Of more than 500 pediatric heart transplants performed each year, 20% are bridged with either a ventricular assist device (VAD) or a total artificial heart.1 Options for pediatric VADs initially consisted primarily of paracorporeal pulsatile devices, but more recently, continuous-flow (CF) intracorporeal devices have emerged as a feasible option for children. Cabrera et al.2 showed that children supported with the HeartMate II (HMII, Thoratec Corp, Pleasanton, CA) left ventricular assist device (LVAD) had comparable outcomes with that of young adults, with 96% of pediatric patients surviving to transplant, recovery, or ongoing support at 6 months. Several case series have also shown the successful use of the HeartWare ventricular assist device (HVAD, HeartWare Inc., Framingham, MA) in pediatric patients.3–9
Application of CF devices to pediatric patients allows for longer-term support with less restriction in activities of daily living. Whether as DT or bridge to transplant (BTT), the use of these CF devices allows for hospitalization discharge and raises issues unique to the outpatient management of pediatric patients with VADs.
In the past decade, more than 1,800 devices have been implanted as DT in adults, resulting in substantial experience with outpatient management of adults with VADs.10,11 Planning for and managing children with VADs in the outpatient setting, however, are different than for adults, and experience is limited. Special considerations are needed for age-appropriate patient training and support, especially as the psychosocial implications of VAD therapy in pediatric patients are not fully understood, and depression, adjustment, and anxiety disorders have been reported.12 Unlike adults, school integration, including community VAD education for teachers and classmates, plays an important role in the outpatient management of pediatric VAD patients. Furthermore, children have unique physiologic difference in hemostasis that varies across the age spectrum and precludes the use of adult antithrombotic guidelines.13
The purpose of this study was to review our single-center, US experience of pediatric patients discharged from the hospital with CF-VADs. We review our discharge process, outpatient management and follow-up plan, and describe patient outcomes.
All pediatric patients <18 years old with intracorporeal CF-VADs implanted at our institution between June 2010 (first HMII implantation in our institution) and July 2014 were identified. Intracorporeal CF devices offered at our institution include the HMII LVAD and the HVAD. Patients implanted with paracorporeal or pulsatile devices were excluded. Medical records were reviewed. This study was approved by our institution’s institution review board.
Ventricular assist device support was offered to patients with severe refractory heart failure caused by cardiomyopathy or congenital heart disease. The acuity of each patient’s clinical presentation and urgency for VAD support was classified according to Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) profiles. Implantation strategy was classified as BTT or DT. For patients bridged to transplant, our institutional practice generally was to list the patient once extubated after VAD placement. Ejection fraction on echocardiogram and laboratory results before implantation, duration of VAD support, duration of intensive care unit (ICU) stay, and duration of hospitalization were collected.
Patients were deemed appropriate for hospital discharge if they met the following criteria: 1) tolerating a stable outpatient medication regimen; 2) completed a comprehensive VAD education program; 3) had at least one caregiver who could be present with the patient at all times and who had also completed VAD training; and 4) demonstrated achievement of all goals identified in an age-appropriate, standardized rehabilitation care plan.14 An example of our comprehensive discharge checklist is available as Supplemental Digital Content (http://links.lww.com/ASAIO/A88). Outpatient medication regimen for all patients included aspirin (target dose 10 mg/kg/day, maximum 325 mg/day) and warfarin (target international normalized ratio (INR) 2.5–3.5). Other medications, such as loop diuretics, angiotensin-converting enzyme inhibitors, carvedilol, aldactone, and antiarrhythmics, were used as clinically indicated. The VAD education program consisted of approximately ten 60–90 minute sessions with a VAD coordinator and included a written examination and hands-on demonstration of VAD care. The rehabilitation care plan included at least two community reintegration excursions to a local shopping center or local pediatric family house under supervision by specially trained physical and occupational therapists.
All discharged patients were followed in our outpatient clinic, as detailed in Table 1. Any changes in device parameters also prompted a clinic visit with an echocardiogram and laboratory testing. We performed echocardiographic ramp testing according to the protocol described in the Columbia Ramp Study.15 All readmissions were reviewed. Admissions for heart transplantation, scheduled cardiac catheterizations, and desensitization therapy were excluded as readmissions. Patients were admitted for subtherapeutic INRs, and bridging with heparin was typically initiated if the INR was less than 2. Serious adverse events were defined according to standardized criteria from INTERMACS.16 Outcomes included death, heart transplantation, transfer of care to another facility, and ongoing VAD support.
Summary statistics are presented as median (range) or mean ± standard deviation, where appropriate. Patient characteristics were compared using Fisher’s exact and Kruskal-Wallis tests. Statistical analysis was performed on STATA/IC 13 (StataCorp LP, College Station, TX) and p values ≤ 0.05 were considered significant.
Between June 2010 and October 2014, 8 HVADs and 9 HMII LVADs were implanted in 17 pediatric patients at our institution. Of these, eight patients (8/17, 47%) were discharged from the hospital, one with HVAD and seven with HMII. Patient characteristics are shown in Table 2. Compared with those discharged, patients not discharged had shorter overall duration of support (47 ± 38 days versus 394 ± 432 days, p = 0.001), and more were classified as INTERMACS profile 1 at the time of implantation (4/8 patients, 56%, versus 0/9 patients, 0%, p = 0.01), transitioned from extracorporeal membrane oxygenation (extracorporeal membrane oxygenation (ECMO); 4/9 patients, 56%, versus 1/8 patients, 13%, p = 0.16), and received biventricular support (3/9 patients, 33%, versus 0/8 patients, 0%, p = 0.09). The length of stay in the ICU postimplant was not significantly different between the two groups (28 ± 20 vs. 25 ± 15, p = 1.0).
Table 3 summarizes the outcomes of the eight discharged patients, and Table 4 details the individual courses. The overall median time to hospitalization discharge postimplant was 49 (26–107) days. In the first 2 years of our VAD outpatient program, the median time to discharge was 53 (39–67) days. In the most recent year of our VAD program, the median time to discharge has decreased to 29 (26–33) days. Patients spent 91 (7–1,217) days outside the hospital, with BTT patients (n = 4) spending 44 (7–112) days as outpatients and DT patients (n = 4) spending 461 (68–1,217) days. All patients were initially discharged to the local Ronald McDonald House (RMH). The median stay at RMH was 35 (22–49) days. Six patients eventually returned home, a median 99 (1–231) miles from our hospital. Two patients who lived greater than 500 miles from our center remained at RMH for the duration of their support. Four patients went back to community schools, two patients elected to be home-schooled, and the two patients who remained at RMH attended the hospital school.
There were 19 total readmissions, with a median 2 (1–5) readmissions per patient. The first readmission occurred 40 (8–177) days from initial hospital discharge. Length of stay for readmissions was 24 (3–100) days. The most common reason for readmission was pump thrombosis (n = 4, 21%), followed by acute gastroenteritis (n = 3, 16%) and arrhythmia (n = 3, 16%).
Adverse events that occurred while patients were outpatients are shown in Table 5. The most common adverse events were major device malfunction (31%, 5 of 16 events) and cardiac arrhythmias (31%, 5 of 16 events). Major device malfunctions included four episodes of device thrombus, occurring in three patients. One patient died from device thrombus—he was implanted as DT and presented to an outside hospital with nausea and abdominal pain. He was in cardiogenic shock at the time of transfer to our facility and died shortly after arrival. One patient had two episodes of device thrombus requiring device exchange; he was our smallest patient discharged with a VAD (body surface area 1.19 m2), and he was actively receiving desensitization therapy with bortezomib and plasmapheresis during these episodes. The third patient with device thrombus had the thrombus detected during a hospitalization for uncontrolled ectopic atrial tachycardia. For these latter two patients, thrombus was suspected because of rising hemolysis laboratories (lactate dehydrogenase (LDH) and plasma-free hemoglobin) and changes in pump parameters; there was no hemodynamic instability before each device exchange. Another patient required device exchange because of a driveline fracture, likely secondary to a recent fall that damaged his system controller. There was also no hemodynamic instability before his device exchange.
Cardiac arrhythmias occurred in two patients. One patient had an implantable cardioverter defibrillator (ICD) implanted before device implantation for a history of ventricular tachycardia, and she was readmitted for two appropriate ICD discharges for ventricular tachycardia. A second patient, the same patient previously mentioned who also had a device thrombus, had ectopic atrial tachycardia leading to changes in pump parameters but produced no symptoms. He was admitted twice for titration of antiarrhythmic therapy and ultimately underwent a successful ablation procedure.
Each of the following adverse events affected a single patient: hemolysis, minor device malfunction (replacement of system controller because of malfunction), major infection, right heart failure, and systemic drug reaction. There were no bleeding or stroke events. Comparisons of our rates of adverse events to both the fifth INTERMACS report (which included 5,358 patients with CF devices) and the series of 28 pediatric patients with HMII described by Cabrera et al. are shown in Table 5. Because of the small number of patients and events, no formal comparison of the event rates was performed.
All discharged patients implanted as BTT were successfully transplanted (n = 4). The median time to transplant was 137 (106–214) days. All four patients are still alive at a median 20 (14–25) months post-transplant. Of the four patients implanted as DT, one died (after 138 days of support), one was successfully transitioned to an adult VAD service (after 319 days of support), and two are still alive and receiving ongoing VAD support (803 and 1,300 days of support).
Since the initiation of Pediatric Registry for Mechanically Assisted Circulatory Support (PEDIMACS) in September 2012, there has been an increase in pediatric VAD use and in centers implanting pediatric VADs.17 The evolution of devices now allows for outpatient management of pediatric patients with VAD support, but the published literature on outcomes of these patients is limited. In an analysis of the INTERMACS registry by Cabrera et al.,2 the average duration of hospitalization for 28 pediatric patients was 1.13 ± 0.85 months, with an average length of support of 10.8 ± 11.5 months. It is likely that many of these patients were managed as outpatients, but discharge outcomes were not reported. In four case reports of pediatric patients supported with the HVAD, two of the six patients were discharged from the hospital and successfully transplanted.4–7 The longest duration of outpatient support was 55 days. In another case series, Miera et al.3 described seven pediatric patients supported with HVAD for up to 136 days although it was not specified whether any of these patients were discharged.
More recently, there have been two reports on pediatric patients discharged with the HVAD. Schweiger et al.8 described 12 pediatric patients across 9 centers in multiple countries who were discharged after a mean hospital stay of 56 ± 28 days. There were 2.5 readmissions per patient, with driveline infections as the most common reason for readmission. Only one patient was implanted as DT. In a single-center report, Sparks et al.9 described four pediatric patients discharged from their institution with the HVAD. There were six readmissions, and all except one were bridge successfully to transplant. One remains with ongoing support. These results, along with the series we report here, reflect the small but growing experience of pediatric outpatient VAD support.
Adult cardiac centers have had significantly more experience with the outpatient management of patients with VADs. In 2000, Morales et al.10 described 44 patients supported with the HeartMate vented electric (VE) (Thoratec, Pleasanton, CA), a paracorporeal pulsatile device. Over an average 103 ± 16 days of outpatient support, there were no deaths, 42 transplantations, and 2 explantations. Since that report, the use of VADs as DT in adults has increased exponentially; in 2013, there were more than 1,000 new implantations for DT.17 Destination therapy in the pediatric population is very infrequently encountered. Our case series is unique, in that we have had four pediatric patients with VADs for DT. All four were initially implanted as bridge to decision: one because of elevated pulmonary vascular resistance precluding transplant, one because of morbid obesity, and two because of instability of their social support system. The patient with morbid obesity was unsuccessful in maintaining weight loss and was eventually transferred to the adult VAD program. The other three patients interestingly all had Becker’s muscular dystrophy, and although this has not been a contraindication to transplant in our program, all three families elected to remain on VAD therapy. Readmissions were more frequent among DT patients than those among BTT patients, although DT patients spent almost 10 times more days as outpatients as the BTT patients. Individual outcomes of BTT and DT patients are presented in Table 4, but because of small numbers, we did not use statistical testing to compare adverse events between DT and BTT patients.
The fifth annual INTERMACS report described adverse event rates for 5,358 patients with CF devices (see Table 5). The most frequent adverse events were bleeding, infection, and arrhythmias (occurring 9.45, 8.01, and 4.66 per 100 patient-months, respectively).11 In contrast, we did not have any bleeding events in our series of pediatric patients, and our infection rate was notably lower. However, device thrombosis was our most common adverse event, with four episodes occurring in three patients. This was higher than we would have anticipated. The first patient with device thrombus was our first pediatric patient with the HMII device, so this was early in our learning curve of VAD management. The second patient with device thrombosis was small, with a body surface area of 1.19 m2. We now selectively use the HVAD device for our smaller patients. Device thrombosis in the third patient occurred in the setting of an uncontrolled arrhythmia with resultant reduced cardiac output. Because of inability to control the arrhythmia with medications, this patient underwent an electrophysiology ablation. This case highlighted the need for aggressive arrhythmia control in pediatric VAD patients.
All of our device thrombus occurred with the HMII device. Recent studies have suggested a higher rate of thrombosis events with the HMII than that reported in earlier studies.18,19 We proactively monitor for the development of thrombosis with frequent laboratory evaluations and echocardiograms, as detailed in Table 1. Any change in device parameters also prompts a clinic visit with echocardiogram and laboratory testing. With vigilant monitoring, three of the four device thrombus events were detected and addressed before any hemodynamic instability.
All eight patients in our series had at least one readmission. Several adult centers have recently published readmission rates after VAD implantation. Among 115 patients followed for 1.4 ± 0.9 years at the Mayo Clinic, there were 224 readmissions in 83 patients.20 The leading causes for readmissions were bleeding (34 patients), infection (25 patients), and thrombosis (15 patients). Tsiouris et al.21 reported on 138 patients from a single center and found that 70% were readmitted in the first 6 months, with 26% readmitted within 30 days. Heart failure (33%) was the most common etiology for readmission, followed by gastrointestinal bleeding (22%). These studies, along with our experience, highlight that readmissions should be anticipated for patients discharged with VAD support.
Although we have discharged almost half of our patients supported with CF devices, the discharge process has been fairly lengthy, with a median postimplant length of stay of 49 days, despite a median ICU stay of 20 days. This is significantly longer than the median length of stay reported by Sparks et al.9 (17 days with interquartile range of 9.8–25 days). However, it is notable that our outpatient VAD program has evolved over the past 5 years, and we were admittedly cautious in the earlier years. The median time to discharge in the first 2 years was 53 days, yet in the most recent year, our median time to discharge has decreased to 29 days. Overall, achieving rehabilitation goals and completing our comprehensive VAD education program have been the rate-limiting steps in our discharge process. It is also notable that we have been able to discharge more patients with the HMII than HVAD device. This was likely because of the higher initial acuity (INTERMACS profile, ECMO use, and biventricular support) and the shorter duration of support for the patients implanted with HeartWare devices. In fact, two patients with HeartWare devices had fulfilled our discharge criteria and were near-discharge but underwent successful heart transplant before discharge actually occurred.
This study has some limitations. It is a retrospective cohort study, and the sample size is small. This is also a single-center study, subject to selection bias. However, with increasing number of centers reporting to centralized databases such as PEDIMACS, we will soon be better able to analyze outcomes of pediatric patients discharged from the hospital with VAD support.
In conclusion, we have demonstrated that supporting pediatric patients with VADs in an outpatient setting is safe and feasible. Ample times should be allocated for discharge preparation, and readmissions are to be expected. Structured follow-up allows for identification of events before serious hemodynamic consequences. As pediatric VAD use increases, we anticipate a growing trend and experience in the outpatient management of pediatric patients with VAD support.
1. Dipchand AI, Kirk R, Edwards LB, et al.International Society for Heart and Lung Transplantation. The Registry of the International Society for Heart and Lung Transplantation: Sixteenth Official Pediatric
Heart Transplantation Report–2013; focus theme: Age. J Heart Lung Transplant. 2013;32:979–988
2. Cabrera AG, Sundareswaran KS, Samayoa AX, et al. Outcomes of pediatric
patients supported by the HeartMate II left ventricular assist device
in the United States. J Heart Lung Transplant. 2013;32:1107–1113
3. 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
4. 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
5. Crews KA, Kaiser SL, Walczak RJ, Jaquiss RD, Lodge AJ. Bridge to transplant with extracorporeal membrane oxygenation followed by HeartWare ventricular assist device
in a child. Ann Thorac Surg. 2013;95:1780–1782
6. 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
7. Niebler RA, Ghanayem NS, Shah TK, et al. Use of a HeartWare ventricular assist device
in a patient with failed Fontan circulation. Ann Thorac Surg. 2014;97:e115–e116
8. Schweiger M, Vanderpluym C, Jeewa A, et al. Outpatient management of intra-corporeal left ventricular assist device
system in children: A multi-center experience. Am J Transplant. 2015;15:453–460
9. Sparks J, Epstein D, Baltagi S, et al. Continuous flow device support in children using the HeartWare HVAD: 1,000 days of lessons learned from a single center experience. ASAIO J. 2015;61:569–573
10. Morales DL, Catanese KA, Helman DN, et al. Six-year experience of caring for forty-four patients with a left ventricular assist device
at home: Safe, economical, necessary. J Thorac Cardiovasc Surg. 2000;119:251–259
11. Kirklin JK, Naftel DC, Kormos RL, et al. Fifth INTERMACS annual report: Risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant. 2013;32:141–156
12. Hollander SA, Callus E. Cognitive and psycholologic considerations in pediatric
heart failure. J Card Fail. 2014;20:782–785
13. Massicotte MP, Bauman ME, Murray J, Almond CS. Antithrombotic therapy for ventricular assist devices in children: Do we really know what to do? J Thromb Haemost. 2015;13(suppl 1):S343–S350
14. Hollander SA, Hollander AJ, Rizzuto S, Reinhartz O, Maeda K, Rosenthal DN. An inpatient rehabilitation program utilizing standardized care pathways after paracorporeal ventricular assist device
placement in children. J Heart Lung Transplant. 2014;33:587–592
15. Uriel N, Morrison KA, Garan AR, et al. Development of a novel echocardiography ramp test for speed optimization and diagnosis of device thrombosis in continuous-flow left ventricular assist devices: The Columbia ramp study. J Am Coll Cardiol. 2012;60:1764–1775
16. INTERMACS adverse events definitions: adult and pediatric
patients. Appendix A. Interagency Registry for Mechanically Assisted Circulatory Support, National Heart Lung and Blood Institute. Contract Award HHSN268201100025C. Available at: http://www.uab.edu/intermacs
. Accessed December 1, 2014
17. Kirklin JK, Naftel DC, Pagani FD, et al. Sixth INTERMACS annual report: A 10,000-patient database. J Heart Lung Transplant. 2014;33:555–564
18. Starling RC, Moazami N, Silvestry SC, et al. Unexpected abrupt increase in left ventricular assist device
thrombosis. N Engl J Med. 2014;370:33–40
19. Moazami N, Milano CA, John R, et al.HeartMate II Investigators. Pump replacement for left ventricular assist device
failure can be done safely and is associated with low mortality. Ann Thorac Surg. 2013;95:500–505
20. Hasin T, Marmor Y, Kremers W, et al. Readmissions after implantation of axial flow left ventricular assist device
. J Am Coll Cardiol. 2013;61:153–163
21. Tsiouris A, Paone G, Nemeh HW, Brewer RJ, Morgan JA. Factors determining post-operative readmissions after left ventricular assist device
implantation. J Heart Lung Transplant. 2014;33:1041–1047
pediatric; ventricular assist device; outpatient outcomes
Supplemental Digital Content
Copyright © 2016 by the American Society for Artificial Internal Organs