Left Ventricular Assist Device Caregiver Experiences and Health Outcomes: A Systematic Review of Qualitative and Quantitative Studies

Abstract

Background

Knowledge synthesis is lacking regarding outcomes and experiences of caregivers of adult patients living with continuous flow left ventricular assist devices (CF-LVAD). The purpose of this systematic review was to summarize qualitative data related to the experience of caregivers of adult patients living with CF-LVAD as well as quantitative data related to health outcomes of caregivers.

Methods and Results

Multiple databases were systematically queried for studies of qualitative experiences and quantitative health outcomes for caregivers of adult CF-LVAD recipients. Search dates were constrained to articles published between 2004 and August of 2018 because CF-LVADs were not routinely implanted before 2004. Two authors independently screened 683 articles; 15 met predetermined inclusion criteria. Eligible articles reported results from 13 studies. Of those, 8 used either qualitative or mixed methods and 5 used quantitative methods. Caregivers were primarily female (81%) and their mean age was 59 years. Qualitative studies revealed 3 overarching themes related to the caregiver role, coping strategies, and LVAD decisions. Quantitative studies revealed caregiver strain peaked between 1 and 3 months after implantation, anxiety and depression were relatively stable, mental health status improved, and physical health status was stable from before to after implantation.

Conclusions

CF-LVAD caregivers experience significant, sustained emotional strain for 3 months after implantation, reporting considerable stress in meeting their personal needs and those of their loved one.

In patients with advanced heart failure (HF) refractory to medical therapy, continuous flow left ventricular assist devices (CF-LVAD) are associated with 73% survival at 2 years,¹ compared with 8%–53% with medical management alone.^2–4 In addition, most patients show improved exercise capacity, less physical and psychological symptom burden, and improved quality of life (QOL) after CF-LVAD implantation.^5,6 Yet, the demands of living with a CF- LVAD require dramatic alterations to daily lifestyle and routines such as bathing, positioning for sleep, sterile dressing changes, frequent medical appointments, and equipment management.7 Owing to the substantial amount of care needed by patients with CF-LVADs to live safely at home, many institutions require that patients have a dedicated caregiver to receive the device.^8,9 Many implanting centers request that the patient have a caregiver available 24 hours a day, 7 days per week after the initial hospital discharge with a CF-LVAD. The caregivers typically assume many responsibilities, including assisting the patient to bathe with the CF-LVAD controller and batteries, medication and equipment management, coordination with the health care team, and frequent transportation to and from the implanting center.¹⁰ Furthermore, the driveline that delivers power to the CF-LVAD exits through the abdomen and requires daily to weekly sterile dressing changes to prevent infection. Caregivers must learn and perform the sterile technique for these dressing changes. Caregivers are integral to improved outcomes in patients with CF-LVADs.¹¹

Research in HF and other chronic illnesses suggests that caregivers may sacrifice their own health to meet the needs of the person for whom they care. Caregivers of persons with chronic illness have poorer health outcomes than noncaregivers, including higher levels of depression and anxiety, fewer healthy behaviors, worse QOL, and high mortality.^12–15 A recent qualitative meta-synthesis by Magid and colleagues¹⁶ described the perception of caregivers of patients with CF-LVADs. Results revealed that caregiving was stressful, with caregivers reporting fears, anxiety, uncertainty, and hypervigilance, concluding that caregivers were at high risk for poor psychological outcomes. Despite this knowledge, and the intensity of caregiving required for patients with CF-LVAD, no intervention programs have been developed to address mental or physical health outcomes for caregivers of patients with CF- LVAD. Thus, urgently needed is a greater understanding of caregiver experiences and the determinants of their health- related outcomes, which will provide essential information to develop targeted interventions and improve outcomes. The purpose of this systematic review is to synthesize the current qualitative literature related to the experiences and quantitative literature related to health outcomes of caregivers of adult patients living with CF-LVADs.

Methods

Design

This systematic review included qualitative studies of caregiver experiences and observational quantitative studies reporting health outcomes of caregivers for adult recipients of CF-LVAD. The investigators identified qualitative studies that used a variety of qualitative approaches to study caregiver experiences, including qualitative descriptive, hermeneutic phenomenology, and grounded theory approaches that typically used semistructured interviews and focus group methodology. Quantitative studies used objective measurement tools to summarize caregiver outcomes, with numerical analyses and statistical methods to determine associations between study variables. Mixed methods studies were defined as those that used both qualitative and quantitative methods. The term caregiver was defined as a person who provides care for a person living with CF-LVAD in the home setting. The caregiver relationship with the patient was either spouse or intimate partner, parent, other family member, or friend. The caregiver must have agreed to be a caregiver for the person living with a CF-LVAD. The selection of studies for review was not defined by the type of relationship between the caregiver and person with CF-LVAD, length of caregiving, CF- LVAD designation (eg, bridge to transplant [BTT] vs destination therapy [DT]), or CF-LVAD type (eg, Heart- Mate II or III [Abott], or HeartWare HVAD [Medtronic]).

The literature search was conducted following the PRISMA¹⁷ recommendations for carrying out a systematic review. Inclusion criteria were (1) published qualitative articles that focused on experiences or observational quantitative articles that focused on health outcomes of caregivers for patients with CF-LVAD, (2) caregivers and patients were 18 years of age or older, (3) articles published in English in peer-reviewed journals, and (4) articles published in 2004 or later. Exclusion criteria were (1) articles in the grey literature such as nondatabased articles, editorials, commentaries, books, conference abstracts, opinion pieces, protocol papers, or review papers, and (2) articles that focused only on patient experiences or outcomes.

Search Strategy

A systematic search was conducted via PubMed, EMBASE, Cochrane, and CINAHL Plus using the following key words: caregiver, carers, carer, care givers, care giver and for the device: VAD, BiVAD, LVAD, ventricle- assist device, ventricle assist device, ventricular assist device, heart-assist device, heart assist device, heart-assist pump, heart assist pump, ventricular-assist device, artificial ventricle, mechanical circulatory support, assist device, circulatory support. The search was limited to articles published 2004 or later, because CF-LVADs were not routinely implanted before 2004. This review focuses only on the type of left ventricular assist device used currently, CF- LVAD.

Search parameters were adapted according to the database requirements, combined and expanded MeSH terms. To complement the search, references from the included articles and systematic reviews found with the search terms were manually searched for additional references. Search results were transferred to Rayyan QCRI,18 a free web- based application for systematic review data management and collaboration. All database searches were performed on August 14, 2018.

Study Selection

With duplicate studies removed, the title and abstract of each article was independently screened and coded by 2 reviewers (MS and JA) according to the inclusion and exclusion criteria. Articles were coded as included or excluded in the Rayyan web-based application. The Rayyan application allows for blinding of collaborators’ decisions, and disagreements are automatically coded in Rayyan as a conflict. Any article coded as a conflict was screened by a third reviewer (AL), who evaluated the article and determined article inclusion or exclusion. Articles included based on the title and abstract screening were imported into a spreadsheet, and the full text of each article was reviewed for a final determination of inclusion or exclusion. In cases where full-text articles did not specify device type (continuous flow vs pulsatile), study authors were contacted by email for clarification. There was no blinding to study author, institution, or journal during the article screening process. A PRISMA flow diagram illustrates the article review process (Fig. 1).

Fig. 1.

PRISMA flowchart.

Search Outcomes

The initial search across all databases resulted in 683 articles; after the removal of duplicates (n = 82), there were 601 articles for review. After title and abstract review, 26 articles were selected for full-text review. In addition, the manual reference search identified 3 additional articles, resulting in a total of 29 articles for full-text review. The type of LVAD device remained unclear in 4 articles. We contacted the corresponding authors for clarification, resulting in the inclusion of 2 and exclusion of 2 articles. A total of 14 articles were excluded after the full-text review, 7 owing to non—CF-LVAD and 7 owing to lack of caregiver outcomes; therefore, 15 articles^8,11,19–32 were included in this systematic review. A summary of the included studies is presented in Table 1 (qualitative and mixed-methods studies) and Table 2 (quantitative studies).

Table 1.

Qualitative and Mixed Methods Studies

Author(s) (date), Study Design, n	Sample Characteristics	Research Question(s) and Methods	Time	Summary of Main Findings
Egerod et al (2012)22  Qualitative descriptive, single center  n = 7	Sample: Close relatives of LVAD patients chosen by the patient to participate  Caregiver characteristics:  Male sex: 1 (14%)  Age: —  Caregiver type: spouse 4 (57%), parent 2 (29%), sibling 1 (14%)  Patient characteristics:  Male sex: 3 (43%)  LVAD designation: BTT 7 (100%)  Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aims:  - to describe the experiences and coping strategies of LVAD patients’ close relatives during BTT and  - to generate theory  Analytic approach:  - grounded theory  Sampling strategy:  - convenience  Interviews:  - paired/grouped  - Mean 87 minutes, range 84–90 minutes	Time since LVAD implantation: Data collected an average of 22.4 months post-transplant (range 0–34 months, 2 patients still awaiting transplantation); patients had LVAD average of 8.3 months pretransplant (range 1–19)	Categories and theoretical ideas:  1. Supporting the Patient  -1.1 Close relatives were there for the patient  -1.2 Close relatives had to be emotionally strong  -1.3 Close relatives had to be practical  -1.4 Close relatives lived vicariously for the patient  2. Suffering and self-preservation  -2.1 Close relatives endured and suffered  -2.2 Close relatives made sacrifices  -2.3 Close relatives eventually needed to help themselves  -2.4 Supporting the patient was a means of self-preservation  Model: Support cascade model (see article)
Marcuccilli et al (2014)25  Qualitative hermeneutic phenomenology, single-center  n = 7	Sample: Family caregivers of LVAD patients living at home for at least 3 months after implant hospitalization  Caregiver characteristics:  Male sex: 1 (14%)  Age: 65 ± 8.3 y  Caregiver Type: spouse 6 (86%), son 1 (14%)  Patient characteristics:  Male sex: —  LVAD designation: DT 7 (100%)  Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aim:  - to explore the experience of caring for an adult with a DT LVAD from the caregivers’ perspective.  Analytic approach:  - thematic analysis  Sampling strategy:  - convenience, clinical registry  Interviews:  -2 interviews per participant  - first interview face to face, one on one  - second interview telephone, one on one  -mean 51 minutes ± 15 (range 27–69)	Time since LVAD implantation: Minimum 3 months living at home post LVAD implant hospital discharge (no mean provided)	Themes/subthemes:  1. Advanced heart failure is a life-changing event  - 1.1 Hope for life renewed  2. Self-doubt about LVAD caregiving improved with time  - 2.1 Learning to care all over again  3. Lifestyle adjustments come with time  - 3.1 A reversal of roles  - 3.2 Self-interests on hold  - 3.3 Keeping it normal  - 3.4 Doing all the driving  - 3.5 Daily dressing changes  - 3.6 Moving forward  4. Persistent worry and stress  5. Caregiving is nota burden—it’s part of life
Marcuccilli et al (2011)31  Hermeneutic phenomenology, single-center  n = 5	Sample: Family caregivers of LVAD patients living at home for at least 3 months following implant hospitalization  Caregiver characteristics:  Male sex: 0 (0%)  Age: 57 ± 17 y  Caregiver type: spouse 3 (60%), parent 1 (20%), adult child 1 (20%)  Patient characteristics:  Male sex: —  LVAD designation:  BTT 4 (80%), DT 1 (20%)  Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aim(s):  - to explore and describe the lifestyle adjustments of caregivers for adults living with long term implantable LVADs  Analytic approach:  - hermeneutic phenomenology  Sampling strategy:  - purposive, convenience  Interview type:  - semistructured, open ended  - range 24–62 minutes	Time since LVAD implantation: Not provided	Themes/subthemes:  1. Caregiving is a 24/7 responsibility  -1.1 Overwhelming experience  -1.2 Hypervigilance  -1.3 Less personal time  -1.4 Supportive role  2. Coping strategies  -2.1 Positive thinking  -2.2 Good family support  -2.3 Developing a routine  -2.4 Acceptance  -2.5 Prayer  -2.6 Faith and hope  -2.7 Religiosity and fatalism  3. Caregiver satisfaction  -3.1 Grateful for witnessing a second chance at life  -3.2 Life transforming event
McIlvennan et al (2016)27  Qualitative descriptive, single center  n = 8	Sample: Bereaved LVAD caregivers  Caregiver characteristics:  Male sex: 2 (25%)  Age: median 64 y (range 54–71 y)  Caregiver type: widow 5 (63%), parent/mother 1 (13%), friend 2 (25%)  Patient characteristics:  Male sex: —  LVAD designation: DT 8 (100%)  Time since LVAD implantation: LVAD patient deceased, with death < 3 years prior (range 1–24 months)  Charlson Comorbidity Index: —	Research aim:  - to explore bereaved caregivers’ perceptions of the death of a loved one with an LVAD  Analytic approach:  - mixed inductive and deductive approach  Sampling strategy:  - convenience, clinical registry  Interviews:  - in-person or by telephone semistructured, one on one  - guided by Ottawa decision support framework  - mean 40 minutes (range 31–65)	Time since LVAD implantation: LVAD patient deceased, with death < 3 years prior (range 1–24 months)	Theme/subthemes:  1. Feelings of confusion in the final days and weeks with their loved ones  -1.1 Process of death with an LVAD  -1.2 Legal and ethically permissible care of patients approaching death with an LVAD  -1.3 Fragmented integration of palliative and hospice care
McIlvennan et al (2015)20  Qualitative descriptive, single-center  n = 17	Sample: Caregivers of alive or deceased LVAD patients  Caregiver characteristics:  Male sex: 1 (6%)  Age: 63 y (35–79 y)  Caregiver type:  spouse/widow 12 (70%), parent/mother 1 (6%), daughter 2 (12%), sibling/sister 1 (6%), friend 1 (6%)  Patient characteristics:  Male sex: — LVAD designation: DT 17 (100%) Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aim:  - to understand the caregivers’ experiences and identify their needs related to decision making surrounding DT LVAD  Analytic approach:  - mixed inductive and deductive approach  Sampling strategy:  - convenience, clinical registry  Interview type:  - semistructured interviews in person or by phone without patient present  - guided by Ottawa Decision Support Framework  - range of 21–72 minutes per interview	Time since LVAD implantation:  <5 years post LVAD implant;  Patient living n = 10  Patient deceased n = 6  Patient declined n = 1	Main theme:   Complex decision-making process  Themes (domains)/subthemes:  1. Decision context  -1.1 Tension: hope and reality  -1.2 Years of poor health  -1.3 A sense of urgency  -1.4 Ineligibility for transplant  2. Decision process  -2.1 Tension: wanting loved one to live and wanting to respect their wishes  -2.2 Influence of relationships with the clinicians  -2.3 Making the decision  3. Decision outcome  -3.1 Tension: Gratitude and a feeling of burden  -3.2 Reflecting on the decision
Blumenthal-Barby et al (2015)28  Cross-sectional, mixed methods with grounded theory (qualitative) and survey (quantitative), single center  n = 15	Sample: Family or significant other caregivers of LVAD patients  Caregiver characteristics:  Male sex: 5(33%)  Age: 59 y (36–74 y)  Caregiver type: —  Patient characteristics:  Male sex: 11 (73%)  LVAD designation: BTT 1 (7%), DT 14 (93%)  Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aim:  - To investigate the decision-making process and informational and decisional needs of LVAD patients and their families, specifically related to:  - the decision and options, knowledge, expectations, and gaps, values and values clarity, and needs for other support and resources  Analytic approach:  - grounded theory  Sampling strategy:  - theoretical, purposive  Interview type:  - in person or by phone  - semistructured with progressive focusing  - guided by Ottawa Decision Support Framework  - mean 35 minutes (range 12–87 minutes)  Quantitative survey  Decisional regret  -Decisional Regret Scale	Time since LVAD implantation: mean 634 days (50–1845) after LVAD implant	Themes:  1. There was no decision to make  2. Decisions were reflexive, made without deliberation  3. Decisions were based on trust in physician  4. Optimistic expectations regarding transplant  5. Values are clear, include life extension, family, and mobility  6. A need to hear from other patients and caregivers  7. A need for an involved and supportive clinical team and caregiver  Model: Knowledge needs fell into 8 domains including lifestyle, rehospitalizations, complications, prognosis (with vs without LVAD), technical (drive-line, battery), the surgery experience, financial, and caregiver lifestyle (see article)  Decisional regret  (n = 13 completed)  - decisional regret score: 1.48 (on 1–5 scale, where 1 indicates no regret)
Kitko et al (2013)21  Qualitative descriptive, single center  n = 10	Sample: Spousal caregivers of LVAD patients currently living at home  Caregiver characteristics:  Male sex: 2 (20%)  Age: 62 6 (52–73 y)  Caregiver type: spouses 10 (100%)  Patient characteristics:  Male sex: 8 (80%)  LVAD designation: DT 10 (100%)  Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aim:  - to describe the experiences of spousal caregivers of patients with stage D heart failure from pre-LVAD to post-LVAD-DT implantation  Analytic approach:  - inductive  Sampling strategy:  - purposive, convenience  Interview type:  - semistructured interviews  - range 45–90 minutes	Time since LVAD implantation: 15.9 months (range 5–38 months) post-LVAD implant	Theme/subthemes:  1. Adaptation within the role as a caregiver  -1.1 Caring for a spouse with heart failure  -1.2 Decision for LVAD implantation made  -1.3 Caring for a spouse with LVAD-DT  Model: Caregiving Transition (see article)
Kirkpatrick et al(2015)29  Concurrent mixed methods with qualitative descriptive (qualitative) and cross-sectional survey (quantitative), multicenter  n = 42	Sample: Primary or secondary caregivers of patients with DT LVAD  Caregiver characteristics:  Male sex: 9 (21%)  Age: 60 y (30–78 y)  Caregiver type: spouse/partner 35 (83%), child 6 (14%), parent 1 (2%)  Patient CHARACTERISTICS:  Male sex: 32 (82%)  LVAD designation: DT 42 (100%)  Time since LVAD implantation: —  Charlson Comorbidity Index: —	Research aim(s):  - to identify burdens and stressors associated with caregiving for patients with DT-LVADs  - to characterize QOL of caregivers of patients with DT-LVADs  Analytic approach:  - modified grounded theory  Sampling strategy:  - convenience, multisite LVAD clinical registries  Interview type:  - semistructured, open ended, telephone  - interview length not provided  Quantitative survey  QOL:  - City of Hope QOL Family Caregiver instrument (modified for LVAD caregivers)	Time since LVAD implantation: Not provided	Themes/subthemes  1. Emotional bind: caregivers during the decision-making process  2. Extent of preparation  3. Postimplant social and psychologic burden  -3.1 Task and time constraints  -3.2 Impact on freedom/independence  -3.3 Everyday psychologic burden  4. Value of psychological and social support  QOL Survey  QOL scores (higher scores are better):  -Psychologic subscale 6.23  -Social subscale 6.67  -Spiritual subscale 7.39  -Physical subscale 7.86  - LVAD <2 years 7.13 ± 1.24  - LVAD ≥2 years 6.98 ± 1.29  - Age <40 y 7.41 ± 1.58  - Age 40–49 y 7.17 ± 0.62  - Age 50–59 y 6.87 ± 1.67  - Age 60–69 y 6.74 ± 1.43  - Age ≥70 y 7.31 ± 1.04  - Male 7.73 ± 0.94  - Female 6.85 ± 1.3  - Advance directive 7.36 ± 1.16  - No advance directive 6.65 ±1.36

BTT, bridge to transplant; DT, destination therapy; LVAD, left ventricular assist device; QOL, quality of life.

Table 2.

Quantitative Studies

			Results
Author(s) (date), Study Design, n	Sample Characteristics	Outcomes	3–4 Weeks After LVAD (n = 33)		3 Months (n = 30)	6 Months (n = 25)
Brouwers et al (2015a, 2015b)30,32  Multicenter (2 medical centers in the Netherlands, 1 medical center in Canada), prospective  n = 33	Sample: Dyads (LVAD partners currently engaged in caregiving)  Caregiver characteristics:  Male sex: 9 (27%)  Age: 54 ± 11 y  Caregiver type: —  Patient characteristics:  Male sex: 24 (73%)  LVAD designation: BTT 33 (100%)  Time since LVAD implantation: Baseline was 3–4 weeks after LVAD implant (just before hospital discharge)  Charlson Comorbidity Index: range = 0 to ≥5, no mean provided  Dropouts: 8 (24%)	3–4 weeks, 3, and 6 months after LVAD implant  2015a  Health status/QOL:  - SF-12 (MCS and PCS)  2015a and 2015b  Anxiety:  - HADS-A  Depression:  - HADS-D  2015b  PTSD:  - Post-traumatic Stress Diagnostic Scale (PTSD scale)	Health Status SF-12 PCS:51.8 ± 1.2  MCS: 40.4 ± 2.2  Anxiety  HADS-A: 7.6 ± 3.5, 15 (48%)  Depression  HADS-D: 6.0 ± 3.8, 12 (39%)  PTSD  PTSD scale: 4 (12%)		Health Status SF-12 PCS: 53.3 ± 2.8  MCS: 47.8 ± 2.2  Anxiety  HADS-A: 7.5 ± 3.1, 12 (44%)  Depression  HADS-D: 5.7 ± 3.0, 9 (33%)  PTSD  PTSD scale: 4 (12%)	Health Status SF-12 PCS: 52.0 ± 3.1  MCS: 46.8 ± 1.9  Anxiety  HADS-A: 6.4 ± 4.9  6 (22%)  Depression  HADS-D: 4.5 ± 3.1, 5 (19%)  PTSD  PTSD scale: 5 (14%)  Change across time  Anxiety prevalence: P < .01  Depression prevalence: P < .01  PTSD prevalence: P < .01  *No P values reported for PCS or MCS
Bidwell et al (2017 & 2018)8,26  Single center, prospective n = 41 (2017, interim analysis)  n = 50 (2018)	Sample: Dyads (LVAD patient and caregiver)  Caregiver characteristics:  Male sex:  2017: 8 (20%)  2018: 9 (18%)  Age:  2017: 54.7 ± 11.4 y  2018: 54.7 ± 12.5  Caregiver type:  2017: spouse 30 (73%), parent 7 (17%), other 4 (10%)  2018: spouse 38 (76%), parent 7 (14%), other 5 (10%)  Patient characteristics:  Male sex:  2017: 35 (85%)  2018: 42 (84%)  LVAD designation:  2017: BTT 25 (63%), BTD 3 (8%), DT 12 (30%)  2018: BTT/BTD 36 (72%), DT 14 (28%)  Time since LVAD implantation: Enrolled 5 days before LVAD implant  Charlson Comorbidity Index: 2.5 ± 1.5  Dropouts:  2017: 4 (10%)  2018: 9 (18%)	5 days before LVAD, 1 and 3 months post-LVAD  n = 41, 2017 QOL:  - EuroQOL visual analog scale (EQ-5D VAS)  Depression:  - Patient Health Questionnaire (PHQ-8)  Anxiety:  - Anxiety subscale of BSI  5 days pre-LVAD, 1, 3, and 6 months post-LVAD  n = 50, 2018  Caregiver strain:  - Multidimensional Caregiver Strain Index (MCSI)	5 days pre-LVAD Quality of Life  EQ-5DVAS: 52.9 ± 8.2  Depression  PHQ-8: 5.6 ± 4.9  Anxiety  BSI: 0.7 ± 0.6  Caregiver Strain  MCSI: 37.0 ± 1.5 (SD not reported)	1 month post-LVAD QOL  EQ-5D VAS: 51.5 ± 9.3  Depression  PHQ-8: 5.1 ± 4.7  Anxiety  BSI: 0.6 ± 0.7  Caregiver Strain  MCSI: 40.6 (SD not reported)  Change across time:  MCSI before LVAD to 1 month:  P < .05	3 months post-LVAD QOL  EQ-5D VAS: 49.8 ± 10.5  Depression  PHQ-8: 6.3 ± 5.5  Anxiety  BSI: 0.6 ± 0.7  Caregiver Strain  MCSI: 40.5 (SD not reported)  Change across time:  QOL before LVAD to 3 months  P =.03  Depression before LVAD to 3 months:  P = .22  Anxiety before LVAD to 3 months:  P = .45	6 months post-LVAD: Caregiver Strain  MCSI: 37.0 (SD not reported)  Change across time:  MCSI 1 month to 6 months:  P < .05  MCSI before LVAD to 6 months:  P = .74
Rossi Ferrario et al (2016)19  Retrospective cross-sectional  n = 16	Sample: LVAD caregivers of patients who died after LVAD implant/hospital discharge  Caregiver characteristics:  Male sex: 3 (19%)  Age: 53.5 ± 16.2 y  Caregiver type: partners 8 (50%), sons/daughters 6 (38%), other 2 (12%)  Patient characteristics:  Male sex: 16 (100%)  LVAD Designation: DT 11 (69%), BTT 2 (12%), BTD 3 (19%)  Time since LVAD implantation: mean survival after LVAD implant 11 ± 9 m  Charlson Comorbidity Index: —  Dropouts: —	34.6 ± 16 months post mortem  Mourning:  - CMQ  Time mourning:  - months  Perception of experience with LVAD:  - locally derived 10-point scale  - Positive and negative aspects of physical, emotional and social support during previous 3 months	34.6 ± 16.0 months post mortem Mourning  CMQ: 14.6 ± 8.5  Time in mourning  34.6 ± 16.0 months  Perception of experience with LVAD  10-point scale 5.33 ± 4.3  Positive aspects: well-being of patient (n = 5), longer survival (n = 5), subjective wellness of caregiver (n = 1)  Negative aspects: difficulty managing infections, driveline, and patient’s incomplete autonomy (n = 8), patient suffering (n = 2)  *No P values reported
Rossi Ferrario et al (2017)24  Single-center, prospective  n = 13	Sample: LVAD patients (n = 27) and caregivers (n = 13) enrolled in inpatient rehabilitation  Caregiver characteristics:  Male sex: 0 (0%)  Age: 60 ± 7 y  Caregiver type: spouse 12 (92%), sibling 1 (8%)  Patient characteristics:  Male sex: 26 (96%)  LVAD designation: DT 7 (26%), BTT 9 (33%), BTD 11 (41%)  Time since LVAD implantation: Admission to inpatient rehabilitation at 37 ± 22 days post-LVAD implant. Discharge from inpatient rehabilitation was 28.5 ± 9.0 days after rehabilitation admission  Charlson Comorbidity Index: —  Dropouts: 2 (15%)	Inpatient rehabilitation admission and 8–12 months after inpatient rehabilitation discharge  Family strain  - FSQ-SF	Inpatient rehabilitation admission Family strain  FSQ-SF: 18.4 ± 4.8			8–12 months post inpatient rehabilitation discharge Family strain  FSQ-SF: 15.5 ± 7.6  Change across time  FSQ-SF admit to after discharge: P = .05.  *FSQ-SF of 13–20 indicates a need for psychological support, so may not be clinically significant
Sample: dyads (LVAD patient and family caregivers)  Primary study longitudinal  Secondary study cross-sectional  Caregiver characteristics:  Male sex:  Primary- 5 (25%)  Secondary- 8 (20%)  Age:  Primary- 49.6 ± 11.9 y  Secondary- 50.0 ± 11.1 y  Caregiver type:  Primary- spouse 10 (50%), sibling 4 (20%), parent 6 (30%)  Secondary- spouse 18 (45%), sibling 5 (13%), parent 17 (40%)  Patient characteristics:  Male sex:  Primary- 18 (90%)  Secondary- 31 (78%)  VAD designation: BTT 100%  Time since VAD implantation: Enrollment during before LVAD implant hospitalization (mean 48 days before implant)  Charlson Comorbidity Index: —  Dropouts: —	Before LVAD, 3 and 6 months after LVAD  n = 20, primary  QOL:  - SF-8 (MCS and PCS)  Caregiver burden:  - Zarit CBI  3 months after LVAD  n = 40, secondary  QOL:  - SF-8 (MCS and PCS)  Caregiver burden:  - Zarit CBI	Primary Before LVAD QOL  PCS: 52.7 ± 7.1  MCS: 38.9 ± 9.3  Caregiver burden  CBI: not reported	Primary 3 months after LVAD QOL  PCS: 49.7 ± 6.5  MCS: 44.2 ± 7.7  Caregiver burden  CBI: median 4.4  CBI ≥ 9:5 (25%)  Change across time:  MCS: before LVAD to 3 months:  P = .03		Primary 6 months after LVAD QOL  PCS: 50.7 ± 6.4  MCS: 46.2 ± 7.4  Caregiver burden  CBI: not reported  Change across time:  MCS: before LVAD to 6 months:  P = .003  *No P values reported for PCS or CBI	Secondary 3 months after LVAD QOL  PCS: 50.3 ± 6  MCS: 44.2 ± 7.8  Caregiver burden  CBI: median 4.4  CBI ≥ 9: 12 (30%)

BSI, Brief Symptom Inventory; BTT, bridge to transplant; CBI, Zarit Caregiver Burden Interview; CMQ, Caregiver Mourning Questionnaire; DT, destination therapy; EQ-5D VAS, EuroQOL visual analog scale; FSQ-SF, Family Strain Questionnaire-Short form; HADS-A, Hospital Anxiety Depression Scale, Anxiety; HADS-D, Hospital Anxiety Depression Scale, Depression; LVAD, left ventricular assist device; MCS, Mental Component Score; MCSI, Multidimensional Caregiver Strain Index; PCS, Physical Component Score; PHQ, Patient Health Questionnaire; PTSD, post-traumatic stress disorder; QOL, quality of life; SF, Short Form.

Data Extraction

Three authors (MS, JA, AL) independently extracted information from articles into 2 evidence tables by study design (qualitative/mixed-methods or quantitative). The evidence tables included study identification (author names, year of publication and article title), study design and participants characteristics (study design, sample, sample size, gender, age, relationship, type of patient device, CF-LVAD designation and time since CF-LVAD implantation), outcomes (all primary and secondary CF- LVAD caregiver outcomes and time points reported), results (categories, themes and subthemes, and models in qualitative studies; questionnaire or direct measure results in quantitative studies), and conclusions and comments (including limitations of the study and authors perspective). Given the relatively small sample size and diversity in study design, the use of a quality assessment tool was not appropriate. Further, to ensure the quality of this systematic review, studies that did not meet the prespecified inclusion/ exclusion criteria were excluded from the review if the study authors did not respond to inquiries for additional information.

Results

The 15 articles included in this review were published between 2011 and 2018. The 2 manuscripts by Bidwell et al^8,26 reported quantitative outcomes from the same sample of caregivers. Similarly, the 2 manuscripts by Brouwers et al^30,32 were drawn from the same sample. Therefore, 13 unique studies were included in this review. Of the 13, 8 were qualitative or mixed methods and 5 were quantitative. Among the qualitative studies, 4 were qualitative descriptive, 2 were hermeneutic phenomenology, and 2 were mixed methods using either qualitative description or grounded theory combined with quantitative surveys. One of the quantitative studies was a multicenter prospective design, 3 were single-center prospective designs, and 1 was a retrospective cross-sectional design. One study described caregiver grief and mourning after the patient died after CF- LVAD implantation. Two studies described outcomes preimplant, 3 studies described outcomes at 3 months after implantation, and 3 studies described outcomes at 6 months after implantation.

Sample Size and Characteristics

Qualitative and Mixed Methods Studies

The qualitative and mixed-methods studies included a total of 111 participants, with a sample size range of 5 to 42. Caregivers were primarily female (81%) with a mean age of 61 years. The most common caregiver-patient relationship was spousal (78%). DT was the most common CF-LVAD designation (88%). Time since CF-LVAD implant in the qualitative studies was 3 months to 5 years. Dropout rates were not reported in qualitative and mixed-methods studies, likely owing to their cross-sectional design. All but one of the qualitative studies was conducted in the United States (88%).

Quantitative Studies

The quantitative studies included a total of 152 participants, with a sample size range of 13 to 50. Caregivers were primarily female (81%) with a mean age of 54 years. The most common caregiver–patient relationship was spousal (64%). BTT was the most common CF-LVAD designation (72%). The duration of follow-up after CF-LVAD implantation ranged from 1.0 to 34.6 months. Only 2 studies assessed outcomes before CF- LVAD implantation. When reported, drop out averaged 19% (range 15%–24%), with reported reasons for dropout including death of the patient and dissolution of the caregiving relationship. The majority (80%) of the quantitative studies were conducted outside the United States in Canada, the Netherlands, Italy, and Japan.

Differences in the proportion of DT (or BTT) patients in the qualitative and quantitative studies may reflect country specific priorities for LVAD caregiver research. In addition, differences in the number of spousal caregivers in the qualitative and quantitative studies may reflect cultural differences in ways that families live together and care for older adults in the United States vs other countries.

Qualitative Findings

Research questions and aims examined in the qualitative studies included describing caregivers’ experiences of the caregiver role, coping strategies, understanding caregivers’ decision-making process and needs; exploring bereaved caregivers’ perceptions of death; and generating theories and models for use in future research and clinical practice. The 2 mixed-methods studies used quantitative surveys of decisional regret²⁸ and QOL²⁹ to augment the qualitative findings (Table 1). In synthesizing the qualitative literature, 3 overarching themes were identified: the caregiver role, coping strategies, and LVAD decisions.

Caregiver Role

The need to adapt to the caregiver role was a common caregiving experience. Kitko et al²¹ described adaptation to caregiving along a continuum. The Caregiving Transition model²¹ highlights important transitional points requiring major adaptations–transition from caring for a spouse with advanced HF, to the CF-LVAD decision, to caring for a spouse with a DT CF-LVAD. Kitko et al²¹ and Kirkpatrick et al²⁹ described that caregivers developed a heightened sense of personal responsibility for the patient’s well-being after CF-LVAD implantation compared with the pre-LVAD time period. The post-LVAD caregiver role was described as a 24/7 multifaceted responsibility, requiring emotional, social, and physical support.^{21,22,25,29,,31} Caregivers described the need to make lifestyle adjustments,²⁵ including role reversals in the home,^25,29 increased household responsibilities,^25,29 development of new daily routines,^21,25,31 and alterations in sex life.²⁹

The caregiver role resulted in disparate feelings of caregiver satisfaction ^, and caregiver burden. ^, Caregivers experienced satisfaction in supporting²² and receiving more time with their loved one.³¹ Nevertheless, caregiving was psychologically burdensome,²⁹ with persistent worry, stress,²⁵ and even guilt over leaving the patient alone.²⁹ The caregiver role required sacrifices of self-interests and personal time, which were not always experienced as burdensome.^22,25,29,31

Coping Strategies

Caregivers described coping mechanisms to manage their 24/7 caregiving tasks.^22,31 Coping strategies included positive thinking, family support, developing routines, acceptance, prayer, faith, hope, religiosity, and fatalism. Egerod et al²² described the concept of caregiver self-preservation, referring to the caregivers’ recognition of the need to help themselves to care for the patient with the CF-LVAD. Self-preservation was described as finding ways to escape the illness environment, by going to work and visiting extended family.²² In contrast, some caregivers described providing support to the CF-LVAD patient as a form of self-preservation, in that supporting the patient helped to fill the caregivers’ personal emotional needs.²²

LVAD Decisions

The decision to have a CF-LVAD implanted is most often made in the context of a serious illness exacerbation after years of living with chronic HF.²⁰ The patient and caregiver may feel as though there is no choice about receiving the CF-LVAD.^20,28 In the decision-making process, patients and caregivers described how they often relied on clinicians’ advice in making their choice about the CF-LVAD implant,^20,28 but they also expressed their interest in hearing from other patients and care- givers.²⁸ The decision to implant was described as almost reflexive and lacking true deliberation.²⁸ Caregivers experienced conflicting feelings of personally wanting the patient to live longer, while also wanting to ensure the patient’s wishes were respected.²⁰ Caregiver reflection on the decision outcome to implant a CF-LVAD tended to evoke a tension between feelings of gratitude vs burdensomeness.²⁰

Quantitative Outcomes

Quantitative outcomes reported for caregivers of patients with CF-LVAD included QOL, health status, psychological status (depression, anxiety, post-traumatic stress disorder), caregiver strain and burden, family strain, and mourning (Table 2). The studies varied considerably in regard to when caregiver outcomes were assessed. QOL, health status, and psychological status were reported before implantation and at 1, 3, and 6 months after implantation. Caregiver strain was reported before implantation and at 1, 3, and 6 months after implantation, whereas caregiver burden was solely reported 3 months after implantation. Family strain was reported on admission to an inpatient rehabilitation center and between 8 and 12 months after discharge from inpatient rehabilitation. Mourning was reported an average of 3 years after the death of the patient with CF-LVAD.

QOL and Health Status

QOL and health status was evaluated in 3 studies. QOL/health status was assessed with the Short Form 12 (SF-12) mental component scores and physical component scores, where higher numbers represented better health status, and with the Euro-QOL visual analog scale, which measures health-related QOL on a 0—100 visual scale, and higher scores represented better QOL. Three investigations noted that after CF-LVAD implantation, caregiver mental health tended to improve, physical health tended to remain stable, and overall health status tended to decline. Brouwers et al³⁰ found that, on average, mental health status (SF-12) improved by about 6 points from 1 month after implantation to 6 months after implantation, whereas physical health status (SF-12) remained relatively stable over the same time period. Kato et al²³ found similar improvements in mental health status (SF-12) at 3 months and 6 months after implantation compared with before implantation, and a small but nonstatistically significant decrease in physical health status (SF-12) from before implantation through 6 months after implantation. In addition, Kato et al showed that the mental health status of spousal caregivers was poorer at 3 months compared with nonspousal caregivers. Conversely, Bidwell et al²⁶ found that compared with before implantation, overall QOL worsened at 1 month after implantation, and worsened again by 3 months after implantation.

A mixed-methods study by Kirkpatrick et al²⁶ evaluated QOL quantitatively concurrently with qualitative interviews. In their cross-sectional study, the authors found greater impairment in psychological and social well-being compared with spiritual and physical well-being. Caregivers who were less than 40 years or more than 69 years of age, male, caregiving for less than 2 years, and those caring for a CF-LVAD patient with an advanced directive tended to report better overall QOL.

Psychological Status

Depression and anxiety were evaluated in 2 studies. Depression was assessed with the Hospital Anxiety Depression Scale, depression subscale and with the Patient Health Questionnaire 8; higher numbers represented more symptoms of depression for both measures. Anxiety was assessed with the Hospital Anxiety Depression Scale, anxiety subscale and the Brief Symptom Inventory anxiety subscale, where higher numbers represented more symptoms of anxiety for each measure. Posttraumatic stress disorder was evaluated at 2 time-points in 1 study, with a 12% prevalence 1 month after CF-LVAD implantation and 14% prevalence at 6 months after implantation.³²

Depression

Brouwers et al³⁰ reported that average depression scores remained relatively stable between 1 month and 3 months after CF-LVAD implantation, with modest improvement at 6 months after implantation. The prevalence of depression declined among caregivers from 1 to 6 months after implantation.³² In contrast, Bidwell et al²⁶ compared depression before CF-LVAD implant with 1 and 3 months after implantation and found initial nonsignificant improvement at 1 month, followed by nonstatistically significant worsening of depression at 3 months. Overall, caregiver depression remains relatively stable for the first 3 months after CF-LVAD implantation, with a decrease in prevalence by 6 months after implantation.

Anxiety

Brouwers et al³⁰ found that average anxiety scores remained stable between 1 month and 3 months after CF- LVAD, with modest improvement by 6 months after CF- LVAD. The prevalence of anxiety decreased between 1 and 6 months after implantation. Similarly, Bidwell et al²⁶ found that average anxiety scores remained relatively stable from before implantation through 3 months after implantation. Overall, caregiver anxiety prevalence is reduced by 6 months after CF-LVAD.

Caregiver Strain and Burden and Family Strain

Caregiver strain was assessed using the Multidimensional Caregiver Strain Index, where higher values represent more strain.⁸ Caregiver burden was assessed with the Zarit Caregiver Burden Interview, where higher scores represent greater burden (mild to moderate caregiver burden is defined as ≥9).²³ The Family Strain Questionnaire-Short Form was used to assess family strain, operationalized as the severity of caregiver stress, with higher scores representing greater strain.²⁴ Bidwell et al⁸ reported that, compared with before implantation, caregiver strain was worse at 1 and 3 months after CF-LVAD implant, returning to preimplant levels by 6 months. Bidwell et al also reported that spousal caregivers had greater strain before implantation compared with nonspousal caregivers. Kato et al²³ reported a 25%–30% prevalence of mild to moderate caregiver burden 3 months after CF-LVAD implant. Ferrario²⁴ found that by 8–12 months after inpatient rehabilitation, family strain improved slightly compared with inpatient rehabilitation admission, yet remained elevated in a range that indicated the caregiver may need psychological evaluation and support. Overall, caregiver strain seems to peak between 1 and 3 months after CF-LVAD implantation.

Caregiver Mourning

One quantitative study examined caregiver mourning after the death of patients with CF- LVAD, using the Caregiver Mourning Questionnaire and a locally developed 10-point scale to capture both positive and negative aspects of the CF-LVAD experience, with higher scores indicating more intense mourning.¹⁹ The average Caregiver Mourning Questionnaire score was 14.6 on a scale of 0–44. The positive aspects of caregiving for a patient with CF-LVAD were the patient’s well-being and longer survival, as well as caregiver wellness. The negative aspects of caregiving were the difficulty managing complications, and the patient’s loss of autonomy and suffering.¹⁹

In a mixed-method study by Blumenthal-Barby et al,²⁸ scores on a decisional regret scale indicated relatively low levels of regret among the primarily DT sample of participants. Significantly, bereaved caregivers described feelings of confusion during the final weeks of their loved one’s life,²⁷ resulting from uncertainty regarding the process of death with an LVAD, and poor integration and understanding of palliative and hospice support services.²⁷

Discussion

This systematic review included 13 studies reporting qualitative themes of caregiver experiences and/or quantitative health outcomes for caregivers of adult recipients of CF-LVAD. Understanding caregiver outcomes before, at the time of, and after CF-LVAD implantation will provide needed information to develop targeted interventions to improve caregiver outcomes. A main finding was that caregivers of patients with CF-LVAD have substantial concerns related to the emotional, social, and physical demands of caregiving. The small number of quantitative studies limited our ability to draw strong conclusions regarding the specific physical and psychological health outcomes that were most affected among caregivers after CF-LVAD implantation.

We created an illustration to highlight key quantitative and qualitative findings from this systematic review, highlighting substantial caregiving demands and caregiver outcomes across time (Fig. 2). Understanding the relationship between caregiving demands and outcomes across time may aid in the development of interventions to address specific concerns at the most critical points in time. On average, discharge from the hospital typically occurs about 3 weeks after CF-LVAD implantation,³³ coinciding with the initiation of peak caregiver burden, which occurs between 1 and 3 months after CF-LVAD implantation.

Fig. 2.

Trajectory of caregiving demands and caregiver outcomes among CF-LVAD caregivers: synthesis of studies included in the systematic review.

Three major themes reflected in the qualitative literature were the caregiver role, coping strategies, and LVAD decisions. The first 3 months after a CF-LVAD implant requires intense modifications to daily life after discharge from the hospital. During this time, the patient and caregiver establish new routines and rearrange household roles. Caregiver household responsibilities increase, while caregivers simultaneously take on caring for the patient’s physical needs. The intensity of adjustment required during this period, combined with caregivers’ reports of wanting to keep things normal and to be strong emotionally for their loved ones, is likely to contribute to persistent worry and daily psychologic burden. Caregivers use Xa range of coping strategies to manage the burden and adjustment. Kitko et al’s²¹ Caregiving Transition model highlights critical transitions as a patient progresses from advanced HF, to making the CF- LVAD decision, to living with a CF-LVAD.

Health care providers should consider caregiving transitions as they plan their approach to care across the initial 3 months after discharge for a CF-LVAD. Approaches to mitigate caregiver burden during this time frame include coordinating appointments across specialties to reduce burdens associated with transportation, offering telemedicine support (video) for trouble-shooting common concerns X and issues related to patient care (eg, dressing changes, equipment management), and encouraging caregivers to establish plans for personal respite.

Qualitative inquiry revealed 3 key domains in the LVAD decision making process: decision context, decision process, and decision outcome.²⁰ Another study examined decision regret quantitatively (93% DT), revealing relatively low levels of decision regret.²⁸ Mcllvennan et al³⁴ recently, with a stepped-wedge clinical trial, evaluated the effectiveness of a shared–decision making intervention on knowledge and values–choice concordance regarding the decision to implant a DT LVAD. The use of decision aids initially led to greater decision conflict for caregivers, highlighting the challenge of supporting patients and caregivers through this complex decision-making process.³⁴ Additional interventional studies designed to improve decision support for patients with a CF-LVAD and their caregivers are warranted to improve values–choice concordance and decrease decision conflict.

Concordant with qualitative studies, the findings from the quantitative review revealed that caregiver strain peaks between 1 and 3 months after CF-LVAD, with at least 25% of caregivers reporting substantial burden 3 months after implantation. However, anxiety and depression tend to be stable over the first 3 months after CF-LVAD implantation, and neither was severely elevated. Anxiety symptoms among caregivers of patients with HF with CF-LVAD²⁶ and caregivers of patients with HF without CF-LVAD¹⁴ suggest minimal difference between the groups (Brief Symptom Inventory anxiety subscale 0.6–0.7 and 0.5, respectively). Comparing depression symptoms (Patient Health Questionnaire-8) between caregivers of patients with HF with (mean range 3.4–4.8)²⁶ and without CF- LVAD (mean range 5.1–6.3)¹⁴ suggests CF-LVAD caregivers experience slightly more depressive symptoms. Scores of 5–9 on the PHS-8 represent mild depressive symptoms, whereas scores of less than 5 represent minimal depression symptoms.³⁵ Caregiver mental health status (SF- 12 mental component scores) showed improvement by 3 months after implant in this review.

The improvement in mental health and lack of significant anxiety or depression somewhat contradicts the qualitative results, which indicated substantial worry, stress, and emotional burden among caregivers. It is possible that the effects of hope and optimism provided by CF-LVAD implantation, and gratitude for a chance to have more time with a loved one, initially offset the contending effects of the emotional, social, and physical demands of caregiving. It is also plausible that the trajectory of mental health status, anxiety, and depression differ between caregivers of patients with BTT v DT CF-LVAD designations, as BTT patients and caregivers retain the hope of cardiac transplantation, whereas DT means a commitment to lifelong CF-LVAD caregiving. Caregivers of DT patients may experience increased levels of emotional burden if they feel as though there is no end in sight. Importantly, the quantitative studies in this review enrolled primarily BTT patients (72%), whereas the qualitative included primarily DT patients (88%). Furthermore, there may be differences in mental health and caregiver burden depending on the type of caregiver–patient relationship. The authors of 2 quantitative studies reported differences in outcomes by caregiver relationship. Bidwell et al (2018) showed spousal caregivers had more preimplant caregiver strain compared with nonspousal caregivers. Kato et al (2018) demonstrated the mental QOL at 3 months after implantation was poorer for spousal caregivers compared with nonspousal caregivers. Thus, longitudinal, quantitative studies evaluating the mental health status and psychological health outcomes of DT CF-LVAD caregivers and spousal caregivers are critical to understanding the impact of long-term CF-LVAD caregiving.

Two studies that quantitatively evaluated caregiver physical health status^23,30 revealed a trend toward stability over the first 3–6 months after CF-LVAD implant, although a mild and nonstatistically significant decrease was noted in 1 study. Both studies recruited caregivers of patients with BTT LVAD designation. Given that caregivers of patients with a DT CF-LVAD designation often provide care for more than 6 months duration, longitudinal studies of caregiver physical health status among the DT CF-LVAD population are warranted to understand the long-term physical health outcomes of this caregiver population.

Notably, in studies of caregivers of patients with other types of chronic medical illnesses, caregiving increased the risk of cardiovascular disease and high mortality.^15,36 Brouwers et al³⁰ compared health status and psychological health outcomes between caregivers of patients with CF- LVAD (BTT designation) and caregivers of patients with an implantable cardioverter defibrillator. CF-LVAD caregivers had better physical health (P < .001), but poorer mental health (P < .001), a higher prevalence of depression (P = .04), and no significant difference in anxiety.³⁰ Compared with caregiving for patient with implantable cardioverter defibrillators, caregiver psychological status after CF-LVAD implant seems to be more negatively affected. On average, this review suggests that caregiver mental health status (SF-12) improves by approximately 6 points after CF-LVAD implantation, and although this reaches statistical significance, clinically significant improvement on the mental component score is estimated as a 9-point change.³⁷ Compared with caregivers of patients with HF without CF-LVAD,¹⁴ caregivers in the present review³⁰ had mean mental components scores that were 9 points lower (worse) 1 month after CF-LVAD implant and similar mental component scores 3–6 months after CF-LVAD implant. Thus, it is important to recognize that, although patients with HF experience improved health outcomes after CF- LVAD implant,²⁶ HF caregiver outcomes seem to deteriorate or stay the same in response to CF-LVAD. Given the increased use of DT CF-LVADs, and the duration and intensity of caregiving required for patients with DT CF- LVAD, additional studies that examine the impact of caregiving on physical and psychological health in this population are needed.

This review reveals several important gaps in the literature regarding the health outcomes of caregivers for patients with CF-LVAD. First, the need for additional quantitative studies describing physical and mental health outcomes is substantiated, particularly among caregivers of patients with a DT LVAD designation. Second, few longitudinal studies have been conducted with CF-LVAD caregivers. The recruitment of caregivers to participate in longitudinal studies across chronic conditions is challenging for those who are already strained by caregiving demands.^38,39 Quantitative studies will be most informative if they include baseline (before CF-LVAD implant) assessment of the outcomes, followed by longitudinal assessment for a minimum of 6 months. Studies with a greater than 6-month duration of follow-up after CF-LVAD implant would provide new knowledge regarding the long-term impact of CF-LVAD on caregiver physical health outcomes: Conducting a needs assessment over this time frame would help to identify what caregivers need most to mitigate physical and emotional challenges. Third, only 2 studies in this review used multisite designs (1 quantitative study and 1 mixed-methods study). Investigators planning CF-LVAD caregiver studies should consider not only longitudinal, but also multisite designs to address gaps in the current scientific literature. Using a multisite approach may improve the feasibility of recruiting larger sample sizes; one option would be to leverage current networks (eg, online recruitment through myl- vad.com). Fourth, only 2 of the included studies quantified CF-LVAD patient comorbidities (eg, comorbidity index).^8,32 The number and type of comorbidities the patient experiences are likely to influence caregiver burden and should be incorporated in future studies. Fifth, the majority of caregivers included were female spouses or partners; thus, the findings of this systematic review may not be applicable to male or nonspousal caregivers. In addition, needed are studies designed to quantitatively compare outcomes between caregivers of patients with CF-LVAD with noncaregivers, caregivers for patients with other high-intensity chronic conditions, and/or caregivers of patients with other implanted devices. Such research will help to contextualize the severity and impact of CF-LVAD caregiving on mental and physical health outcomes.

A strong body of qualitative literature exists, conducted with caregivers several months to years after CF-LVAD implantation. Future qualitative research would be most informative if caregivers were followed prospectively over time to advance understanding of the scope of the caregiving process, and to provide essential insights into caregiver needs at specific time points along the caregiving trajectory.

Limitations

This review included only 5 quantitative studies that evaluated outcomes of CF-LVAD caregivers, each limited to 50 or fewer participants and which included mostly patients with a BTT LVAD designation. The small number of quantitative studies with relatively small sample sizes and, and only 2 of which measured outcomes both before and after CF-LVAD implantation, precluded the use of meta-analysis and limited the conclusions that could be drawn regarding the influence of caregiving on the long-term physical and mental health outcomes of caregivers. In addition, none of the studies identified whether caregivers lived in the same household as the patient. Where a caregiver resides is likely to be an important influence on caregiver health outcomes. Many of the qualitative studies, which were conducted months to years after CF-LVAD implant or after patient death, provided unique perspectives that were not realized in the quantitative studies conducted during the caregiving period.

Conclusions

Caregivers of patients with a CF-LVAD experience significant, sustained emotional burden and strain following implantation for up to 3 months, with beginning resolution at 6 months. Caregiver physical health status remained stable after CF-LVAD implantation, while mental health status tended to improve. Future quantitative research should be designed with larger sample sizes, enrollment of patients/ caregivers with DT and BTT designations, exploration of whether switching from one designation to another (eg, DT to BTT) adds to the caregivers emotional burden and strain, preimplantation and postimplantation measurement of outcomes, and longitudinal follow-up of caregivers for 6 months or more to characterize the caregiving process and evaluate its impact on long-term health outcomes.