Cardiac Rehabilitation in Patients With Left ... - IngentaConnect

Scientific Review

Cardiac Rehabilitation in Patients With Left Ventricular Assist Device A SYSTEMATIC REVIEW AND META-ANALYSIS Toufik Mahfood Haddad, MD; Alok Saurav, MD; Aiman Smer, MBBCh; Muhammad S. Azzouz, MD; Abhilash Akinapelli, MBBS; Mark A. Williams, PhD; Venkata M. Alla, MD

Purpose: Exercise-based cardiac rehabilitation (EBCR) has been demonstrated to improve functional capacity in heart failure (HF). However, there are limited data on the effect of EBCR in patients with advanced HF and left ventricular assist devices (LVADs). This meta-analysis sought to evaluate the effects of EBCR on functional capacity in patients with LVAD. Methods: PubMed, Web of Science, CINAHL, and Cochrane Library databases were searched for randomized studies assessing the impact of EBCR in patients following LVAD implantation compared with standard therapy (ST). Using pre-defined criteria, appropriate studies were identified and selected. Data from selected studies were extracted in a standardized fashion and a meta-analysis was performed using a random-effects model with DerSimonian Liard weighting. Analysis employed weighted mean difference (WMD) as the effect size and intention-to-treat (ITT) principle. Study quality, publication bias, and heterogeneity were assessed. Results: Six trials with a total of 183 patients (EBCR: 125; ST: 58) were identified. Mean age was 51 years and 83% were males. The initiation of EBCR varied from LVAD implantation during the index hospitalization to 10 mo post-LVAD implantation. The median rehabilitation period ranged from 6 to 10 wk. Exercise-based cardiac rehabilitation was associated with improved · peak oxygen uptake (VO2) in all trials. Quantitative analysis was performed on 3 randomized studies involving 61 patients (EBCR = 39, ST = 22). Exercise-based cardiac rehabilitation was asso· ciated with significantly greater peak VO2 (WMD: 3.00 mL/kg/ min; 95% CI: 0.64-5.35, P = .001). Similarly, 6-minute walk distance (6MWD) showed significantly greater improvement in the EBCR group than in the ST group (WMD: 60.06 m; 95% CI, 22.61-97.50, P = .002). Heterogeneity was low among the included trials. Exclusion sensitivity and per-protocol analysis demonstrated results consistent with ITT analysis. None of the included studies reported serious adverse events related to EBCR, which supports the safety of EBCR after LVAD implantation. Conclusion: This systematic review and meta-analysis demonstrated that EBCR following LVAD implantation is associated with greater improvement in functional capacity compared with · ST as reflected by improved peak VO2 and 6MWD. However, Author Affiliation: Division of Cardiology, Creighton University School of Medicine, Omaha, Nebraska. Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s website (www.jcrpjournal.com). The authors declare no conflicts of interest. Correspondence: Toufik Mahfood Haddad, MD, Division of Cardiology, 3006 Webster St, Omaha, NE 68131 (Toufikmahfoodhaddad@creighton. edu). Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/HCR.0000000000000254

390

given the small number of patients, further research into the clinical impact of EBCR in LVAD patients is necessary. Key Words: cardiac rehabilitation • heart failure • left ventricular assist device • meta-analysis

L

eft ventricular assist devices (LVADs) are being increasingly used to support patients with advanced heart failure (HF), both as a bridge to recovery and as destination therapy.1,2 It is well-known that LVADs improve survival, functional capacity, and quality of life (QOL) in patients with advanced HF.3,4 However, compared with recipients of heart transplants, many patients with LVADs continue to experience exercise intolerance.5 Exercise-based cardiac rehabilitation (EBCR) has been shown to be safe and effective in patients with HF.6,7 Exercise-based cardiac rehabilitation · improves peak oxygen uptake (VO2), functional capacity, and QOL and reduces HF symptoms and hospitalizations.6-8 However, there are very limited data on the safety and efficacy of EBCR in patients with LVADs. We therefore systematically reviewed the literature and performed a meta-analysis to evaluate the impact of EBCR in LVAD recipients.

METHODS LITERATURE SEARCH A systemic literature review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.9 Two authors separately searched PubMed, Web of Science, CINAHL, and Cochrane Library databases for studies assessing the effect of cardiac rehabilitation in patients following LVAD implantation from January 1966 through May 2016. We used the following key words in various combinations: heart assist device; ventricular assist device; assistive device; and cardiac rehabilitation. Additional details of search terms and strategy are provided in the Supplemental Appendix (Supplemental Digital Content 1, available at http://links.lww.com/JCRP/A52). The bibliographies of selected manuscripts and review articles were also manually searched for additional studies that were not identified in the original search. Titles and abstracts were screened to identify studies for full-text review. Studies assessing the effect of EBCR in patients following LVAD implantation and fulfilling the pre-defined inclusion and exclusion criteria were selected for inclusion in the review and quantitative analysis. Search strategy and results are depicted in Figure 1. STUDY SELECTION Studies were included in the systematic review and meta-analysis if they (1) assessed the impact of EBCR in

Journal of Cardiopulmonary Rehabilitation and Prevention 2017;37:390-396

www.jcrpjournal.com

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

Figure 1. PRISMA flowchart. Abbreviations: CINAHL, Cumulative Index to Nursing and Allied Health Literature; MeSH, Medical Subject Headings; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

LVAD recipients; (2) included a standard therapy (ST) group for comparison; and (3) enrolled adult subjects (>18 years of age). Exclusion criteria included (1) non-English language publication; (2) published abstract without fulltext publication; (3) studies assessing the effect of cardiac rehabilitation on LVAD recipients without an ST group; (4) studies assessing the effect of different treatment strategies of cardiac rehabilitation without an ST group; and · (5) studies lacking endpoint measures such as peak VO2 and 6-minute walk distance (6MWD). Standard therapy was defined as LVAD recipients who were not given any individualized/structured exercise prescription apart from advice to walk regularly by their physicians (30-45 min/d). The EBCR group was defined as participation in a structured cardiac rehabilitation program with a clearly defined protocol. DATA EXTRACTION AND OUTCOME ASSESSMENT Two independent authors performed the search in duplicate, selected studies for inclusion and extracted data in a standardized fashion. Differences were resolved by consensus or after review with a third author. The following data from selected studies were extracted: (1) general characteristics, including study design, sample size, year of publication; (2) patient characteristics, including mean age, gender; (3) duration between LVAD insertion and study start date; (4) definition of outcomes and duration of follow-up;

www.jcrpjournal.com

(5) type of LVAD used; and (6) summary findings. Quality assessment was performed using the Newcastle-Ottawa Scale method10 for nonrandomized controlled studies and Jadad scale11 for randomized controlled studies. STATISTICAL ANALYSIS Data from selected studies were extracted and a meta-analysis of randomized trials was performed comparing EBCR versus ST using the random effects model with DerSimonian Liard weighting.12 Analysis employed weighted mean difference (WMD) as the effect size and intention-to-treat (ITT) principle. Publication bias was analyzed visually using funnel plot. The Cochrane Q statistic was calculated and used to determine the heterogeneity of included studies for each endpoint. I2 index values of 25% to 50%, 51% to 75%, and >75% were considered as low, moderate, and high heterogeneity, respectively. An exclusion sensitivity analysis was performed when necessary and a 2-sided P value of < .05 was considered statistically significant for all analyses.

RESULTS A total of 6 studies involving 183 patients (EBCR = 125; ST = 58) were included in the review.13-18 The most common indication for LVAD was bridging to heart transplantation

CR in Patients With LVADs


391

and the most common device used was a second-generation axial continuous-flow pump. Study characteristics are summarized in the Table. The mean age was 51 years (EBCR = 51 years; ST = 51 years) and 83% of subjects were males. Exercise-based cardiac rehabilitation was associated with · improved peak VO2compared with ST in all but 2 trials14,17 and showed QOL improvement in all trials except one.15 The duration of follow-up was variable across the studies, ranging from 1 to 18 mo. Initiation of exercise rehabilitation ranged from initiation during the index hospitalization (for LVAD implantation) to 10 mo post-LVAD implantation and the median rehabilitation period varied from 6 to 10 wk. Cardiac rehabilitation training session frequency ranged from 3 to 5 times weekly. Quantitative analysis was performed on 3 randomized studies involving 61 patients (EBCR = 39; ST = 22).13-15 Exercise-based cardiac rehabilitation was associated with · significantly greater peak VO2 (WMD = 3.00 mL/kg/min; 95% CI, 0.64-5.35, P = .001) (Figure 2). Similarly, 6MWD improved to a greater extent in the EBCR group than in the ST group (WMD = 60.06 m; 95% CI, 22.61-97.50, P = .002) (Figure 3). Quantitative analysis was not performed on QOL measures as the tools used to evaluate the QOL varied between studies (Minnesota Living with Heart Failure Questionnaire [MLWHFQ], Kansas City Cardiomyopathy Questionnaires, and 36-item Short Form Health Survey). Heterogeneity was low among the included trials with an I2 · value of 25.6% in the analysis of VO2 and 0% in the analysis of 6MWD. Exclusion sensitivity and per-protocol analysis demonstrated results consistent with main (ITT) analysis.

DISCUSSION Our analysis suggests that EBCR results in significant improvements in functional capacity as measured by 6MWD · and peak VO2 in patients with advanced HF who undergo LVAD implantation. In our review, the mean baseline · peak VO2 and 6MWD of patients were 13 mL/kg/min and 380 m, respectively. In patients who were prescribed EBCR, · the peak VO2 and 6MWD increased to 17 mL/kg/min (23% increase from baseline) and 490 m (28% increase), respectively, compared with 13.8 mL/kg/min and 429 m in the · control group. The baseline values of peak VO2 and 6MWD are consistent with published data on LVAD recipients. In a systematic review of patients receiving continuous-flow LVADs, the average baseline 6MWD ranged from 180 to 200 m and increased to 320 to 350 during 3- to 6-mo follow-up,19 highlighting the dramatic improvements in functional capacity with LVAD implantation. The additional improvements noted in our study provide evidence that patient participation in EBCR after LVAD implantation can provide substantial improvements in functional capacity above the dramatic improvements that occur with LVAD implantation alone. To the best of our knowledge, this is the first systematic review/meta-analysis assessing the benefits of EBCR in LVAD recipients. In the pivotal HF-ACTION study (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) randomized controlled trial, 2331 stable outpatients with systolic HF were randomized to supervised exercise program or usual care. During a median follow-up of 30 mo, patients randomized to EBCR had an 11% reduction in the primary endpoint of all-cause mortality or hospitalization and a 15% reduction in the secondary endpoint of cardiovascular mortality or HF hospitalization after adjustment for other predictors.6,7 The majority of these patients had New York Heart Association class II or III symptoms. Base· line VO2 and 6MWD in the ST and EBCR groups were 14.5

392

mL/kg/min and 373 m and 14.4 mL/kg/min and 366 m, respectively. At 3 mo, EBCR was associated with an increase · in peak VO2 of 0.6 mL/kg/min (15%) and an increase in 6MWD of 20 m compared with 0.2 mL/kg/min and 5 m, respectively, in the usual care group.6 Despite the modest improvements in measures of functional capacity, significant clinical benefits were noted. The relative improvements noted in our analysis appear larger than those observed in the HF-ACTION study. This is likely due to the following factors: (1) our analysis included a small number of patients, which might have resulted in an overestimation of the impact of EBCR; (2) patients in our study had more advanced · HF and worse functional capacity (peak VO2) at baseline; thus, larger percentage improvements are likely to be observed even with small changes; and (3) patients receiving LVADs or transplant generally get more intense teaching and closer follow-up, which may result in better adherence to pharmacologic and exercise prescriptions compared to general HF populations. Of note, only 30% of those in the HF-ACTION trial exercise-arm exercised at or above the target number of minutes per week.20 Unfortunately, none of the studies in our review assessed the impact of EBCR on major clinical outcomes such as mortality or hospitalization although it has previously been shown that even modest improvements in functional capacity can predict improvements in long-term clinical outcomes.21 The improvements noted with EBCR in our analysis appear comparable to those observed with other common interventions in advanced systolic HF such as cardiac resynchronization therapy. In a substudy of COMPANION (The Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure),22 trial patients who underwent cardiac resynchronization therapy had an average improvement of 1.26 mL/kg/min (from a baseline of 12.7 mL/kg/min) in · peak VO2 and an increase of 43.0 m (baseline of 292 m) in 6MWD. While heart transplantation is associated with larger improvements in functional capacity ranging from · 3 to 4.5 mL/kg/min increase in VO2,23-25 the continuing organ donor shortage precludes this option for the majority of patients with advanced systolic HF. Left ventricular assist devices therefore represent a significant advancement in the therapy of patients with stage D HF with reduced ejection fraction.26 Significant technological advances in LVAD design have improved the durability and safety of these devices, resulting in increased use of LVAD as destination therapy. However, despite these advancements, QOL and exercise tolerance remain lower than that reported in recipients of heart transplants.5,27 The persistent functional impairment, despite adequate hemodynamic support after LVAD implantation, may be related to other peripheral mechanisms such as skeletal muscle deconditioning and apoptosis associated with advanced HF.28 Derangement of skeletal muscle structure and metabolic function is associated with exercise intolerance and fragility. These changes correlate poorly with central hemodynamics.29 Recruitment of skeletal muscle fibers during exercise is known to increase oxygen transport and uptake by skeletal muscles, upregulation of anabolic enzymes, and decrease in the catabolic catalysts. These favorable changes lead to an increase in peak muscle strength and could contribute to the additive benefits of EBCR in LVAD recipients.30 Furthermore, none of the included studies in our review reported serious adverse events related to EBCR. In 1 recent retrospective study, Marko et al18 identified only 1 training-related adverse event in 41 LVAD recipients undergoing a total of 1600 training sessions. Thus, participation in EBCR shortly after LVAD implantation appears safe, despite the limited nature of the available evidence.31

Journal of Cardiopulmonary Rehabilitation and Prevention 2017;37:390-396

www.jcrpjournal.com


www.jcrpjournal.com

CR in Patients With LVADs


393

13

26 18 73 55 ± 13 Second-generation CF

Bridge to HT or destination 1–6 mo 6 wk 3 KCCQ 9.8 EBCR improves functional capacity and health status in LVAD patients 5 NA

Total patients, n EBCR patients, n Male, % Age,c y Type of LVAD

Indication Rehabilitation start Rehabilitation period Sessions per week, n Quality of life (QOL) Impact of EBCR on QOL Authors’ conclusions

Bridge to HT