Individual patient meta-analysis of exercise training effects on systemic brain natriuretic peptide expression in heart failure

Abstract

Background: Brain natriuretic peptide (BNP) predicts exercise performance and exercise training may modulate BNP and its N-terminal portion (NT-pro-BNP), we therefore conducted an individual patient analysis of exercise training effects on BNP and NT-pro-BNP. Aims: To use an individual patient meta-analysis to relate changes in BNP, NT-pro-BNP, and peak VO₂; to link these changes to volume parameters of exercise training programmes (intensity etc.); and to identify patient characteristics likely to lead to greater improvements in BNP, NT-pro-BNP, and peak VO₂. Design: Individual patient meta-analysis. Methods: A systematic search was conducted of Medline (Ovid), Embase.com, Cochrane Central Register of Controlled Trials, and CINAHL (until July 2008) to identify randomized controlled trials of aerobic and/or resistance exercise training in systolic heart failure patients measuring BNP and/or NT-pro-BNP. Primary outcome measures were change in BNP, NT-pro-BNP, and peak VO2. Subanalyses were conducted to identify (1) patient groups that benefit most and (2) exercise programme parameters enhancing favourable changes in primary outcome measures. Results: Ten randomized controlled studies measuring BNP or NT-pro-BNP met eligibility criteria, authors provided individual patient data for 565 patients (313 exercise and 252 controls). Exercise training had favourable effects on BNP (−28.3%, p < 0.0001), NT-pro-BNP (−37.4%, p = < 0.0001), and peak VO₂ (17.8%, p < 0.0001). The analysis showed a significant change in primary outcome measures; moreover, change in BNP (r = −0.31, p < 0.0001) and NT-pro-BNP (r = −0.22, p < 0.0001) were correlated with peak VO₂ change. Conclusion: Exercise training has favourable effects on BNP, NT-pro-BNP, and peak VO₂ in heart failure patients and BNP/NT-pro-BNP changes were correlated with peak VO₂ changes.

Keywords

Exercise training heart failure meta-analysis natriuretic peptides

Introduction

Exercise therapy provides many benefits to people with chronic heart failure (CHF), while some information exists on the effects of exercise training on BNP levels in heart failure patients,¹ evidence may be equivocal due to the magnitude of the standard deviation measured for both brain natriuretic peptide (BNP) and the N-terminal portion (NT-pro-BNP).²^–⁴ Moreover a recent systematic review has shown both BNP and NT-pro-BNP to be modulated after exercise training although this conclusion was based on post-training data only and assumptions that participant groups were matched at baseline.⁵ We therefore sought to establish, via an individual patient meta-analysis whether exercise training is able to modulate systemic BNP, NT-pro-BNP and functional capacity levels in heart failure patients. We also sought to examine the relationship between changes in BNP/NT-pro-BNP and peak VO₂, patient characteristics and the volume parameters of exercise training regimes.

Methods

Following the acceptance for publication of a systematic review evaluating the effects of exercise training on BNP and NT-pro-BNP,⁵ a collaborative group was coordinated from Bond University, Australia. Prospective data collection was agreed and a common dataset of collected variables was sent to the included study authors.

Search strategy

Potential studies were identified by a systematic review librarian (ST). The search was conducted of Medline (Ovid) (1950−January 2010), Embase.com (1974−January 2010), Cochrane Central Register of Controlled Trials and CINAHL (1981−January 2010). The search strategy included a mix of MeSH and free text terms for the key concepts heart failure, exercise training, and brain natriuretic peptide and these were combined with a sensitive search strategy to identify randomized controlled trials. Reference lists of papers found were scrutinized for new references. All identified papers were assessed independently by two reviewers.

Inclusions

Randomized, controlled trials of exercise training in heart failure patients, for a minimum of 12 weeks, with a primary or secondary outcome measure of BNP or NT-pro-BNP were included.

Exclusions

Animal studies, review papers and non-randomized controlled trials were excluded. Studies that did not have desired outcome measures (BNP or NT-pro-BNP) or participants who were non-heart failure or healthy in either treatment or control groups were also excluded. Authors were contacted to provide missing data or to clarify if data was duplicated in multiple publications by the same author or research group, and incomplete data or data from an already included study resulted in exclusion. Studies of physical therapy other than aerobic or strength exercise and acute exercise rather than training response studies were excluded.

Studies included in the review

We examined conference proceedings and the latest editions of relevant journals not yet available on electronic databases. Initial screening identified 126 potential reports; 94 were not randomized trials of exercise training in heart failure patients, four were animal studies, and nine were drug studies, leaving 19 full manuscripts for review. Two groups had reported information from the same trial in more than one publication, one study was excluded because only some patients had congestive heart failure. Five studies had used other types of physical therapy, e.g. electrical stimulation that were considered heterogeneous compared to exercise training. One study was excluded because of selective (incomplete) outcome reporting, leaving ten eligible trials. Two studies had conducted more than one similar trial. Individual patients were therefore combined into one dataset for these trials, yielding eight non-overlapping datasets meeting the eligibility criteria.

Selection and validity assessment

All principal investigators provided anonymized individual patient data for originally randomized patients. Datasets included age, gender, medication [beta-blocker, angiotensin-converting enzyme (ACE) inhibitor, angiotensin receptor blocker (ARB), spironolactone, diuretic] brand and dosage, ischaemic vs. non-ischaemic causes, New York Heart Association (NYHA) functional class, left ventricular ejection fraction, peak oxygen consumption, and exercise adherence.

Included datasets were randomized parallel group controlled trials assessing effects of exercise training on BNP and/or NT-Pro-BNP levels. Patients with left ventricular ejection fraction >50% or peak VO₂ >24 ml O₂/kg/min were removed from the database as these patients were deemed atypical of CHF patients. Datasets were checked for completeness and consistency with original publications. Queries were resolved by communication with principal investigators.

Data synthesis

Data relating to BNP and NT-pro-BNP, heart failure patient characteristics, and exercise training protocols were reviewed. Primary outcome measures following exercise training were post-exercise change in BNP, NT-pro-BNP, and peak VO₂. Secondary outcome was exercise programme energy expenditure. Measures of study quality were also recorded such as interstudy variation in age, gender, functional status e.g. NYHA score, left ventricular ejection fraction (LVEF), and cardiac medications.

Subgroup meta-analysis of dichotomous end points

The effect of exercise training was also evaluated in pre-specified subgroups: ischaemic vs. non-ischaemic causes, gender, beta-blocker use, age (<60 and ≥60 years), body mass index (BMI) (<27 vs. ≥27), peak oxygen uptake (<14 ml/kg/min vs. ≥14 ml/kg/min), LVEF (<34% vs. ≥34%). The continuous variables were dichotomized. The BMI and LVEF criteria were selected because previous randomized trials and meta-analyses of heart failure exercise training studies have shown these to be mean values.⁶⁻⁸ Subanalyses were also conducted according to change in peak VO₂ as outlined by Balady et al.⁹ So three categories were used, those who showed more than a + 5% change (positive responders) in peak VO₂, neutral responders < ± 5%, and negative responders showed more than a 5% fall in peak VO₂.

Exercise programme parameters

Where possible we calculated, using established methods,¹⁰ work completed during exercise training (kcal) and mean weekly calorie expenditure, in order to determine a dose−response relationship with change in BNP or NT-pro-BNP, disease severity, and clinical characteristics of CHF participants.

Statistical analysis

SPSS version 17.0 was used to conduct meta-analyses for BNP, NT-pro-BNP, and peak VO₂ following exercise training. As BNP and NT-pro-BNP data were not normally distributed, BNP and NT-pro-BNP were reported as median and interquartile range, all remaining data were reported as mean and standard deviation. We assessed inter- and intragroup differences in BNP and NT-pro-BNP using non-parametric statistics (Wilcoxon signed ranks). We analysed differences between baseline and post-intervention using repeated-measures ANOVA. If there was significant interaction from the ANOVA, paired t-tests calculated intragroup significance. Appropriate correlation coefficients were calculated for change (normally distributed for BNP and NT-pro-BNP) in primary outcome measures. We used a 5% level of significance and a 95% confidence interval to report outcome measure changes.

Results

Study characteristics

Ten randomized, controlled studies (eight discrete datasets, overlapping data was not duplicated) met our eligibility criteria, with an aggregate number of 565 subjects (313 exercise participants and 252 controls)³⁻^4,11⁻¹⁸ were included. Five studies measured BNP, six measured NT-pro-BNP, and one study measured NT-pro-BNP and BNP. BNP data from 230 patients, NT-pro-BNP data in 466 patients, and both BNP and NT-pro-BNP in 133 patients were available.

Participant characteristics

Exercise and control group participants were matched at baseline, except in one study.¹⁴ There were 451 male participants (80%) and all participants were NYHA class I−III. Participants were treated according to medication guidelines (see Table 1).

Table 1.

Baseline clinical, demographic, and pharmacological characteristics of exercise and control group patients

Characteristic	Exercise (n = 313)	Control (n = 252)	p
Male	256 (82)	195 (77)	0.20
Age (years)	60.2±12.0	62.7 ± 10.4	0.02
New York Heart Association Classification (%)
I	3	4	0.47
II	59	55
III	38	41
Body mass (kg)	78.6 ± 12.7	77.7 ± 13.7	0.45
Body mass index (kg/m²)	27 ± 3.3	27.2 ± 3.8	0.55
LVEF	34.7 ± 8.2	34.6 ± 8.6	0.96
Peak VO₂ (ml/kg/min)	14.3 ± 3.6	14.3 ± 3.9	0.84
Ischaemic cardiomyopathy	62.4	62.1	0.96
BNP (pg/ml), median (IQR)	181 (262)	190 (331)	0.47
NT-pro-BNP (pg/ml), median (IQR)	1114 (1297)	1253 (1286)	0.11
Beta blocker use	242 (77)	194 (77)	0.97
ACE inhibitor use	150 (48)	112 (44)	0.75
ARB use	75 (24)	46 (18)	0.31
Spironolactone use	73 (23)	56 (22)	0.88
Diuretic	188 (60)	160 (64)	0.67

Values are mean ± SD or n (%) unless otherwise indicated. ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; IQR, interquartile range; LVEF, left ventricular ejection fraction.

Exercise programme parameters

All but one study utilized cycling as the exercise mode.¹⁸ Training frequency was 2–7 weekly sessions, intensity 50–95% of peak VO₂, 30–50 minutes per session, and 3–9 month duration. Energy expenditure calculations were possible in all but one study,¹¹ only the aerobic, not resistance, activity was calculated in one study.¹⁴ Mean weekly energy expenditure was 457 ± 135 kcal/week and when multiplied by completed exercise sessions, the exercise participants were estimated to have expended 430 ± 127 kcal/week or 94% of target.

Change in BNP and NT-pro BNP

Compared to the control group exercise training significantly lowered BNP (−28.3%, p < 0.0001) and NT-pro-BNP (−37.4%, p = < 0.0001) and increased peak VO₂ (17.8%, p < 0.0001). Baseline and post-intervention changes in BNP and NT-pro BNP in both group patients can be seen in Table 2.

Table 2.

Baseline and post-intervention changes in peak oxygen consumption (VO₂), brain natriuretic peptide (BNP), and N-terminal portion of BNP (NT-pro-BNP) in exercise and control group patients

	Exercise	Control
Peak VO₂ (ml/kg/min (mean ± SD)
Baseline	14.0 ± 3.5	14.1 ± 3.9
Post	16.5 ± 4.1	14.3 ± 3.8
p-value	<0.0001	0.02
% change	17.8	1.4
BNP (pg/ml), median (IQR)
Baseline	181 (262)	190 (331)
Post	100 (181)	162 (313)
p-value	<0.0001	0.77
% change	−28.3	−2
NT-pro-BNP (pg/ml), median (IQR)
Baseline	1114 (1297)	1253 (1286)
Post	593 (876)	1117 (1352)
p-value	<0.0001	0.06
% change	−37.4	−7.4

IQR, interquartile range.

Sensitivity analyses

We conducted sensitivity analyses, removing studies where we were unable to calculate exercise energy expenditure. After sensitivity analysis (removal of the two studies) significant changes in both BNP and peak VO₂ remained; however, the increment of change was unaltered.

Correlates of baseline primary outcome measures in all patients

Baseline measures of peak VO₂ and BNP (r = −0.33, p < 0.001) and NT-pro-BNP (r = −0.32, p < 0.001) were both inversely correlated.

Correlates of change in primary outcome measures in the exercise training group

Correlation coefficients showed inverse relationships between change in peak VO₂ and change in BNP (r = −0.31, p < 0.0001) and NT-pro-BNP (r = −0.21, p = 0.001): see Figures 1 and 2. Change in BNP was also related to exercise energy expenditure (r = −0.19, p = 0.04). Body mass index was correlated with both change in BNP (r = 0.26, p = 0.003) and NT-pro-BNP (r = 0.16, p = 0.01). Other correlation coefficients can be seen in Table 3.

Figure 1.

Correlation of change in peak oxygen consumption (VO₂) and N-terminal portion of brain natriuretic peptide (NT-pro-BNP).

Figure 2.

Correlation of change in peak oxygen consumption (VO₂) and brain natriuretic peptide (BNP).

Table 3.

Correlation between change in primary outcome measures and other markers of prognosis in patients undertaking exercise training

	Change in BNP (pg/ml)		Change in NT-pro-BNP (pg/ml)		Change in peak VO₂ (ml/kg/min)
	r	p	r	p	r	p
Change in peak VO₂ (ml/kg/min)	−0.31	<0.0001	−0.21	<0.0001	N/A	N/A
Exercise energy expenditure (kcal/week)	−0.19	0.04	−0.006	0.93	0.45	<0.0001
Exercise intensity (% maximum)	−0.12	0.19	0.06	0.33	0.32	<0.0001
Exercise frequency (sessions per week)	0.12	0.17	0.08	0.19	0.28	<0.0001
Body mass index (kg/m²)	0.26	0.003	0.16	0.01	−0.02	0.81

BNP, brain natriuretic peptide; NT-pro BNP, N-terminal portion of BNP; VO₂, oxygen consumption.

Training response

The exercise training group (n = 284) was further divided into 227 patients (80%) who exhibited a positive training response (5% or greater increase in peak VO₂), 37 patients (13%) with ‘neutral’ response, and 20 patients (7%) showing a negative response. Those patients with a positive training response showed a greater reduction in both BNP and NT-pro-BNP compared to both neutral and negative responders (all p < 0.003).

Subanalyses

Subanalyses were conducted for age, gender, beta-blocker use, heart failure aetiology, baseline peak VO₂, left ventricular ejection fraction, body mass index, and mean weekly calorie expenditure (see Table 4). Those with LVEF <34% were likely to show greater improvements in BNP (p = 0.003) and NT-pro-BNP (p = 0.002). Those aged <60 years were likely to attend more exercise sessions (93 vs. 88%, p = 0.03). There were no differences between gender and use of beta-blockers in any primary outcome measure. Those with BMI <27 kg/m² showed greater improvements in BNP (p = 0.04). The mean weekly exercise energy expenditure of 457 kcal per week was sufficient to elicit significant changes in peak VO₂ (p < 0.0001) and BNP (p = 0.02).

Table 4.

Subanalysis of predictors of change prognosis in patients undertaking exercise training

Characteristic	Predictor		p
Age	<60 years	≥60 years
Change in BNP (pg/ml)	−55 ± 107 (n = 46)	−70 ± 144 (n = 92)	0.54
Change in NT-pro-BNP (pg/ml)	−520 ± 557 (n = 142)	−494 ± 799 (n = 117)	0.77
Change in peak VO₂ (ml/kg/min)	2.76 ± 2.44 (n = 142)	2.28 ± 2.25 (n = 143)	0.08
Exercise adherence (% sessions)	93 ± 16 (n = 149)	88 ± 16 (n = 137)	0.03
Left ventricular ejection fraction	<34%	≥34%
Change in BNP (pg/ml)	∼−100 ± 140 (n = 107)	−34 ± 119 (n = 178)	0.003
Change in NT-pro-BNP (pg/ml)	−652 ± 818 (n = 65)	−440 ± 574 (n = 73)	0.002
Change in peak VO₂ (ml/kg/min)	2.52 ± 2.3 (n = 92)	2.52 ± 2.4 (n = 162)	0.99
Exercise adherence (% sessions)	89 ± 14 (n = 109)	92 ± 17 (n = 172)	0.13
Cardiomyopathy	Ischaemic	Non-ischaemic
Change in BNP (pg/ml)	−54 ± 129 (n = 169)	−85 ± 139 (n = 98)	0.19
Change in NT-pro-BNP (pg/ml)	−543 ± 712 (n = 88)	−490 ± 616 (n = 50)	0.55
Change in peak VO₂ (ml/kg/min)	2.46 ± 2.37 (n = 94)	2.36 ± 2.26 (n = 148)	0.74
Exercise adherence (% sessions)	90 ± 17 (n = 105)	90 ± 16 (n = 163)	0.98
Peak VO₂	>14 ml/kg/min	<14 ml/kg/min
Change in BNP (pg/ml)	−44 ± 102 (n = 50)	−89 ± 143 (n = 78)	0.06
Change in NT-pro-BNP (pg/ml)	−440 ± 558 (n = 136)	−584 ± 781 (n = 123)	0.09
Change in peak VO₂ (ml/kg/min)	2.37 ± 2.57 (n = 133)	2.65 ± 2.15 (n = 152)	0.31
Exercise adherence (% sessions)	92 ± 17 (n = 145)	89 ± 15 (n = 141)	0.21
Body mass index	<27 kg/m²	≥27 kg/m²
Change in BNP (pg/ml)	−91 ± 148 (n = 77)	−42 ± 89 (n = 51)	0.04
Change in NT-pro-BNP (pg/ml)	−579 ± 780 (n = 117)	−470 ± 559 (n = 125)	0.21
Change in peak VO₂ (ml/kg/min)	2.53 ± 2.37 (n = 137)	2.3 ± 2.28 (n = 130)	0.42
Exercise adherence (% sessions)	88 ± 17 (n = 138)	92 ± 15 (n = 130)	0.02
Energy intake	<457 kcal/week	≥457 kcal/week
Change in BNP (pg/ml)	NA	NA
Change in NT-pro-BNP (pg/ml)	−429 ± 657 (n = 230)	−619 ± 527 (n = 74)	0.02
Change in peak VO₂ (ml/kg/min)	1.8 ± 2.2 (n = 282)	4.3 ± 2.2 (n = 58)	<0.001

Values are mean ± SD. BNP, brain natriuretic peptide; NT-pro BNP, N-terminal portion of BNP; VO₂, oxygen consumption.

Discussion

Ten published studies, with an aggregate of 565 subjects allowed an individual patient meta-analysis of the effect of exercise training on BNP and NT-pro-BNP in CHF patients. At baseline BNP and NT-pro-BNP values were both predictors of peak VO₂. Analyses of BNP and NT-pro-BNP demonstrated overall reductions and an improvement in peak VO₂ following exercise training. Statistical significance of these changes remained after conducting sensitivity analyses.

Participants were generally well matched at baseline for clinical, demographic and pharmacological characteristics. Previous data from exercise training studies in heart failure patients suggest the majority are male.^6,8 One study used NYHA class I, as well as class II and III patients. Mean LVEF was 34%.

Both BNP and NT-pro-BNP showed an inverse relationship with peak VO₂ at baseline. Baseline BNP and NT-pro-BNP can be considered additional criteria for selecting exercise training candidates and appear surrogates of peak VO₂ in evaluating exercise training response and prognosis in heart failure. Recent work has shown NT-pro-BNP to predict peak VO₂ in heart failure patients.¹ Change in BNP was 28% vs. 2% in the exercise vs. control groups, respectively. Similarly the change in NT-pro-BNP was 37 vs. 7% in the exercise vs. control groups, respectively. Both groups significantly improved their functional capacity (peak VO₂) from baseline, although the magnitude of change was only 1.4% in the control group compared to 18% in the exercise group. One could speculate that the improvement in functional capacity in the control group could be attributed to habituation to the exercise test, degree of effort spent,¹⁹ or effects of optimal medication. A net improvement of 16.6% in peak VO₂ is consistent with previous reports.⁸ Our data suggested that beta-blockade did not attenuate exercise training effects on BNP, NT-pro-BNP and functional capacity. Rosenberg reported that 6 months of ACE inhibition (enalapril), but not beta-blockade (carvedilol), lowers BNP levels in heart failure patients,²⁰ while other work reported that metoprolol causes plasma BNP/NT-pro-BNP to rise.²¹

Change in both BNP and NT-pro-BNP were correlated with change in peak VO₂ and BNP change was correlated with mean weekly exercise energy expenditure. Data from the heart failure action study (HF-ACTION) supports the notion that NT-pro-BNP is a strong predictor of peak VO₂.¹ The two exercise training programmes yielding the lowest mean weekly energy expenditures did not show significant changes in BNP,^11,14 while our analyses (Table 4) suggest about 460 kcal/week may be required to change BNP. The work by Wisloff et al.¹⁸ reported that high intensity intermittent exercise was effective in significantly reducing NT-pro-BNP while a comparative group performing continuous moderate intensity exercise of isocaloric energy expenditure did not alter NT-pro-BNP levels. We calculated that mean weekly exercise energy expenditure of the participants of Wisloff’s¹⁸ study was 505 kcal/week. It may be that the very small sample size (n = 9 for each group) and the relatively short study duration (12 weeks) may have precluded NT-pro-BNP changes in the moderate exercise intensity group. Our data did not suggest that 12 weeks was as effective as 39 weeks exercise training programme in eliciting BNP and NT-pro-BNP changes.

Previous work has suggested that reductions in BNP following exercise training may be due to blunted neurohormonal responses at rest.³ It has been shown that exercise training ameliorates sympatho-vagal balance^22,23 and reduces sympathetic outflow, plasma catecholamines, angiotensin II, vasopressin, and brain natriuretic peptides at rest.²⁴ Patients with ejection fraction <34% showed a greater improvement in both BNP and NT-pro-BNP, this is unsurprising as natriuretic peptides are released in response to myocardial stretch and those with lower ejection fraction have greatest improvement potential. Body mass index showed a positive correlation with change in both BNP and NT-pro-BNP, therefore lower body mass may be related to greater lowering of natriuretic peptides. Future short-term research might aim at linking changes in BNP and NT-pro-BNP with improved myocardial function, body composition, and blunted neurohormonal responses at rest. Longer studies, with larger sample size, may be able to demonstrate a link between exercise-induced BNP/NT-pro-BNP decrease and reduced morbidity, healthcare costs, and mortality in CHF patients.

Conclusions

Data from published studies, suggest exercise training has a favourable effect on BNP and NT-pro-BNP in heart failure patients; moreover, the changes in natriuretic peptides were related to functional capacity change. Baseline BNP and NT-pro-BNP can be considered additional criteria for selecting exercise training candidates and appear surrogates of peak VO₂ in evaluating exercise training response and prognosis in heart failure. Exercise intensity and weekly minimum exercise energy expenditure around 460 kcal may determine the magnitude of change in BNP and the N-terminal portion of BNP in heart failure patients.

Footnotes

Funding

Ulrik Wisloff received funding from the KG Jebsen Foundation and Norwegian Research Council granting bodies. Jon Butterworth received equipment from ALERE, Australia.

References

Felker

Whellan

Kraus

Clare

Zannad

Donahue

. N-terminal pro-brain natriuretic peptide and exercise capacity in chronic heart failure: data from the Heart Failure and a Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) study. Am Heart J 2009; 158(4 Suppl): S37–S44.

Berent

von Duvillard

Crouse

Auer

Green

Sinzinger

. Short-term residential cardiac rehabilitation reduces B-type natriuretic peptide. Eur J Cardiovasc Prev Rehabil 2009; 16(5): 603–608.

Passino

Severino

Poletti

Piepoli

Mammini

Clerico

. Aerobic training decreases B-type natriuretic peptide expression and adrenergic activation in patients with heart failure. J Am Coll Cardiol 2006; 47(9): 1835–1839.

Sarullo

Gristina

Brusca

Milia

Raimondi

Sajeva

. Effect of physical training on exercise capacity, gas exchange and N-terminal pro-brain natriuretic peptide levels in patients with chronic heart failure. Eur J Cardiovasc Prev Rehabil 2006; 13(5): 812–817.

Smart

Steele

. Systematic review of the effect of aerobic and resistance exercise training on systemic brain natriuretic peptide (BNP) and N-terminal BNP expression in heart failure patients. Int J Cardiol 2010; 140(3): 260–265.

Piepoli

Davos

Francis

Coats

. Exercise training meta-analysis of trials in patients with chronic heart failure (ExTraMATCH). Br Med J 2004; 328(7433): 189–189.

Smart

Haluska

Jeffriess

Marwick

. Predictors of a sustained response to exercise training in patients with chronic heart failure: a telemonitoring study. Am Heart J 2005; 150(6): 1240–1247.

Smart

Marwick

. Exercise training for patients with heart failure: a systematic review of factors that improve mortality and morbidity. Am J Med 2004; 116(10): 693–706.

Balady

Arena

Sietsema

Myers

Coke

Fletcher

. Clinician’s Guide to cardiopulmonary exercise testing in adults: a scientific statement from the American Heart Association. Circulation 122(2): 191–225.

10.

Astrand

Rodahl

(eds) Textbook of work physiology. Sydney: McGraw-Hill, 1986.

11.

Butterfield

Faddy

Davidson

Ridge

. Exercise training in patients with stable chronic heart failure: effects on thoracic impedance cardiography and B-type natriuretic peptide. J Cardiopulm Rehabil Prev 2008; 28(1): 33–37.

12.

Giallauria

Cirillo

Lucci

Pacileo

De Lorenzo

D’Agostino

. Left ventricular remodelling in patients with moderate systolic dysfunction after myocardial infarction: favourable effects of exercise training and predictive role of N-terminal pro-brain natriuretic peptide. Eur J Cardiovasc Prev Rehabil 2008; 15(1): 113–118.

13.

Giallauria

De Lorenzo

Pilerci

Manakos

Lucci

Psaroudaki

. Reduction of N terminal-pro-brain (B-type) natriuretic peptide levels with exercise-based cardiac rehabilitation in patients with left ventricular dysfunction after myocardial infarction. Eur J Cardiovasc Prev Rehabil 2006; 13(4): 625–632.

14.

Jonsdottir

Andersen

Sigurosson

Sigurdsson

. The effect of physical training in chronic heart failure. Eur J Heart Fail 2006; 8(1): 97–101.

15.

Malfatto

Branzi

Osculati

Valli

Cuoccio

Ciambellotti

. Improvement in left ventricular diastolic stiffness induced by physical training in patients with dilated cardiomyopathy. J Card Fail 2009; 15(4): 327–333.

16.

Meyer

Schwaab

Gorge

Scharhag

Herrmann

Kindermann

. Can serum NT-proBNP detect changes of functional capacity in patients with chronic heart failure? Z Kardiol 2004; 93(7): 540–545.

17.

Passino

Severino

Gabutti

Prontera

Clerico

. C-type natriuretic peptide expression in patients with chronic heart failure: effects of aerobic training. Eur J Cardiovasc Prev Rehabil 2008; 15(2): 168–172.

18.

Wisloff

Stoylen

Loennechen

Bruvold

Rognmo

Haram

. Superior cardiovascular effect of aerobic interval training versus moderate continuous training in heart failure patients: a randomized study. Circulation 2007; 115(24): 3086–3094.

19.

Meyer

Scharhag

Kindermann

. Peak oxygen uptake. Myth and truth about an internationally accepted reference value. Z Kardiol 2005; 94: 255–265.

20.

Rosenberg

Gustafsson

Remme

Riegger

Hildebrandt

. Effect of beta-blockade and ACE inhibition on B-type natriuretic peptides in stable patients with systolic heart failure. Cardiovasc Drugs Ther 2008; 22(4): 305–311.

21.

Davis

Richards

Nicholls

Yandle

Frampton

Troughton

. Introduction of metoprolol increases plasma B-type cardiac natriuretic peptides in mild, stable heart failure. Circulation 2006; 113(7): 977–985.

22.

Dimopoulos

Anastasiou-Nana

Sakellariou

Drakos

Kapsimalakou

Maroulidis

. Effects of exercise rehabilitation program on heart rate recovery in patients with chronic heart failure. Eur J Cardiovasc Prev Rehabil 2006; 13(1): 67–73.

23.

Giallauria

De Lorenzo

Pilerci

Manakos

Lucci

Psaroudaki

. Long-term effects of cardiac rehabilitation on end-exercise heart rate recovery after myocardial infarction. Eur J Cardiovasc Prev Rehabil 2006; 13(4): 544–550.

24.

Gademan

Swenne

Verwey

van der Laarse

Maan

van de Vooren

. Effect of exercise training on autonomic derangement and neurohumoral activation in chronic heart failure. J Card Fail 2007; 13(4): 294–303.