Introduction
Mechanical circulatory support (MCS) for pediatric patients has advanced enormously, since its first reported use in the 1950s by Gibbon, Kirklin, and Lillehei in the form of a cardiopulmonary bypass machine. Kirklin and others were the first group to develop a pump oxygenator with a much smaller priming volume. The rapid evolution of cardiopulmonary bypass continued for the next 20 years, including the refinement of deep hypothermic circulatory arrest in infants by Barratt-Boyes and Casteneda in the 1970s. Over the ensuing decades, continued iterative improvements in cardiopulmonary bypass technology and operative techniques contributed to a significant reduction in mortality associated with pediatric cardiac surgery and paved the way for expanded indications for ventricular support in children with heart failure. While MCS is a remarkably diverse field with a multitude of devices, ventricular assist devices (VADs) are now a key component in the management of advanced pediatric heart failure. The early use of adult VADs in older, adult-sized adolescents demonstrated encouraging outcomes, but the development of the Berlin Heart EXCOR (Berlin Heart AG, Berlin, Germany) was the pivotal event that allowed for mechanical support in children of all sizes. Nonetheless, key technical challenges specific to pediatrics remain to this day, including pump design and the management of anticoagulation. Our understanding of the outcomes of MCS has also continued to grow, as there are now over 700 device placements at more than 40 hospitals documented in the Pediatric Registry for Mechanically Assisted Circulatory Support (PediMACS) database, allowing for detailed evaluation of this important group of patients.
Heart failure in children
Although, worldwide, the major etiologies of heart failure in children include rheumatic heart disease, Chagas disease, cardiomyopathies, and congenital heart disease (CHD), in the United States, the most common causes of heart failure in children are dilated cardiomyopathy and CHD. Epidemiological studies estimate that pediatric dilated cardiomyopathy has a case incidence of ~ 400 cases per year in the US, an estimated case prevalence of 6000 pediatric patients, and a median survival ranging from 7 to 18 years. It is important to note that not all of these patients have severe heart failure. According to the Organ Procurement and Transplantation Network, roughly 300 pediatric patients are listed annually for heart transplantation due to severe heart failure from dilated cardiomyopathy. CHD constitutes a much more heterogeneous group of patients as the probability of heart failure varies greatly by specific lesion ( Fig. 18.1 ).
Single-ventricle lesions are a large contributor to the incidence of heart failure in the CHD population. It is estimated that over 2000 children are born annually with single-ventricle CHD, with approximately 300 CHD children undergoing heart transplantation. Using national datasets, investigators have estimated that 14,000 pediatric heart failure hospitalizations occur annually with a 7.4% unadjusted mortality rate. Of these, 60%–70% are attributed to CHD. Despite the greater prevalence of heart failure admissions for CHD, the vast majority of VADs (80%) are implanted in patients with cardiomyopathy. High mortality associated with pediatric end-stage heart disease stems from directly related sequelae, such as low cardiac output, respiratory failure, malignant arrhythmias, stroke, thromboembolism, irreversible end-organ dysfunction, and infection. This mortality has continued to improve over time with better medical care and advancing surgical repairs, including the use of MCS.
Heart transplantation has become the standard of care for children with end-stage heart disease. Although survival has steadily improved after pediatric heart transplantation, donor availability leads to prolonged waitlist times, and mortality while listed continues to be a major challenge, with mortality rates approaching 20%. However, recent studies have demonstrated a 50% reduction in waitlist mortality in pediatric patients receiving VAD support. Zafar and colleagues found VADs to be an independent predictor of survival, whereas weight < 10 kg, CHD diagnosis, extracorporeal membrane oxygenation (ECMO), mechanical ventilation, and renal dysfunction were all independent predictors of mortality on the waitlist. Evaluating the impact of age on this phenomenon, Law and colleagues showed that the VAD waitlist survival advantage was most profound in adolescent (> 11 years) patients as compared with medical therapy. This is likely an effect of the current stages of device development, with older patients being supported with third-generation VADs while pulsatile support or ECMO are the primary available options for smaller children. In this context, the National Heart, Lung, and Blood Institute (NHLBI) is funding development of the Jarvik 2015, a continuous-flow pediatric-specific device, which is beginning a clinical trial in patients 8–15 kg.
Current devices for pediatric cardiac support
Extracorporeal Membrane Oxygenation
ECMO remains a common method of MCS for pediatric patients due to institutional familiarity and ability to rapidly initiate support. This familiarity has developed over 30 years, but improved outcomes have been limited, with nearly half of cardiac ECMO patients dying prior to discharge. Although individual institutions have shown success and ECMO equipment has improved significantly, widespread success nationally as bridge-to-transplant support has not been realized. In addition, the incidences of mechanical, hemorrhagic, and neurologic complications all exceed 30%.
ECMO was initially used in pediatrics for primary respiratory failure with considerable success, in contrast to adult ECMO for respiratory failure, in which a National Institutes of Health trial was halted due to the absence of a survival benefit. The successful use of ECMO in pediatric respiratory disease is important for two reasons: (1) this finding highlighted the need for pediatric-specific investigations of new devices and (2) successful neonatal respiratory ECMO likely altered the course of development of mechanical cardiac support in children. With clinicians encouraged by its success in the area of respiratory support, ECMO for cardiac support became increasingly common despite its challenges, since it was all that was available for children with acute heart failure.
Initial reports of pediatric ECMO described its application primarily in postcardiotomy patients. Survival rates for all pediatric patients supported for cardiac indications were nearly 50%, with significant complications of bleeding, thrombosis, and neurologic injury in over 30% of cases. In 1992, Del Nido and colleagues described the use of ECMO in 33 patients for cardiac support after cardiac surgery. Eleven of these patients were placed on support following cardiac arrest after an average of 65 minutes of cardiopulmonary resuscitation. Patients supported in this “rescue” fashion had a 64% early and 55% long-term survival rate despite the long duration of cardiopulmonary resuscitation. In addition, survival in this group was similar to that of the entire group. These findings led most centers with significant volumes of pediatric cardiac surgery to develop ECMO “rapid deployment” units.
In cardiac surgery, ECMO has been used to maintain cardiorespiratory function until other cardiopulmonary derangements have been adequately treated, as a bridge to transplantation or as a bridge to long-term device placement. Despite advancing VAD technology, ECMO remains the best support option for patients in cardiogenic shock requiring rapid support for cardiac arrest or to provide cardiopulmonary support as long as a pulmonary component is necessary. If a VAD is initially used, extracorporeal centrifugal pumps, such as the Pedi/CentriMag (Abbott, Lake Bluff, IL) or Jostra ROTAFLOW (MAQUET, Inc., Rastatt, Germany), lend themselves to the “splicing” of an oxygenator into the VAD circuit tubing if respiratory failure ensues. Other novel techniques have been used to place a gas exchange membrane within an existing VAD circuit. The most frequent potential contraindications to ECMO support are severe neurologic injury, prematurity and small size (< 2 kg), active bleeding, irreversible disease, and noncardiac major structural and chromosomal abnormalities. In general, successful arterial-venous ECMO support is usually for periods of less than 2 weeks, whether being used as a bridge to recovery or transplantation. Furthermore, ECMO support at time of heart transplantation is well recognized as one of the highest risk factors for posttransplant mortality, with 1- and 5-year survival reported at 61% and 35%, respectively, significantly lower than that of other pediatric heart transplant recipients whose median survival is > 10 years.
Left Ventricular Assist Devices
In addition to pulmonary function, ECMO supports both the right and left heart using a single cannulation strategy of systemic venous and arterial cannulation. Therefore, in the current era, most centers still rely on ECMO for all patients needing rapid or postcardiotomy support.
Nonetheless, it is clear that unchanged survival rates since 1985, high incidences of complications, and the inability to reliably and safely support children for longer than 2 weeks leave significant room for improvement. Over the same period of development of ECMO technology for children, ever-improving VAD technologies were being developed for long-term adult cardiac support. This technology has been carefully integrated into the care of children with heart failure as we await the development of better pediatric-specific devices.
Temporary Support Strategies
Regardless of device, temporary support has come to mean a quick and simple cannulation strategy for stabilization of cardiac function in (1) a patient thought to have an inflammatory etiology of heart failure (i.e., myocarditis or acute graft rejection) with an expectation to recover within 2 weeks and (2) as support for someone in cardiogenic shock as a bridge to a more durable device. Furthermore, temporary VAD support may provide a period to assess neurologic status or etiology of heart failure prior to placement of a durable VAD or performance of a heart transplant. Typically, this has been accomplished with extracorporeal, centrifugal pumps such as the Centri/PediMag or the ROTAFLOW connected to temporary bypass cannulas placed centrally and off cardiopulmonary bypass. It is the cannulation strategy rather than the choice of pump that best defines the type of support, as Berlin Heart cannulas placed on cardiopulmonary bypass can be used with extracorporeal, centrifugal pumps for extended periods of time as a bridge to transplant. This strategy also allows transition to other devices (e.g., Berlin Heart EXCOR) externally at the bedside and for pump exchange if thrombosis develops. This strategy is becoming increasingly popular, particularly in smaller patients presenting challenges with anticoagulation during their postoperative inflammatory state and for single-ventricle patients with pulmonary artery bands, large aortopulmonary collateralization, or shunted physiology. In these patients, the centrifugal mechanics of the pump make it ideal for rapidly accommodating changes in preload and flow without constant adjustment to the device parameters. Additionally, an oxygenator can be spliced into the circuit relatively easily to simulate ECMO, as can an ultrafiltration/dialysis system to provide renal support. As these techniques in management and others are honed, evidence is accumulating that these historically “short-term” devices can safely support patients for longer periods of time than previously thought.
Two other less commonly used temporary options are delivered percutaneously and are therefore less invasive. These include the TandemHeart (CardiaAssist, Pittsburgh, PA) and the Impella (Abiomed, Danvers, MA). Both devices require large vessels to accommodate the catheter dimensions, generally necessitating a body surface area (BSA) > 1.3 m 2 . The Impella is placed across the aortic valve under echocardiographic guidance to augment cardiac output with axial, continuous-flow support, while the TandemHeart has a venous cannula percutaneously placed through the atrial septum into the left atrium with arterial output to the femoral artery. Although little data exist regarding these devices in pediatrics, a single center reported that 10 Impella devices with a median implant time of 8 days (range, 1–21) supported eight patients (median age, 17 years; range, 6.5–25 years), with all discharged from the intensive care unit. Five patients had posttransplant rejection or allograft vasculopathy causing heart failure, and four patients also required ECMO. Because the pediatric MCS field includes many children with multiple prior sternotomies, there was initial enthusiasm for the TandemHeart percutaneous system, but the fact that most movement, even for daily care (i.e., chest roentgenogram), often causes a change in location requiring repositioning has limited its use in the pediatric community.
Durable Strategies
As “adult” VADs became available over the past 20 years, they have been increasingly used in adolescents. Although covered in detail elsewhere in this text (see Chapter 10 ), durable VADs can be grouped broadly into two separate categories, pulsatile and continuous-flow pumps. Pulsatile pumps were the first VADs developed but have since given way to second- and third-generation continuous-flow pumps in the great majority of cases, with the exception of the Total Artificial Heart (SynCardia, Tuscon, AZ) and the pediatric-specific Berlin Heart EXCOR. In the PediMACS registry, among patients receiving a durable VAD, 94% of patients < 20 kg receive a pulsatile, extracorporeal VAD, while 85% of those > 20 kg get an intracorporeal, continuous-flow VAD, which in the past several years has increased to approximately 95%.
Continuous-flow devices
The HeartMate II (Thoratec) is an axial, continuous-flow pump approved for adults that can provide up to 10 L/min of cardiac output. When the HeartMate II pump was compared in a randomized fashion with the pulsatile HeartMate XVE (Thoratec) in a destination therapy population, it was found to have a significantly greater rate of “survival free from disabling stroke and reoperation to repair or replace the left VAD (LVAD) (62% vs. 7%),” with a 1-year survival of 68% versus 58% for the pulsatile pump. With higher survival, better durability, and more functional capacity leading to a better quality of life, this device initiated the transition from pulsatile to continuous-flow technology in adult devices. This pump, along with others, has been used successfully in pediatric patients, although due to the device size, it is best suited for pediatric patients with a BSA ≥ 1.4 m 2 (~ 40–45 kg). There are reports of pediatric patients discharged home on this device. In a series of pediatric patients implanted with a HeartMate II, 95% were transplanted, recovered, or stable on the device at 6 months postimplantation, with 21% experiencing a bleeding complication, 7% suffering a stroke, and 11% developing sepsis. In comparison to the HeartMate II, the HeartWare HVAD (Medtronic, MN) has a smaller device profile and is the most commonly used durable continuous-flow device in pediatric patients. More recently, the HeartMate 3 (Abbott, Lake Bluff, IL) has been used in children, making the HeartMate II more of historical significance in pediatric MCS.
The HeartWare HVAD, the most common intracorporeal continuous-flow device for children, was implanted in 192 pediatric patients (< 19 years old) in the pediatric portion (PediMACS) of the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) from 2012 to 2017. Within this review, the median BSA was 1.5 m 2 (range, 0.6–2.9 m 2 ), with 12 children weighing ≤ 20 kg, of whom 53% had CHD. The median duration of support was 2.8 months, with fewer infections (27% vs. 44%) and device malfunction/pump thrombosis (11% vs. 19%) compared with matched patients aged 19–30 years and a similar rate of major bleeds (23% vs. 23%) and stroke (10% vs. 12%) as the older patients. Importantly, there was no survival difference between the pediatric patients and young adults, while the rate of discharge on the device was 48% in children ≥ 20 kg. However, at 6 months, 59% of the pediatric patients were transplanted, with 32% alive on device compared with only 17% of the young adults transplanted with 77% alive on the device. These excellent outcomes were corroborated by a global report of 205 pediatric patients, with an 89% positive outcome at 1 year. With outcomes comparable to those in adults, the HVAD is a great device option in children over 20 kg, with success also in some carefully selected patients below 20 kg.
With its Food and Drug Administration (FDA) approval for adults in late 2017, the HeartMate 3 has become an increasingly popular off-label intracorporeal continuous-flow device for pediatric patients. Its complete magnetic levitation technology provides frictionless rotor movement, with the goal of reducing hemolysis and thrombosis. Due to the magnetic levitation, it is slightly larger than the HVAD, although it has a shorter inflow cannula, which can be beneficial in pediatric patients with smaller ventricular cavities. It is also thought to decompress the systemic ventricle better, which may be essential in supporting certain patient cohorts such as Fontan patients. Through the authors’ personal communication (April 02, 2019, verbal communication with David Morales), we are aware of 32 pediatric implants of the HeartMate 3 and know data on 14 at six centers, primarily for dilated cardiomyopathy (93%). The median BSA was 1.8 m 2 (1.3–2.4 m 2 ), with support ranging from 5 to 315 days. Only one patient died on the device, and 36% were discharged home with currently no reports of stroke or pump thrombosis. The HeartMate 3 shows promise to be another excellent device to support pediatric patients, although longer follow-up and determination of the lower size limits must still be established.
Pulsatile devices
The total artificial heart (TAH) is the only currently available pulsatile MCS device for adults and it has also been employed in pediatric patients. The original TAH was conceived by Robert Jarvik in 1982 as the Jarvik-7 and has also had several other names, including the Symbion, Cardiowest, and currently the SynCardia TAH. It was first approved for bridge to transplantation in 2004. Since the development of the Freedom driver (~ 6 kg portable driver), the number of implants has tripled, with over 40% of the > 2000 implants in the past 7 years. The TAH has offered life-sustaining support to a number of pediatric patients who could not be adequately supported with a VAD. Issues with multiple congenital defects requiring repair prior to LVAD support (i.e., intraventricular shunts, semilunar valve insufficiency, etc.), primary cardiac arrhythmias, and ventricular clots are uniquely resolved by replacement of the ventricles and atrioventricular valves with the device. Furthermore, the TAH is suited for specific patient populations, including those with chronic rejection after orthotopic heart transplantation (as the TAH allows for cessation of immunosuppression), late failing Fontan circulations, or chronic biventricular failure in the setting of an LVAD; cancer patients; or those with multiple defects that would require repair prior to VAD placement. The device has biventricular, pneumatic compression chambers, which are implanted in the thoracic cavity in place of the native ventricles. The TAH is capable of providing supraphysiologic cardiac output with the differentiating ability to do so with a minimal central venous pressure that is nearly impossible to achieve with a VAD or even with a newly transplanted heart. This has allowed for recovery of other end organs (e.g., kidney or liver) thought not to be salvageable. The 70 cc TAH is available for use in patients with a BSA > 1.7 m 2 . Because of improving results and increasing use in patients with congenital defects and in pediatric patients, a 50 cc device was developed and is being evaluated in an FDA trial for use in patients with a BSA between 1.2 and 1.85 m 2 or for those in whom fit can be demonstrated by virtual implantation. This is the first time virtual fit has been allowed to be a criterion by the FDA, which is a result of work done by a few programs in the pediatric VAD community. This software has shown that the device could be fit in 33% more patients than using BSA-derived criteria alone. As shown through personal communication with SynCardia (December 03, 2018, email communication with David Morales), the new device has been implanted > 70 times, with a particular increase in congenital defect (4%–9% of implants), pediatric (4%–13%), and female (12%–70%) patients, making it a promising option for a selected pediatric population.
The Berlin Heart EXCOR is a paracorporeal, pulsatile, pneumatically compressed volume displacement pump. It is available in 10-, 15-, 25-, 30-, 50-, and 60-mL blood chamber sizes ( Fig. 18.2 ). Use of customized polyurethane valves has allowed for the manufacturing of smaller blood pump sizes than have been used in “adult” pumps. Furthermore, a bedside driver unit, which is under development, will be available in 2019 and would allow better mobilization and the potential for pediatric patients to be discharged home with the EXCOR for the first time ever in North America. Its first reported successful use as a bridge to transplantation occurred in 1990, with subsequent worldwide experience in more than 2000 patients at > 100 centers. The longest known support is 877 days before successful transplantation. The experience at the Berlin Heart Institute (Berlin, Germany) has been extensively reported. The initial outcomes of 28 patients are displayed in Fig. 18.3 , with 13 dying and 13 bridging to transplantation with 5 early posttransplantation deaths. By 2000, survival had improved significantly, with a 74% survival rate to transplant or discharge. The authors also attributed this improvement to better decision-making, which entailed earlier implantation prior to irreversible organ failure, newer cannulas, switching from atrial cannulation to apical cannulation of the left ventricle, standardization of anticoagulation regimens, and a decrease in biventricular assist support.
The device was first used in North America in 2001 and was widely used by 2006. The initial multicenter experience with the EXCOR reported on 73 patients who had a 77% positive outcome. In 2008, the FDA approved the device for an Investigational Device Exemption (IDE) trial, which enrolled 48 patients in two study arms for smaller and larger patients. Several single-center reports describe many successful outcomes with great variation in the need for LVAD versus biventricular assist device (BiVAD) support, and ultimately, the IDE study confirmed the Berlin Heart EXCOR to be better than ECMO. Enrolled contemporaneously with the IDE trial, the outcomes of all 204 children who underwent EXCOR implantation in North America (including those ineligible for the IDE trial) were reported as the real-world application of the device. The investigators described a successful outcome in 75% of patients, with only one death due to device malfunction. Predictors of early mortality (< 2 months) included lower weight, higher bilirubin, and BiVAD support. The most common cause of death was stroke in 33% of patients, most commonly thromboembolic. More recently, posttransplant survival analysis has shown no difference between patients bridged with a Berlin Heart EXCOR and patients with no device, with 94% 30-day, 90% 1-year, and 72% 5-year survival. Similar outcomes were noted in CHD patients bridged to transplantation, and findings in both series suggested that patients who would have been high-risk transplant candidates (i.e., ventilated, renal insufficiency) were made better candidates with VAD therapy.
Pediatric Device Initiatives
With increased use of the Berlin Heart in North America, the U.S. Department of Health and Human Services identified pediatric MCS as a critical unmet need for pediatric patients. Because of the relatively small numbers of children requiring support compared with the adult market, industry investment was lacking. This prompted the development of a $22 million NHLBI initiative, the Pediatric MCS Program, to promote research and development of mechanical support devices specifically for small children. These awards (2004–2009) led to significant scientific advances culminating in the development of five novel pediatric circulatory support devices. These include a smaller version of the Jarvik 2000 (Jarvik Heart, NY) continuous-flow pump; a smaller pulsatile pump based on the Pierce-Donachy design of the Thoratec PVAD (Penn State Pediatric VAD; Pennsylvania State University–Hershey Medical Center, Hershey, PA); an adult catheter-based pump modified for extracardiac, intrapericardial use (Cleveland Clinic PediPump; Cleveland, OH); an implantable, fully magnetically levitated mixed flow pump (PediaFlow; University of Pittsburgh, Pittsburgh, PA); and a paracorporeal integrated pump oxygenator (Ension pCAS; Ension, Inc., Pittsburgh, PA). Although all of these pumps saw successful animal experimentation and $23 million additional dollars were allocated for further funding, only the Jarvik 2015 is still undergoing testing and is currently in a safety-only FDA trial. The final device design is an intracorporeal, axial continuous-flow device the size of an AA battery for use in children 8–20 kg.
In addition to device development, the NHLBI has recognized the importance of rigorous data collection after approval of a device. INTERMACS has been a pivotal development for the adult MCS field, as has the pediatric-specific PediMACS portion, which now has over 700 device implants (January 12, 2019, verbal communication with PediMACS executive committee). Another recently established network, Advanced cardiac therapies improving outcomes network (ACTION), is a learning network of over 25 pediatric centers focused on improving the quality of care and outcomes for pediatric MCS through real-time sharing of data and management strategies. Together, these networks and registries will be pivotal for researchers looking to optimize VAD care for pediatric patients, especially since individual center volume per year is low (< 10 per year).
Bridge-to-transplantation
An increasing percentage of pediatric heart transplants (currently over 40%) are being performed in patients bridged with VAD support, and current practice is demonstrating excellent waitlist survival while on VAD support with improvement in other end-organ function prior to transplantation. All of this is translating into excellent posttransplantation survival. Comparing posttransplantation outcomes between patients bridged to transplantation with VADs and those with no pretransplantation VAD support, Sutcliffe and colleagues found no difference in 1-year posttransplantation survival (96% vs. 93%), infection (81% vs. 79%), or rejection (71% vs. 74%). Also, they noted that the VAD cohort had greater requirements for inotrope and ventilator support, along with increased liver and renal dysfunction at device implantation. By transplant, both liver and renal dysfunction had normalized in VAD patients and there was substantial decrease in the number of patients requiring inotropic support (94%–33%) and mechanical ventilation (44%–11%). Further evidence for VAD support improving renal function was shown in a PediMACS review of 150 patients with renal dysfunction (estimated glomerular filtration rate < 90 mL/min/1.73 m 2 ), which found that estimated glomerular filtration normalized in 67% of patients by 1 month postimplantation. In patients with renal dysfunction, an average improvement from 62.4 mL/min/1.73 m 2 to 107.7 ml/min/1.73 2 was associated with VAD support. This is important since renal dysfunction is highly associated with poor posttransplantation survival. This renal improvement is seen as early as 1 week postimplantation and generally reaches a plateau at about 1 month.
Discharge from the hospital prior to transplantation may be an indicator of overall wellness and therefore may be associated with better posttransplantation survival. Additionally, hospital discharge has been associated with a lower cost of care while bridging patients to transplantation, which makes a continuous-flow device more desirable in a patient where discharge is possible. Furthermore, the authors are aware of impending reports demonstrating that a period of time, generally ~ 2 months from time of VAD implantation, allows for improvement in numerous end-organ functions (i.e., respiratory, renal, functional, etc.) and is associated with improved posttransplantation survival as compared with less than 2 months of support. This is of critical importance, since patients “limping” to transplantation with other end-organ dysfunction consistently demonstrate worse posttransplantation survival, if they reach transplantation at all. Overall, VADs of all types have continued to improve patient survival to transplantation and are now being examined for the potential to improve posttransplantation survival as well.
Management of pediatric patients receiving cardiac assist device therapy
Cardiac support in children presents particular challenges beyond those seen in adults. Because of their size, devices must be smaller than those used in adults, which adds to the biomechanical complexity. In addition, whereas the majority of adults (> 99%) in need of these services have normal anatomical cardiac and arteriovenous anatomy, the anatomy in children who are candidates for VAD support is frequently complex, since over 20% of VADs are placed in children with CHD. To compound this, the number of children requiring these devices is far smaller than the number of adults, resulting in a prolonged learning curve for each device. This is why ACTION, a multiinstitutional learning network, is such an important addition for improving VAD therapy in the field.
Indications for Mechanical Circulatory Support in Children
As mentioned previously, the general indications for cardiac support in children are primarily bridge to transplantation and, occasionally, bridge to recovery (6%) and, even less commonly, destination therapy (2%). Although chronic care in children is limited, the fact that 50% of children with commonly used continuous-flow devices are discharged home means the line between bridge to candidacy and chronic support is becoming less clear. Although the use of the HeartMate 3 in adults has shown equivalent survival at 2 years to heart transplantation, minimization of complications remains an important hurdle for chronic VAD therapy. There is occasionally a role for destination therapy in certain pediatric cohorts (i.e., Duchene muscular dystrophy, cancer patients, obesity, etc.) who are thought not to be good transplant candidates and in patients with complex CHD for whom heart transplantation is not an option secondary to chronic medical conditions. It is possible in this group that destination therapy will improve end-organ function sufficiently to allow select patients to eventually undergo transplantation.
Acute pediatric heart failure most commonly arises postcardiotomy, induced by prolonged ischemic cardiac arrest, ventricular distention, residual defects, or as an exacerbation of preexisting (but often undiagnosed) myocardial dysfunction. Other cohorts include patients with structurally normal hearts presenting with acute myocarditis, decompensated cardiomyopathy, acute graft rejection, and those with CHD with long-standing myocardial dysfunction who decompensate secondary to a stressor (i.e., respiratory infection, arrhythmia, etc.). Because of the short-term nature of successful ECMO support, initiation of MCS was historically reserved for those patients in extremis with fulminant cardiovascular collapse and impending cardiac arrest, and often only after cardiac arrest. However, with improving devices and MCS experience, it has become clear that implanting devices prior to major patient decompensation allows for the best survival while on circulatory support. This is best demonstrated by the near 50% mortality of INTERMACS level I patients in critical cardiogenic shock undergoing VAD implantation compared with all other levels ( Fig. 18.4 ).
Early initiation of mechanical support may avoid the excessive use of pharmacologic agents (catecholamines) that increase myocardial oxygen demands while also increasing the workload of the heart (systemic and pulmonary vascular resistance). These additional stresses can compound the ongoing myocardial injury. Mechanical support reduces myocardial work while increasing myocardial oxygen delivery. These benefits are enhanced by the frequent ability to wean the patient from pharmacologic support and mechanical ventilation and improve other end-organ function. Conditions that still benefit from the rapid initiation of ECMO with transition to temporary or durable support include patients with impending arrest not amenable to the timeframe required for VAD support, cardiac arrest, and those children presenting with heart failure and significant difficulty to ventilate.
Timing of Support and Device Selection
The field of VAD support has conclusively decided and demonstrated that the early implementation of VAD support, before end-organ deterioration, leads to significantly better outcomes, and this has been corroborated in children by multiple studies. Nonetheless, evidence of end-organ injury should be considered a relative indication for VAD placement and should be a cause for caution when considering heart transplantation, as it is associated with worse posttransplantation survival. Furthermore, significant cardiogenic shock should be considered a contraindication to durable devices, and these patients should undergo temporary VAD support to improve their candidacy for durable support.
Overall, indications for initiation and type of mechanical support are becoming better defined, as displayed in Fig. 18.5 . ECMO would seem warranted for those patients who acutely collapse or have the need for pulmonary support. A temporary device strategy is warranted in those who have a condition thought to be rapidly reversible over the course of days (e.g. arrhythmia, postcardiotomy or posttransplantation, and myocarditis). In some institutions, this might be best performed with ECMO, while in others, a greater familiarity with temporary VADs might tip the balance toward this approach. The projected inability to wean from support in days (< 2 weeks) or a chronic heart failure patient who is worsening in the absence of contraindications (e.g., severe neurologic injury and unrecoverable concomitant illness) should be considered an indication for durable VAD placement. Clinicians should consider VAD support in severe heart failure patients in whom another end-organ system begins to be threatened because of poor perfusion (i.e., the need for intubation due to cardiac failure; escalating or high doses of inotropic support; poor peripheral perfusion as evidenced by acidosis or base deficit, low mixed venous saturation, and cool extremities; and/or the onset of end-organ failure [renal or hepatic failure]).
