2025 年 89 巻 4 号 p. 479-484
Background: In Japan, caregiver presence 24 h/day, 7 days a week is required for patients with left ventricular assist devices (LVADs) during the first 6 months after discharge, with ongoing cohabitation recommended thereafter. This study evaluated the incidence of LVAD pump stoppages during home care, the role of caregivers in preventing adverse events, and the need for continuous caregiver support.
Methods and Results: A retrospective analysis was conducted on 264 patients who underwent LVAD implantation between 2010 and 2023 and were managed at home. In all, 116 power loss incidents were documented, with 65 leading to pump stoppages. Notably, no stoppages occurred in patients using the EVAHEART or HeartMate 3 devices, which are equipped with backup battery systems. Of the 65 stoppages, 83% were resolved by patients and only 6% required caregiver intervention. The Zarit Burden Interview revealed a mean caregiver burden score of 30.1, comparable to that of caregivers for patients with severe brain damage.
Conclusions: The burden experienced by caregivers of LVAD patients is substantial, but with the advent of advanced devices like the HeartMate 3, the need for caregiver support 24 h/day, 7 days a week may be reconsidered. The findings of this study suggest that continuous caregiver presence may not be essential for all LVAD patients, potentially easing the burden on caregivers.
A significant challenge to the management of severe heart failure in Japan is the critical shortage of heart transplant donors. Consequently, many patients who require heart transplantation must rely on long-term support from an implantable left ventricular assist device (LVAD) while waiting for a suitable donor.1 Since April 30, 2021, the use of the HeartMate 3 (HM3) as destination therapy has been covered by insurance, leading to an increase in the number of patients who receive an LVAD without the need for a subsequent heart transplant.2,3
As a condition for heart transplantation, the Japanese Society of Cardiology’s guidelines stipulate that “at least one, preferably 2 or more, immediate adult family members or a spouse must provide support” to the patient after the procedure.4 Furthermore, in Japan, there is an unwritten rule that for a patient to be eligible for home-based LVAD therapy, a caregiver must be present with them 24 h/day, 7 days a week (24/7). Although, in Japan, destination therapy regulations permit patients to live independently from their caregivers 6 months after discharge, and similar allowances have been extended to bridge-to-transplant therapy as of April 2024,5 LVAD patients are still prohibited from being alone during the initial 6-month post-discharge period.
In a previous study we reported that approximately 10% of patients referred to the Osaka University Hospital Heart Center for the treatment of severe heart failure declined heart transplantation or LVAD therapy because they did not have a caregiver who could provide continuous 24/7 support.6 The stringent caregiver requirements for LVAD therapy in Japan effectively deprive many patients with severe heart failure of equitable access to optimal treatment. Although the critical role of caregiver support in the management of severe heart failure, particularly in the case of long-term LVAD therapy, is well recognized,7–9 few countries uniformly prohibit implantable LVAD therapy for patients unable to secure a caregiver, and Japan remains unique in requiring 24/7 caregiver support.6
The aim of the present study was to assess the need for 24/7 caregiver attendance for the continued safety of home-based LVAD therapy. We evaluated the frequency of LVAD pump stoppage and stroke episodes during the home care period and the extent to which caregivers were able to prevent adverse events resulting from these episodes. In addition, an anonymous survey was conducted to examine adherence to the 24/7 caregiver requirement, and caregiver burden was measured using the Zarit Burden Interview (ZBI).
This study was approved by the Institutional Review Board of Osaka University Graduate School of Medicine, Japan (Approval no. 13369, May 11, 2017) and was conducted in compliance with the Declaration of Helsinki. The requirement for written informed consent from participants was waived due to the anonymous nature of data collection.
Between January 2010 and December 2023, 320 patients were implanted with an LVAD at the Heart Center, Osaka University Hospital. Of these patients, 264 who were discharged and managed with an LVAD at home were included in the present study. We investigated the incidence and characteristics of LVAD power supply failures and stroke events during the observation period, which extended to May 2024. In addition, in November 2018, 60 pairs of LVAD outpatients and their caregivers who attended the Osaka University Hospital Heart Center outpatient clinic were asked to complete an anonymous questionnaire regarding their daily lives and caregiving experiences. A caregiver burden survey, using the Japanese version of the ZBI,10 was also conducted, with 43 of the 60 patient-caregiver pairs responding to the survey.
The characteristics of the 264 patients at the time of LVAD implantation are summarized in Table 1. The mean (±SD) age of the patients was 44.1±15.0 years, and 70.8% were male. The leading cause of heart failure was idiopathic cardiomyopathy (79.5%), followed by ischemic cardiomyopathy (9.5%) and post-myocarditis cardiomyopathy (4.2%). The devices implanted included the HeartMate II (HMII) in 74 patients, HM3 in 63 patients, Jarvik2000 (J2K) in 38 patients, HVAD in 33 patients, DuraHeart (DH) in 31 patients, and EVAHEART (EH) in 25 patients.
Patient Characteristics at LVAD Implantation
| Age (years) | 44.1±15.0 |
| Sex | |
| Male | 187 (70.8) |
| Female | 77 (29.2) |
| Diagnosis | |
| Idiopathic cardiomyopathy | 210 (79.5) |
| Ischemic cardiomyopathy | 25 (9.5) |
| Post-myocarditis cardiomyopathy | 11 (4.2) |
| OthersA | 18 (6.8) |
| Implanted devices | |
| HeartMate II | 74 (28.0) |
| HeartMate 3 | 63 (23.9) |
| Jarvik2000 | 38 (14.4) |
| HVAD | 33 (12.5) |
| DuraHeart | 31 (11.7) |
| EVAHEART | 25 (9.5) |
Data are given as the mean±SD or n (%). LVAD, left ventricular assist device. AArrhythmogenic right ventricular cardiomyopathy 4, drug induced cardiomyopathy 4, cardiac sarcoidosis 4, valvular cardiomyopathy 2, left ventricular noncompaction 2.
LVAD Power Loss Episodes
During the observation period, there were 116 episodes of power loss at home. These included incidents of battery depletion, bilateral battery disconnection, driveline disconnection, and controller malfunction. The distribution of episodes of power loss according to device is shown in Figure 1A; specifically, 27, 13, 2, 9, 16, and 49 episodes occurred in patients implanted with a DH, HVAD, EH, HMII, J2K, and HM3 device, respectively. Of the 116 episodes, 65 resulted in pump stoppage: 27, 13, 9, and 16 episodes in patients with DH, HVAD, HMII, and J2K devices, respectively (Figure 1B). Notably, no pump stoppages due to power loss occurred with the EH or HM3 devices, which have controllers equipped with a built-in backup battery. Of the 65 pump stoppages, 83% were resolved by the patients themselves and 6% were resolved by caregivers; in 11% (n=7) of cases, the pump failed to restart, resulting 5 in fatal events (Figure 2). Details of these 7 cases of pump failure are provided in Table 2. All 7 patients who experienced pump failure had one or more designated caregivers at the time of the incident; in 4 cases, the caregiver was physically present but unable to prevent the critical event, in 2 cases the caregiver was not present, and, in the remaining case, it was suspected the caregiver was not present. The devices involved in these fatal cases were DH (1 patient), HVAD (2 patients), HMII (2 patients), and J2K (2 patients), with no fatalities occurring in patients implanted with the EH or HM3 devices.
Device-specific distribution of (A) power loss episodes and (B) pump stoppage episodes due to power loss during the observation period.
Pie chart showing who restarted the pump in the case of 65 pump stoppages.
Critical Events Due to Pump Stoppage
| Patient no. |
Age (years), sex |
Device | Cause of pump stoppage |
Place where event occurred |
Did the patient had a designated caregiver? |
Was the caregiver present at the time of the event ? |
Outcome |
|---|---|---|---|---|---|---|---|
| 1 | 61, F | Jarvik2000 | Battery exhaustion |
Out of home (Pachinko parlor) |
Yes | Yes | Patient died |
| 2 | 42, M | HeartMate II | Battery exhaustion |
Out of home (Pachinko parlor) |
Yes | Unknown | Recovered |
| 3 | 32, M | DuraHeart | Driveline disconnection |
Out of home (Train station) |
Yes | Yes | Brain damage |
| 4 | 57, F | HeartMate II | Bilateral battery disconnection |
Home | Yes | Yes | Patient died |
| 5 | 30, M | Jarvik2000 | Controller failure | Out of home (Restaurant) |
Yes | Yes | Patient died |
| 6 | 59, M | HVAD | Unknown | Home | Yes | No | Patient died |
| 7 | 52, F | HVAD | Controller failure | Home | Yes | No | Patient died |
F, female; M, male.
Stroke Episodes
During the observation period, 51 patients experienced ≥1 stroke events. A significant number of stroke events occurred during hospitalization, mainly in patients with device infections. During home LVAD therapy, 22 episodes of stroke were recorded (Table 3); the devices implanted in patients who experienced a stroke event were the EH (n=8), J2K (n=5), DH (n=4), HMII (n=4), and HVAD (n=1). No stroke episodes at home were recorded in patients implanted with the HM3. At the time when the symptoms occurred, the first contact with medical staff was made by the patient themselves in 5 (22.7%) cases, by the caregiver in 16 (72.7%) cases, and by a colleague in 1 (4.5%) case. The leading type of stroke was cerebral hemorrhage (n=8; 36.4%) followed by cerebral infarction (n=7; 31.8%), subarachnoid hemorrhage (n=5; 22.7%), and transient ischemic attack (n=2; 9.1%).
Stroke Events
| No. patients with stroke | 51 |
| No. stroke episodes at home | 22 |
| Mean (±SD) age (years) | 48.1±15.1 |
| Male : female ratio | 14 : 8 |
| Mean (±SD) LVAD-supported daysA | 521±450 |
| Device | |
| EVAHEART | 8 |
| Jarvik2000 | 5 |
| DuraHeart | 4 |
| HeartMate II | 4 |
| HVAD | 1 |
| HeartMate 3 | 0 |
| Who made the first contact with medical staff when the symptoms occurred? | |
| The patient | 5 (22.7) |
| Caregiver | 16 (72.7) |
| Colleague | 1 (4.5) |
| Type of stroke | |
| Cerebral bleeding | 8 (36.4) |
| Cerebral infarction | 7 (31.8) |
| Subarachnoid hemorrhage | 5 (22.7) |
| Transient ischemic attack | 2 (9.1) |
| Treatment | |
| Emergency rehospitalization | 22 (100) |
| Conservative treatment | 18 (81.8) |
| Endovascular treatment | 2 (9.1) |
| Surgery | 2 (9.1) |
| Outcomes | |
| Recovered | 14 (63.6) |
| Neurological deficit (mRS <3) | 2 (9.1) |
| Neurological deficit (mRS ≥3) | 1 (4.5) |
| Death | 5 (22.7) |
Unless indicated otherwise, data are given as n (%). AAt the time of the stroke event. mRS, modified Rankin scale.
All patients with a stroke event were readmitted as emergencies. Eighteen patients (81.8%) underwent conservative treatment, 2 (9.1%) underwent endovascular treatment, and the remaining 2 (9.1%) underwent surgery (craniotomy for the removal of a hematoma or cerebral decompression). Fourteen (63.6%) patients recovered without any symptoms, 2 (9.1%) recovered with mild neurological deficit (modified Rankin score <3), and 1 (4.5%) developed significant neurological deficit (modified Rankin score ≥3). The remaining 5 patients died.
Anonymous Questionnaire SurveyThe results of the anonymous survey regarding adherence to the 24/7 caregiver attendance rule revealed that 30%, 44%, and 26% of respondents reported adhering to the rule “always”, “generally”, and “half the time”, respectively (Figure 3A). When asked “Who accompanies the patient when they go on a 1- to 2-h outing?”, 62% indicated “the caregiver”, 7% reported “someone other than the caregiver”, and 31% responded “the patient goes out alone” (Figure 3B). Of the 7 patient-caregiver pairs who responded to the question about commuting to work, 71% of patients reported commuting alone and 29% reported commuting with the caregiver (Figure 3C). Regarding the question “Does the caregiver feel more stressed than before LVAD implantation?”, 14% responded “very much”, 28% “a lot”, 28% “somewhat”, 21% “a little”, and 7% “not at all” (Figure 3D).
Results of the anonymous survey of patient-caregiver pairs (n=43 unless indicated otherwise).
ZBI
The results of the ZBI indicated a mean (±SD) burden score of 30.1±12.0 among the caregivers of LVAD patients at home (Table 4). This exceeds the threshold score of 24, which is considered indicative of a risk of developing depressive symptoms.11
Zarit Burden Interview (Japanese Version) Scores
| Mean (±SD) score | |
|---|---|
| Total ZBI score | 30.1±12.0 |
| Personal strainA | 14.6±6.9 |
| Role strainB | 6.0±3.9 |
ZBI, Zarit Burden Interview. ABurden caused by the care itself. BBurden caused by the caregiver’s inability to maintain his or her previous lifestyle because of the caregiver’s initiation of caregiving.
This study provides critical insights into the management of LVAD patients at home, particularly regarding the need for continuous caregiver attendance. One of the most striking findings of this study was the complete absence of LVAD pump stoppages and related fatal episodes in patients using the HM3 and EH devices. Furthermore, no stroke episodes have been recorded thus far for patients with the HM3 being managed at home. This contrasts sharply with other devices, where such incidents, including 7 critical events due to pump stoppage and 5 deaths due to stroke, were observed. In recent years, HM3 has become the mainstream implantable LVAD, with significant reductions in events such as pump stoppages and severe stroke.12,13 The superior safety profile of the HM3, likely due to its built-in backup battery and advanced design features, underscores the potential for this newer technology to reduce the burden on both patients and caregivers. The findings of the present study challenge the current “24/7 caregiver attendance rule,” particularly for patients using more advanced LVADs like the HM3. The results of the anonymous survey demonstrated varying levels of adherence to the 24/7 caregiver rule, with only 30% of caregivers always complying with this requirement. In addition, ZBI scores indicated a considerable burden on caregivers, with many at risk of developing depressive symptoms.
The findings of this study provide a nuanced perspective on the role of caregivers in LVAD management, a topic that has been extensively explored in the literature. Previous studies have consistently emphasized the vulnerability of LVAD patients to device-related complications.14 However, our finding that a significant majority of pump stoppages were managed independently by the patients themselves suggests that the reliance on caregivers may decrease as patients gain confidence and competence in managing their devices. This finding resonates with the broader literature on LVAD patients, which highlights the potential for patient autonomy with adequate training and support,15,16 suggesting that the caregiver’s role may evolve from one of constant vigilance to a more supportive, on-demand presence.
The variability in adherence to the 24/7 caregiver attendance rule in our survey highlights the practical challenges faced by many families. This is consistent with studies that document the immense physical, emotional, and financial strain placed on caregivers of LVAD patients.17 Caregivers often report high levels of stress, anxiety, and depression, which are exacerbated by the constant need to be vigilant and the fear of potential adverse events.18 The burden is particularly acute in Japan, where cultural expectations and the absence of a robust support system for caregivers can intensify these pressures.19
The global perspective on LVAD caregiving also offers valuable insights. In contrast to Japan, many countries do not impose such stringent caregiver requirements,20 and studies from these regions suggest that less restrictive caregiving arrangements do not necessarily lead to worse outcomes. For example, in the US and Europe, where the emphasis is often placed on patient autonomy and self-management,21,22 there has been no significant increase in adverse events reported in patients with less intensive caregiver involvement. This raises important questions about the balance between ensuring patient safety and recognizing the rights and capacities of patients to manage their own health. It also suggests that Japan may benefit from re-evaluating its caregiving policies, potentially adopting a more individualized approach that takes into account the specific needs and capabilities of each patient-caregiver dyad.
The ZBI is a tool used to assess the subjective burden experienced by caregivers, particularly those involved in the long-term care of patients with chronic and complex conditions.23 In the present study, the ZBI scores for LVAD caregivers averaged 30.1±12.0, a finding that is particularly noteworthy when considered in the broader context of caregiver burden across different medical conditions. A ZBI score above 24 is generally associated with a significant risk of developing depressive symptoms and other stress-related disorders.11 Compared with caregivers of other chronic conditions, the burden faced by LVAD caregivers is comparable or even higher. For instance, caregivers of patients with dementia typically report ZBI scores ranging from 25 to 30, reflecting the intense and often unpredictable nature of dementia care.24 Similarly, caregivers of stroke survivors have reported ZBI scores in the range 25–28, particularly when the stroke results in significant physical or cognitive impairment.25 The ZBI sore of caregivers of patients on home enteral nutrition was reported to be 36.5.26 These conditions are known for placing heavy demands on caregivers, both physically and emotionally, often leading to burnout and other negative outcomes.
Study LimitationsThis study has several limitations that should be acknowledged. First, the retrospective design of the study may introduce bias, particularly in the reporting of LVAD power loss episodes and caregiver responses. The anonymous nature of the survey also limits the ability to verify the accuracy of the reported adherence to the 24/7 caregiver rule. In addition, the sample size, although substantial, may not fully represent the broader population of LVAD patients in Japan, particularly those treated in smaller or less-specialized centers. Furthermore, the study’s observational nature precludes establishing a causal relationship between caregiver presence and patient outcomes.
The findings of this study highlight the evolving landscape of LVAD therapy, where advances in LVAD technology, particularly with devices like the HM3, have significantly enhanced the safety of home-based care. The absence of fatal episodes due to pump stoppages and stroke in patients using the HM3 device demonstrates that modern LVADs can operate reliably even without the constant presence of a caregiver.
As the safety of home LVAD therapy continues to improve, it is both reasonable and necessary to reconsider and potentially relax the caregiver attendance rules. This shift would not only align Japan’s practices with those of other countries but also ensure that severe heart failure patients have broader access to life-saving LVAD therapy without being hindered by excessively demanding caregiver requirements.
During the preparation of this manuscript, the authors used DeepL to assist with English language editing.
This research received no grants from any funding agency in the public, commercial, or non-profit sectors.
The authors declare no conflicts of interest.
This study was approved by the Institutional Review Board of Osaka University Graduate School of Medicine (Approval no. 13369, May 11, 2017).
The deidentified participant data will not be shared.
