Mechanical circulatory support with continuous flow devices is increasingly common as a treatment for advanced heart failure in pediatric patients. The HeartWare HVAD is a small centrifugal continuous flow device, designed to be implanted within the pericardium. It was approved by the FDA for use as bridge to transplantation in adults in 2012.1 There have been limited, but encouraging, reports of use of this device to successfully bridge pediatric patients to transplant as well.2–4 Successful use of the HeartWare HVAD to provide right ventricular and biventricular support has been described in adult patients,5–9 albeit with lower survival compared with left ventricular (LV) support.10 To our knowledge, this is the first description of biventricular support with HeartWare HVAD in pediatric patients.
We describe our experience using the HeartWare HVAD to provide biventricular support to 3 patients < 21 years old at Lucile Packard Children’s Hospital from June 1, 2013 to December 31, 2014. We report clinical course and outcomes with an emphasis on functional recovery as measured by the treatment intensity score (TIS). In addition, we compare our experience with biventricular support with our experience using the HeartWare HVAD for isolated LV support in 5 patients < 21 years old.
Detailed patient characteristics are described in Table 1 for each of the biventricular assist device (BiVAD) patients as well as summary characteristics for the five patients receiving LV support alone with HeartWare left ventricular assist device (LVAD).
BIVAD Patient 1
An 18 year old male with arrhythmogenic right ventricular cardiomyopathy and chronic biventricular failure was initially admitted to the pediatric cardiovascular intensive care unit (CVICU) for recurrent ventricular tachycardia resulting in multiple discharges from his AICD. On hospital day 9, he developed intractable ventricular tachycardia and was emergently placed on extracorporeal membrane oxygenation (ECMO). After stabilization, he was transitioned to HeartWare BiVAD.
VAD Cannula Position
LVAD inflow cannula was placed on the LV apex. Right ventricular assist device (RVAD) inflow ring was placed on the diaphragmatic surface of the right ventricle (RV). RVAD outflow graft was narrowed from 10 to 6 mm. Initial settings achieved flows of 4 LPM (LVAD) and 3.2 LPM (RVAD).
Intraoperative course was complicated by bleeding necessitating transfusion of multiple blood products. A few hours following return from the operating room, he required mediastinal exploration to control bleeding and relieve cardiac tamponade. Following chest closure, VAD settings were adjusted to achieve flows of 3.5 LPM. Anticoagulation management is detailed in Table 2.
Inflammatory Response in the Setting of Marginal Support
The initial postoperative course was complicated by respiratory failure, elevated central venous pressure, and transaminitis concerning for inadequate RVAD support. He also developed an inflammatory response in the absence of infection, with fever, elevated white blood cell count (peak 48.6 K/μl POD 5, normal 4–11 K/μl) and elevated C-reactive protein (CRP; 26.9 mg/dl POD 5, normal < 0.9 mg/dl). He briefly required vasopressor support for low systemic vascular resistance.
Interventions: Banding of RVAD Outflow Compromised RV Support
Postoperative echocardiography demonstrated right ventricular dilatation with poor LV filling. Despite progressive increases in RVAD rate, there was no improvement in echocardiographic findings or clinical cardiac output. On POD 4, the patient underwent diagnostic cardiac catheterization, which revealed elevated filling pressures that only modestly decreased with an increase in RVAD rate (Table 3). Therefore on POD 5, the patient returned to the OR for removal of the RV outflow cannula band. VAD rates were adjusted to achieve flows of 5.5 LPM. This resulted in substantial improvement in clinical cardiac output. With minor adjustment in VAD rate, flows improved to 6 LPM, and patient was extubated 1 week following band removal.
Recovery, Mobilization, and Transplantation
The patient continued to steadily improve; he was weaned from all respiratory support by day 30 of VAD support and mobilized out of bed the following day (day 31). A suitable donor was found and he underwent successful orthotopic heart transplantation after 40 days of VAD support. His post-transplant course was complicated by one episode of focal moderate grade 2/1R treated with outpatient steroid pulse 5 months post-transplant, but has otherwise been uneventful.
BIVAD Patient 2
A 19 year old male, 9 years status postorthotopic heart transplantation with recently diagnosed graft coronary artery disease and biventricular dysfunction, was admitted to the pediatric CVICU with a pleural effusion and fluid overload. After initial stabilization, he developed worsening heart failure with fluid overload and renal dysfunction despite initiation of milrinone therapy.
On hospital day 38, he underwent HeartWare LVAD implantation as bridge to retransplantation. The inflow cannula was placed on the LV apex. He weaned from bypass with adequate right heart function on inhaled nitric oxide (iNO) and vasoactive infusions with LVAD flow of approximately 5 LPM. Intraoperative course was complicated by hemorrhage necessitating large volume transfusion of blood products and anti-inhibitor coagulant complex (FEIBA); however, bleeding was controlled on arrival in the CVICU.
Subsequent RV Failure and RVAD Implantation
Following LVAD implantation, he initially demonstrated stable but moderately decreased RV function, and was able to tolerate weans in inotropic support and slow discontinuation of iNO. However, renal function did not recover and he was initiated on hemodialysis on POD 3. He was extubated on POD 4. On POD 5, he developed RV failure and HeartWare RVAD was placed. Due to the expected long duration of support and difficulty with rehabilitation, HeartWare HVAD was selected. RVAD inflow ring was placed on the diaphragmatic surface of the RV. This required extensive resection of RV muscle and tricuspid valve apparatus. RVAD outflow graft was not banded or narrowed given prior experience with patient 1.
Initial setting achieved flows of 6 LPM (RVAD) and 5.2 LPM (LVAD). There was significant intraoperative and postoperative bleeding necessitating transfusion of multiple blood products. Anticoagulation management is detailed in Table 2. He was extubated on POD 4 from RVAD implantation and began upright in bed mobilization on POD 5. On POD 8, he developed pericardial effusion with tamponade and returned to the operating room for mediastinal exploration and evacuation of hematoma.
Inflammatory Response in the Setting of Marginal Support
He had ongoing respiratory insufficiency requiring BiPAP, and over the next several days developed signs of systemic inflammation in the absence of infection with fevers, elevated CRP (peak >60 mg/dl POD 18), and capillary leak with pulmonary edema and pleural effusion, briefly necessitating vasopressor support for low systemic vascular resistance.
On POD 16, he required emergency intubation for respiratory failure. He underwent diagnostic cardiac catheterization on POD 18, which revealed elevated filling pressures. Filling pressures decreased with increase in LVAD rate resulting in flows of ~7 LPM (Table 3).
Throughout his course, the patient was continued on immune suppressive therapy for his transplanted heart. Due to evidence of increased rates of infection in patients with VADs, immunosuppression targets were lowered. Maintenance daily mycophenolate dose was 950 mg/m2, and goal tacrolimus trough is 4–6 ng/ml.
Recovery and Mobilization
The patient steadily improved during the first weeks following implantation. He was extubated 3 days following the catheterization (POD 20) and was weaned from all respiratory support by day 31 of BiVAD support. He was mobilized out of bed on day 34 and was ambulatory by day 41 of support. He was transferred from the CVICU to the acute care ward on day 63 of BiVAD support.
Additional Complications and Interventions
His course was complicated by renal failure, requiring renal replacement therapy for more than 2 months post-VAD implantation. He was able to stop dialysis on day 68 of BiVAD support. Although discharge from the hospital was considered, the patient was significantly deconditioned and required ongoing hospitalization on the acute care ward for physical and nutritional rehabilitation.
This patient had significant restrictive physiology and fluid overload. On day 145 of BiVAD support, he developed respiratory failure secondary to large pleural effusions. Subsequent diagnostic cardiac catheterizations (support days 146 and 160) revealed elevated wedge pressures (27 mm Hg), and did not significantly improve with changes in VAD settings. Optimally managing his fluid status to prevent pulmonary edema and pleural effusions while maintaining adequate preload to support VAD output and renal perfusion was challenging and eventually required reinitiation of renal replacement therapy on day 170 of BiVAD support.
He underwent repeat orthotopic heart transplantation on day 205 of BiVAD support. Initial post-transplant course was complicated by hemorrhage, respiratory failure, and myocardial stun requiring VA ECMO. He subsequently recovered and was discharged to a rehabilitation facility on post-transplant day 95. He continues to require renal replacement therapy with peritoneal dialysis as of this writing.
BIVAD Patient 3
A 14 year old male with repaired tetralogy of Fallot, and biventricular heart failure was admitted in cardiogenic shock. By hospital day two, he had developed hepatic and renal dysfunction in the setting of ongoing severe biventricular dysfunction and atrial tachycardia. He was urgently intubated and cannulated for VA ECMO. After stabilization of hepatic and renal function, he was transitioned from ECMO to HeartWare HVAD.
VAD Cannula Position
LVAD inflow cannula was placed on the LV apex, and he was weaned from bypass with LVAD in place. Initially the RV appeared to have adequate function, although it was extremely dilated. Soon after separating from bypass, RV function deteriorated and decision was made to reinstitute bypass and place a HeartWare RVAD. RVAD inflow ring was placed on the diaphragmatic surface of the RV. RVAD outflow graft was pexied to the lower end of the sternum and was not banded or narrowed. Initial settings achieved flows of 3 LPM.
Intraoperative course was complicated by hemorrhage necessitating large volume transfusion of blood products and anti-inhibitor coagulant complex (FEIBA). Bleeding was controlled on arrival in the CVICU. Anticoagulation management is detailed in Table 2. He was extubated on POD 5 and began upright, in bed mobilization on POD 8. He developed tamponade on POD 10, for which he underwent emergent chest exploration and evacuation of hematoma in the CVICU.
Inflammatory Response in the Setting of Marginal Support
He had ongoing respiratory failure and systemic inflammation in the absence of infection, with fevers, elevated white blood cell count (peak 35.6 K/μl POD 10) and elevated CRP (peak 30.7 mg/dl POD 12). In addition, he demonstrated low systemic vascular resistance and required frequent adjustments of his VAD settings to maintain adequate clinical cardiac output, as well as initiation of esmolol to slow his native heart rate and optimize VAD filling.
On POD 13, he developed ventricular tachycardia associated with decreased VAD filling necessitating defibrillation. Later the same day, he underwent diagnostic cardiac catheterization, which revealed moderately elevated filling pressures (right atrial pressure 20 mm Hg, pulmonary capillary wedge pressure 17 mm Hg), which showed moderate improvement with changes in VAD settings (Table 3).
Recovery, Mobilization, and Transplantation
The patient slowly improved. He was extubated to BiPAP 5 days following the catheterization (POD 18). He was mobilized out of bed the following day (POD 19). A suitable donor was found and he underwent successful orthotopic heart transplantation on day 22 of VAD support. His initial post-transplant course was complicated by acute renal failure requiring dialysis. Renal function recovered, and he was discharged from the hospital on post-transplant day 49.
Of the five LVAD patients supported, four had dilated cardiomyopathy and one had restrictive cardiomyopathy (Table 1). They ranged in age from 6 to 19 years and in size from 21 to 65 kg (body surface area [BSA]: 0.8–1.8 m2). Median support duration was 39 days (range: 10–155 days). Two patients required surgical exploration for control of bleeding. One patient was discharged home with HeartWare HVAD. Four of the patients were successfully transplanted and subsequently discharged from the hospital. The patient who died was a 6 year old (BSA: 0.8 m2) who had an undefined underlying metabolic disorder resulting in dilated cardiomyopathy. HeartWare HVAD was implanted urgently for LV support after the patient became unstable during cardiac catheterization as part of pretransplant evaluation. He suffered a basal ganglia stroke on POD 27 and a second basal ganglia stroke on POD 29, both thought to be secondary his metabolic disorder and subsequently progressed to multiorgan dysfunction leading to withdrawal of support.
In this cohort of eight patients supported with HeartWare HVAD, 30 day survival was 100%, and overall survival was 88% with 88% of patients transplanted within the study period. All three of the patients requiring BiVAD support were alive post-transplant at the end of the study period.
Treatment Intensity Score
The TIS is a composite scoring system that measures functional status and intensity of therapy in pediatric heart failure patients.11 It is comprised of five measures describing activity level, nutritional support, respiratory support, heart failure treatment, and acuity of care. The TIS provides a standardized measure of functional recovery and improvement in heart failure symptoms. Treatment intensity score was determined on the day before institution of VAD support the 3rd and 7th days of support, and every 7 days thereafter for the duration of support. Lower scores represent lower intensity of support (Table 4). High TIS was the norm at the onset of support. The LVAD patients demonstrated rapid improvement in functional status following VAD implantation; the BiVAD patients’ TIS required more time to improve (Figure 1). Four of five LVAD patients were extubated by POD 3. Of the four LVAD patients supported at least 2 weeks, all were mobilizing out of bed, and 3 were ambulating by POD 14 (Figure 2). The BiVAD patients were extubated on POD 4, 5, and 12; first mobilization out of bed occurred on POD 19, 31, and 33. One BiVAD patient ambulated during the period of support.
Pediatric patients with biventricular heart failure can be successfully bridged to transplantation with the HeartWare HVAD with acceptable morbidity and the potential for significant recovery. In this small series, none of the BiVAD patients suffered a significant neurologic event or stroke during the support course. The reported rate of stroke in adults on HeartWare LVAD is 5.7% for hemorrhagic stroke and 7.1% for ischemic stroke.1
Anticoagulation therapy consisted of heparin infusion, which was transition to warfarin as clinically appropriate as determined by cardiovascular surgery, intensive care, and heart failure teams. Therapeutic ranges for anticoagulation were determined by the clinical team and varied over the course of support (Table 2). In general, goal heparin activity level (HAL) was 0.3–0.6 units/ml. International normalized ratio (INR) goal was variable. For patient 1, INR goal was 2.5–3.5 until POD 25 when it was changed to 1.5–2.5. For patients 2 and 3, INR goal was 1.8–2.5. Antiplatelet therapy was used only in the first patient and thromboelastography was not routinely monitored.
Significant bleeding resulting in tamponade and the need for emergent surgical evacuation of hematoma occurred in each of the HeartWare BiVAD patients. In one patient, bleeding occurred within hours following device implantation, but in the other two, the bleeding occurred after therapeutic levels of anticoagulation had been achieved and the patients had begun mobilizing from bed. For the HeartWare LVAD, two of five patients at our institution (40%) experienced bleeding requiring reoperation. In comparison, a recent report of 60% Berlin BiVAD patients at our institution experienced INTERMACS defined major bleeding (reoperation or transfusion above a defined volume of packed red blood cells). None of the Berlin BiVAD patients at our institution had tamponade. A recent multicenter report on Berlin BiVAD patients demonstrated a 46% incidence of major bleeding, not significantly different from the incidence in Berlin LVAD patients.12 Overall incidence of major bleeding in Berlin Excor investigational device exemption (IDE) and compassionate use patients was 45%.13 Adult HeartWare LVAD literature reports a bleeding rate of 30% with a reoperation rate of 14–23%.1,10 Furthermore, in a series of 17 adult patients receiving biventricular support with two HeartWare HVADs, six (35%) experienced bleeding requiring surgical exploration.9 At this point it is unclear if the high rate of bleeding in the BiVAD patients is related to device positioning, anticoagulation, mobilization, hepatic congestion, or other factors. Additional studies are clearly warranted to address this issue.
Implantation of the HeartWare HVAD on the right atrium,14 diaphragmatic surface of the RV,7 and free wall of the RV have been previously described.5,8–10 In this case series, RVAD was implanted on the diaphragmatic surface of the RV with the cannula directed toward the right ventricular outflow tract. With this configuration, we were able to successfully support these patients, but optimizing support was challenging. In subsequent cases in adult patients, the right atrium was successfully used for the inflow, and we would consider this configuration in future cases.
There have been reports in the literature of patients with normal pulmonary vascular resistance supported with BiVADs developing pulmonary overcirculation with an inability to decongest the lungs.5,9 Given this concern, the RVAD outflow cannula was banded in the first patient. Instead of the intended effect, this resulted in poor LVAD filling with a low clinical cardiac output state. Ultimately, the patient returned to the operating room for removal of RVAD outflow cannula clip, leading to significantly improved clinical cardiac output. Based on this experience, the RVAD outflow cannula was not banded or narrowed for either of the subsequent patients.
The HeartWare BiVAD patients uniformly required diagnostic cardiac catheterization to optimize VAD settings. High filling pressures were commonly encountered and were usually improved with appropriate changes in VAD settings (Table 3). The HeartWare HVAD is both preload and afterload sensitive and requires fine control of blood pressure and fluid balance. All three patients required pharmacologic arterial vasodilation for afterload reduction. However, this was complicated by a systemic inflammatory state with elevated inflammatory markers and low systemic vascular resistance, requiring vasopressor support in two of three patients. Peak inflammatory markers coincided with timing of diagnostic cardiac catheterization for VAD optimization. In the absence of infection, systemic inflammatory response raises concern for inadequate VAD support.
The TIS was used to objectively assess functional status following VAD implantation. Rapid improvement in functional status following HVAD implantation for isolated LV support was seen; however, functional recovery in patients requiring biventricular support was more modest. All three patients were extubated and receiving adequate nutrition within 2 weeks of BiVAD implantation. Mobilization and physical rehabilitation were slower to improve in the BiVAD patients. It has been reported that adult patients are able to recover and to be discharged safely from the hospital with ongoing intracorporeal biventricular support.9 Further experience with biventricular support in children and adolescents will be necessary to explore the potential for out of hospital support.
This report is limited by its retrospective nature, small sample of patients receiving biventricular support with HeartWare HVAD, and limitation to single center. Although successful support and bridge to transplantation is possible, there are significant challenges including high rates of bleeding and limited physical rehabilitation. Further study of surgical technique, anticoagulation strategies, and rehabilitation plans are needed to determine the optimal approach to support in children and adolescents with severe biventricular heart failure.
1. Aaronson KD, Slaughter MS, Miller LW, et al; Use of an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation. Circulation 2012.125: 3191–3200.
2. Miera O, Potapov EV, Redlin M, et al: First experiences with the HeartWare ventricular assist system in children. Ann Thorac Surg 2011.91: 1256–1260.
3. D’Alessandro D, Forest SJ, Lamour J, Hsu D, Weinstein S, Goldstein D: First reported use of the heartware HVAD in the US as bridge to transplant in an adolescent. Pediatr Transplant 2012.16: E356–E359.
4. Padalino MA, Bottio T, Tarzia V, et al: HeartWare ventricular assist device
as bridge to transplant in children and adolescents. Artif Organs 2014.38: 418–422.
5. Strueber M, Meyer AL, Malehsa D, Haverich A: Successful use of the HeartWare HVAD rotary blood pump for biventricular support. J Thorac Cardiovasc Surg 2010.140: 936–937.
6. Hetzer R, Krabatsch T, Stepanenko A, Hennig E, Potapov EV: Long-term biventricular support with the heartware implantable continuous flow pump. J Heart Lung Transplant 2010.29: 822–824.
7. Loforte A, Monica PL, Montalto A, Musumeci F: HeartWare third-generation implantable continuous flow pump as biventricular support: Mid-term follow-up. Interact Cardiovasc Thorac Surg 2011.12: 458–460.
8. McGee EC Jr, Ahmad U, Tamez D, et al: Biventricular continuous flow VADs demonstrate diurnal flow variation and lead to end-organ recovery. Ann Thorac Surg 2011.92: e1–e3.
9. Krabatsch T, Potapov E, Stepanenko A, et al: Biventricular circulatory support with two miniaturized implantable assist devices. Circulation 2011.124: S179–S186.
10. Wu L, Weng YG, Dong NG, et al: Outcomes
of HeartWare ventricular assist system support in 141 patients: A single-centre experience. Eur J Cardiothorac Surg 2013.44: 139–145.
11. May LJ, Ploutz M, Hollander SA, et al: A novel pediatric
treatment intensity score: Development and feasibility in heart failure
patients with ventricular assist devices. J Heart Lung Transplant 2015.34: 509–515.
12. Zafar F, Jefferies JL, Tjossem CJ, et al: Biventricular Berlin Heart EXCOR pediatric
use across the United States. Ann Thorac Surg 2015.99: 1328–1334.
13. Almond CS, Morales DL, Blackstone EH, et al: Berlin Heart EXCOR pediatric ventricular assist device
for bridge to heart transplantation in US children. Circulation 2013.127: 1702–1711.
14. Krabatsch T, Stepanenko A, Schweiger M, et al: Alternative technique for implantation of biventricular support with HeartWare implantable continuous flow pump. ASAIO J 2011.57: 333–335.