The last decade has witnessed significant advancements in ventricular assist devices (VADs) for the adult population. Continued refinement of device technology as well as clinical management has led to improved outcomes of VAD support, resulting in a rapid increase in the number of patients supported with VADs (1). In stark contrast to the adult population, options for mechanical circulatory support (MCS) in children, particularly infants and small children, are limited by a paucity of devices designed for pediatric use (2). In fact, extracorporeal membrane oxygenation (ECMO) used to be the only MCS option in this group of patients. This frustrating reality, however, is starting to change. Owing to widespread acceptance of the Berlin Heart EXCOR (Berlin Heart, The Woodland, TX), which is equipped with a variety of pump and cannula sizes, even infants can now benefit from longterm VAD support. In addition, the Infant Jarvik 2000 (Jarvik Heart Inc., New York, NY), an implantable, continuous-flow VAD for small children, will be analyzed for safety and efficacy in its first randomized trial (3). Finally, the era of pediatric MCS has begun. In this chapter, we will review VAD devices available for the pediatric population.

In recent years, VADs have become an integral component of the standard of care provided to adults with end-stage heart failure. The most significant change to impact patient management strategy is the emergence of implantable, continuous-flow VADs such as the HeartMate II (Thoratec Corporation, Pleasanton, CA) and the HeartWare HVAD (HeartWare Inc., Framingham, MA). The efficiency of these devices, coupled with low-morbidity profiles, has contributed to a change in indications for device placement. Early utilization of such a device is now considered a reasonable option in preference to escalating medical management. Nowadays, VADs are being used not only for bridging the patient to heart transplantation (bridge to transplant, BTT) but also as destination therapy (DT), which is currently offered only to those candidates deemed unsuitable for transplantation. In fact, the number of patients who undergo VAD placement for DT has increased, now accounting for about 41% of all adult VAD implantations (1). It is suggested that with continuing evolution of VAD therapy, the medical community is approaching the point where VADs may be considered the primary alternative to cardiac transplantation (4,5).

Although the pediatric population has not yet experienced this shift in the MCS paradigm, there has been a growing interest in pediatric MCS in recent years for the following reasons: First, the number of children with endstage heart failure far exceeds the availability of suitable pediatric cardiac donors. The number of pediatric heart transplants worldwide has been stagnant at approximately 300 to 400 per year (6,7). By analyzing 15 million pediatric hospitalizations using the Healthcare Cost and Utilization Project Kids Inpatient Database, Rossano et al. have shown that heart failure-related hospitalizations occur in 11,000 to 14,000 children annually in the United States, with an overall mortality of 7% (8). It is important to note that admissions did not greatly increase over the 10-year study period (1997-2006), but the total mean length of stay increased from 13.8 to 19.4 days. This data possibly indicates a temporal increase in the proportion of sicker children. Secondly, an increment in the number of children listed for heart transplantation would support this view, which has inevitably resulted in prolongation of wait times on the heart transplant list in the setting of unchanged organ supply (6). While ECMO has traditionally been used for bridging the patient to transplantation, the outcome of such an intervention is generally poor. Inherently temporary, ECMO support cannot be instituted indefinitely until a suitable organ is found. This limitation has gained greater significance in recent years where waitlist duration has been steadily increasing. Using the ELSO Registry, Almond and associates (9) found that a third of pediatric heart transplant recipients on ECMO support as a BTT die during the same admission. Five- and ten-year posttransplantation survival is also worse in patients on ECMO prior to transplant than in patients who were on a VAD or not on MCS (10).

At present, there is no universal consensus with regard to indications for instituting VAD support. In general, VAD therapy for any patient is considered when the proposed benefits of VAD outweigh the risks. Traditionally, VAD support is indicated when severity of heart failure extends beyond the capability of maximal medical management. As with all interventions, a practitioner’s individual threshold in deciding whether to provide or withhold a therapy is influenced by the degree of confidence in that modality to consistently produce positive outcomes. Since the risk-to-benefit ratio is largely determined by the patient’s clinical status, institutional experience, and type of devices available, the decision needs to be made on a case-by-case basis by an experienced multidisciplinary team. The Fifth Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) report found that preoperative clinical status of a VAD candidate has an appreciable impact on postimplant outcomes (5). There are ongoing discussions about extending partial VAD support to include patients with INTERMACS profile 4 to 7 (1 being worst, 7 being best) who may derive some benefit (11,12). The trend of early VAD institution has so far resulted in considerably improved outcomes after VAD implantation (1).

Pediatric physicians, contrary to adult physicians, tend to be more judicious when evaluating indications for VAD use in children. Globally, most pediatric heart centers have less experience with such devices as compared with adult centers. This is particularly true for small children, who can only be supported with pulsatile, extracorporeal VADs (e.g., Berlin Heart EXCOR) that are associated with significantly higher risk profiles than implantable, continuous-flow devices (13). Devices such as these present a challenge to clinicians with respect to determining the optimum timeframe within which to commence therapy, as waiting too long to implant the EXCOR sharply increases the risk of mortality, while initiating EXCOR support too soon increases the risk of device-related morbidities. The ideal timeframe for implanting the device is, unfortunately, debatable and dependent on several variables (14). Further complicating matters, there are several contraindications to VAD support in children including, but not limited to, extreme prematurity, low body weight (<2.0 kg), preexisting neurologic injury, and congenital anomalies or major chromosomal aberrations with poor prognosis. In addition, multisystem end-organ injury presents a relative contraindication, but does not necessarily preclude patients from consideration of VAD support if end-organ recovery is predicted with improved cardiac output. Again, individual assessment of patient risk profile is of critical importance.

At Texas Children’s Hospital (TCH), the application of the Berlin Heart EXCOR is limited to critically ill children with cardiogenic shock or end-organ injury (INTERMACS profile 1 or 2) (15). The reality is that the vast majority of these patients who undergo Berlin EXCOR implantation are already on ECMO support or mechanically ventilated. In older children who can accommodate an adult-sized implantable device (e.g., HeartMate II or HeartWare HVAD), TCH initiates VAD support sooner, typically at INTERMACS profile 2 or 3. Lower risk profiles of these implantable devices, compared to an extracorporeal pulsatile VAD, justify such an early initiation. Certain clinical circumstances have to be carefully considered before VADs are implanted in children: Congenitally malformed hearts have anatomic variations, such as abnormal size and location of the aorta or unusual location or shape of the ventricle, that pose technical challenges, specifically with regards to placement of inflow and outflow cannulas. In addition, previous surgical palliation may jeopardize the provision of MCS due to further anatomic and circulatory physiologic disorder, such as systemic-to-pulmonary artery shunts or disconnected vena cava and pulmonary artery after the Glenn or Fontan operations. Furthermore, unique pathophysiologic features of heart failure must be taken into account when considering VAD support for patients with single-ventricle physiology. For instance, device selection for the failing Fontan circulation requires careful assessment of the etiology, which can be multifactorial. If systemic ventricular dysfunction is the primary impediment to the Fontan circulation, placement of a VAD to support the failing systemic ventricle would suffice (16). A total artificial heart may be necessary, however, when the Fontan circulation is restricted at multiple levels (e.g., ventricular dysfunction, atrioventricular valve insufficiency, and high pulmonary vascular resistance) (17). In addition, psychosocial and family counseling should be offered, particularly if the possibility of home discharge with an implantable VAD arises, to optimize outpatient management.