ Procurement and Transplant Network (OPTN) patients. Survival of children supported with the Berlin Heart EXCOR Pediatric transplant compared with OPTN data. Kaplan Meier curves and log-rank P values for each comparison are shown. A, EXCOR (n=106) and OPTN (n=1021) P=0.75; (B) EXCOR (n=106) and OPTN status 1A (n=840) P=0.91; (C) EXCOR (n=106) and OPTN on ventilator (n=126) P=0.21; (D) EXCOR (n=106) and OPTN on ECMO (n=63) P< required cardiopulmonary resuscitation within 48 hours before EXCOR implant also had a significantly greater risk of death compared with those who did not (30.0% versus 9.4%; nonsignificant in the univariate analysis Fisher exact test P=0.08). No other factor evaluated proved to be a significant risk factor for post-transplant morality (Table 2). To specifically look at the role of perioperative bleeding in mortality for the CHD group, we looked at the distribution of major bleeding events among EXCOR patients. A major bleeding event occurred in 47 of the 106 EXCOR patients. Of the total, 23 were patients with CHD, and 13 of these (56.5%) had a bleeding event compared with 34 (41.0%) of the remaining 83 non-chd patients. This difference between the 2 groups was not significant (P=0.18). Adverse Events and Impact on Mortality Serious adverse events, as defined by Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) criteria 7 recorded during EXCOR support that were evaluated as risk factors for post-transplant mortality, are summarized in Table 3. Renal dysfunction while on device was the only significant (Fisher exact test P=0.14) event that was more frequent in EXCOR nonsurvivors versus survivors at 12 months after transplant. Occurrence of other serious adverse events, including major bleeding or infection, stroke, or pump change as a result of thrombus, did not have a significant impact on post-transplant mortality. Causes of Post-transplant Mortality A breakdown of the causes of post-transplant mortality in EXCOR (n=12) and OPTN (n=90) patients is shown in Table 4. The EXCOR patients had several major serious adverse events during the IDE trial, ranging from 0 (n=2) to 16 (n=1) events. The most prevalent cause of death after transplant in the EXCOR group was rejection, which accounted for 50.0% of the mortality and was a significantly more frequent (P=0.05) cause of mortality than in the OPTN group (12.2%). The next greatest cause of death in the EXCOR group was cerebrovascular events (16.7%). Infection, nonspecific graft failure, respiratory failure, or unknown causes equally accounted for 8.3% of causes of mortality with EXCOR support. Of the 23 EXCOR patients with CHD, 13 had a major bleed, and 2 of these (15.4%) died after transplant. No major bleeding occurred in 4 of the remaining 10 EXCOR patients with CHD (40.0%) who died after transplant. Differences between these 2 EXCOR patients with CHD were nonsignificant (P=0.34). Discussion The use of VAD to bridge pediatric patients to cardiac transplantation has steadily increased during the past decade. 4 Our study focused on the outcomes of the patients who were bridged effectively to transplantation and demonstrates that the outcomes of patients after transplantation are similar to those not requiring support with a VAD. We found that

4 Eghtesady et al Berlin Heart Outcomes After Transplant vs OPTN S27 Figure 2. Cumulative mortality in EXCOR and Organ Procurement and Transplant Network (OPTN) patients. Survival of children supported with the Berlin Heart EXCOR Pediatric transplant compared with OPTN data. Kaplan Meier curves and log-rank P values for each comparison are shown. A, EXCOR (n=106) and OPTN congenital heart disease (CHD)/status 1A (n=388) P=0.22; (B) EXCOR CHD (n=23) and OPTN CHD/status 1A (n=388) P=0.17; (C) EXCOR (n=106) and OPTN status 1B (n=103) P=0.06; (D) EXCOR CHD (n=23) and OPTN CHD (n=470) P=0.09. pediatric patients supported with EXCOR have post-transplant survival similar to OPTN status 1A patients supported either with inotropic therapy or on ventilator support. This finding is consistent with previous observations in adult studies comparing post-transplant survival between VAD and non-vad ventilator or inotrope-dependent patients. 8 More importantly, children supported with EXCOR have more favorable outcomes compared with those bridged to transplantation with ECMO, which in the past was the primary option for bridging children to transplant. 5 This implies that ECMO support results in inferior outcomes after transplant, consistent with previous reports by Almond and colleagues. 9,10 A recent analysis of a large data set consisting of adult OPTN patients confirmed that all other support pathways besides VAD support, including inotropes, right VAD, and ECMO, increase the risk of post-transplant death. 11 The data presented here show that patients supported with EXCOR have worse outcome than status 1B patients, but the difference is small from clinical perspective. Currently, all pediatric patients supported with a VAD are considered status 1A. The small difference between EXCOR and status 1B patients suggests that perhaps with continued improvements in EXCOR and other pediatric VAD outcomes, status listing for pediatric patients supported with a VAD could change in the future. The majority of post-transplant deaths in EXCOR patients occur early after surgery, suggesting that the perioperative risk of surgery contributes to these outcomes. This is of interest because prior investigations in the adult setting have shown that extracorporeal VADs like EXCOR are associated with higher mortality within the first 6 months after transplantation, an effect not observed with intracorporeal VADs (even when adjusted for era effect). 12 This time-dependent hazard has also been reported for adult patients supported with extracorporeal devices that survived to 5 years after transplant. 13 Our study duration of 1 year does not allow analysis of such longer-term effects, which could be important if overall longevity of grafts or need for retransplantation is affected by prior VAD support. In this data set, rejection was the most significant cause of death after transplantation for EXCOR patients. In contrast, post-transplant deaths in the OPTN group were accounted for by multisystem organ failure, graft failure (either nonspecific or primary), and infection, which were absent or a less prevalent cause of EXCOR deaths. The absence of primary graft dysfunction among the EXCOR patients is significant because previous adult studies have shown that paracorporeal and VAD use in adult patients can lead to an increased incidence in primary graft dysfunction

5 S28 Circulation September 10, 2013 Table 2. EXCOR Patients: Univariate Variable Tests Associated With Mortality After Transplant Implant and Preimplant Variables Overall Mortality (12/106) n Died (%) Fisher Exact Test P Value Cardiac diagnosis 0.02* Cardiomyopathy/myocarditis 83 6 (7.2) Congenital heart disease 23 6 (26.1) Ventricle type ventricle 7 1 (14.3) 2 ventricles (11.1) EXCOR pump size, ml (14.3) 25/ (12.3) 50/ (8.3) Device type 0.35 LVAD 67 6 (9.0) BVAD 39 6 (15.4) Prior stroke within 30 d 1.00 Yes 7 0 (0.0) No (12.1) ECMO >10 d before implant 0.46 Yes 5 1 (20.0) No (10.9) CPR within 48 h 0.08 Yes 10 3 (30.0) No 96 9 (9.4) Infection within 48 h 0.73 Yes 27 2 (7.4) No (12.7) Active infection at the 0.14 time of implant Yes 6 2 (33.3) No (10.0) Total number of site 1.00 implants ( ) 1 5 implants 6 0 (0.0) >5 implants (12.0) Age y 31 4 (12.9) >1 y 75 8 (10.7) Weight (kg) categories 0.10 <5 6 0 (0.0) (20.0) > (6.7) Sex 1.00 Female 52 6 (11.5) Male 54 6 (11.1) Race 0.12 White 66 5 (7.6) Missing for 2 Nonwhite 38 7 (18.4) (continued) Table 2. Continued Implant and Preimplant Variables Overall Mortality (12/106) n Died (%) Fisher Exact Test P Value Hemodynamic Support 0.93 ECMO 33 3 (9.1) Ventilator 45 6 (13.3) Neither 28 3 (10.7) INTERMACS profile (14.0) 0.54 INTERMACS profile (8.2) 0.35 GFR categories (age adjusted) 0.10 <30% predicted 0 0 (0.0) 30% to 89% predicted 17 4 (23.5) >89% predicted 89 8 (9.0) Total bilirubin, mg/dl 0.52 < (10.5) Missing for (15.2) Univariate analysis of implant and preimplant variables associated with mortality after transplant in children supported with EXCOR Pediatric transplant compared with data from OPTN. BVAD indicates biventricular assist device; CPR, cardiopulmonary resuscitation; ECMO, extracorporeal membrane oxygenation; GFR, glomerular filtration rate; INTERMACS, Interagency Registry for Mechanically Assisted Circulatory Support; LVAD, left ventricular assist device; and OPTN, Organ Procurement and Transplant Network. *Statistically significant values. Findings related to rejection after VAD support are often ascribed to potential sensitization from support, although there has been some controversy. 14,17 19 Using the OPTN adult data, Arnaoutakis et al 14 demonstrated that device type (continuous flow versus pulsatile, with higher sensitization rates in latter) or levels of panel-reactive antibody activity (>25% defined as high) were not associated with worse mortality. Furthermore, they found that panel-reactive antibody activity did not affect rejection rates in the first year after transplant. The incidence of rejection (both early and late) for pediatric heart transplant patients has decreased in the current era, especially during the first post-transplant year. 20 This report is the first to suggest increased incidence of rejection-related death after transplant in pediatric heart transplant candidates supported with EXCOR. In the adult setting, VAD support has not been associated with increased incidence of fatal episodes of rejection after transplant. 21 Sensitization in adults has been mostly reported in the setting of long-term support. 19 Most of the EXCOR patients who died from rejection after transplant were supported for a short duration. It is possible, therefore, that this select group represents a presensitized cohort with unique attributes. Unfortunately, our study is limited by a lack of relevant and important data related to risk factors for posttransplant rejection such as preformed reactive antibodies. Pending further investigation, careful analysis of the sensitization status of the pediatric patients should be noted before VAD implantation. In our study, univariate analysis suggested cardiopulmonary resuscitation within 48 hours of device implant as a significant risk factor for post-transplant mortality. The overall

6 Eghtesady et al Berlin Heart Outcomes After Transplant vs OPTN S29 Table 3. EXCOR Patients: Events Occurring While on EXCOR Device Support Events Overall Mortality (12/106) n Died (%) Fisher Exact Test P Value Any serious adverse event 0.68 Yes (11.0) No 15 2 (13.3) Major bleeding event 1.00 Yes 47 5 (10.6) No 59 7 (11.9) Stroke event (ischemic/hemorrhagic CVA) 0.72 Yes 23 3 (13.0) No 83 9 (10.8) Renal dysfunction event 0.14 Yes 6 2 (33.3) No (10.0) Major infection event 0.55 Yes 55 5 (9.1) No 51 7 (13.7) Pump change for thrombus 0.13 Yes 59 4 (6.8) No 47 8 (17.0) Adverse events in the children who died while supported with the Berlin Heart EXCOR as a bridge to transplant are listed (n=12). CVA indicates cerebrovascular accident. numbers, however, for this, as well as preimplant liver and kidney dysfunction, were too small to reach definitive conclusions. Similar limitation related to the number of events could have accounted for the lack of significance for prior prolonged ECMO support on outcomes, despite a trend in numbers that would suggest otherwise. Alternatively, patients who transitioned from ECMO to VAD or who were salvaged by the device after cardiopulmonary resuscitation are a select group of patients who by definition had to survive to transplantation. Residual end-organ dysfunction, however, could impact the outcome of the survivors. Recently, Engin et al 21 provided a retrospective review of 14 critically ill adult patients, 11 of whom had received EXCOR support, demonstrating improved end-organ function after VAD implantation and before heart transplantation. Therefore, additional studies are needed to address the potential impact of EXCOR on end-organ recovery. Unequivocally, our data demonstrate that diagnosis of CHD continues to be a significant risk factor for less than optimal post-transplant outcomes even when patients are bridged to transplantation with EXCOR. Fan et al 22 had previously demonstrated that CHD diagnosis is a risk factor for in-hospital mortality after VAD implantation in pediatric patients. As well, Almond et al 3 had shown that only a third of the patients with CHD bridged to transplantation with ECMO survive to hospital discharge after transplantation. Clearly, in this cohort of patients, EXCOR when feasible is superior to ECMO. For many of these patients, because of patient size or other anatomic constraints, ECMO is still often used as the primary alternative. Combining postimplant mortality Table 4. Causes of Mortality After transplant in EXCOR and OPTN Patients Cause of Death EXCOR n (% of 12) OPTN n (% of 90) Rejection 6 (50.0) 11 (12.2) Cerebrovascular accident 2 (16.7) 11 (12.2) Infection 1 (8.3) 12 (13.3) Nonspecific graft failure 1 (8.3) 9 (10.0) Respiratory failure 1 (8.3) 6 (6.7) Unknown 1 (8.3) 2 (2.2) Graft failure: technical 0 (0.0) 1 (1.1) Hemorrhage 0 (0.0) 5 (5.6) Homicide 0 (0.0) 1 (1.1) Multisystem organ failure 0 (0.0) 15 (16.7) Primary graft failure 0 (0.0) 12 (13.3) Renal failure 0 (0.0) 2 (2.2) Sudden cardiac death 0 (0.0) 1 (1.1) Trauma 0 (0.0) 1 (1.1) Summary of post-transplant deaths in children supported with EXCOR Pediatric VAD compared with data from OPTN. Rejection in EXCOR (50.0%) significantly higher than OPTN (12.2%) P= OPTN indicates Organ Procurement and Transplant Network; and VAD, ventricular assist device. during the waitlist period and mortality after transplantation, children with CHD carry a substantial burden of mortality in the pathway to transplantation as amply demonstrated in International Society of Heart and Lung Transplantation database. Are these results a reflection of additional surgeries done in the hopes of avoiding transplantation and resulting in less than optimal candidates for VAD support (eg, worse renal function), timing of referral for VAD support (eg, after cardiac arrest and initiation of ECMO), or technical challenges posed by underlying complex anatomic substrate? Understanding these nuances would have significant implications in developing strategies (eg, new techniques) to improve outcomes for this cohort (and likely increasing numbers in the future) of patients. That only 2 of the rejection deaths occurred in patients with CHD as the primary diagnosis suggests that the risk of mortality in this subpopulation is not driven simply by increased incidence of sensitization or preformed antibodies. The steep drop-off on the Kaplan Meier curves would suggest the risk primarily relates to perioperative mortality and the likelihood these patients have had multiple prior operations with complex anatomy and complex physiologies. Further studies looking specifically at the cause of death among the patients with CHD would be valuable as the first step in improving outcomes for these patients. There are several other limitations to our study, one of which is that we could not do further subgroup analysis of the patients. For example, although the EXCOR device is clearly superior to support with ECMO, superiority to ventilatory support was not seen. It is conceivable that a certain cohort of patients requiring ventilatory support would be further helped by being on the EXCOR device because such support would allow the children to rehabilitate more effectively than those who remain on the ventilator. Similarly, we do not know how many EXCOR patients were or were not on a ventilator at the

7 S30 Circulation September 10, 2013 time of transplant; thus, we could not separate any increased risk of post-transplant mortality in EXCOR patients on or not on a ventilator at the time of transplant. We also could not evaluate donor organ quality that might be relevant when comparing EXCOR 1A patients with nonsupported 1A patients. Finally, our analysis was limited to 1 year after transplant; it is conceivable that not enough time has been allotted for observation of effects for some other important factors that require longer follow-up for manifestation (eg, coronary vasculopathy). In conclusion, our study shows that the EXCOR device support leads to post-transplant survival outcomes similar to those seen among patients not requiring support. Furthermore, post-transplant survival is improved with an EXCOR device compared with post-transplant survival in those bridged with ECMO. Patients with CHD who are bridged to transplantation have an increased risk of mortality after transplantation, regardless of the mode of support. Finally, in the IDE trial, a potential significant cause of mortality for VAD-supported patients after transplantation was related to rejection, a finding that deserves further detailed investigations. With increasing experience at multiple centers and the potential development of data registries and use of the device in nontrial setting, further insights will be gained into the use of EXCOR as bridge for transplantation and its impact on outcomes after transplantation. The evolution of support technology in the pediatric population, including the upcoming continuous flow devices, will inevitably lead to a paradigm shift in the field of pediatric heart failure and transplantation and likely all aspects of pediatric cardiology and cardiac surgery. Acknowledgments We gratefully thank Dr Giv Heidari-Bateni, MD, MPH, for preparation of the Kaplan Meier curves in this manuscript. Sources of Funding This study was supported, in part, by Berlin Heart Inc (The Woodlands, TX) and a grant funding the trial (RO1 FD ). Disclosures United Network for Organ Sharing, the contractor for the Organ Procurement and Transplant Network (OPTN), supplied the data reported in this article. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of, or interpretation by, the OPTN or the Government of the United States. C. Tjossem, BS, an author on this manuscript is an employee of Berlin Heart Inc, The Woodlands, TX. The other authors report no conflicts. References 1. FDA approval (h100004) of the Berlin heart EXCOR pediatric ventricular assist device. Available at PressAnnouncements/ucm htm. Accessed August 4, Fraser CD Jr, Jaquiss RD, Rosenthal DN, Humpl T, Canter CE, Blackstone EH, Naftel DC, Ichord RN, Bomgaars L, Tweddell JS, Massicotte MP, Turrentine MW, Cohen GA, Devaney EJ, Pearce FB, Carberry KE, Kroslowitz R, Almond CS; Berlin Heart Study Investigators. Prospective trial of a pediatric ventricular assist device. N Engl J Med. 2012;367: Almond CS, Singh TP, Gauvreau K, Piercey GE, Fynn-Thompson F, Rycus PT, Bartlett RH, Thiagarajan RR. Extracorporeal membrane oxygenation for bridge to heart transplantation among children in the United States: analysis of data from the Organ Procurement and Transplant Network and Extracorporeal Life Support Organization Registry. Circulation. 2011;123: Morales DL, Zafar F, Rossano JW, Salazar JD, Jefferies JL, Graves DE, Heinle JS, Fraser CD Jr. Use of ventricular assist devices in children across the United States: analysis of 7.5 million pediatric hospitalizations. Ann Thorac Surg. 2010;90: ; discussion Almond CS, Buchholz H, Massicotte P, Ichord R, Rosenthal DN, Uzark K, Jaquiss RD, Kroslowitz R, Kepler MB, Lobbestael A, Bellinger D, Blume ED, Fraser CD Jr, Bartlett RH, Thiagarajan R, Jenkins K. Berlin Heart EXCOR Pediatric ventricular assist device Investigational Device Exemption study: study design and rationale. Am Heart J. 2011;162: e6. 6. 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: Alba AC, Rao V, Ivanov J, Ross HJ, Delgado DH. Usefulness of the INTERMACS scale to predict outcomes after mechanical assist device implantation. J Heart Lung Transplant. 2009;28: Jaski BE, Kim JC, Naftel DC, Jarcho J, Costanzo MR, Eisen HJ, Kirklin JK, Bourge RC; Cardiac Transplant Research Database Research Group. Cardiac transplant outcome of patients supported on left ventricular assist device vs. intravenous inotropic therapy. J Heart Lung Transplant. 2001;20: Almond CS, Morales DL, Blackstone EH, Turrentine MW, Imamura M, Massicotte MP, Jordan LC, Devaney EJ, Ravishankar C, Kanter KR, Holman W, Kroslowitz R, Tjossem C, Thuita L, Cohen GA, Buchholz H, St Louis JD, Nguyen K, Niebler RA, Walters HL 3rd, Reemtsen B, Wearden PD, Reinhartz O, Guleserian KJ, Mitchell MB, Bleiweis MS, Canter CE, Humpl T. Berlin Heart EXCOR pediatric ventricular assist device for bridge to heart transplantation in US children. Circulation. 2013;127: Morales DL, Almond CS, Jaquiss RD, Rosenthal DN, Naftel DC, Massicotte MP, Humpl T, Turrentine MW, Tweddell JS, Cohen GA, Kroslowitz R, Devaney EJ, Canter CE, Fynn-Thompson F, Reinhartz O, Imamura M, Ghanayem NS, Buchholz H, Furness S, Mazor R, Gandhi SK, Fraser CD Jr. Bridging children of all sizes to cardiac transplantation: the initial multicenter North American experience with the Berlin Heart EXCOR ventricular assist device. J Heart Lung Transplant. 2011;30: Karamlou T, Gelow J, Diggs BS, Tibayan FA, Mudd JM, Guyton SW, Slater MS, Song HK. Mechanical circulatory support pathways that maximize post-heart transplant survival. Ann Thorac Surg. 2013;95: ; discussion Patlolla V, Patten RD, Denofrio D, Konstam MA, Krishnamani R. The effect of ventricular assist devices on post-transplant mortality an analysis of the United network for organ sharing thoracic registry. J Am Coll Cardiol. 2009;53: Bittner HB, Lehmann S, Rastan A, Garbade J, Binner C, Mohr FW, Barten MJ. Outcome of extracorporeal membrane oxygenation as a bridge to lung transplantation and graft recovery. Ann Thorac Surg. 2012;94: ; author reply Arnaoutakis GJ, George TJ, Kilic A, Weiss ES, Russell SD, Conte JV, Shah AS. Effect of sensitization in us heart transplant recipients bridged with a ventricular assist device: Update in a modern cohort. J Thorac Cardiovasc Surg. 2011;142: , 1245 e DeNofrio D, Rho R, Morales FJ, Kamoun M, Kearns J, Dorozinsky C, Rosengard BR, Acker MA, Loh E. Detection of anti-hla antibody by flow cytometry in patients with a left ventricular assist device is associated with early rejection following heart transplantation. Transplantation. 2000;69: John R, Pagani FD, Naka Y, Boyle A, Conte JV, Russell SD, Klodell CT, Milano CA, Rogers J, Farrar DJ, Frazier OH. Post-cardiac transplant survival after support with a continuous-flow left ventricular assist device: impact of duration of left ventricular assist device support and other variables. J Thorac Cardiovasc Surg. 2010;140: Drakos SG, Kfoury AG, Kotter JR, Reid BB, Clayson SE, Selzman CH, Stehlik J, Fisher PW, Merida M 3rd, Eckels DD, Brunisholz K, Horne BD, Stoker S, Li DY, Renlund DG. Prior human leukocyte antigen-allosensitization and left ventricular assist device type affect degree of postimplantation human leukocyte antigen-allosensitization. J Heart Lung Transplant. 2009;28: Kirsch L, Timmermans T, Van Caenegem O, Gurne O, Noirhomme P, Jacquet LM, Latinne D, Poncelet AJ. Allosensitization in bridge to

8 Eghtesady et al Berlin Heart Outcomes After Transplant vs OPTN S31 transplant Novacor left ventricular assist device patients: analysis of longterm outcomes with regard to acute rejection and chronic allograft vasculopathy. Eur J Cardiothorac Surg. 2008;34: ; discussion Malickaite R, Rucinskas K, Staneviciene A, Miniauskas S, Maneikiene V, Zuoziene G, Sirvydis V. Sensitisation and post-transplant course after the implantation of ventricular assist device. Interact Cardiovasc Thorac Surg. 2009;8: ; discussion Ameduri RK, Zheng J, Schechtman KB, Hoffman TM, Gajarski RJ, Chinnock R, Naftel DC, Kirklin JK, Dipchand AI, Canter CE. Has late rejection decreased in pediatric heart transplantation in the current era? A multi-institutional study. J Heart Lung Transplant. 2012;31: Engin C, Ayik F, Oguz E, Eygi B, Yagdi T, Karakula S, Ozbaran M. Ventricular assist device as a bridge to heart transplantation in adults. Transplant Proc. 2011;43: Fan Y, Weng YG, Huebler M, Cowger J, Morales D, Franz N, Xiao YB, Potapov E, Hetzer R. Predictors of in-hospital mortality in children after long-term ventricular assist device insertion. J Am Coll Cardiol. 2011;58: