SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

The authors sought to conduct a systematic review comparing the effects of exercise training in heart failure patients taking β-blockers vs those not. A systematic search of exercise training trials in chronic heart failure patients that compared groups who took β-blocker medication or compared selective and nonselective β-blockers during exercise training was conducted. Eight prospective studies met the criteria for the quantitative synthesis, which included data from 236 participants. The increment in peak oxygen consumption (VO2) was greater in exercising vs control participants, with a mean difference (MD) of 1.27 mL/kg/min (95% confidence interval [CI], 0.85–1.70; P<.00001). In exercising patients, the increment in peak VO2 was greater in the group taking β-blocker vs those taking placebo (MD, 1.66 mL/kg/min; 95% CI, 0.36–2.97; P=.01). In exercising patients, there was no difference in the increment of peak VO2 between nonselective β-blocker and selective β-blockers groups (MD, −0.09 mL/kg/min; 95% CI, −1.54–1.36; P=.09). Minnesota Quality of Life Score was significantly better in the exercise group vs sedentary control group (both groups taking β-blockers) (MD, −11.3; 95% CI, −15.9 to −6.8; P<.00001). Our analysis demonstrated that β-adrenergic blocker therapy did not reduce exercise capacity or exercise training adaptations and quality of life in heart failure patients.

The use of β-adrenergic–blocking agents as prescribed medication is well established in people with chronic heart failure (HF).1β-Blockers lower systolic blood pressure and heart rate relative to cardiac output and increased stroke volume. Cardioselective β-blockers inhibit β1 receptors, whereas nonselective β-blockers inhibit both β1 and β2 receptors, but the net result of β-blockers is reduced ventricular contractility during systole. For this reason, physicians initially considered the possibility that β-blockers may be harmful to people with HF, although trials in the 1990s suggested this was not the case.2

HF is related to many pathophysiologic abnormalities that occur in order to maintain cardiac output to the body’s organs and tissues, this is especially challenging during physical exertion. Meta-analyses have shown that exercise training improves the clinical status of HF patients by increasing peak oxygen consumption (VO2),3,4 improving the percentage of left ventricular ejection fraction (LVEF),5 lowering systemic brain natriuretic peptide (BNP),4,6 and improving quality of life,7 but it is possible that β-blockers may attenuate exercise training adaptations. Previous work has examined whether exercise-induced adaptations are attenuated by β-blockers in people with coronary heart disease8 and HF.9 Later work also examined whether exercise training adaptations were superior with selective vs nonselective β-blockers.10 The published work to date that has studied the interaction between β-blockers and exercise training adaptations in HF patients have small study populations, limiting the robustness of the findings. Many commonly presumed adverse β-blocker effects are not supported by published randomized clinical trials11 and as such many more patients may be eligible to use these drugs and compliance rates may be improved if the effects on exercise capacity are better understood.12

As β-blocker therapy is the cornerstone of pharmacologic management of stable HF patients and exercise training is the nonpharmacologic equivalent, it is important to understand the relationship between these therapies. We therefore conducted a systematic search of the literature to identify randomized controlled studies of exercise training in HF patients who were and were not taking β-blockers. We subsequently conducted a meta-analysis of studies that met our inclusion criteria. The primary aim of this work was to establish whether β-blockers attenuated physical training adaptations in HF patients. The secondary aim was to establish whether selective or nonselective β-blockers were more beneficial to those people with HF undertaking exercise training. Finally we sought to establish whether quality of life was better in those individuals with HF who were or were not taking β-blockers during exercise training.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

Search Strategy

Studies were identified through a MEDLINE search strategy (1985–2011) and the Cochrane Controlled Trials Registry (1966–2011). The search strategy included a mix of MeSH and free text terms for the key concepts β-blocker, β-blockade, exercise training, HF, cardiomyopathy, left ventricular dysfunction, maximal oxygen consumption, and peak VO2. These terms were combined with a sensitive search strategy to identify prospective randomized or controlled trials. We scrutinized a relevant reference list of papers for new references. We included studies if ejection fraction was <40%. Two included study authors were contacted and provided further data.

Study Selection

The search was specified to prospective randomized or controlled parallel trials of exercise training in chronic HF patients that compared groups who took β-blocker medication or compared selective and nonselective β-blockers during exercise training. Records identified 85 papers through database searching. Three additional records from the reference list were included. We initially removed 61 overlapping abstracts and irrelevant articles that did not match the inclusion criteria, leaving 27 possible studies. After screening, we excluded 7 cohort studies, 3 animal based studies, and 2 review papers. Seven studies of acute exercise testing responses or exercise training intervention <2 weeks were excluded. There were no language restrictions. Two reviewers (HI and NS) matched the identified studies to inclusion/exclusion criteria and a third reviewer (JF) adjudicated over any disputes.

Outcomes Measures

We recorded the percentage of exercise sessions and adverse events (those events causing temporary or permanent withdrawal and any death or hospitalization). We also measured the relationship between estimated total work and percentage of peak VO2.

Data Synthesis

We archived data on types of β-blocker, age, sex, peak VO2, and ventricular ejection fraction. Energy expenditure on aerobic training was calculated using established equations.21 We also included data on duration of exercise, frequency, mode of training, and training intensity. Total percentage change in peak VO2 was calculated for measuring patients training adaptation. We analyzed the change in peak VO2 across 3 comparative groups: exercise vs sedentary (all participants taking β-blockers), β-blockade vs placebo (all participants exercised), and selective vs nonselective β-blockade (all participants exercising).

Assessment of Study Quality

We assessed study quality with regard to randomization methods, recruitment, outcome reporting, any incomplete outcome data, and study design such as drug blinding and allocation concealment. Only three studies were prospective randomized controlled trials.13–15 We also assessed study quality according to the Jadad and colleagues16 and Physiotherapy Evidence Database (PEDro)17 scales.

Statistical Analyses

We conducted a sample size calculation based on a 2004 systematic review of 81 studies that quantified expected peak VO2 changes in HF patients with exercise training at 16.8%±8%.4 We found a sample size of 38 (19 in each group) was needed to give 80% power and 5% significance for change in peak VO2. Revman 5.1 (The Nordic Cochrane Centre, Copenhagen, Denmark) was used to conduct meta-analyses. Data used were continuous and were reported as mean and standard deviation. Revman 5.1 enabled calculation of postintervention change from baseline for standard deviation using change in mean values, number of patients, and P value for each group. Mean difference in these data from baseline were analyzed. Pearson correlation coefficient was used to examine relationships between exercise delivery parameters (eg, intensity and duration) and change in primary outcome measures. An Egger plot was produced to identify sources of publication bias.18 We used a 5% level of significance and a 95% confidence interval to report change in outcome measures. We analyzed baseline vs postintervention change in peak VO2 across 3 comparative groups to eliminate the effects of poor matching of study groups at baseline: exercise vs sedentary (all participants taking β-blockers), β-blockade vs placebo (all participants exercised), and selective vs nonselective β-blockade (all participants exercising). We conducted a sensitivity analysis for peak VO2 (exercise vs sedentary control—both groups taking β-blockers). We achieved this by removing studies that were not prospective or randomized to see whether the point estimate remained statistically significant. Other sensitivity analyses were not justified due to study exclusions. We also conducted a sensitivity analysis that included only studies that used cycle ergomtery for exercise testing.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

Of the 15 studies assessed for eligibility, 7 were excluded and 8 prospective studies6,9,10,13–15,19,20 met the criteria for the quantitative synthesis (Figure 1). Participants from the studies totaled 236. Of these, 147 took β-blockers and completed exercise training, 46 took β-blockers in the nonexercise group, and 29 were placebo participants. The peak VO2 of participants was between 12.9 mL/kg/min to 20.9 mL/kg/min, LVEF ranged from 15% to 33%, age ranged from 51±14 to 68.6±13.4 years, and the majority were men (Table I). Only 1 death was reported in a sedentary control patient and 1 episode of worsening HF was reported in an exercising patient.

image

Figure 1.  Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 flow diagram.

Download figure to PowerPoint

Table I. Trials Reporting Effects of β-Blockade of Exercise Training Adaptations in Heart Failure Patients
Study, YearPatients, No.AgeMale/FemalePeak VO2, mL/kg/minBaseline LVEF, %Medication, %
  1. Abbreviations: ACE, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; LVEF, left ventricular ejection fraction, NA, not applicable; VO2, peak oxygen consumption.

Curnier, 2001β-blocker and exercise (n=16)51.3±9.1NA20.3±6.133.9±7.5ACE/ARB 81 Diuretic NA
No β-blocker and exercise (n=18)55.2±12.8NA17.8±3.829.1±6.5ACE/ARB 94 Diuretic NA
Demopoulos, 1997Carvedilol and exercise (n=8)59±13Male 75%12.9±1.423±9ACE/ARB 100 Diuretic NA
Carvedilol no exercise (n=8)61±8Male 70%13.1±2.623±6ACE/ARB 100 Diuretic NA
Propanolol and exercise (n=7)58±11Male 75%12.4±1.024±7ACE/ARB 100 Diuretic NA
Forissier, 2001Exercise and no β-blockade (n=14)53±12Male 79%18.1±425±6ACE/ARB 100 Diuretic 100
Selective β-blockade with exercise (n=11)54.6±14Male 100%21.5±628±6ACE/ARB 100 Diuretic 100
Carvedilol and exercise (n=13)55.6±16Male 69%20±926±8ACE/ARB 100 Diuretic 100
Fraga, 2007Carvedilol and exercise (n=15)57±2.6Male 53%13.2±1.127±1.5ACE/ARB 80/20 Diuretic 87
Carvedilol no exercise (n=12)53±2.5Male 75%13±0.926±2ACE/ARB 83/17 Diuretic 83
Levinger, 2005β-blocker with exercise (n=8)57.3±11.1Male 100%14.4±2.835.4±6.3ACE/ARB 75 Diuretic 28
β-blocker no exercise (n=7)56.7±10.1Male 100%14.9±1.034.0±8.8ACE/ARB 86 Diuretic 29
Nishi, 2011β-blocker with exercise (n=33)51±14Male 88%16.3±4.318±4ACE/ARB 91 Diuretic 97
β-blocker without exercise (n=12)52±16Male 83%17.4±2.715±5ACE/ARB 92 Diuretic 92
Pietila, 1999Celiprolol and exercise (n=20)59.1±1.9Male 90%17.2±1.333.3±1.8ACE/ARB 65 Diuretic 50
Placebo and exercise (n=11)59.6±2.3Male 91%20.9±1.233.6±1.8ACE/ARB 82 Diuretic 45
Pozehl, 2003Carvedilol with exercise (n=16)66.9±9.6Male 87%17.62±3.328.3±7.0ACE/ARB 100 Diuretic 100
Carvedilol no exercise (n=7)68.6±13.4Male 87%16.20±3.930.9±8.6ACE/ARB 100 Diuretic 100

Exercise Training Parameters

Types of training were aerobic and combination of aerobic and anaerobic activity, except one study of resistance training. Modes of training were cycle, walk, calisthenics, step-board, treadmill, and strength training. Duration of the study was from 1 month to 6 months with frequency of training between 3 and 6 days. The intensity of training was prescribed by peak VO2 (50%–85%), heart rate reserve (30%–50%), 40 % to 90% maximum strength, below ventilatory threshold, and 60% to 85% of maximum heart rate. Exercise training varied between 15 and 60 minutes in a session. Estimated total work from exercise training was between 439 Kcal/wk and 977 Kcal/wk, although the calculations were based on only the aerobic portion of exercise program (See Table II).

Table II. Exercise Program Characteristics of Included Randomized Controlled Trials
Study, YearTraining StudyDuration, Session DurationFrequency, d/wkIntensityModeEstimated, Kcal/wk% Change Peak VO2
  1. Abbreviations: BB, β-blocker; bpm, beats per minute; CON, sedentary control; HR, heart rate; Ex, exercise; NA, not applicable; VO2, oxygen consumption; VT, ventilatory threshold.

Curnier, 2001Aerobic4 wk5–6HR at a VT >5 bpm or HR at a VT <5 bpmStationary electromagnetic bicycle977Ex 22 Con 17
Demopoulos, 1997Aerobic12 wk (15–60 min)450% peak VO2Cycle660Ex/carvedilol 24 Con 2 Ex/propranolol 24
Forissier, 2001Aerobic4-wk (20–40 min)3–5Below ventilatory thresholdCycle, calisthenics exercise, walkGroup 1=576 Group 2=586 Group 3=548Ex/No BB 17 Ex/nonselective BB 13 Ex/carvedilol 17
Fraga, 20071=Aerobic 2=Resistance4 mo Aerobic (25–40 min) Resistance (20 min)310% below anaerobic thresholdStretching, cycle, strengthening exercise439Ex 21 Con −8
Levinger, 2005Resistance8 wk3Wk 1–6=40%–60% of maximum strength Wk 7–12=80–90% maximum strengthResistance trainingNAEx 23 Con 1
Nishi, 2011Aerobic3 mo (40–60 min)3 to 51.30–50% heart rate reserve 2.11–13 on rate of perceived exertion (Borg) scaleWalk, cycle, calisthenics exercise464Ex 16 Con 1
Pietila, 19991=Aerobic 2=Anaerobic6 mo (30 min)660–85% maximum HRWalk, stepboard exercise or ergometer training, aerobic exercise, circuit muscle training965Ex +13 Con −3
Pozehl, 20031=Aerobic 2=Resistance 12 wk Aerobic (20 min) (50 min)3 + 3 (home)1=Rate of perceived exertion (Borg scale 12–14) 2=60–85% peak VO2Treadmill, stationary bikes, rowers, arm ergometers, strength training, stretching540Ex −3.4 Con 3.3

Change in Peak VO2, VE/VCO2, and Quality of Life

Four studies were included in the analysis of exercise training groups that compared exercise training vs control, the increment in peak VO2 was greater in exercising participants (mean difference [MD], 1.27 mL/kg/min; 95% confidence interval [CI], 0.85–1.70; P<.00001) (Figure 2). Two studies measured the effect of β-blocker and placebo in exercising participants, and the increment in peak VO2 was greater in the former group (MD, 1.66 mL/kg/min; 95% CI, 0.36–2.97; P=.01) (Figure 3). Two studies measured the comparison between nonselective β-blocker and selective β-blockers, and there was no difference in the increment of peak VO2 between groups (MD, −0.09 mL/kg/min; 95% CI, −1.54–1.36; P=.09) (Figure 4). The Minnesota Living With HF Score (quality of life questionnaire) was significantly better in the exercise group vs sedentary control (both groups exercising) (MD, −11.3; 95% CI, −15.9 to −6.8; P<.00001) (Figure 5). Minute ventilation/carbon dioxide output (VE/VCO2) slope was lower in the postintervention exercise group vs sedentary control (both groups exercising) (MD, −3.4; 95% CI, −4.8 to −1.5; P=.0002) (Figure 6). A sensitivity analysis for peak VO2 (exercise vs sedentary control—both groups taking β-blockers) meant the point estimate was no longer statistically significant (Figure 7). A sensitivity analysis including only studies that used cycle ergomtery for exercise testing produced an MD in peak VO2 of 1.94 mL/kg/min (95% CI, 1.26–2.62; P=.00001) (Figure 8).

image

Figure 2.  Change in peak VO2 for exercise vs sedentary control (both groups taking β-blockers). SD indicates standard deviation; CI, confidence interval.

Download figure to PowerPoint

image

Figure 3.  Change in peak VO2 for β-blockade (BB) exercise vs placebo exercise (both groups exercise). SD indicates standard deviation, CI, confidence interval.

Download figure to PowerPoint

image

Figure 4.  Change in peak VO2 for selective vs nonselective β-blockade and exercise (both exercise). SD indicates standard deviation; CI, confidence interval.

Download figure to PowerPoint

image

Figure 5.  Mean change in Minnesota Score for exercise vs sedentary control (both groups exercising). SD indicates standard deviation; CI, confidence interval.

Download figure to PowerPoint

image

Figure 6.  Mean change in minute ventilation/carbon dioxide output (VE/VCO2) slope for exercise vs sedentary control (both groups exercising). SD indicates standard deviation; CI, confidence interval.

Download figure to PowerPoint

image

Figure 7.  Sensitivity analysis for peak oxygen consumption (VO2) (exercise vs sedentary control— both groups taking β-blockers) including only prospective, randomized, controlled studies. SD indicates standard deviation; CI, confidence interval.

Download figure to PowerPoint

image

Figure 8.  Sensitivity analysis for peak oxygen consumption (VO2) (exercise vs sedentary control— both groups taking β-blockers) including studies using only cycle ergomtery for exercise testing. SD indicates standard deviation; CI, confidence interval.

Download figure to PowerPoint

Adverse Events

Overall, there were no serious adverse events from included studies. Only one death was reported, from the study of Pozehl,14 and the patient was from a nonexercise group. Four patients did not complete their exercise training due to exacerbation of HF, claudication, acute myocardial infarction, and diagnosis of cancer (Table III).

Table III. Exercise Adherence, Withdrawals, and Adverse Events from Included Studies
StudySessions Attended, %Intermittent WithdrawalComparator WithdrawalDeath or Hospitalization
Demopoulos100Not reportedNilNot reported
ForrisierNot reportedNot reportedNot reportedNo adverse side effects
FragaNot reportedNot reportedNot reportedNot reported
Levinger1003 of 8 not performed some exercises3 of 7No death or hospitalization
Nishi739 of 24 not complete the studyNilNo death. 2 patients not completed due to exacerbation of heart failure (1) and claudication (1)
Pietila815 of 31 not attend sessionsUnclear (1 patient had non–insulin-dependent diabetes)No death. 1 person acute myocardial infarction
Pozehl99.01 of 16 unable to exercise1 of 71 patient (control) died of myocardial infarction, 1 patient (exercise group) diagnosed with cancer
Curnier100Not reportedNot reportedNo adverse event

Study Quality

We examined several aspects of study quality of the included studies. Four studies were stratified based on New York Heart Association (NYHA) class II and III. Median PEDro score was 5, and the scores ranged between 4 (two studies) and 7 (1 study). Eligibility criteria were specified although these were not used to calculate the score (Table IV). A funnel (Egger) plot of the analysis showed moderate evidence of publication bias with 3 of 6 included studies falling outside of the funnel.

Table IV. Study Quality of Included Studies
StudyAllocation Concealment (Drug)Blinding (Drug)Incomplete Outcome DataRandomization MethodSelective Outcome ReportingScore /5PeDro/10
  1. Abbreviations: BB, β-blocker; con, control group; ex, exercise group; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; PeDro, Physiotherapy Evidence Database; VO2, peak oxygen consumption.

DemopoulosUnclearUnblindedUnclear bias on data result on ex with two groups of BB (missing data)UnclearNo14
ForrisierUnclearUnblindedNot reportedNo randomization Stratified based on NYHA class, history of heart failure, LVEF <35%, ex 3.No14
FragaUnclearUnblindedNot reportedRandomization ex 1: con 1, Stratified based on NYHA class, LVEF <35%, peak VO2 <20 mL/kg/minNo25
LevingerUnclearUnblindedIntention to treatRandomization ex 1: con 1, stratified based on LVEF=34.7±7.2 and exercise capacityNo36
NishiUnclearUnblindedIntention to treatRandomization ex 1: con 1, stratified based on LVEF <25%Yes26
PietilaPlacebo-controlledSingle blindIntention to treatRandomization ex: con 1, stratified to NYHA classNo57
PozehlUnclearUnblindedIntention to treatRandomization ex 1: con 1, method repeated measures designNo36
CurnierUnclearUnblindedNot reportedRandomization ex 1: con 1, stratified to NYHA classNo25

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

Our analysis is the first to pool data from several small prospective randomized or controlled trials that have examined whether β-blockers attenuate the effects of exercise training in HF patients. Our results suggest that β-blockade compared with placebo actually enhances improvements in cardiorespiratory fitness via exercise training intervention; moreover, neither selective nor nonselective β-blockade appeared to attenuate exercise training–induced adaptations for HF patients. Unanimous improvements in all outcome measures suggest that β-blockade, unless contraindicated by comorbid medical conditions, facilitates exercise training–induced adaptations and physicians and patients alike should not be concerned by unsubstantiated suggestions that β-blockade worsens depression and exercise capacity.

Our primary outcome measures were change in peak VO2 and baseline mean was 18.2 mL/kg/min, a net increase of 1.27 mL/kg/min is equivalent to 7%, which is larger than the 4% observed in the Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) study, the largest randomized controlled trial of exercise training in HF participants to date.21 In the HF-ACTION study, use of β-blockade was a baseline covariate predictor of survival, which is perhaps not surprising considering the link between peak VO2 and mortality in HF patients.22 The study by Sarullo and colleagues also highlights lower VE/VCO2 slope to be a covariate predictor of survival.22 Our analyses also suggested that exercising patients taking β-blockade exhibited significantly lower (better) VE/VCO2 slopes. Our work also showed that there were no differences between classifications of β-blockers on the magnitude of exercise training adaptations, despite previous reports that β-adrenoreceptor–blocking agents with intrinsic sympathomimetic activity (ISA) have lesser effects on exercise capacity.23 The physiological rationale is that β-blockers with ISA have smaller reductions in heart rate compared with β-blockers without ISA.8 Other effects of ISA are reduced skeletal and cardiac cardiotoxicity and inhibited myocardial apoptosis.24 It seems that the absence of intrinsic sympathomimetic activity is the vital component for optimal results.

Part of the justification for this work was to prevent withdrawal from β-blocker treatment, especially in older patients with congestive HF, since it has been reported to be twice that previously reported in younger patients.12 As with our analyses, the work of Baxter and colleagues found that there was no evidence of a negative impact on symptoms or exercise capacity in patients who tolerated bisoprolol.12 Another reason for poor uptake of, and compliance with, β-blockers is fear of depression. Our analyses showed that quality of life (depression data were not reported in included studies) was significantly better in patients exercising and taking β-blockers, compared with sedentary patients taking β-blockers. Depression should not be the reason for reluctance in prescribing β-blockers to cardiovascular patients.25

Limitations

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

The limitations of this work were the small sample sizes of the included studies. This is perhaps explained by the ethical difficulties posed by withholding medication that is the cornerstone of HF therapy.26 Some analyses were not warranted, that may later become possible as further studies are published, because an insufficient volume of data exists. Mean baseline peak VO2 for this analysis is higher than that reported in previous meta-analyses.4 A possible explanation is that included studies have a mean age younger than 60 years, although two previous meta-analyses3,4 both found mean age of HF exercise training study participants to be 58 to 59 years, despite the mean age of patients with HF in the wider community being higher.27

Conclusions

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References

Our data suggest that irrespective of β-blocker classification, β-blockers optimize exercise training adaptations in HF patients. Physicians should therefore not withhold β-blocker therapy for patients with HF because of fear of reduced quality of life or depressed exercise capacity. Moreover, patients should not cease taking β-blocker medication over concern about limiting their quality of life or exercise capacity.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Limitations
  7. Conclusions
  8. References
  • 1
    Gordon NF, Duncan JJ. Effect of beta-blockers on exercise physiology: implications for exercise training. Med Sci Sports Exerc.1991;23:668676.
  • 2
    Packer M, Coats AJ, Fowler MB, et al.Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med.2001;344:16511658.
  • 3
    Piepoli MF, Davos C, Francis DP, et al.Exercise training meta-analysis of trials in patients with chronic heart failure (ExTraMATCH). BMJ.2004;328:189.
  • 4
    Smart N, Marwick TH. Exercise training for patients with heart failure: a systematic review of factors that improve mortality and morbidity. Am J Med.2004;116:693706.
  • 5
    Haykowsky MJ, Liang Y, Pechter D, et al.A meta-analysis of the effect of exercise training on left ventricular remodeling in heart failure patients: the benefit depends on the type of training performed. J Am Coll Cardiol.2007;49:23292336.
  • 6
    Nishi I, Noguchi T, Iwanaga Y, et al.Effects of exercise training in patients with chronic heart failure and advanced left ventricular systolic dysfunction receiving beta-blockers. Circ J.2011;75:16491655.
  • 7
    Zuurmond WW, Gootjes JR, Tol-Verhagen CV, et al.A simple standardized method for VAS measurements in terminally ill patients. J Palliat Med.2005;8:490491.
  • 8
    Wilmore JH, Ewy GA, Freund BJ, et al.Cardiorespiratory alterations consequent to endurance exercise training during chronic beta-adrenergic blockade with atenolol and propranolol. Am J Cardiol.1985;55:142D148D.
  • 9
    Demopoulos L, Yeh M, Gentilucci M, et al.Nonselective beta-adrenergic blockade with carvedilol does not hinder the benefits of exercise training in patients with congestive heart failure. Circulation. 1997;95:17641767.
  • 10
    Forissier JF, Vernochet P, Bertrand P, et al.Influence of carvedilol on the benefits of physical training in patients with moderate chronic heart failure. Eur J Heart Fail.2001;3:335342.
  • 11
    Lama PJ. Systemic adverse effects of beta-adrenergic blockers: an evidence-based assessment. Am J Ophthalmol.2002;134:749760.
  • 12
    Baxter AJ, Spensley A, Hildreth A, et al.Beta blockers in older persons with heart failure: tolerability and impact on quality of life. Heart.2002;88:611614.
  • 13
    Fraga R, Franco FG, Roveda F, et al.Exercise training reduces sympathetic nerve activity in heart failure patients treated with carvedilol. Eur J Heart Fail.2007;9:630636.
  • 14
    Pozehl B, Duncan K, Krueger S, et al.Adjunctive effects of exercise training in heart failure patients receiving maximum pharmacologic therapy. Prog Cardiovasc Nurs.2003;18:177183.
  • 15
    Pietila M, Malminiemi K, Huupponen R, et al.Celiprolol augments the effect of physical exercise on insulin sensitivity and serum lipid levels in chronic heart failure. Eur J Heart Fail. 2000;2:8190.
  • 16
    Jadad AR, Moore RA, Carroll D, et al.Assessing the quality of reports of randomized clinical trials: is blinding necessary?Control Clin Trials.1996;17:112.
  • 17
    Moseley AM, Herbert RD, Sherrington C, et al.Evidence for physiotherapy practice: a survey of the Physiotherapy Evidence Database (PEDro). Aust J Physiother.2002;48:4349.
  • 18
    Egger M, Davey Smith G, Schneider M, et al.Bias in meta-analysis detected by a simple, graphical test. BMJ1997;315:629634.
  • 19
    Curnier D, Galinier M, Pathak A, et al.Rehabilitation of patients with congestive heart failure with or without beta-blockade therapy. J Card Fail.2001;7:241248.
  • 20
    Levinger I, Bronks R, Cody DV, et al.Resistance training for chronic heart failure patients on beta blocker medications. Int J Cardiol.2005;102:493499.
  • 21
    O’Connor CM, Whellan DJ, Lee KL, et al.Efficacy and safety of exercise training in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA.2009;301:14391450.
  • 22
    Sarullo FM, Fazio G, Brusca I, et al.Cardiopulmonary exercise testing in patients with chronic heart failure: prognostic comparison from peak VO2 and VE/VCO2 slope. Open Cardiovasc Med J.2010;4:127134.
  • 23
    Ades PA. Cardiac effects of beta-adrenoceptor blockade with intrinsic sympathomimetic activity during submaximal exercise. Br J Clin Pharmacol.1987;24(suppl 1):29S33S.
  • 24
    Cruickshank JM. Are we misunderstanding beta-blockers. Int J Cardiol.2007;120:1027.
  • 25
    Verbeek DE, van Riezen J, de Boer RA, et al.A review on the putative association between beta-blockers and depression. Heart Fail Clin.2011;7:8999.
  • 26
    Hunt SA, Baker DW, Chin MH, et al.ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to revise the 1995 Guidelines for the Evaluation and Management of Heart Failure). J Am Coll Cardiol.2001;38:21012113.
  • 27
    Smart N. Exercise training for heart failure patients with and without systolic dysfunction: an evidence-based analysis of how patients benefit. Cardiol Res Pract.2010 Sep 30;2011:pii: 837238.