Current Heart Failure Reports

, Volume 9, Issue 3, pp 236–243

Mechanical Circulatory Support in Children: Bridge to Transplant Versus Recovery


    • The Heart InstituteCincinnati Children’s Hospital Medical Center
  • David L. Morales
    • The Heart InstituteCincinnati Children’s Hospital Medical Center
Decompensated Heart Failure (MM Givertz, Section editor)

DOI: 10.1007/s11897-012-0103-y

Cite this article as:
Jefferies, J.L. & Morales, D.L. Curr Heart Fail Rep (2012) 9: 236. doi:10.1007/s11897-012-0103-y


The number of children and adolescents suffering from heart failure is increasing dramatically. Some of these patients will progress to need advanced therapies in the form of mechanical circulatory support (MCS). Over the past few years, increased attention has been focused on clinical use of existing devices as well the development of pediatric-specific ventricular assist devices (VADs). As in adult populations, these devices offer unique opportunities to successfully support children as a bridge-to-transplant, but increasing data suggest that bridge-to-recovery and bridge-to-destination are also viable options in select pediatric populations. Herein, we will review existing approaches as well describe future potential MCS options.


Heart failurePediatricAdolescentAssistMechanicalSupport


Advancements in the recognition and management of infants, children, and adolescents with genetically triggered or acquired myocardial dysfunction and heart failure have been realized over the past decade. As a result, the number of children surviving with heart failure has been increasing dramatically. The number of children hospitalized for heart failure in the U.S. is reported to be between 11,000 and 14,000 with an overall mortality of 7 % [1]. This ever-growing number of pediatric heart failure patients cannot be fully addressed by heart transplantation alone secondary to the invariable limitation in donor organ supply. Importantly, the annual number of pediatric heart transplants performed worldwide has remained stagnant for over a decade. This has necessitated ongoing suboptimal medical management for worsening heart failure in the form of ventilatory support and inotropic therapy with consideration of extracorporeal membrane oxygenation (ECMO) and emergent listing for cardiac transplantation in extreme cases. This approach has resulted in a ceiling for complete care resulting in increased morbidity, including end-organ damage, and avoidable pre- and post-transplant mortality. Limitations in availability to advanced therapies beyond ECMO such as mechanical circulatory support (MCS) have resulted in pediatric patients succumbing to complications of heart failure. New advances in mechanical support technology and increasing interest in pediatric populations have resulted in exciting changes in the management of advanced heart failure in younger patients. Herein, the current state of MCS in children, including bridge to transplant (BTT) or recovery, and future directions are described.

History of Mechanical Circulatory Support

The concept of MCS has been evolving since the inception of cardiopulmonary bypass in 1953 [2]. Subsequently, increasing need for support of cardiogenic shock in postcardiotomy patients spurred the development of an artificial ventricle in 1963 supporting an adult for 4 days [3]. In 1969, Cooley et al. [4] reported the successful use of a total artificial heart to bridge an adult patient to cardiac transplantation. The subsequent development of devices has been a developing field, with initial efforts directed at pulsatile devices with ultimate approval of several types of devices by the 1990s as a BTT. The results of the Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial noted a 48 % reduction in mortality with use of left ventricular assist devices (LVAD) when compared to medical therapy for patient with chronic heart failure. This demonstrated the advantage of alternative therapies for advanced heart failure and resulted in U.S. Food and Drug Administration (FDA) approval of the HeartMate XVE (Thoratec Corporation, Pleasanton, CA) for permanent destination therapy (DT) in 2003 [5]. In 2006, the Interagency Registry of Mechanically Assisted Circulatory Support (INTERMACS) was developed in an effort to centralize data collection and track the progress of device use in the United States. Recent shifts from pulsatile devices to continuous-flow devices (CFD) have resulted in rapidly expanding use of ventricular assist devices (VAD). The approval of the HeartMate II (Thoratec Corporation) device for DT in 2010 has has resulted in a dramatic increase in VAD use for transplant-ineligible adult patients [6]. This explosion has favorably impacted the face of advanced heart failure management given the fixed number of organs available for transplant juxtaposed to the increasing numbers of heart failure patients. In 2011, a milestone was realized when more ventricular assist device (VADs) operations were performed than heart transplants. In the current era, 1-year survival is greater than 80 %, with CFD solidifying their use.

Pediatric investigation has been slow to proceed secondary to multiple factors, including varying sizes of patients necessitating different size pumps, technological limitations, and lack of funding interest. Increasing need for durable support strategies to replace ECMO resulted in the recently completed Berlin EXCOR (Berlin heart, Inc., Berlin, Germany) device trial. The results of this trial, detailed below, led to FDA approval of this extracorporeal, pulsatile VAD in December 2011.

Current Technologies for Pediatric Support

Short-Term Support

In many pediatric institutions, ECMO remains the mainstay of therapy for acute circulatory support. The familiarity and availability of ECMO affords widespread use in most pediatric institutions. Biventricular and respiratory support can be provided via central or peripheral cannulation. ECMO is typically used when both cardiac and respiratory support is required. The use of ECMO for cardiac failure only is not logical unless a patient is arresting or has had progression of cardiac failure to the degree that they can no longer be mechanically ventilated. The latter usually is a failure to recognize the opportunity to mechanically support a patient earlier. ECMO often is ineffective at significantly and reliably unloading the heart, which is paramount to providing the best opportunity at myocardial recovery or as a BTT (Fig. 1).
Fig. 1

Approach to MCS at Cincinnati Children’s Hospital Medical Center, including strategies for short-term and long-term support (NovaLung® manufactured byNovalung GmbH (Heilbronn, Germany), Syncardia® TAH manufactured by SynCardia Systems, Inc. (Tucson, AZ). Rotaflow® manufactured by MAQUET Medical Systems (Wayne, NJ), PediMag® manufactured by Thoratec Corporation (Pleasanton, CA), TandemHeart® manufactured by CardiacAssist, inc. (Pittsburgh, PA), EXCOR® manufactured by Berlin Heart, Inc. (Berlin, Germany), HeartMate II® manufactured by Thoratec Corporation (Pleasanton, CA), HeartWare® manufactured by HeartWare (Framingham, MA). MCS mechanical circulatory supports; ECMO extracorporeal membrane oxygenation; PHTN pulmonary hypertension; CF continuous flow; Periph peripheral; Vent Press ventricular pressure; Tx transplantation; HTx heart transplant; req. required; VAD ventricular assist device; TAH total artificial heart; BSA body surface area; Pt patient; DCM dilated cardiomyopathy; CHD congestive heart disease)

The approach to temporary cardiac support at most centers has been through the use of centrifugal VADs. These devices have many benefits making them desirable for use in children, including simplicity of implantation, widespread availability, accommodation of all patient sizes, and low cost [7]. These devices are implanted when the etiology of heart failure is acute (temporary) in that we expect the heart to recover from the process within 2 weeks. These include etiologies such as myocarditis or acute graft failure. These devices also allow us to support children who present acutely or as a transfer who clearly need some type of support but we are unclear if their heart failure is a temporary or chronic process. While on the device and well supported, transplant candidacy, neurological status, and recoverability can be assessed. This bridge to decision allows these patients an opportunity to recover or to be bridged to a long-term device. Most centers in the U.S. would utilize either the Biomedicus BP-50 (Medtronic, Minneapolis, MN), the Levitronix Centrimag (Levitronix GmbH, Zurich, Switzerland), or the MAQUET Rotaflow Centrifugal pump (MAQUET Medical Systems, Wayne, NJ). In addition, percutaneous extracorporeal support may be considered using the TandemHeart device (CardiacAssist, Inc., Pittsburgh, PA), which we have successfully used in pediatric patients. Other possibilities include the Impella 2.5 (ABIOMED,Inc., Danvers, MA) and standard intraaortic balloon pump counterpulsation, both of which have been used in adolescents but have no current applications in neonates and small children secondary to device size.

Long-Term Support

As the result of a multi-institutional investigational device exemption (IDE) study from May 2007 to December 2011, the Berlin Heart EXCOR became the only pediatric-specific long-term device available currently for infants and children in the U.S. The IDE trial cohort was a selected cohort of 48 patients; however, the entire IDE study included, in addition to the trial patients, patients who did not meet criteria or who were implanted in non-IDE sites. This entire cohort represents all 204 patients implanted with the EXCOR between May 2007 and December 2010 in the U.S. The median age and weight were 19 months and 10 kg, respectively. One third of patients (68) met IDE inclusion criteria compared to two thirds of patients (136) who were implanted under compassionate-use (CU) criteria, the former with a significant survival advantage of 92 % compared to 66 %. CU patients were more likely to be younger, smaller, have congestive heart disease (CHD), be on ECMO pre-implant, have severe renal dysfunction, and have an elevated bilirubin level (P < 0.01). In the multivariable analysis including all patients, early mortality (< 7 days) was associated with lower weight, biventricular assist device (BIVAD) support, and elevated bilirubin whereas late mortality (>7 days) was associated with elevated bilirubin and renal dysfunction. Neurological dysfunction occurred in 30 % of patients regardless of cohort and was the leading cause of death. For all 204 non-selected consecutive patients, 75 % were successfully bridged to transplant or recovery.

The selection of devices in children is changing. Increased use of the HeartMate II device was recently presented in a retrospective analysis of the INTERMACS registry. There were 26 pediatric patients identified (18 male) of which 7 (27 %) were implanted in a pediatric hospital. Survival to transplantation, ongoing support, or recovery at 6 months was 95 % for these patients, which was similar to a comparative group of young adults. However, a higher incidence of device malfunctions, arrhythmias, and bleeding requiring surgery was noted in the pediatric group [8•]. Use of the HeartWare (HeartWare, Framingham, MA) device is also being reported. Miera et al. [9•] recently described use of the HeartWare device as a bridge to cardiac transplantation in seven children ranging in age from 6–16 years. All but one child was successfully transplanted with the remaining child currently on support. No thromboembolic complications were reported.

Indications for Long-Term Devices in Children

Bridge to Transplantation

MCS has been used for many years as a BTT technology in children but with varying degrees of success. Many centers rely on ECMO but associated mortality rates remain high [10, 11]. The use of ECMO limits the effective time for safe support to cardiac transplant to less than 2 weeks. As such, different support strategies enlisting ECMO or other short-term devices as a bridge to more durable long-term devices have emerged [7]. Previous reports have documented favorable results with paracorporeal pneumatic devices but size limitations restricted use in some patients [12, 13]. The recent completion of the Berlin EXCOR study and approval on December 16, 2011 ushered in a new era in LVAD therapy for children as a bridge to transplant. The EXCOR was approved for pediatric patients with uni- or biventricular heart failure who are candidates for heart transplant. There were no weight, age, or diagnosis restrictions. The 10-, 25-, 30-, 50-, and 60 cm3 pumps all were approved. The protocol for a postmarket study required by the FDA has been submitted by Berlin Heart. It is important to remember that the EXCOR is a humanitarian device exemption (HDE)–approved device; therefore an active institutional review board (IRB) is required. Most notable is that this approval does not require a center implanting the EXCOR to be a transplant center.

Bridge to Recovery

Pediatric populations offer a different substrate to assess for myocardial recovery with use of MCS. Although bridge to possible recovery has long been recognized as an indication for MCS in children, most pediatric institutions would employ ECMO as a first line of support [14, 15]. Our decision tree for implementation of MCS facilitates the possibility of early recovery in acute clinical scenarios but typically does not involve ECMO except in special circumstances such as cardiac arrest or late presentation. We use short-term VADs wherever appropriate because the short-term VAD has theoretical advantages over ECMO owing to avoidance of an oxygenator and, more importantly, direct unloading effect of the failing ventricle. Recently presented at the International Society for Heart and Lung Transplantation (ISHLT) were preliminary data regarding the use of the short-term VADs applied in 31 occasions (left [or systemic] ventricle in 27, right ventricle in 2, and both in 2). The type of VAD used was Biomedicus® in 22, RotaFlow® in 7, and TandemHeart® in 2. Median age and weight were 6 years (range: 2 days to 20 years) and 19 kg (range: 2.7 to 93 kg). The etiology of heart failure was an acute process (ie, myocarditis) in 18 (58 %) or an acute exacerbation of chronic condition (ie, cardiomyopathy) in 13. Median support time was 6 (3 to 19) days in total. A positive outcome was achieved in 90 % (successful bridge to either recovery, transplantation, or another bridge to a long-term device). In particular, 83 % (15/18) of those with an acute etiology recovered, with a median support time of 6 (3 to 18) days. In contrast, recovery was rarely (15 % [2/13]) observed in those with acute or chronic causes, with the remaining majority (77 % [10/13]) requiring either bridge to transplant or bridge to a long-term device. Death on the device occurred in 3 (10 %); all smaller than 3 kg, for whom no long-term devices were available. Overall, hospital discharge occurred in 83 % of the patients. We concluded that short-term VAD support is a useful therapy for pediatric heart failure, especially when myocardial recovery is anticipated in the setting of adequate ventricular unloading.

Only limited experience with device removal in chronic cardiomyopathy phenotypes has been reported in children. We successfully explanted a HeartMate II device while at Texas Children’s Hospital from a 14-year-old patient with end-stage heart failure through a subcostal incision after 9 months of support. At 1 year, he remains active and asymptomatic on medical therapy [13, 16•, 17]. However, it is increasingly recognized that our younger populations should be evaluated more aggressively for elective device removal. Traditional approaches to VAD implantation in children has been predicated on listing for cardiac transplantation. This was a requirement of the recent Berlin EXCOR study [18••]. With increasing use of VADs in children, opportunities to consider explant must be a part of the decision-making process. Numerous adult reports document that some patients recover sufficient myocardial function to result in LVAD removal and avoidance of transplantation [1922]. A recent report retrospectively analyzed HeartMate II BTT and DT trials that included 1,108 patients [1]. The results suggested that those patients with the highest likelihood of recovery were those less than 40 years of age with nonischemic cardiomyopathy of less than 1-year duration. Given that our patients meet all these criteria, the authors suggest that this bodes well for our patients and consideration for VAD explantation. However, regardless of age of the patients, it must be remembered that VAD therapy is not a true “bridge to recovery.” Rather, these patients have been restaged to a more functional class with improved myocardial function and less symptoms. Life-long surveillance for worsening clinical condition with ongoing medical therapy is critical to success.

Unique Challenges

During the Berlin EXCOR trail, the planning for the PumpKIN (Pumps for Kids, Infants, and Neonates)protocol, and the creation of the PediMacs database, it has become obvious that a thoughtful approach to the long-term developmental, neurological, and quality-of-life outcomes in these pediatric VAD patients is required. Although many adult reports characterize quality of life and psychosocial effects of VADs, children represent a completely different cohort, especially because quality-of-life measures must be assessed on not only the patient but also their parents and siblings. This is a large initiative for both PediMacs and the PUMP’s for Kids Infants and Neonates (PUMPKIN) trial.

Numerous studies have documented the important effects of cardiopulmonary bypass on neurocognition in children requiring corrective or palliative surgery [23, 24]. Children undergoing repair at less than 3 months of age requiring cardiopulmonary bypass (CPB) may still have impaired development up to 2 years after surgical intervention [24]. However, limited data are available regarding cognitive outcomes in children supported with VADs or those undergoing cardiac transplantation. Neurobehavioral performance is of great clinical importance in adult populations because this may influence the implantation of a VAD and is a parameter that is often followed postimplantation to assess for improvement [25]. Petrucci et al. [26] reported data from 11 centers evaluating end-stage heart failure patients at 1, 3, and 6 months after HeartMate II implantation as a BTT [26]. Interestingly, there were improvements in visual memory, executive functions, visual special perception, and processing speed. Stein and colleagues [27•] recently presented data regarding cognitive outcomes in pediatric patients supported with VADs as a BTT. Overall intellectual functioning was the same for VAD patients when compared to a retrospective cohort of transplant recipients but was reduced when compared to healthy controls. Recognizing that opportunities exist to alter neurodevelopmental outcomes in children, careful attention must be given to ongoing surveillance and appropriate management as the window for intervention is typically longer in pediatrics than in adults.

Impact of Device Use in Children

Recently, a retrospective review was presented as preliminary data detailing hospitalization cost (HC) in U.S. hospitals reported in the Pediatric Health Information System from January 2006 to June 2011 exploring cost for pediatric VAD placement [28•]. In this study, 237 VADs were placed in 236 patients (< 18 years), of which 127 (54 %) were men. The median age at admission was 3.2 years (0–17.5 years). The median length of stay was 62 days (1 day–1.3 years). 130 (55 %) survived to heart transplantation (HTx), 17 (7 %) were explanted to recovery, 36 (15 %) were discharged on VAD support,and 54 (23 %) died on the device. The mean hospital cost and cost/day was $693,000 ± 466 ($21–$2,845,000) and $12.6 K ± 21.0(0.3-$298,000). Multivariable regression revealed that ECMO support increased the hospital costs by $114,000 (P = 0.049), while HTx increased the hospital cost by $355,000 (P < 0.0001). However, the mean cost to support children with VADs was noted to actually decrease in the last 5 years.

Effective Delivery of Care

The effective use of MCS in pediatric populations requires a thoughtful approach integrating numerous health care providers for the patient. This has been realized in successful adult LVAD programs. However, the care of children necessitates a broader care delivery target that encompasses the family as well. Significant effort is required to ensure effective communication at all stages of device consideration, which may involve multiple family members and caregivers. Preimplant evaluations demand thorough documentation of the family dynamics and support structure. The results of these evaluations are critical given that pediatric patients receiving long-term devices are also being evaluated for HTx and now, in some cases, chronic therapy. The approach at many pediatric centers is based in part on experiences seen in adult centers but has been modified to include additional components more specific to children (Fig. 1).

Most pediatric LVAD centers do not currently discharge patients on support. However, given the recent use of the HeartMate II and Syncardia devices in children, we have faced the challenge of effective care delivery in an outpatient setting. The successful transition from inpatient to outpatient care requires an ongoing investment from all parties involved [20]. Initial discussions with families and health care providers in the preimplantation meetings should include the concept that discharge from the hospital is one of the primary goals. The development of an outpatient team dedicated to outpatient management must be constructed before discharge. In preparation for leaving the hospital, ongoing patient and family education must be documented including device management and alarm recognition, wound care, and need for compliance with ongoing medical follow-up. An integrated multidisciplinary clinic involving pediatric cardiologists and cardiac surgeons, VAD coordinators, and social work is essential. In addition, access to dedicated imaging services, typically echocardiography, with VAD-specific protocols is critical to surveillance and assessment of critical data including right ventricular size and function, aortic valve insufficiency, and indicators of possible successful device recovery. We also involve other subspecialties such as hematology and infectious disease based on individual patient need.

Devices in Development

Pediatric populations offer a heterogeneous substrate for device use. However, the greatest limitation to expansive use of MCS in children has been the availability of appropriately sized devices. The completion of the Berlin EXCOR study marked an important landmark in pediatric heart failure given the focus on use of smaller, size-specific devices allowing for durable neonatal, infant, and small child support. Given that adult investigations are moving away from pulsatile devices, increased emphasis has been placed on the development of CF pediatric devices. Older children can derive benefit from adult-sized intracorporeal devices such as the Thoratec HeartMate II® LVAD, the HeartWare® VAD, and the SyncCardia TAH. All of these companies also have significant research and development of even smaller VADs including the Syncardia 50/50 cm3 pump, the Thoratec HeartMate X, Thoratec HeartMate III, and the HeartWare HVAD. Unfortunately, smaller children, infants, and neonates cannot be supported with such devices secondary to size mismatch. Within the next 5 years, we may hopefully have the ability to implant intracorporeal VADs that can be used in small children and infants. The PUMPKIN program funded by the National Heart, Lung, and Blood Institute (NHLBI) is promoting the development of two intracorporeal VADs for small children and infants. These are the Jarvik Pediatric 2000 (Jarvik Heart, Inc., New York, NY [Figs. 2 and 3) and PediaFlow (LaunchPoint Technologies Inc.; Goleta, CA [Fig. 4]). Industry is also increasingly invested in developing smaller devices. The CirCulite LVAD (CircuLite, Saddle Brook, NJ), which is the size of a AA battery, has a NHLBI grant to help convert their adult partial flow LVAD to a pediatric device.
Fig. 2

Infant VAD (Jarvik Pediatric 2000; manufactured by Jarvik Heart, Inc. [New York, NY], used with permission). (VAD ventricular assist device)
Fig. 3

VAD in hand (Jarvik Pediatric 2000; manufactured by Jarvik Heart, Inc. [New York, NY], used with permission). (VAD ventricular assist device)
Fig. 4

PediaFlow (LaunchPoint Technologies Inc. [Goleta, CA], used with permission)

Future Directions

Given the rapid expansion of pediatric and adolescent patients with advanced heart failure and the increasing availability of adult VAD technologies, future use of mechanical support in younger populations will only continue to increase. Current approaches to VAD implantation in children has typically been focused on patients with advanced INTERMACS severity. Most patients are hospitalized in the intensive care unit, requiring mechanical ventilatory support and multiple inotropes. In addition, evidence of end-organ dysfunction is usually evident at time of elective implantation, resulting in potentially less benefit to the patient. We foresee a more timely approach to use of VADs in pediatrics with increasing use in more stable and less symptomatic patients with better preserved end-organ function. This has already started to occur in select pediatric programs as well as in other units. This would parallel the current approaches in adult patients. In addition, enhanced reporting directed at pediatric patients is in development. The PediMACs database, an independent branch of the INTERMACS database, will house data for all pediatric VADs including temporary devices, allowing for improved reporting of device use in children and facilitating impactful research efforts.

An area of focus that has not been explored to any great degree is the use of MCS in complex congenital heart disease and heart failure. The Berlin EXCOR study did not include single ventricle physiology patients. Anecdotal experience of use with device in single ventricles has been poor at most institutions. As such, there are no ongoing multicenter trials addressing this growing cohort. Recognizing that congenital surgical advances have greatly enhanced survival, discussion of expanding use of MCS into this population must be considered. There are currently more adults over the age of 18 years living with congenital heart disease than under the age of 18 years. Many of these patients face a high likelihood of myocardial dysfunction and worsening heart failure in their future. Recently reported is the successful use of a HeartMate II as a systemic ventricular assist device (SVAD) in a failing fontan patient who was discharged home and then successfully transplanted [29•]. Availability of devices, as a BTT or possibly DT, will be essential to the success of pediatric and adult congenital programs.

Although no current indications exist for DT in pediatric or adolescent populations, certain populations may lend themselves to consideration. Patients with complex congenital heart disease, particularly those with irreversible pulmonary hypertension, may be one such population if they are not suitable HTx-lung transplant candidates. The use of VADs in these patients and with primary pulmonary vascular disease may be a bridge to candidacy because pulmonary vascular remodeling has occurred in these types of patients resulting in successful cardiac transplantation. This bridge to candidacy also may be applied in obese adolescents who, after VAD placement, enter a hospital-based weight program that includes surgical intervention. Other populations that would not be typically considered for cardiac transplant include those with progressive neuromuscular disease. One such cohort are those boys with Duchenne muscular dystrophy (DMD). Historically, most centers in the United States have not considered these patients for transplant secondary to their limited functional state and limited lifespan. However, recent advances in pulmonary care have resulted in improved longevity. Most boys with DMD now die secondary to heart failure. This may be an ideal group to consider use of DT in adolescent populations.


Given the growing population of young patients with advanced heart failure and with continued improvements in device size and availability, opportunities for VAD use as BTT, bridge to recovery, and bridge to destination (BTD) in the pediatric and adolescent population will expand and dramatically change the current approach to these populations. Pediatric patients may offer insight into myocardial recovery that cannot be seen in adults and serve as fertile ground for future investigation in this area. The evolution of MCS strategies over the next decade will add substantially to the treatment armamentarium and bring transformative changes in our management of children and adolescents with heart failure.


J. L. Jefferies: none. Dr. David Morales serves as director of the Reference and Training Center established by Berlin Heart, Inc.; in recognition, Berlin Heart, Inc. provides compensation to Baylor College of Medicine (Dr. Morales receives no direct personal compensation).

Copyright information

© Springer Science+Business Media, LLC 2012