Title: Ventricular Assist Devices: A Review of Psychosocial Risk Factors and Their Impact on Outcomes

Authors: Courtenay R. Bruce, JD, MA,1,2 Estevan Delgado, BA,1 Kristin Kostick, PhD,1 Sherry Grogan, RN, PMH NP-BC,3 Guha Ashrith, MD, FACC, 3-5 Barry Trachtenberg, MD, FACC,3-5 Jerry D. Estep, MD, FACC,3-5 Arvind Bhimaraj, MD, MPH, FACC, 3-5 Linda Pham, MSSW, LCSW, 3 Jennifer S. Blumenthal-Barby, PhD1

Affiliations: 1 Center for Medical Ethics & Health Policy; Baylor College of Medicine; Houston, TX; USA

2 Houston Biomedical Ethics Program; Methodist Hospital; Houston, TX; USA

3 Methodist DeBakey Heart and Vascular Center; Houston Methodist Hospital; Houston, TX; USA

4 Methodist DeBakey Cardiology Associates; Houston Methodist Hospital; Houston, TX; USA

5 Weill Cornell Medical College; New York, NY; USA

Correspondence: Courtenay Rose Bruce Center for Medical Ethics & Health PolicyBaylor College of MedicineOne Baylor Plaza, MS: BCM 420Houston, TX, 77030(713) 798-4929crbruce@bcm.edu

Word count: 3,231

1

ABSTRACT

Background: Psychosocial contraindications for ventricular assist devices (VADs) remain particularly

nebulous and are driven by institution-specific practices. Our multi-institutional, multidisciplinary

workgroup conducted a review with the goal of addressing the following research question: How are pre-

operative psychosocial domains predictive of or associated with post-operative VAD-related outcomes?

Answers to this question can contribute to the development of treatment-specific (contra)indications for

patients under consideration for mechanical devices.

Methods: We identified 5 studies that examined psychosocial factors and their relationship to post-

operative VAD-related outcomes.

Results: Our results suggest that three psychosocial variables are possibly associated with VAD-related

outcomes: depression, functional status, and self-care. Of the few studies that exist, the generalizability of

findings is constrained by a lack of methodological rigor, inconsistent terminology, and a lack of

conceptual clarity.

Conclusion: This review should serve as a call for research. Efforts to minimize psychosocial risk pre-

device placement can only be successful insofar as VAD programs can clearly identify who is at risk for

suboptimal outcomes.

2

INTRODUCTION

Psychosocial risk factors impact patient survival and graft success following cardiac transplantation.

Poor perioperative physical functioning, psychiatric disorders, poor social support, use of avoidant coping

strategies, poor self-efficacy, and low optimism have all been identified as factors that could potentially

impact post-transplant outcomes.1–3 Whether and how psychosocial considerations should be weighed in

the context of mechanical circulatory support devices is less clear, however, particularly when the

intended device strategy is destination therapy (DT).4,5

Professional guidelines recommend that all candidates for mechanical circulatory support be screened

for psychosocial risk prior to device placement. However, the use of psychosocial criteria as

contraindications for placement is variable and unstandardized across settings, primarily because far less

is known about the role of psychosocial risk factors for mechanical support devices rather than cardiac

transplantation.6,7 Understanding factors affecting mechanical support device outcomes in particular can

help to tailor (contra)indications that are specific to patients being considered for this intervention,

whether as DT or bridge-to-transplant (BTT).

This review aims to assimilate studies that identify pre-operative psychosocial risk factors and their

impact on post-operative ventricular assist device (VAD)-related outcomes. We approached this review

with the goal of addressing the following research question: How are pre-operative psychosocial domains

predictive of or associated with post-operative VAD-related outcomes? Answers to this question will

help to develop inclusion and exclusion criteria for treatment candidacy that takes into consideration the

distinct trajectories of mechanical support device outcomes in relation to transplant or other end-stage

heart failure interventions.

METHODS

Our workgroup consisted of members from the fields of heart failure cardiology, nursing, bioethics,

decision science, social work, psychiatry, medical anthropology, and epidemiology. We adhered to the

3

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to conduct

our review.8

Conceptual Bases

Our workgroup worked iteratively to develop a coherent and comprehensive definition of

“psychosocial considerations.” To inform its development, we reviewed existing conceptual models of

“quality of life” and “psychosocial,” revealing that specific dimensions are often labeled differently by

different authors.9 For instance, the term “quality of life” often refers to health status, physical

functioning, psychosocial adjustment, well-being, life satisfaction, and happiness. Within these models,

physical functioning is often conceptualized as falling under the rubric of psychosocial considerations,

and there is considerable variation concerning dimensions and proxies for “physical functioning.”

“Physical” domains may refer to pathophysiological changes, functional deficits, or perceived health

status.10,11

Therefore, definitional variability and inconsistency for analogous concepts across studies and the

limitations of existing conceptual models made it difficult to develop an operational definition for

“psychosocial considerations.” In order to be consistent with other workgroup’s practices and

professional guidance statements in heart failure,1 we ultimately opted to define “psychosocial

considerations” to encompass five domains to guide us during data abstraction:

1. Physical functioning (which we refer to more precisely as “functional status”), 2. Psychological functioning (e.g., psychiatric illness; behavioral disorders; neurocognitive

functioning), 3. Overall quality of life considerations (defined to include subjective well-being, which means how

happy or satisfied someone is with life as a whole)4. Behavioral functioning (e.g., compliance, substance use/abuse), 5. Social functioning (e.g., social adjustment; stability; social support)1

These domains have the added benefit of being largely consistent with existing tools for psychosocial

assessments used during pre-transplant and pre-VAD placement. The Stanford Integrated Psychosocial

Assessment for Transplant (SIPAT) tool is considered a psychometrically rigorous instrument that

4

provides a comprehensive list of psychosocial factors frequently assessed for transplant and mechanical

circulatory support candidates, which are similar to the psychosocial factors as those numbered above. 1,12

We used post-operative “outcomes” that are collected by Interagency Registry for Mechanically Assisted

Circulatory Support (INTERMACS) as a conceptual basis to inform our review of post-device placement

outcomes.13 “Outcomes” was defined to include mortality or morbidity, including infection, re-

hospitalization, or post-VAD placement perceptions of quality of life.

Search Strategy

We searched PubMed, PsychINFO, and SCOPUS databases using the following search terms: [left]

ventricular assist device [OR] mechanical circulatory support [OR] VAD [AND] patient selection [OR]

contraindications [OR] social support, [OR] neurocognitive/neurocognition, [OR] substance abuse or

dependence or use, [OR] alcohol, [OR] psychopathology/psychology, [OR] personality traits or disorder,

[OR] compliance/adherence, [OR] anxiety, [OR] depression [OR] quality of life [OR] functional status.

Searches were conducted by at least two independent reviewers (CB and JBB). We also manually

searched reference lists, and we reviewed the bibliographies of all articles that fulfilled the inclusion

criteria to capture all potentially relevant articles. After our pilot search, we contacted experts to validate

the scope of our review.

Inclusion Criteria and Data Abstraction

Our search criteria included adult (non-pediatric) studies, conducted in the United States and

published in English, which examined pre-operative psychosocial factors and their relationship to VAD-

related outcomes. We excluded studies that did not: (a) present information on pre-operative psychosocial

factors, (b) include VAD candidates or patients in the study, (c) assess the relationship between

psychosocial factors and outcomes, or (d) expressly use concrete, definitive psychosocial criteria to

exclude patients from being considered for VAD placement. Abstract-only publications, editorials,

reviews, and commentaries containing no data were excluded. We also excluded studies that were

published prior to 2001. Our rationale for extending as far back as 2001 (during a time when pulsatile

5

pumps were used) is that a review examining the relationship between psychosocial considerations and

outcomes would likely be under-inclusive if it focused exclusively on continuous pumps.14–18 We did not

restrict our search on the basis of study design other than the ways specified above. The search was

conducted during February-May 2014, with most database searches occurring during March. Figure 1

illustrates our search results.

At least two of us (CB and JBB) independently appraised all of the studies that met our inclusion

criteria. These two authors examined abstracts that met criteria. We (CB, JE, GA, and BT) used

standards developed by the Evidence-Based Medicine Working Group to score the methodological rigor

of the studies.19 We resolved discrepancies by using a third investigator to reconcile the discrepancy.

Table 1 provides a description of the studies included in this review. Table 2 illustrates the

methodological rigor of the studies, provides overall quality scores, and defines the criteria used to

evaluate methodological rigor.

RESULTS

Functional Status

Two studies examined the relationship between functional status and post-device placement

outcomes, yet only one found a significant relationship between pre-device placement status and

increased risk of post-operative death, perhaps owing to the different instruments and methodologies used

to measure functional status. Dunlay (2014) defined functional status (e.g., “frailty”) as a “state of

increased vulnerability to adverse outcomes.” 20,21 Frailty was assessed for 99 VAD-DT patients using the

deficit index (31 impairments, disabilities, and comorbidities) pre-VAD placement. The association

between frailty, mortality, and re-hospitalization was calculated using Kaplan-Meier curves. The authors

found stepwise increase in 1-year mortality with increasing deficit index. Patients who were not frail had

mortality rates of 16.2%, whereas those who were frail had mortality rates of 39.9%. The authors

concluded that pre-device placement frailty is associated with increased risk of death. 20

6

In contrast, Flint (2013) assessed pre-operative functional status, using the Kansas City

Cardiomyopathy Questionnaire and the Minnesota Living with Heart Failure Questionnaire (rather than

the deficit index) for 1,125 clinical trial participants who received the HeartMate II as destination therapy

(n=635) or bridge to transplant (n= 490).22 The primary outcome measure was overall survival, which

was analyzed using Kaplan-Meir scores. Pre-operative health status scores did not correlate with overall

post-operative mortality. Their findings suggest that pre-operative functional status has limited

association with outcomes after VAD implantation.

Psychological Functioning

Although there are studies that conduct pre- and post-surgery comparisons in neurocognitive status,

we were not able to identify a study that examined the relationship between pre-operative neurocognitive

status and post-operative outcomes.23–25 However, there are studies that have found connections between

pre-operative psychological disorders and outcomes after device placement.

Gordon (2013) studied the correlation between infection and mortality in a prospective study

involving 11 clinical centers and 150 patients between 2006 and 2009, finding an association between

depression and infection.26 Patient history of depression was collected as baseline data during a pre-

operative chart review. Thirty-three out of 150 patients (22%) experienced VAD infections. The authors

found that a pre-operative history of depression was a significant predictor of a VAD infection, resulting

in a 3-fold increased risk of post-operative VAD infection.26,27

Overall Quality of Life

Patients’ perception of quality of life is often treated as a post-device placement secondary endpoint

in the clinical trials that have been conducted.28–30 Studies have examined determinants of post-VAD

placement quality of life in ways that extend beyond the clinical trials. For instance, Dew, Wray, Grady,

and Molzahn have found significant multivariate relationships between post-VAD placement perceptions

of quality of life and depressive symptoms for VAD patients and transplant recipients. 31–37 These authors

7

concluded that depression relates to patients’ overall psychological state and their perceptions of quality

of life with a VAD.

We could only find one study that documented an association between quality of life and outcomes

(i.e., survival). In this study, Grady and colleagues (2004) took a nonrandom sample of 78 patients who

received a HeartMate I who had quality-of-life data at 1,2,3,6,9, or 12 months after device placement. 32

Although this study centers on pulsatile technologies--limiting its applicability to continuous-flow

technologies-- a significant contribution of this work is that the authors found that ambulation and self-

care was negatively associated with mortality after device placement.

We were unable to categorize the findings related to ambulation and depression as part of the physical

functioning or psychopathology domains, respectively, because the studies were, by their own definition,

quality of life studies, making it difficult for us to draw out distinctions where the researchers themselves

did not.

Social Functioning

We did not find a study that examined whether and how social functioning and support affects

VAD-related outcomes. Patients who have strong support networks may be better equipped to contribute

to achieving good outcomes compared to patients with weak or no supportive networks, but this has yet to

be directly tested. Several quality-of-life studies included a social functioning domain, but few examine

mechanisms by which social support impacts outcomes,1 how different sources of social support outside

of significant partners (e.g. wife or husband) contribute to patient care, or track the impact of social

support factors after the first few weeks post-implantation.

Behavioral Functioning

There are studies looking at patient adherence to heart failure self-care recommendations, but they are

not VAD-specific.38 Poston et al. examined the link between alcohol use and VAD-related infection. 39

The authors performed a retrospective review of 123 VAD patients. Descriptive statistics were generated

8

and Pearson’s correlation coefficient was used for testing association. Variables that were tested included:

length of VAD support, rate of infection, age, body mass index, length of ICU stay, and presence of

comorbidities (e.g., pulmonary disease, tobacco or alcohol use, hypertension, prior stroke). Alcohol use

was not defined, making it difficult to tell whether “alcohol use” included any drinking rather than

distinguishing occasional slips from problematic drinking.

Of the 123 patients who underwent VAD placement, 53% remained free of infection during the

support period. The length of time that patients received VAD support was much longer in those with

infection compared to those patients that had no infection during VAD support (132 v. 48 days), but the

variables that were significant predictors of infection in a univariate analysis, including history of alcohol

use, all lost their significance when the length of VAD support was controlled.39 In the multivariate

analysis, the only significant predictors of mortality included infection during device support and device-

related infection.

DISCUSSION

We were unable to provide definitive evidence-based psychosocial contraindications, owing to

the incomplete and muddied status of the outcomes research. In reviewing these studies, we identified a

recurring problem associated with a lack of conceptual clarity for quality of life and physical function,

resulting in an uncertainty about how these factors influence psychosocial dimensions. This uncertainty is

not something unique to our examination of these concepts, but rather a recurring problem encountered

by researchers who have tried to identify and measure components of quality of life or subjective well-

being, which is inherently comprised of conceptually distinct but interdependent and interrelated sub-

dimensions.15 While they are conceptually distinct dimensions from a researcher’s point of view in trying

to measure influences of one on the other (e.g., impacts of functionality on psychosocial well-being),

from a subjective point of view these dimensions are experienced by patients as interdependent aspects of

a larger intuitive, global sense of these domains. Patients evaluate their well-being from an overall

perspective rather than distinct domains.40

9

This review should, at a minimum, serve as a call for conceptual clarity. Although patients

subjectively experience these dimensions as interdependent aspects, taking into account overall

experience, this does not mean that the research community should avoid attempting to create conceptual

distinctions encompassed by the subjective whole. Until this is done, it is difficult for the VAD

community to define evidence-based psychosocial contraindications to VAD placement, and the quality

of the studies on subjective well-being or quality of life with a VAD can be questioned.

While it may be difficult to draw definitive conclusions based on these studies, particularly to the

extent that the generalizability of the findings is constrained by key factors including a lack of

methodological rigor, inconsistent terminology, and lack of conceptual clarity, our findings have

significant implications for clinical practice and research agendas for several reasons.

Where evidence is lacking or weak, there are ethical implications for patient selection processes

because decision-making processes must be transparent, consistent, and grounded in evidence. 41 Patient

selection processes should be treatment-specific, rather than assuming carryover of indications and

contraindications from other, even similar, treatments (like transplant). We were particularly surprised by

the lack of studies on the impacts of social functioning (e.g., social support and adjustment) given that

many VAD programs use social support as a screening tool for device placement. Patients with

mechanical support devices face challenges that are significantly different from those of transplant

patients in that they often experience greater periods of dependency on caregivers for device maintenance,

hygiene, and logistics of ambulation. Further, VAD-DT patients in particular are often older or in less

stable condition than transplant patients. The relative severity of VAD patients’ health problems might

constitute a different contingency for psychosocial factors.

Any efforts to minimize psychosocial risk pre-device placement can only be successful insofar as

VAD programs can clearly identify who is at risk for suboptimal outcomes.42 Yet, our review suggests

there are little data available on the impact of most psychosocial domains on VAD-related clinical

outcomes. This means that the VAD community might be focusing on incorrect sub-populations for high

10

psychosocial risk by borrowing from what is known about suboptimal outcomes from the transplant (not

device) literature, even using the same psychosocial screening instruments for device-only and transplant

patients. VAD-DT candidates may warrant different psychosocial evaluation criteria and different

screening tools than transplant patients or VAD-BTT candidates because the evidence bases may differ.

We can provide some preliminary recommendations for psychosocial considerations that should be

considered in patient selection processes for device placement, although we caution that the

recommendations are only preliminary. Mechanisms remain tenuous until comprehensive studies

evaluating all relevant psychosocial factors and their relationships to outcomes are conducted and

mechanisms are well-understood.

First, our work suggests that there three specific psychosocial factors (depression, functional status,

and self-care) that may be worthy of being an integral part of the screening process, but further evidence

is needed to confirm these relationships and their degree of importance and interrelationships. 20,26,32 Based

on our review, one can hypothesize that depression, functional status, and self-care are linked somehow to

increased morbidity (i.e., infection and readmission) post-device placement. Infection and recurring

admissions, in turn, are associated with increased mortality. Depression may influence patients’ self-

hygiene and adherence tendencies, which, in turn, can impact VAD-related outcomes. If this hypothesis

holds true, instruments for assessing psychosocial risk in transplant candidates should be modified for

device-only candidates by shifting the emphasis of the SIPAT from psychopathology and substance abuse

to more of an emphasis on functional status and self-care.12 Future research should be geared toward

understanding how psychosocial disorders and behavioral functioning causes patients to do something or

inhibit them from doing something that affects morbidity.

Additionally, it is likely that functional status could be used as a prognostic indicator during the

screening process, although more work is needed to determine the degree to which physical factors more

generally overlap with or should be distinguished from psychosocial factors. A particular challenge is that

“frailty,” “functional status,” and “physical functioning” have been defined in numerous ways, in the

11

transplant literature and elsewhere, making it very difficult to draw conclusions between studies.

Functional status often includes dimensions related to: functional capacity, functional performance,

functional capacity utilization, and functional reserve. Further confounding matters is that functional

status can be viewed from subjective or objective points of view, much like quality of life. Flint and

Dunlay likely approached their studies conceptualizing functional status differently, resulting in the

selection of different instruments and methodologies. The degree to which these dimensions are

conceptually distinct and separately measured depends heavily on the focus of inquiry.

We anticipate criticism that the reason why there might so few studies available on pre-operative

psychosocial domains and relationships to post-operative VAD-related outcomes is because of inherent

methodological limitations of pre- and post-comparison studies. That is, device candidates might be

declined for device placement on the basis of psychosocial contraindications, making it difficult to make

comparisons pre- and post-operative comparisons because there may be too few patients with pre-device

placement psychosocial risk profiles that actually received devices to be able to draw any comparisons.

Such an interpretation, however, does not fully account for the fact that psychosocial contraindications for

device placement is variable and unstandardized across settings, making it difficult to assert with any

confidence whether and how candidates with intermediate to high-risk psychosocial profiles are declined

or accepted for VAD placement. One of the main purposes of the paper is to show that we do not yet have

psychosocial contraindications, and so by logical extension, the range of values for predicting outcomes is

not yet exclusive or truncated. Additionally, we tried to account for this limitation by excluding studies

that expressly used psychosocial contraindications as exclusionary criteria for VAD placement.

CONCLUSIONS

This review should serve as a call for research among bench scientists, outcomes researchers,

cardiologists, and multidisciplinary personnel in VAD programs. As a preliminary step, we hope to

encourage the development of consensus statements surrounding conceptual and operational definitions

of psychosocial domains. Where there is considerable variation in definitions and methodologies, it

12

makes it difficult to build a comprehensive knowledge base about the relative weight each psychosocial

consideration should be afforded.37,40,43,44 The lack of definitional consistency has further implications for

the selection of measurements used to evaluate each domain, since how a concept is defined greatly

impacts the measurement of it.

We also hope to encourage the development of a research agenda aimed at providing consistent and

transparent patient selection processes that are treatment-specific. One approach that could maximize

transparency and relevancy is to create a research agenda prioritizing VAD-DT studies. During this time,

patients that are VAD-BTT candidates may be evaluated using psychosocial selection criteria for

transplant on grounds that they will progress to transplant. Until a solid, high-quality evidence base is

created, patients and providers considering mechanical device implantation may legitimately claim that

selection processes on the basis of psychosocial risk are neither evidence-based nor adequately tailored to

preferred treatment options.

13

FUNDING AND DISCLOSURES

This project was funded as part of an award from the Patient-Centered Outcomes Research Institute (CDR-1306-01769). There are no disclosures or conflicts of interest to report.

14

FIGURE LEGEND

Figure 1 = Search Results and Data Abstraction

Table 1= Description of studies included in review

Table 2 = Quality assessment of strong (1), moderate (0.5), and weak (0) relevant study components using the EPHPP instrument

15

FIGURE 1

Figure 1. Search results and data abstraction

16

489 Potentially Relevant Studies Identified in Databases for Retrieval:

453 PubMed11 Scopus25 PsychINFO

478 Records (abstracts) Retrieved from Databases and Screened

74 Full-Text Studies Retrieved for More Detailed Evaluation

5 Full-Text Studies Included in Review

11 Duplicates Removed

404 Records Excluded After Reviewing Abstracts:

Non-StudiesNon-VAD StudiesStudies Before 2001Non-U.S. StudiesPediatric Studies

69 Studies Excluded After Full-Text Search:

No assessment of pre-operative psychosocial factors & post-operative outcomes

TABLE 1

Table 1. Description of studies included in review

AuthorsRef. Study design Study population and setting Study SettingSample size

Assessment instruments and tools Outcomes Limitations

Dunlay et al, 201420

Prospective cohort

Patients undergoing LVAD as DT (HeartMate II & HeartWare LVAD)

Mayo Clinic, 2007 through June 2012

99 Pre-operative frailty assessment using the deficit index

Increased risk of post-operative DT LVAD mortality as a function of pre-operative frailty

Single center study

Flint et al, 201322

Retrospective analysis of a prospective clinical trial

Patients who received a HeartMate II for either DT or BTT

Multi-site trail, 2005 through 2009

1,125 Kansas City Cardiomyopathy Questionnaire, Minnesota Living with Heart Failure Questionnaire, ongoing clinical and physical assessments,

Pre-operative health status scores did not correlate with overall post-operative mortality

Homogeneous heart failure population

Gordon et al, 201326

Prospective cohort

Patients scheduled for HeartMate II, HeartMate I, Thoratec Implantable Ventricular Assist Device, VentrAssist, Novacor, Thoratec paracorporeal BiVAD, Thoratec HeartMate I/Abiomed RVAD, and Thoratec HeartMate II/Abiomed RVAD implantation

11 site study, 2006 through 2008

150 Clinical infection assessment, ongoing clinical and physical assessments, chart review

A history of depression was an independent predictor of VAD infection, even for the newly introduced HeartMate II patients

Small sample size may have limited the identification of additional risks for VAD infection

Grady et al, 200432

Prospective cohort

Patients who received either a HeartMate VE LVAD or HeartMate Implantable Pneumatic LVAD as DT

10 site study, August 1994 through August 1999

78 Quality of Life Index, Rating Question Form, Heart Failure Symptoms Checklist, Sickness Impact Profile, LVAD Stressor Scale, Jalowiec Coping Scale, Chart Review

Decreased ambulation and self-care were significantly associated with the post-operative risk of dying

One study site non-U.S., high percentage of dropouts

Poston et al, 200339

Retrospective cohort

Patients who received VADs (HeartMate, Thoratec & Novacor)

University of Pittsburgh Medical Center, 1987 through 2000

123 Chart review History of alcohol abuse significantly predictive in univariate analysis but not when controlling for length of VAD support

Single center study

17

18

TABLE 2

Table 2. Quality assessment of strong (1), moderate (0.5), and weak (0) relevant study components using the Effective Public Health Practice Project (EPHPP) instrument*

AuthorsRef. Selection bias† Design‡ Con-

founders§ Blinding||Data collection methods#

Withdrawals & dropouts**

Overall Quality Score

Dunlay et al, 201420

0.5 0.5 0 Unknown 0.5 1 2.5 of 6.0

Flint et al, 201322

0.5 0.5 0 0 0 Unknown 1.0 of 6.0

Gordon et al, 201326

0.5 0.5 0 0 0 1 2.0 of 6.0

Grady et al, 200433

0 0.5 0 0 0.5 0 1.0 of 6.0

Poston et al, 200339

0.5 0.5 0.5 0 1 1 3.5 of 6.0

*Balshem et al, 200419

†Strong: Very likely to be representative of the target population and greater than 80% participation rate, Moderate: Somewhat likely to be representative of the target population and 60-70% participation rate, Weak: All other responses or not stated

‡Strong: Randomized controlled trail (RTC) & Controlled Clinical Trial (CCT), Moderate: Cohort analytic, case-control, cohort, or an interrupted time series, Weak: All other study designs or design not stated

§Strong: Controlled for at least 80% of confounders, Moderate: Controlled for 60-79% of confounders, Weak: Confounders not controlled for, or not stated

||Strong: Blinding of outcome assessor & study participants to intervention status and/or research question, Moderate: Blinding of either outcome assessor or study participants, Weak: Outcome assessor and study participants are aware of intervention status and/or research question

#Strong: Tools are valid and reliable, Moderate: Tools are valid but reliability not described, Weak: No evidence of validity or reliability**Strong: Follow-up rate of >80% of participants, Moderate: Follow-up rate of 60-79% of participants. Weak: Follow-up rate of <60% of

participants or withdrawals and dropouts not described.

19

REFERENCES

1. Cupples S, Dew MA, Grady KL, De Geest S, Dobbels F, et al. Report of the Psychosocial Outcomes Workgroup of the Nursing and Social Sciences Council of the International Society for Heart and Lung Transplantation: present status of research on psychosocial outcomes in cardiothoracic transplantation: review and recommendations for the field. J Heart Lung Transplant. 2006 Jun;25(6):716–25.

2. Eshelman AK, Mason S, Nemeh H, Williams C. LVAD destination therapy: applying what we know about psychiatric evaluation and management from cardiac failure and transplant. Heart Fail Rev. 2009 Mar;14(1):21–8.

3. Dew MA, DiMartini AF, Steel J, De Vito Dabbs A, Myaskovsky L, Unruh M, et al. Meta-analysis of risk for relapse to substance use after transplantation of the liver or other solid organs. Liver Transplant. 2008 Feb;14(2):159–72.

4. Entwistle JWC, Sade RM, Petrucci RJ. The ethics of mechanical support: the need for new guidelines. Ann Thorac Surg. 2011 Dec;92(6):1939–42.

5. Bruce CR. A review of ethical considerations for ventricular assist device placement in older adults. Aging Dis. 2013 Apr;4(2):100–12.

6. Feldman D, Pamboukian SV, Teuteberg JJ, Birks E, Lietz K, Moore SA, et al. The 2013 International Society for Heart and Lung Transplantation Guidelines for mechanical circulatory support: executive summary. J Heart Lung Transplant. 2013 Feb;32(2):157–87.

7. Brouwers C, Denollet J, de Jonge N, Caliskan K, Kealy J, Pedersen SS. Patient-reported outcomes in left ventricular assist device therapy: a systematic review and recommendations for clinical research and practice. Circ Heart Fail. 2011 Nov;4(6):714–23.

8. PRISMA [Internet]. [cited 2014 Mar 19]. Available from: http://www.prisma-statement.org/index.htm

9. Bakas T, McLennon SM, Carpenter JS, Buelow JM, Otte JL, Hanna KM, et al. Systematic review of health-related quality of life models. Health Qual Life Outcomes. 2012;10:134.

10. Ferrans CE, Zerwic JJ, Wilbur JE, Larson JL. Conceptual model of health-related quality of life. J Nurs Scholarsh. 2005;37(4):336–42.

11. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA. 1995 Jan 4;273(1):59–65.

12. Maldonado JR, Dubois HC, David EE, Sher Y, Lolak S, Dyal J, et al. The Stanford Integrated Psychosocial Assessment for Transplantation (SIPAT): a new tool for the psychosocial evaluation of pre-transplant candidates. Psychosomatics. 2012 Apr;53(2):123–32.

13. UAB - Interagency Registry for Mechanically Assisted Circulatory Support - Home [Internet]. [cited 2014 Mar 19]. Available from: https://www.uab.edu/medicine/intermacs/

20

14. Park SJ, Milano CA, Tatooles AJ, Rogers JG, Adamson RM, Steidley DE, et al. Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy. Circ Heart Fail. 2012 Mar 1;5(2):241–8.

15. Grady KL, Warner Stevenson L, Pagani FD, Teuteberg J, Pamboukian SV, Birks E, et al. Beyond survival: Recommendations from INTERMACS for assessing function and quality of life with mechanical circulatory support. J Heart Lung Transplant. 2012 Nov;31(11):1158–64.

16. Rosenberger EM, Fox KR, DiMartini AF, Dew MA. Psychosocial factors and quality-of-life after heart transplantation and mechanical circulatory support. Curr Opin Organ Transplant. 2012 Oct;17(5):558–63.

17. Maciver J, Ross HJ. Quality of life and left ventricular assist device support. Circulation. 2012 Aug 14;126(7):866–74.

18. Hallas C, Banner NR, Wray J. A qualitative study of the psychological experience of patients during and after mechanical cardiac support. J Cardiovasc Nurs. 2009 Feb;24(1):31–9.

19. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011 Apr;64(4):401–6.

20. Dunlay SM, Park SJ, Joyce LD, Daly RC, Stulak JM, McNallan SM, et al. Frailty and outcomes after implantation of left ventricular assist device as destination therapy. J Heart Lung Transplant. 2014 Apr;33(4):359-65.

21. Afilalo J, Karunananthan S, Eisenberg MJ, Alexander KP, Bergman H. Role of frailty in patients with cardiovascular disease. Am J Cardiol. 2009 Jun 1;103(11):1616–21.

22. Flint KM, Matlock DD, Sundareswaran KS, Lindenfeld J, Spertus JA, Farrar DJ, et al. Pre-operative health status and outcomes after continuous-flow left ventricular assist device implantation. J Heart Lung Transplant. 2013 Dec;1;32(12):1249–54.

23. Slaughter MS, Pagani FD, McGee EC, Birks EJ, Cotts WG, Gregoric I, et al. HeartWare ventricular assist system for bridge to transplant: combined results of the bridge to transplant and continued access protocol trial. J Heart Lung Transplant. 2013 Jul;32(7):675–83.

24. Zimpfer D, Wieselthaler G, Czerny M, Fakin R, Haider D, Zrunek P, et al. Neurocognitive function in patients with ventricular assist devices: a comparison of pulsatile and continuous blood flow devices. ASAIO. 2006 Feb;52(1):24–7.

25. Slaughter MS, Sobieski MA, Gallagher C, Dia M, Silver MA. Low incidence of neurologic events during long-term support with the HeartMate XVE left ventricular assist device. Tex Heart Inst J. 2008;35(3):245–9.

26. Gordon RJ, Weinberg AD, Pagani FD, Slaughter MS, Pappas PS, Naka Y, et al. Prospective, multicenter study of ventricular assist device infections. Circulation. 2013 Feb 12;127(6):691–702.

21

27. Holman WL, Park SJ, Long JW, Weinberg A, Gupta L, Tierney AR, et al. Infection in permanent circulatory support: experience from the REMATCH trial. J Heart Lung Transplant. 2004 Dec;23(12):1359–65.

28. Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med. 2001 Nov 15;345(20):1435–43.

29. Miller LW, Pagani FD, Russell SD, John R, Boyle AJ, Aaronson KD, et al. Use of a continuous-flow device in patients awaiting heart transplantation. N Engl J Med. 2007 Aug 30;357(9):885–96.

30. Slaughter MS, Rogers JG, Milano CA, Russell SD, Conte JV, Feldman D, et al. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med. 2009 Dec 3;361(23):2241–51.

31. Grady KL, Meyer P, Mattea A, White-Williams C, Ormaza S, Kaan A, et al. Improvement in quality of life outcomes 2 weeks after left ventricular assist device implantation. J Heart Lung Transplant. 2001 Jun;20(6):657–69.

32. Grady KL, Meyer PM, Dressler D, Mattea A, Chillcott S, Loo A, et al. Longitudinal change in quality of life and impact on survival after left ventricular assist device implantation. Ann Thorac Surg. 2004 Apr;77(4):1321–7.

33. Grady KL, Meyer P, Mattea A, Dressler D, Ormaza S, White-Williams C, et al. Predictors of quality of life at 1 month after implantation of a left ventricular assist device. Am J Crit Care. 2002 Jul;11(4):345–52.

34. Grady KL, Meyer PM, Dressler D, White-Williams C, Kaan A, Mattea A, et al. Change in quality of life from after left ventricular assist device implantation to after heart transplantation. J Heart Lung Transplant. 2003 Nov;22(11):1254–67.

35. Wray J, Hallas CN, Banner NR. Quality of life and psychological well-being during and after left ventricular assist device support. Clin Transplant. 2007 Oct;21(5):622–7.

36. Molzahn AE, Burton JR, McCormick P, Modry DL, Soetaert P, Taylor P. Quality of life of candidates for and recipients of heart transplants. Can J Cardiol. 1997 Feb;13(2):141–6.

37. Dew MA, Kormos RL, Winowich S, Stanford EA, Carozza L, Borovetz HS, et al. Human factors issues in ventricular assist device recipients and their family caregivers. ASAIO. 2000 Jun;46(3):367–73.

38. Marti CN, Georgiopoulou VV, Giamouzis G, Cole RT, Deka A, Tang WHW, et al. Patient-reported selective adherence to heart failure self-care recommendations: a prospective cohort study: the Atlanta Cardiomyopathy Consortium. Congest Heart Fail. 2013 Feb;19(1):16–24.

39. Poston RS, Husain S, Sorce D, Stanford E, Kusne S, Wagener M, et al. LVAD bloodstream infections: therapeutic rationale for transplantation after LVAD infection. J Heart Lung Transplant. 2003 Aug;22(8):914–21.

22

40. Sandau KE, Hoglund BA, Weaver CE, Boisjolie C, Feldman D. A conceptual definition of quality of life with a left ventricular assist device: results from a qualitative study. Heart Lung J Crit Care. 2014 Feb;43(1):32–40.

41. OPTN: Organ Procurement and Transplantation Network [Internet]. [cited 2014 Mar 19]. Available from: http://optn.transplant.hrsa.gov/resources/bioethics.asp?index=5

42. Gilotra NA, Russell SD. Patient selection for mechanical circulatory support. Heart Fail Rev. 2013 Jan;18(1):27–34.

43. Dew MA, Kormos RL, Winowich S, Nastala CJ, Borovetz HS, Roth LH, et al. Quality of life outcomes in left ventricular assist system inpatients and outpatients. ASAIO. 1999 Jun;45(3):218–25.

44. Marcuccilli L, Casida J, Peters RM. Modification of self-concept in patients with a left-ventricular assist device: an initial exploration. J Clin Nurs. 2013 Sep;22(17-18):2456–64.

23