Secondary Logo

Journal Logo

Pediatric Circulatory Support

Rehospitalization Patterns in Pediatric Outpatients with Continuous-Flow VADs

Hollander, Seth A.*; Chen, Sharon*; Murray, Jenna M.; Lin, Aileen; McBrearty, Elizabeth; Almond, Christopher S.*; Rosenthal, David N.*

Author Information
doi: 10.1097/MAT.0000000000000505
  • Free


As continuous-flow ventricular assist devices (CF-VADs) are used increasingly in children and adolescents, more pediatric patients will be supported as outpatients as a bridge to transplant (BTT), destination therapy (DT), or bridge to decision (BTD).1–3 Even with excellent postimplantation survival, complications in children are common and include pump malfunction, thrombosis/bleeding events, infection, and right heart failure, all of which may result in hospital readmission.4,5 Repeated hospitalizations in children with chronic conditions represent a significant and potentially modifiable burden to the patient and health care system, including decreased quality of life and scholastic performance, interruptions in normal childhood development, and high utilization of health care resources.6–11

Whereas rehospitalization rates in CF-VAD patients are well described in adults, reports in children are limited to small series where rehospitalizations are not analyzed in detail.5,12–15 Our group recently published our outpatient experience with our first eight patients over a period of 2,413 outpatient days; however, since that report, we have more than doubled the number of patients discharged on CF-VAD support and total outpatient days.16 Moreover, discharges since that analysis have included smaller and younger children who may pose novel management challenges. The purpose of the current study was to investigate the incidence, indications, and outcomes of hospitalizations in CF-VAD patients at a single center and to better characterize the demographics and implications of hospitalization in these patients with a specific focus on admission and discharge diagnoses as well as interventions performed while hospitalized. We hypothesized that rehospitalization in pediatric patients supported with CF-VADs is common, with the most common indications for hospitalization being thrombus and device malfunction. We also hypothesized that device-related procedures were more common in DT patients owing to their longer duration of support.


Study Cohort and Data Source

We performed a retrospective review of all patients who received CF-VADs at Lucile Packard Children’s Hospital, Stanford between December 6, 2010 and November 5, 2016 and survived to hospital discharge. Patients who did not survive their implantation hospitalization or were transplanted before discharge were excluded.

Antithrombosis Prophylaxis, Discharge Criteria, and Outpatient Surveillance

All patients received standard antithrombosis prophylaxis, which included intravenous (IV) heparin started in the first few postoperative days, followed by daily warfarin targeting an internationally normalized ratio of 2.5–3.5. All patients were also started on aspirin (10 mg/kg/day) orally once daily before discharge. Before discharge, all patients were required to complete a comprehensive discharge program, including a standardized rehabilitation plan, VAD education program, and two community reintegration excursions. Our outpatient surveillance schedule has been detailed previously.16

Clinical Data Collection

Baseline demographics, including age, weight, gender, race, and cardiac diagnosis, as well post-VAD hospitalization data were manually extracted from the electronic medical record. Hospitalizations occurring after the date of discharge from the implantation hospitalization through August 1, 2016 were analyzed. All hospitalizations, both scheduled and unscheduled, as well as those for both cardiac and noncardiac indications were included, including those to outside institutions. Hospitalizations specifically for the purpose of transplantation were not included.

For the purposes of data collection, all patients were assigned a single admitting diagnosis, defined as the diagnosis reported in the original admission note. If a differential diagnosis was listed, the diagnosis reported as most likely was considered the admission diagnosis. The discharge diagnosis was assigned according to the primary diagnosis listed on the discharge summary. Secondary diagnoses were not analyzed.


The primary outcomes studied were time to first rehospitalization and number of hospitalizations after the implantation hospitalization discharge through censoring, which occurred at the time of transplantation, death, or transfer to another center, whichever came first. Patients who did not experience any of these outcomes were censored event-free if they reached the end of the data collection period (January 8, 2016). For patients who were transplanted when already hospitalized, their transplant date was considered their discharge date and they were censored at that time. Admitting and discharge diagnoses were also analyzed, as were interventions performed while hospitalized. Summary statistics are reported as number (%) or median (interquartile range [IQR]). For the purposes of comparison, patients designated as either BTD or DT were treated as one group, with comparisons between the BTD/DT and BTT groups performed using Fisher’s exact test for categorical and Wilcoxon rank-sum test for continuous variables as appropriate. Time to first rehospitalization after discharge from the transplant hospitalization was depicted with a Kaplan–Meier curve, with comparisons of freedom from rehospitalization between groups performed using the log-rank test.

Data were collected and stored in RedCAP (version 6.9.7, Vanderbilt University, Nashville, TN), a web-based application designed to support data capture for research studies.17 Statistics were performed using Microsoft Excel (version 14.4.8, Microsoft Corporation) and Stata (version 12.1, STATA Corp., College Station, TX). This study was approved by the Stanford University Institutional Review Board.


Patient Characteristics

Patient characteristics are displayed in Table 1. Twenty-nine patients received CF-VADs during the study period, eight (28%) and two (7%) of whom were transplanted and died before discharge, respectively. The remaining 19 (66%) patients were discharged on device therapy. All discharged patients (100%) had a diagnosis of cardiomyopathy and received left ventricular support only. Eight (42%) of the patients were supported with a Thoratec HeartMate 2 left ventricular assist device (LVAD) and 11 (58%) were supported with a HeartWare HVAD. The goals of VAD therapy were BTT in 12 (63%), DT in five (26%), and BTD in two (11%). Six (32%) patients were female, five (83%) of whom were in the BTT group (p = 0.2). The median age at discharge was 15 (IQR 11, 17) years and was not significantly different between the BTT and DT/BTD groups (p = 0.08). Three (16%) patients had a diagnosis of Becker muscular dystrophy and one (5%) had severe autism. One DT patient (5%) was discharged on both an HVAD and IV milrinone therapy.

Table 1.:
Baseline Demographics (n = 19 Patients, 52 Hospitalizations)


There were 5,101 (median 93, IQR 38, 226) total follow-up days, including 921 (median 68, IQR 35, 101) follow-up days in the BTT group and 4,180 (median 292, IQR 71, 1,437) in the DT/BTD group (p = 0.03). There were a total of 52 rehospitalization events, 21 (40%) in the BTT group and 31 (60%) in the BTD/DT group (p = 0.07). The overall rates of hospitalizations per year were 8.3 in the BTT and 2.7 in the DT/BTD group, which were not significantly different (p = 0.1). Fifty (96%) of the hospitalizations originated at the implanting center, one (2%) hospitalization originated at an outside hospital before transfer to the implanting center, and one (2%) hospitalization included a patient death from hemorrhagic stroke at an outside hospital. Sixteen (84%) patients had at least one hospitalization.

In the total cohort, there were a median of two (IQR 1, 3) hospitalizations per patient, which was the same as for the BTT group alone. In the BTD/DT group, there were a median of two (IQR 1, 10) hospitalizations per patient. There were 577 total hospital days, including 192 intensive care unit (ICU) days, with a median length of stay for all hospitalizations of 4 (IQR 2, 15) days. Twenty-two (42%) admissions were directly to the ICU and 28 (54%) of the admissions included an ICU stay.

The median time to first hospitalization after implantation discharge was 14 (IQR 8, 58) days (Figure 1). Sex, device type, or goal of therapy (BTT versus DT/BTD) did not predict time to first rehospitalization by log-rank test (p = NS). Overall, patients spent 89% of their postimplant discharge days outside of the hospital.

Figure 1.:
Freedom from rehospitalization (n = 19).

Forty-eight (92%) of the hospitalizations resulted in survival to discharge, three (6%) patients died while hospitalized, and one (2%) patient remained hospitalized at the end of the data collection period. Two patients who were hospitalized for VAD-related indications received their transplants while already hospitalized, both of whom survived to discharge.

Indications for Rehospitalization

Six (12%) admissions were for scheduled procedures all related to pretransplant preparation, including cardiac catheterization (one, 17%), plasmapheresis (one, 17%), line placement (one, 17%), and IV immunoglobulin infusion for human leukocyte antigen (HLA) alloantibody desensitization (three, 50%). For the remaining 46 unscheduled admissions, the most common admitting diagnoses were suspected infection in 13 (28%) followed by suspected device thrombosis in 8 (17%) (Figure 2). Two admissions for scheduled procedures were prolonged beyond their intended length of stay for confirmed infection and pump thrombosis, occurring in one patient each. Including these but not the one patient who remained hospitalized at the end of the data collection period (n = 47), the most common diagnoses at discharge were confirmed infection in 13 (28%) and pump thrombosis in 6 (13%) (Figure 3).

Figure 2.:
Admitting diagnoses for unscheduled admissions (n = 46).
Figure 3.:
Discharge diagnoses (n = 47*).

Admissions for Suspected Infection

Of the 13 admissions for suspected infection, a positive blood culture was discovered in only 1 (8%). Two (15%) admissions were for surgical wound infections, one of the driveline site and one of the sternotomy. The remaining 10 (77%) of the admissions were for suspected or confirmed viral infections, most commonly gastroenteritis (4, 40%).

Admissions for Suspected Device Thrombus

There were eight admissions in three patients for suspected device thrombus. All of these patients were supported with a HeartMate 2 LVAD. The reasons for suspicion were elevated lactate dehydrogenase (LDH) in five (63%), increased power readings in two (25%), and both rising LDH and power in one (13%). Of these, thrombosis was confirmed in five (63%). An additional patient admitted for suspected infection was discovered to have pump thrombosis before discharge. Of the four patients with at least one admission for pump thrombosis, one (25%) has died, one (25%) eventually was transplanted successfully, one (25%) has had a device exchanged, and one (25%) had two device exchanges followed by conversion to a HeartWare HVAD.

Interventions Performed

Interventions performed are listed in Table 2. Thirty-one (60%) hospitalizations included at least one intervention. The most common intervention was administration of IV antibiotics in 13 (25%) hospitalizations, followed by an echocardiographic ramp study, a protocolized echocardiogram performed at staged degrees of VAD support to evaluate for pump thrombosis occurring in 12 (23%) hospitalizations.18 Cardiac catheterization occurred in nine (17%) hospitalizations. Seven (13%) admissions in 3/19 (16%) patients included VAD-related surgery, including five pump replacements, one driveline/controller replacement, and one wound exploration. One (2%) patient was placed on extracorporeal membrane oxygenation in the setting of pump thrombosis and did not survive to discharge.

Table 2.:
Interventions Performed During Rehospitalization Events (n = 52)

Patient Outcomes

Forty-eight (92%) hospitalizations resulted in survival to discharge. Of the three (6%) patients who died in the hospital, two (67%) died from hemorrhagic stroke in the setting of appropriate anticoagulation levels and one (33%) died from complications of pump thrombosis in the setting of suspected medication nonadherence. One (2%) patient remained hospitalized at the end of the data collection period. Overall, in the BTT group, six (50%) have been transplanted, one (8%) died, and five (42%) are still on device awaiting transplantation. In the BTD group, one (50%) has died and one (50%) has chosen DT and remains on device. In the DT group, two (40%) patients died, one (20%) was subsequently transplanted at another center, one (20%) was transitioned to the adult heart failure program and died shortly thereafter, and one (20%) remains alive on device therapy. There were no explants.


The current study is unique in its detailed description of rehospitalization patterns in pediatric patients supported with CF-VADs, demonstrating that this population can be discharged with acceptable readmission rates and significant time spent out of hospital. Most patients will be rehospitalized at least once between VAD and transplant, often within 2 weeks of implant hospital discharge, with the most common indications for admission being suspected infection and suspected pump thrombosis. The current study also demonstrates that, consistent with adult studies, device-related complications necessitating surgical intervention are relatively common, particularly in DT patients who are supported for longer periods of time.4

As with adults, rehospitalization in pediatric patients supported with CF-VADs is common.19,20 Forest et al.12 reported a readmission rate of 80% in 71 adults supported with CF-VADs with 35% patients admitted within 30 days of index hospitalization. Hasin et al.13 reported a readmission rate of 1.64 per patient-year of follow-up in 115 patients supported with axial-flow LVADs, noting a higher readmission rate in the first 6 postimplant months. Our finding of a higher readmission rate, 3.7 per patient-year, is likely explained by a more conservative approach to outpatient management of children on CF-VADs resulting in a lower threshold for admission as well as fewer total DT patients in our study. With fewer DT patients, there were fewer total outpatient follow-up days as well as an overrepresentation of the early postdischarge period during which time most rehospitalizations occur. When analyzed separately, the DT/BTD cohort in our study was rehospitalized at a rate of 2.7 per patient-year, similar to that reported for adults.12,13,20

The indications for admission in our study also differ from those reported in adults. Whereas bleeding (mostly gastrointestinal) is a common reason for hospital admission in adults, it occurs infrequently in children supported by CF-VADs and did not occur at all in our cohort. Similarly, heart failure exacerbations, the most common admission indication reported by Tsiouris et al.21 and the second most common reported by Hasin et al.13 in adults, were encountered less frequently in our study. Conversely, suspected infection and suspected device thrombosis were the most common reasons for hospital admission in our cohort, a finding consistent with prior adverse event and rehospitalization indications reported in children with CF-VADs.5,22 Despite these differences, we found that children supported with CF-VADs spend on average 89% of their postimplant time out of the hospital, which is similar to that reported for adults.12

The current study also builds upon the growing experience with outpatient CF-VAD support in children. In 2012, Schweiger et al.22 reported a readmission rate of 0.02 per patient-month in 12 patients from nine centers, although the study was limited by its small sample size and relatively short follow-up time (median follow-up 56 days, maximum 95 days). The most common reason for hospital admission was driveline infection. More recently, Miera et al.15 published a series of 13 children from four centers supported as outpatients with CF-VADs all with a body surface area of <1 m2, reporting 92% survival and readmission occurring only in three patients over a median support time of 111 days with the longest duration of support being 551 days. This study was also limited not only by small sample size but also by the lack of standardized of care owing to several centers contributing data. This year, the Pediatric Interagency Registry for Mechanical Support (PediMACS) reported that 61% of the children discharged with CF-VADs were readmitted at least once, finding that anticoagulation management and device malfunction/alarm were the most common reasons for hospital admission.3,5 Interestingly, rehospitalization in and of itself is not considered an adverse event in PediMACS and no further analysis was performed on this cohort.

Our study contributes to both smaller single-center and larger registry studies with its more detailed description of rehospitalization characteristics in a larger cohort all from a single institution, and with a notably longer follow-up period, including two patients with >1,000 follow-up days. The results of this study underscore the need for providers to consider infection, pump malfunction/thrombus, and right heart failure in VAD-supported patients after discharge. It also emphasizes the importance of the partnership between VAD programs and emergency departments, who will be encountering pediatric VAD patients with greater frequency as the population of pediatric VAD outpatients increases. Lastly, the frequency of complications that occur during the outpatient phase of care emphasizes the need for sophisticated home monitoring programs to identify and rectify potential VAD complications as early as possible.

The current study has several limitations, including its small sample size and retrospective design. Moreover, discerning DT and BTD indications can be difficult in clinical practice, particularly with children. We minimized this limitation as much as possible by cohorting these two groups together for the analysis. Nevertheless, this study demonstrates that most CF-VAD patients will be hospitalized at least once after initial discharge during their course of support, most within 14 days of implant hospitalization discharge, primarily for suspected infectious or thrombotic complications. Despite this, most hospitalizations are short in duration and result in survival to discharge. DT patients for whom support time can be quite long are likely to require surgical revisions to their device, including pump replacement or driveline exchange. Overall, outcomes after discharge in patients with CF-VADs is acceptable, with excellent survival and nearly 90% of postindex hospitalization days spent out of the hospital. As the number of children supported with CF-VADs as outpatients increases, larger studies are required to better delineate the patterns of rehospitalization in this population in an effort to reduce rehospitalization rates, improve patient survival, and maximize quality of life for those on long durations of support.


1. Conway J, VanderPluym C, Jeewa A, Sinnadurai S, Schubert A, Lorts A. Now how do we get them home? Outpatient care of pediatric patients on mechanical circulatory support. Pediatr Transplant 2016.20: 194202.
2. Kirklin JK, Naftel DC, Pagani FD, et al. Sixth INTERMACS annual report: A 10,000-patient database. J Heart Lung Transplant 2014.33: 555564.
3. Rossano JW, Lorts A, VanderPluym CJ, et al. Outcomes of pediatric patients supported with continuous-flow ventricular assist devices: A report from the Pediatric Interagency Registry for Mechanical Circulatory Support (PediMACS). J Heart Lung Transplant 2016.35: 585590.
4. Najjar SS, Slaughter MS, Pagani FD, et al.; HVAD Bridge to Transplant ADVANCE Trial Investigators: An analysis of pump thrombus events in patients in the HeartWare ADVANCE bridge to transplant and continued access protocol trial. J Heart Lung Transplant 2014.33: 2334.
5. Rosenthal DN, Almond CS, Jaquiss RD, et al. Adverse events in children implanted with ventricular assist devices in the United States: Data from the Pediatric Interagency Registry for Mechanical Circulatory Support (PediMACS). J Heart Lung Transplant 2016.35: 569577.
6. Midgley DE, Bradlee TA, Donohoe C, Kent KP, Alonso EM. Health-related quality of life in long-term survivors of pediatric liver transplantation. Liver Transpl 2000.6: 333339.
7. Burdelski M, Nolkemper D, Ganschow R, et al. Liver transplantation in children: Long-term outcome and quality of life. Eur J Pediatr 1999.158(suppl 2): S34S42.
8. Pereira AA, Bhattacharya R, Carithers R, Reyes J, Perkins J. Clinical factors predicting readmission after orthotopic liver transplantation. Liver Transpl 2012.18: 10371045.
9. Medicare MPACRttCpgei. Published June 2007.
10. Vannatta K, Zeller M, Noll RB, Koontz K. Social functioning of children surviving bone marrow transplantation. J Pediatr Psychol 1998.23: 169178.
11. Miller BD, Wood BL. Childhood asthma in interaction with family, school, and peer systems: A developmental model for primary care. J Asthma 1991.28: 405414.
12. Forest SJ, Bello R, Friedmann P, et al. Readmissions after ventricular assist device: Etiologies, patterns, and days out of hospital. Ann Thorac Surg 2013.95: 12761281.
13. Hasin T, Marmor Y, Kremers W, et al. Readmissions after implantation of axial flow left ventricular assist device. J Am Coll Cardiol 2013.61: 153163.
14. 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: 569573.
15. Miera O, Kirk R, Buchholz H, et al. A multicenter study of the HeartWare ventricular assist device in small children. J Heart Lung Transplant 2016.35: 679681.
16. Chen S, Lin A, Liu E, et al. Outpatient outcomes of pediatric patients with left ventricular assist devices. ASAIO J 2016.62: 163168.
17. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009.42: 377381.
18. 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: 17641775.
19. Rose EA, Gelijns AC, Moskowitz AJ, et al.; Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Study Group: Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med 2001.345: 14351443.
20. Slaughter MS, Rogers JG, Milano CA, et al.; HeartMate II Investigators: Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009.361: 22412251.
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: 10411047.
22. 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: 453460.

pediatric; ventricular assist; hospitalization

Copyright © 2017 by the ASAIO