Secondary Logo

Journal Logo

Assessing Frailty in Patients Undergoing Destination Therapy Left Ventricular Assist Device: Observations from Interagency Registry for Mechanically Assisted Circulatory Support

Cooper, Lauren, B.*†; Hammill, Bradley, G.; Allen, Larry, A.§¶; Lindenfeld, JoAnn; Mentz, Robert, J.†‡; Rogers, Joseph, G.†‡; Milano, Carmelo, A.‡#; Patel, Chetan, B.†‡; Alexander, Karen, P.†‡; Hernandez, Adrian, F.†‡

doi: 10.1097/MAT.0000000000000600
Adult Circulatory Support
Free

Frailty and heart failure share common pathways with symptoms that often coexist. Assessment of frailty may inform patient selection for left ventricular assist device (LVAD) therapy. Using Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) data of destination therapy (DT) LVAD patients from January 1, 2012, to March 31, 2014, we examined preimplantation provider-assessed frailty and gait speed testing and the association with 1 year postimplantation outcomes. Of 2,469 patients, 227 (9.2%) had provider-assessed frailty. Only 320 (13.0%) completed gait speed testing, whereas 1,047 (42.4%) were “too sick” to perform the test. Provider-assessed frail and nonfrail patients had similar distributions of INTERMACS profiles and similar median gait speeds. One year mortality was higher for patients with provider-assessed frailty versus nonfrail (24.6% vs. 18.9%; p = 0.01) and for those too sick to complete gait speed testing versus completed testing (22.0% vs. 15.9%). There was an association between provider-assessed frailty and mortality, although it was not clinically significant after adjustment (hazard ratio [HR]: 1.38 [95% confidence interval {CI}: 0.97–1.95]). Useful information regarding frailty on postimplant mortality is gained from provider assessment of frailty or knowing gait speed could not be performed. Development of frailty measures better suited for DT LVAD candidates may help in distinguishing between a frailty phenotype and a more reversible from heart failure–related vulnerability.

From the *Inova Heart and Vascular Institute, Falls Church, Virginia; Department of Medicine, Division of Cardiology, Duke University Medical Center, Durham, North Carolina; Duke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina; §Department of Medicine, Division of Cardiology, University of Colorado School of Medicine, Aurora, Colorado; Colorado Cardiovascular Outcomes Research Consortium, Denver, Colorado; Department of Medicine, Division of Cardiovascular Medicine and Vanderbilt Heart and Vascular Institute, Vanderbilt University Medical Center, Nashville, Tennessee; and #Department of Surgery, Division of Cardiovascular and Thoracic Surgery, Duke University Medical Center, Durham, North Carolina.

Submitted for consideration October 2016; accepted for publication in revised form April 2017.

This project has been funded in whole or in part with federal funds from the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health (NIH), Department of Health and Human Services (DHHS), under Contract number HHSN268201100025C.

The data for this study came from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS), NHLBI, Contract number HHSN268200548198C, 2010. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH).

Disclosures: Dr. Cooper was supported by grant T32HL069749-11A1 from the NIH. Dr. Allen’s time and the analytics for this study were supported by K23HL105896 from NHLBI of the NIH. He receives grant funding from NIH and PCORI, and consultancy fees from Novartis, Janssen, and St. Jude. Dr. Mentz received grant funding from Amgen, AstraZeneca, Bristol-Myers Squibb, Glaxo SmithKline, Novartis, Otsuka, Thoratec, and ResMed, and consultancy fees from Luitpold. Dr. Alexander received research support from Gilead, Sanofi-Aventis, and Regeneron, and consultancy fees from CytRx. Dr. Hernandez received research support from the American Heart Association, Amgen, AstraZeneca, Bayer, Bristol-Myers Squibb, GlaxoSmithKline, Merck, and Portola; and serving as a consultant or scientific advisor to Amgen, AstraZeneca, Bayer, Eli Lilly, Gilead, GlaxoSmithKline, Janssen, Merck, MyoKardia, Novartis, Ortho McNeil-Janssen, Pfizer, Pluristem, and Sensible. No other disclosures were reported.

Correspondence: Lauren Cooper, Inova Heart and Vascular Institute, 3300 Gallows Road, Falls Church, VA, 22042. Email: Lauren.Cooper@inova.org.

Over the last decade, more than 15,000 patients have received a durable left ventricular assist device (LVAD) with an increasing number of implantations as destination therapy (DT).1 For patients with very advanced heart failure who are not candidates for heart transplantation, DT LVAD therapy offers improved survival more than medical therapy.2,3 However, mortality remains high, complications are common, and this therapy requires complex long-term care and management.4–6 Thus, appropriate patient selection is a critical component in the success of this therapy, and characteristics beyond the severity of heart failure may play a role in the outcomes after device implantation.

Frailty, a syndrome characterized by poor reserve and decreased resiliency, is recognized as an important predictor of cardiovascular outcomes.7,8 Frailty is of particular interest in patients with advanced heart failure, as the syndromes of frailty and heart failure share common pathways via inflammation and metabolic stress, and common symptoms of fatigue, exercise intolerance, and cachexia.9 It has been suggested that some of the frailty phenotype may be modifiable in the advanced heart failure population if the left ventricular dysfunction can be reversed with mechanical circulatory support.10 Thus, assessment of frailty may help with patient selection for LVAD therapy in predicting patients who may have increased mortality and adverse events after LVAD implantation.

The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) database provides multicenter data on patients undergoing LVAD implantation, including comorbidities, nutrition status, provider-assessed frailty, and gait speed testing, a commonly used test to assess frailty.11 This database offers a unique opportunity to study postimplantation outcomes and factors, including frailty, that are associated with those outcomes. Our objectives were to: 1) describe provider-indicated frailty and compare patients with and without provider-indicated frailty; 2) describe the collection and results of gait speed testing in this population; and 3) describe the association of these frailty measures with patient characteristics and outcomes.

Back to Top | Article Outline

Methods

Data Source

The INTERMACS database is a United States national registry that includes patients who undergo implantation of a U.S. Food and Drug Administration-approved durable mechanical circulatory support device since June 23, 2006.12 Interagency Registry for Mechanically Assisted Circulatory Support is sponsored by the National Heart, Lung, and Blood Institute, and participation in the registry is mandated by the Joint Commission for medical centers implanting mechanical circulatory support devices for DT. The institutional review board of each participating center may waive authorization and informed consent for participation in INTERMACS; otherwise, patients are consented for participation in INTERMACS. As of October 6, 2015, there were 164 active sites. The database includes information about preimplantation patient characteristics, including assessments of function and quality of life, and also includes details about the implant hospitalization and the implant itself. Short-term and long-term outcome data are also collected, including information about adverse events, repeat hospitalizations, and death. Data are monitored by the data coordinating center at the University of Alabama for completeness and accuracy.13

Back to Top | Article Outline

Study Population

The study population for this analysis was derived from all patients who received an LVAD from January 1, 2012, to March 31, 2014, entered into the INTERMACS Registry (N = 6,522), as frailty data were included in the INTERMACS registry starting in 2012. Patients who did not receive a DT LVAD (N = 3,627), who had missing information on the frailty indicator (N = 829), or who received a right ventricular assist device (N = 178) were excluded. The final analysis population included 2,469 patients receiving a DT LVAD.

Back to Top | Article Outline

Definitions and Outcomes

The primary variables of interest were provider-assessed frailty, defined as clinician assessment before LVAD implantation, and objective measures of functional capacity, specifically the gait speed test. The INTERMACS preimplant forms include a list of possible concerns and contraindications, and providers must denote if each condition is present (yes/no). A checkbox for frailty (yes or no) must have been marked for patients to be included in this analysis. Frailty was included in this list of specific conditions that also included advanced age, musculoskeletal limitations to ambulation, cardiothoracic issues, nutritional/gastrointestinal problems, vascular issues, oncology/infection issues, and psychosocial issues.14

Interagency Registry for Mechanically Assisted Circulatory Support data include preimplant exercise function, including gait speed, 6 minute walk test (6MWT) distance, and maximum oxygen consumption (VO2). Reasons for incomplete functional data included the following: 1) patient was too sick; 2) other; or 3) unknown. Patients who completed gait speed testing were divided into groups based on gait speed: < 0.8 and ≥ 0.8 m/s. Gait speed slower than 0.8 m/s has been shown to increase risk of adverse outcomes, and the cutpoint of 0.8 m/s is endorsed by the International Academy of Nutrition and Aging.11

Interagency Registry for Mechanically Assisted Circulatory Support also contains quality of life assessments including Kansas City Cardiomyopathy Questionnaire (KCCQ) and European Quality of Life (EQ-5D). For the quality of life questionnaires, we specifically examined the EQ-5D self-care and mobility measures, and the KCCQ fatigue measure. Reasons for incomplete functional data included the following: 1) patient was too sick; 2) patient preference (i.e., too tired, too stressed/anxious/depressed, too busy, too much trouble/do not want to be bothered/not interested, cannot concentrated, unwilling, unable to read English/illiterate); or 3) an administrative reason (i.e., urgent/emergent implant, no time to administer; coordinator too busy or forgot; unable to contact patient within the window).

In addition, INTERMACS contains data about patients’ demographics, medical history, laboratory values, device type, and discharge time and disposition, as well as outcomes. Outcomes of interest included mortality and rehospitalization in the first year postimplant. Adverse events of interest included neurologic events, infections, or major bleeding.

Back to Top | Article Outline

Statistical Analysis

We first described and compared the baseline characteristics of study groups defined by those with provider-assessed frailty and defined by gait speed test results. Characteristics were reported using frequencies and percentages for categorical variables and medians with 25th and 75th percentiles for continuous variables. We tested for differences between study groups using general χ2 tests for categorical variables and Wilcoxon rank-sum test for continuous variables.

To characterize the subjective assessment of frailty by healthcare providers, logistic regression was used to estimate the age–sex–adjusted associations of baseline characteristics with provider-assessed frailty in the study population. Each characteristic is included as a covariate, along with age and sex, in separate models. Covariates were chosen based on clinical significance.

Postimplant length of stay, as well as observed rates of adverse events, rehospitalization, and death were reported. Rates of rehospitalization were only calculated among those discharged alive with the LVAD in place. All outcomes were estimated using the cumulative incidence function to account for competing risks of LVAD explant and to account for mortality for the nonmortality outcomes. Differences between groups were evaluated using Gray tests. The association of frailty or gait speed testing with postimplant mortality outcomes was modeled using Cox regressions to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).

We performed a stepped regression analysis of 1 year mortality. First, we evaluated the association between provider-assessed frailty and mortality. We then adjusted that association for the contribution of age, sex, and INTERMACS profile on that association. Finally, we added inability to perform the gait speed test in the model. These predictors were chosen based on perceived clinical significance. This analysis only included patients who completed the gait speed test and patients who were too sick to complete the gait speed test.

All analyses for this study were performed at Duke University, using SAS version 9.3 (SAS Institute Inc., Cary, NC). This study was approved by the institutional review board of the Duke University Health System.

Back to Top | Article Outline

Results

Of 2,469 patients included in this analysis, 227 (9.2%) had provider-assessed frailty (Figure 1). Only 320 of 2,469 patients (13%) performed gait speed testing. Of those who completed the gait speed test, we observed similar frequencies between the provider-frail and nonfrail groups (13.7% vs. 12.9%). A higher percentage of patients in the provider-frail group did not complete the gait speed test because of being too sick (55.5% vs. 41.1%) compared with the nonfrail group. Of the 320 patients who completed gait speed testing, 174 (54.4%) had gait speed < 0.8 m/s. Among the 31 patients with provider-assessed frailty who completed the gait speed test, 20 (64.5%) had gait speed < 0.8 m/s.

Figure 1

Figure 1

Back to Top | Article Outline

Patient Characteristics Associated with Frailty

The provider-assessed frail group had a higher percentage of patients over the age of 70, whereas the nonfrail group had a higher percentage of patients under the age of 60 (Table 1). A higher percentage of provider-assessed frail patients had two or more hospitalization in the year preceding LVAD implant compared with nonfrail patients, but a similar distribution of New York Heart Association (NYHA) Class and INTERMACS profiles. Among those with EQ-5D data, patient-reported mobility measures from the EQ-5D survey were similar between provider-assessed frail and nonfrail patients, but using the EQ-5D self-care measures, more provider-assessed frail patients reported problems with self-care (Table 2). Kansas City Cardiomyopathy Questionnaire fatigue measures were similar between groups. Among patients with exercise and functional testing before LVAD implantation, there was no significant difference in 6MWT distance or gait speed between provider-frail and nonfrail groups.

Table 1

Table 1

Table 2

Table 2

A higher percentage of younger patients were in the group that was too sick to complete the gait speed test than in the other gait speed groups (Table 1). Those too sick to perform the gait speed test were more frequently INTERMACS profile 1 or 2. Those unable to complete the gait speed test more frequently reported difficulty with mobility and self-care in the EQ-5D survey and had worse KCCQ fatigue measures compared with those able to complete the test (Table 2). Overall, among patients who completed the gait speed test, those with gait speed < 0.8 m/s were similar to those with gait speed ≥ 0.8 m/s, although the slower gait speed group had a lower percentage of males (77.6% vs. 86.3%), a lower median prealbumin (19 mg/dl vs. 22 mg/dl), and a higher rate of multimorbidity (28.2% vs. 22.6%).

After adjusting for age and sex, variables that were associated with provider-assessed frailty include older age, lower body mass index (BMI), lower prealbumin, and other concerns or contraindications, including advanced age, musculoskeletal limitations, malnutrition, and multimorbidity (Table 3). Worse EQ-5D self-care score was associated with higher odds of frailty classification in an age- and sex-adjusted analysis (odds ratio [OR]: 1.67 [95% CI: 1.28–2.19]; p < 0.001). However, higher self-reported EQ-5D mobility scores and KCCQ fatigue measures were not associated with classification of frailty.

Table 3

Table 3

Back to Top | Article Outline

Association of Frailty with 1 Year Outcomes

Adverse outcomes of interest included in-hospital length of stay and mortality, 1 year mortality, rehospitalization, neurologic dysfunction, infection, and bleeding. Patients with provider-assessed frailty had a higher rate of death during index hospitalization compared with nonfrail patients (10.6% vs. 7.5%; p = 0.10). In addition, provider-assessed frail patients had longer postimplant hospitalizations (median [interquartile range {IQR}] days: 21 [14–33] vs. 19 [14–28] days; p = 0.03). In the first year after LVAD implantation, 24.6% of provider-assessed frail patients and 18.9% of nonfrail patients died (p = 0.01) (Table 4). The provider-assessed frail and nonfrail patients had similar rates of 1 year rehospitalization (78.5% vs. 77.4%; p = 0.24), neurologic events (3.0% vs. 2.1%; p = 0.55), and infections (9.0% vs. 10.9%; p = 0.53). One year mortality was similar for patients with gait speed < 0.8 and ≥ 0.8 m/s (15.2% vs. 16.6%), but was higher for patients who were too sick to complete gait speed testing (22.0%).

Table 4

Table 4

In stepped regression models, provider-assessed frailty was associated with 1 year mortality, but this association did not reach clinical significance (HR: 1.38 [95% CI: 0.97–1.95]) (Table 5). Adding in age, sex, and INTERMACS profile lessened the association between provider-assessed frailty and mortality. Advanced age and INTERMACS profiles 1 (critical cardiogenic shock) and 2 (progressive decline) were associated with 1 year mortality. The addition of inability to perform gait speed testing did not substantially change the effect estimate of provider-assessed frailty. Adding an indicator for inability to perform gait speed testing because of illness was not significant in the model for mortality (HR: 1.31 [95% CI: 0.93–1.85]); however, it lessened the effect associated with the INTERMACS profiles in the model, suggesting a correlation between those variables.

Table 5

Table 5

Back to Top | Article Outline

Discussion

In this contemporary cohort of 2.469 DT LVAD patients with frailty information, almost 10% were considered to be frail by their providers before LVAD implantation, and these patients had higher mortality. Although slow gait speed is a common marker of frailty, in this population, over 40% of patients were too sick to even perform gait speed testing, and mortality was significantly higher for those who were unable to complete the test. Among those who completed the gait speed test, there was no significant difference in gait speed between provider-frail and nonfrail patients, and mortality among patients who performed gait speed testing was not significantly different regardless of the gait speed. Moreover, adding a variable for inability to complete gait speed testing to the multivariable model lessened the strength of the INTERMACS profile on 1 year mortality. These results suggest that existing objective performance-based assessments of frailty may not be relevant for assessing frailty in this advanced heart failure population. Rather, the inability to perform the test may, in itself, be a marker for poor outcomes after LVAD implantation, as patients who were too sick to perform the gait speed test had worse heart failure and significantly higher rate of mortality at 1 year compared with patients who performed the test, regardless of the actual gait speed.

Advanced heart failure patients have limitations beyond those of less severe heart failure patients; thus, objective physical assessments that have been used to assess frailty in other heart failure populations may not be feasible in the advanced heart failure population.15 Accordingly, two prior single-center studies describing the impact of frailty on post-LVAD outcomes did not use performance-based measures to assess frailty. Dunlay et al.16 used the Deficit Index to show that worsening frailty was associated with an increased risk of mortality, whereas Chung et al.17 demonstrated that decreased hand grip strength was associated with increased mortality and postoperative complications. Adding to these prior studies, and recognizing that objective data may not be available to assess frailty in some end-stage heart failure patients, we examine subjective assessments of frailty in DT LVAD patients. Without objective data to define frailty, providers may use subjective assessments or gestalt—a so-called “eyeball test”—to identify frailty. Consistent with prior studies, provider-assessed frailty in this study had important associations with postimplant outcomes. Frail patients had longer postimplantation index hospitalization and higher incidence of postimplant mortality underscoring the importance of appropriate recognition and accurate assessment of frailty.

Provider-assessed frailty in this DT LVAD population appears to track some domains that define frailty, including shrinking, slowness, exhaustion, low physical activity, and weakness.9 For example, low BMI and low prealbumin were associated with the indicator of frailty, as was patient-reported difficulty with self-care, while patient-reported measurements of mobility and fatigue did not identify frail patients. In addition, advanced age and multimorbidity which contribute to the phenotype of frailty were both associated with provider-assessed frailty in this study. Notably, disease severity was not associated with provider-assessed frailty, highlighting the distinction between frailty and end-stage illness, such that providers are able to recognize frailty in patients along the spectrum of heart failure severity. Given performance-based measures of frailty are rarely feasible in the DT LVAD population, they provide limited ability to identify frailty. Future research should focus on identifying the best tools to assess for frailty in this population, either with the development of novel frailty assessments or the validation of known frailty assessment tools. In addition, it is unknown whether interventions, such as optimizing nutrition, medical or hormonal therapy, conditioning and physical therapy, or even temporary mechanical circulator support, can improve frailty before LVAD implantation, thus exploration in this area is warranted. Furthermore, additional research is needed to identify if improvements in frailty can mitigate post-LVAD risks. Finally, a better understanding of how to use frailty assessment in clinical decision making is needed.

Back to Top | Article Outline

Limitations

Our results should be taken in the context of several limitations. First and foremost, we were limited by the data available in this national registry. There was a high degree of missingness for many baseline variables. It is unknown whether these data were not collected, unavailable, or simply not entered. In addition, other relevant outcomes such as falls and level of disability were not included in the registry. Furthermore, we were limited by the quality and accuracy of data entered into the database. The high degree of missing data in this robust registry highlights the need for accurate and complete data collection. This is of utmost importance for future research and clinical care of these patients.

Second, we recognize that subjective measures of frailty have the potential for inaccuracies and bias. Third, this is a retrospective cohort study, and thus there is a possibility that unmeasured confounders influenced the outcome results. In addition, there was likely a selection bias in the study population. The population only included patients who underwent LVAD implantation, and so we were unable to study frail patients with end-stage heart failure who were not offered or declined this therapy. This may explain why the prevalence of frailty in our study population was less than the prevalence in a chronic heart failure population.18,19 And finally, we were unable to account for regional or center-specific differences in patient care which may have influenced outcomes. Despite these limitations, INTERMACS is the only national registry of durable mechanical circulatory support devices, so this is the most comprehensive multicenter data of frailty in DT LVAD patients.

Back to Top | Article Outline

Conclusion

Although end-stage heart failure and frailty share common features, frailty may mitigate some of the benefits of LVAD therapy in frail advanced heart failure patients, thus recognition of frailty is an important component of patient selection for LVAD therapy. Performance-based measures of frailty are difficult to obtain in this critically ill population, but inability to perform these assessments, often associated with worse heart failure severity, is a meaningful marker of worse outcomes after LVAD implantation. Future research should focus on assessment of frailty in this advanced heart failure population in ways to provide specific information which will enable better prediction of outcomes.

Back to Top | Article Outline

References

1. Kirklin JK, Naftel DC, Pagani FD, et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015.34: 1495–1504.
2. Rose EA, Gelijns AC, Moskowitz AJ, et al; Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Study Group: Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med 2001.345: 1435–1443.
3. Slaughter MS, Rogers JG, Milano CA, et al; HeartMate II Investigators: Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009.361: 2241–2251.
4. Holman WL, Kormos RL, Naftel DC, et al. Predictors of death and transplant in patients with a mechanical circulatory support device: A multi-institutional study. J Heart Lung Transplant 2009.28: 44–50.
5. Rosenbaum AN, John R, Liao KK, et al. Survival in elderly patients supported with continuous flow left ventricular assist device as bridge to transplantation or destination therapy. J Card Fail 2014.20: 161–167.
6. Kirklin JK, Naftel DC, Pagani FD, et al. Sixth INTERMACS annual report: A 10,000-patient database. J Heart Lung Transplant 2014.33: 555–564.
7. Bergman H, Ferrucci L, Guralnik J, et al. Frailty: An emerging research and clinical paradigm–issues and controversies. J Gerontol A Biol Sci Med Sci 2007.62: 731–737.
8. Afilalo J, Alexander KP, Mack MJ, et al. Frailty assessment in the cardiovascular care of older adults. J Am Coll Cardiol 2014.63: 747–762.
9. Fried LP, Tangen CM, Walston J, et al; Cardiovascular Health Study Collaborative Research Group: Frailty in older adults: Evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001.56: M146–M156.
10. Flint KM, Matlock DD, Lindenfeld J, Allen LA. Frailty and the selection of patients for destination therapy left ventricular assist device. Circ Heart Fail 2012.5: 286–293.
11. Abellan van Kan G, Rolland Y, Andrieu S, et al. Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) Task Force. J Nutr Health Aging 2009.13: 881–889.
12. Kirklin JK, Naftel DC, Stevenson LW, et al. INTERMACS database for durable devices for circulatory support: First annual report. J Heart Lung Transplant 2008.27: 1065–1072.
13. Kirkland JK. Interagency Registry for Mechanically Assisted Circulatory Support Manual of Operations and Procedures. 2014. Available at: https://www.uab.edu/medicine/intermacs/images/protocol_4.0/protocol_4.0_MoP/INTERMACS_MOP_Protocol_4_0_V1_0__06-3-2014.pdf. Page 39–40. Accessed May 17, 2017.
15. Chiarantini D, Volpato S, Sioulis F, et al. Lower extremity performance measures predict long-term prognosis in older patients hospitalized for heart failure. J Card Fail 2010.16: 390–395.
16. Dunlay SM, Park SJ, Joyce LD, et al. Frailty and outcomes after implantation of left ventricular assist device as destination therapy. J Heart Lung Transplant 2014.33: 359–365.
17. Chung CJ, Wu C, Jones M, et al. Reduced handgrip strength as a marker of frailty predicts clinical outcomes in patients with heart failure undergoing ventricular assist device placement. J Card Fail 2014.20: 310–315.
18. Jha SR, Ha HS, Hickman LD, et al. Frailty in advanced heart failure: A systematic review. Heart Fail Rev 2015.20: 553–560.
19. Goldwater DS, Pinney SP. Frailty in advanced heart failure: A consequence of aging or a separate entity? Clin Med Insights Cardiol 2015.9 (suppl 2): 39–46.
Keywords:

frailty; left ventricular assist device; destination therapy; outcomes

Copyright © 2018 by the American Society for Artificial Internal Organs