Bereaved Caregiver Perspectives on the End-of-Life Experience of Patients With a Left Ventricular Assist Device

Importance  For patients and their loved ones, decisions regarding the end of life in the setting of chronic progressive illness are among the most complex in health care. Complicating these decisions are increasingly available, invasive, and potentially life-prolonging technologies such as the left ventricular assist device (LVAD).

Objective  To understand the experience of bereaved caregivers and patients at the end of life who have an LVAD.

Design, Setting, and Participants  Semistructured, in-depth interviews were conducted between September 10 and November 21, 2014, with 8 bereaved caregivers of patients with an LVAD who were recruited from a single institution. Data were analyzed from December 13, 2014, to February 18, 2015, using a mixed inductive and deductive approach.

Main Outcomes and Measures  Themes from semistructured interviews.

Results  The 8 caregivers (6 females) described 3 main themes that coalesced around feelings of confusion in the final weeks with their loved ones: (1) the process of death with an LVAD, (2) the legal and ethically permissible care of patients with an LVAD approaching death, and (3) fragmented integration of palliative and hospice care.

Conclusions and Relevance  Despite increasing use of LVADs in patients with advanced heart failure, bereaved caregivers of patients with an LVAD describe a high level of confusion at the end of life. There remains a need for the health care community to develop clear guidance on the management of patients with an LVAD at the end of life. Future work will focus on the educational process and the ideal timing and reiteration of such information to patients and families.

Patients with advanced illnesses, as well as their loved ones, face a growing array of decisions about invasive and often life-prolonging technologies.^1,2 One such technology, the left ventricular assist device (LVAD), is becoming an increasingly available treatment option for people dying from end-stage heart failure.³ Use of this innovative technology has expanded from its original indication as a bridge to transplantation to also include destination therapy, in which patients live with the device for the remainder of their lives. Compared with medical therapy alone, LVADs offer the potential to prolong survival and increase quality of life. However, LVAD therapy carries significant risks and burdens, and mortality remains high. At 1 year, approximately 10% of patients have a stroke, 20% have a serious device-related infection, 30% have major bleeding, more than 50% are rehospitalized, and 20% die.⁴ Significant focus has been placed on expanding LVAD programs that support surgical device implantation and subsequent patient management, with relatively less thought on the inevitable end-of-life process awaiting patients whose LVAD is destination therapy.⁵

The process of dying with an LVAD is complicated for both patients and caregivers. Decisions at the end of life in the setting of an LVAD are arguably some of the most challenging in health care. Patients have gone through the decision to undergo implantation, recovery after the surgery, and the work of day-to-day living with the LVAD.^6,7 For many patients, the LVAD stands between life and death: most patients will die within 20 minutes of an LVAD being shut off.⁸ Thus, device discontinuation and death can be particularly dramatic. To illustrate the complexities of end of life in a patient with an LVAD, we explored bereaved caregivers’ perceptions of the death of a loved one with an LVAD.⁵

Box Section Ref ID

We conducted a qualitative, descriptive study using in-depth, semistructured interviews⁹ with caregivers of patients who died with an LVAD. Owing to the lack of existing data on this topic, a qualitative study design was chosen to gain a comprehensive understanding of caregivers’ experiences with a patient with an LVAD at the end of life. The study was approved by the University of Colorado Institutional Review Board. Written or verbal informed consent was obtained from all study participants; those who provided verbal consent did so via the telephone. The interviewer read the consent form verbatim and mailed a copy to the participant. Participants were not compensated for their time.

Caregivers of patients with LVADs who had previously received care at the University of Colorado Hospital were eligible for the study. The University of Colorado Hospital is a tertiary medical center with a moderate-sized mechanical circulatory support and transplantation program. Patients with LVADs are cared for by the mechanical circulatory support team, which includes 6 board-certified advanced heart failure and transplant cardiologists (including L.A.A.), 2 nurse practitioners (including C.K.M.), and 3 mechanical circulatory support coordinators. The study team and mechanical circulatory support coordinators identified caregivers of patients with an LVAD who died and had been observed at the University of Colorado Hospital between September 10 and November 21, 2014. Participants were identified as caregivers if they provided the primary caregiving during the period that the patient had an LVAD and were involved in their loved one’s care at the end of life. We excluded caregivers if the patient’s death had occurred more than 3 years prior, to avoid recall bias. We interviewed a convenience sample of bereaved caregivers who were willing to participate in the study. Interviews were conducted until no new themes emerged. The principal investigator (C.K.M.) directly recruited caregivers via the telephone using her established clinical relationships. Twelve caregivers were identified and approached and 8 caregivers agreed to participate.

Interviews were conducted either in person at the caregiver’s home or by telephone. An interview guide, based on the Ottawa Decision Support Framework,¹⁰ was developed (eAppendix in the Supplement). Given the potential for intense emotional responses, a member of the study team (C.N.) with a social work background, expertise in qualitative research, and no clinical relationship with the patients or caregivers performed the interviews. All caregivers were interviewed alone to allow them to speak as freely as possible regarding their experience. The interviewer composed field notes after each interview. Consistent with study goals and established qualitative research methods,^11-13 these in-depth, semistructured interviews used broad, open-ended questions to elicit personal thoughts and experiences surrounding the end of life in a patient with an LVAD. The interviews were digitally recorded, professionally transcribed, and reviewed for accuracy. The digital recordings were destroyed after transcription. Caregivers also completed a short survey with demographic questions.

Interviews were entered into ATLAS.ti, version 7.0.¹⁴ Data were analyzed from December 13, 2014, to February 18, 2015, using a mixed inductive and deductive approach.^15-17 Two members of the study team reviewed the interviews separately, meeting regularly to reach consensus on the identification and meaning of coded textual data. Through an iterative process, themes were reviewed and discussed to ensure their completeness and contextual authenticity. A multidisciplinary team–based approach was used to add depth to interpretation. Furthermore, using a multidisciplinary team representing the various facets of the end-of-life process ensured that one point of view did not dominate interpretation of themes. The final codebook incorporated both codes derived from data and formal domain codes defined a priori from the Ottawa Decision Support Framework, consistent with our mixed inductive and deductive approach.^15-17 A process of open coding was applied to the remainder of the interviews, along with team discussion and data reimmersion, followed by axial coding in which we established links between the a priori and in vivo codes to provide a conceptual and thematic description.¹⁸ Confirming and disconfirming cases were discussed until consensus was reached. The entire study team determined that thematic saturation was reached when additional interview data created little or no change to the codebook and no new patterns or themes emerged.⁹

We used several strategies to ensure trustworthiness of the findings.^19-21 To ensure transferability, we have described our sample in adequate detail so that others can determine how different populations compare. To ensure dependability, the primary coder kept an audit trail of detailed records during the analytic phase of the project. To ensure credibility and confirmability, we triangulated our findings using a multidisciplinary study team consisting of an advanced heart failure and transplant cardiologist, an advanced heart failure and transplant nurse practitioner, 2 palliative care physicians, and 2 qualitative experts. In addition, themes were presented to a qualitative and palliative care group to seek input regarding trustworthiness.

Between September 10 and November 21, 2014, a total of 8 bereaved caregivers of patients with LVADs were interviewed. All patients were men and ranged in age from 36 to 75 years. Six patients had LVADs as destination therapy. Two patients initially received an LVAD as a bridge to transplantation; however, both patients became ineligible for transplantation (1 owing to substance abuse and 1 owing to comorbidities) and their LVAD indication was transitioned to destination therapy. Etiologic factors of heart failure included 5 ischemic cardiomyopathies, 2 toxin-induced cardiomyopathies, and 1 hypertrophic cardiomyopathy. Four of the patients died in the hospital, 3 died at home, and 1 died in a long-term acute care facility. Four patients had palliative care or hospice services at the end of life.

Interview durations ranged from 31 to 65 minutes (mean, 40 minutes). The median age of caregivers was 64 years and 6 were female. Five caregivers were widows of the patients; other caregivers included 2 friends and 1 mother. Time from the patient’s death to the interview ranged from 1 month to 2 years. Additional demographic data are provided in Table 1. There was a wide range of case histories represented by these patients (Table 2). Overall, caregivers described 3 main themes that centered around feelings of confusion in the final days and weeks with their loved ones: (1) confusion about the process of death with an LVAD, (2) confusion about the legal and ethically permissible care of patients approaching death with an LVAD, and (3) confusion about the fragmented integration of palliative and hospice care.

Seven of the caregivers reported feeling surprised that their loved one was approaching the end of life. This sense of surprise led to feelings of being overwhelmed with the process of death with an LVAD. Many caregivers expressed feelings of anxiety during the days and weeks leading up to the death of their loved one. One caregiver described her confusion after the clinical team discussed with the patient and family that he was likely to die in the coming days and discussed the process around that:

“But it just seemed that the rug got pulled out from under us. And we kinda didn’t understand what was going on, and to what extent it all meant.”

Specifically, caregivers felt unprepared for what to expect regarding the progression toward death and how to handle their loved one should the patient die at home. One caregiver explained that he found his friend slumped in a chair. He dialed 911, and when emergency medical personnel arrived, the caregiver reported that there was a large amount of confusion and misunderstanding about the LVAD itself. The medical personnel were unclear if the patient could receive cardiopulmonary resuscitation or how to know when the patient was dead. It was the caregiver who explained the LVAD to the emergency medical personnel, and he reported that he had to stop them from performing resuscitation efforts:

“I had to go through the litany that people that help with LVADs learn…‘No, he doesn’t have a pulse’; ‘You cannot take his blood pressure with that cuff.’”

Five caregivers expressed confusion about knowing how the patient would die. In the case of a patient who expressed wanting to turn off his LVAD at the end of life, the family felt that even though he wanted to, he was unable to die because the LVAD kept him alive.

“[My daughter] said, ‘You know, Mom, he probably can’t die even though he wants to. Because everything is keeping him going.’”

In addition to being confused about the progression of death, caregivers expressed many different beliefs about the legal and ethical principles that surround deactivation of the LVAD. All the caregivers felt confused and uneducated about the option to turn off the LVAD, and 5 caregivers stated that they were uncomfortable with that option:

“When we talked about that pamphlet, and they put in there that you had that option [unplugging the LVAD], we were both kind of stunned and we talked about it afterwards. We talked about how we never even thought about that.”

Six caregivers viewed turning off the LVAD as suicide and 1 caregiver even went so far as to call it murder. One caregiver described how her husband requested several times that the LVAD be turned off; however, she continued to tell him that it was not his decision, but it was “the Lord’s decision”:

“They don’t actually turn it off until certain signs aren’t there, ‘cause as long as there’s a chance, you know…it would be considered murder.”

Resources available at the end of life, specifically palliative care and hospice, were a major source of confusion for caregivers, who expressed feelings of abandonment by the LVAD team and uneasiness with transferring care to unfamiliar health care professionals:

“He was bleeding again. And I didn’t know what to do. I didn’t know if we could take him to the ER. And I really didn’t understand palliative care, in its entirety. It was kind of that abandonment feeling.”

Four caregivers had a loved one who received palliative care or hospice services. Three of these caregivers stated that they were uncomfortable with hospice and palliative care teams because of their lack of knowledge of the LVAD:

“I remember now, looking at those [flow rate numbers] and feeling alarmed that nobody at [hospice] knew anything about those numbers.”

Patients with chronic progressive illness face an increasing array of decisions about invasive and potentially life-prolonging technologies.² Furthermore, there is a paucity of literature describing end-of-life processes and decisions from a caregiver’s perspective. Our study explores the perspectives of bereaved caregivers surrounding the end of life in a patient with an LVAD, possibly the most invasive technology available at the end of life. We found that caregivers report a high level of confusion at the end of life: about the nature of the death, about the medical decisions to be made, and about the palliative care and hospice services provided. Although some of the concerns regarding palliative care and hospice are not unique to patients with an LVAD, most caregivers expressed confusion regarding death in patients with an LVAD specifically.

The concerns reported in these interviews are not unexpected; many in the health care community continue to grapple with the complex process of death in a patient with an LVAD. More controversial is the issue of deactivating an LVAD.¹ In a survey of physicians about withdrawing LVAD support, 18 (17%) had refused a request to deactivate an LVAD in a patient approaching death and 114 (60%) believed that a patient should be dying before LVAD support should be withdrawn.²² Furthermore, the 2013 International Society for Heart and Lung Transplantation Guidelines for mechanical circulatory support include only 1 recommendation related to the end of life: an ethics committee should be consulted if there is disagreement about discontinuing LVAD support.²³ Owing to the complexity of these issues and lack of guidance from the health care community, it is not surprising that caregivers are left with feelings of distress and confusion about the end-of-life process.

With approximately 2000 LVADs implanted annually in the United States—almost half of them indicated as destination therapy³—the need for specialized end-of-life care is increasing. There remains a need for programs specific to hospice or palliative care to manage these patients. For patients who do transition to hospice care, their caregivers have reported feeling uncomfortable with hospice professionals because they did not fully understand the LVAD.⁸ Furthermore, patients with an LVAD and their families have developed close relationships with the LVAD team and expressed feeling abandoned because palliative care or hospice professionals became their primary health care team. There is also an unmet need for educating patients and families on what to expect at the end of life with this technology. As an LVAD community, our focus should not only be implant and medical management but also the aftereffects of implantation and death with an LVAD.

With any novel, major medical advance comes potentially unforeseen challenges and unintended consequences.²⁴ Improved education of patients, loved ones, and health care professionals about the complexities this technology presents at the end of life is warranted.²⁵ More important, the time and attention given to these end-of-life complexities during the period before implantation is relatively minimal and must be expanded. Even when health care professionals who are comfortable with this material address end-of-life care, there appears to be a mismatch between information shared and assimilation.²⁶

To address some of these issues, our LVAD program has begun to make advancements in end-of-life care for patients with an LVAD. First, we have partnered with a local hospice and palliative care program that provides both in-home and inpatient services. This collaboration has required extensive education and teamwork to provide specialized care and options to patients with an LVAD. Second, the palliative care team has played a more active role throughout the continuum of care for patients with an LVAD, consulting with the patient starting before implantation and for the duration of the patient’s life. Last, advanced care directives are prepared that specifically address device-related end-of-life processes for each patient considering an LVAD.

Continued improvement in end-of-life care for patients with an LVAD is needed, especially for patients with an indication for destination therapy. Physicians must anticipate issues and explicitly acknowledge that dying with an LVAD implanted as destination therapy is expected.^5,27 Furthermore, physicians have an obligation to better prepare patients and families for this reality. Many of the caregivers used the words confusion and abandonment to describe their feelings during the end-of-life process. Similarly, feelings of abandonment at the end of life have been described in patients with cancer and chronic obstructive pulmonary disease as well as their loved ones.²⁸ A first step is to develop clear consensus among the health care community on how to approach management at the end of life for patients with an LVAD. This consensus should then be communicated and practiced to avoid the confusing situations that patients and families are currently experiencing. Future initiatives should focus on education for patients and their loved ones, including the process of death with an LVAD, the decisions patients may face at the end of life, and the availability of palliative care or hospice. In addition, the ideal timing and reiteration of such information needs to be further explored.

There are several limitations to our study. First, this study was performed at a single academic center with a relatively homogeneous convenience sample (all patients were male and 6 of the caregivers were female); however, to our knowledge, this is the first study to focus on the unique population of bereaved caregivers of patients with an LVAD. Second, caregivers were interviewed after their loved one had died, not during the end-of-life process. Last, there was wide variability in the amount of time between implantation of the LVAD and death, with some patients living for years with the LVAD and others only days.

Despite the increasing use of LVADs in patients with advanced heart failure, bereaved caregivers of these patients describe a high level of confusion at the end of life, including expectations of the death process and the ethical and legal principles of deactivating an LVAD. There remains a need to educate and prepare patients and caregivers for the end of life with an LVAD, including more specialized hospice and palliative care programs to manage the complex needs of these patients.

Accepted for Publication: December 23, 2015.

Corresponding Author: Colleen K. McIlvennan, DNP, ANP, Section of Advanced Heart Failure and Transplantation, Division of Cardiology, University of Colorado School of Medicine, 12631 E 17th Ave, Mail Code B130, Aurora, CO 80045 (colleen.mcilvennan@ucdenver.edu).

Published Online: March 21, 2016. doi:10.1001/jamainternmed.2015.8528.

Author Contributions: Dr McIlvennan had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: McIlvennan, Jones, Allen, Matlock.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: McIlvennan, Jones, Allen, Swetz, Matlock.

Statistical analysis: McIlvennan.

Obtained funding: Allen, Matlock.

Administrative, technical, or material support: McIlvennan, Allen, Matlock.

Study supervision: Jones, Allen, Swetz, Matlock.

Conflict of Interest Disclosures: None reported.