International Journal of Cardiology 179 (2015) 269–274

Contents lists available at ScienceDirect

International Journal of Cardiology

journal homepage: www.elsevier.com/locate/ijcard

Cardiovascular effects of high-intensity interval aerobic training

combined with strength exercise in patients with chronic heart failure. A
randomized phase III clinical trial
Christina Chrysohoou a,⁎, Athanasios Angelis a, George Tsitsinakis a, Stavroula Spetsioti b, Ioannis Nasis b,
Dimitris Tsiachris a, Panagiotis Rapakoulias c, Christos Pitsavos a, Nikolaos G. Koulouris c,
Ioannis Vogiatzis b,c, Dimitris Tousoulis a
a
First Cardiology Clinic, Hippokration Hospital, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece
b
Physical Education and Sports Science, National and Kapodistrian University of Athens, Athens, Greece
c
First Department of Respiratory Medicine, Pulmonary Rehabilitation Unit, Sotiria Hospital, National and Kapodistrian University of Athens, Greece

a r t i c l e i n f o a b s t r a c t

Article history: Background: The aim of this work was to evaluate the effect of high-intensity interval exercise (i.e., 30 s at 100% of
Received 4 October 2014 max workload, followed by 30 s at rest, 45 min 3 days/week working-out schedule for 12 weeks) on left ventric-
Accepted 5 November 2014 ular function and aortic elastic properties among chronic heart failure (CHF) patients.
Available online 6 November 2014 Methods: This study is a phase III clinical trial. Of the 100 consecutive CHF patients (NYHA classes II–IV, ejection
fraction b 50%) that were randomly allocated, 72 completed the study (exercise training group, n = 33, 63 ±
Keywords:
9 years, 88% men, and control group, n = 39, 56 ± 11 years, 82% men). All patients underwent cardiopulmonary
Rehabilitation
Heart failure
stress test, non-invasive high-fidelity tonometry of the radial artery, pulse wave velocity measurement using a
Aortic elastic properties SphygmoCor device and echocardiography before and after the completion of the training program.
Quality of life Results: Both groups reported similar medical characteristics and physical activity status. General mixed
effects models revealed that the intervention group reduced pulse wave velocity by 9% (p = 0.05); Emv/Vp
by 14% (p = 0.06); E to A ratio by 24% (p = 0.004), E to Emv ratio by 8% (p = 0.05), MLHFQ score by 66%
(p = 0.003) and the depression score by 19% (p = 0.5); increased augmentation index by 29%; VTI by 4%
(p = 0.05), 6-minute-walk distance up to 13% (p = 0.05), peak oxygen uptake by 28% (p = 0.001) and peak
power by 25% (p = 0.005). There were no significant changes in the control group.
Conclusion: Interval high-intensity aerobic training, combined with strength exercise, seems to benefit aortic di-
latation capacity and augmented systolic pressure in parallel with improvement in left ventricular diastolic func-
tion and quality of life.
© 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction those patients, arterial dysfunction was not related with augmented
systolic pressure, as the observed decrease in late systolic augmented
Heart failure constitutes a complex health problem which involves pressure may be due to the impaired LV systolic function [6,7].
both cardiac function and arterial circulation. This is due to the interac- Regular exercise enhances muscular function and exercise capacity
tive functions of the left ventricle (LV) and peripheral arterial system and promotes the body's ability to utilize oxygen [8,9]. It also improves
(ie, ventricular/vascular coupling) [1]. Earlier studies, including the Sys- the capacity of the blood vessels to dilate in response to exercise, left
tolic Hypertension in the Elderly (SHEP) and Survival and Ventricular ventricular diastolic function and neurohormonal activation [10–13].
Enlargement (SAVE) studies, have illustrated adverse outcomes in pa- Despite those facts, the recommendation for systematically performed
tients with LV dysfunction related with increased arterial stiffness, as exercise among heart failure patients has been poorly implemented in
assessed by a pulse pressure measurement [2–5]. Interestingly in daily clinical practice; while there is lack of studies evaluating the
role of exercise on aortic function in those patients. Furthermore, up
to our knowledge, most studies of rehabilitation in heart failure patients
Abbreviations: LV, left ventricle; QoL, quality of life; CHF, chronic heart failure; VE, min- have used moderate intensity continuous training (50%–70% of VO2max)
ute ventilation; VO2max, oxygen uptake; VCO2, carbon dioxide production; MLHFQ, or repetitions of high-intensity intervals (80%–95% of VO2max) of a rela-
Minnesota Living with Heart Failure Questionnaire; HR, heart rate; SVI, stroke volume
index; ITT, Intention-to-Treat; ZDRS, Zung Depression Rate Scale.
tively long duration (2–5 min) [14–16].
⁎ Corresponding author at: 46 Paleon Polemiston St., Glyfada, 166 74 Attica, Greece. The aim of this study was to evaluate the effect of high intensity, in-
E-mail address: chrysohoou@usa.net (C. Chrysohoou). termittent and short duration exercise in a 12 week training program on

http://dx.doi.org/10.1016/j.ijcard.2014.11.067
0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.
270 C. Chrysohoou et al. / International Journal of Cardiology 179 (2015) 269–274

aortic elastic properties, left ventricular function and quality of life minute walking test, 6MWT) was recorded, whereas peak values for oxygen uptake
(VO2max), minute ventilation (VE) and carbon dioxide production (VCO2max) were also re-
(QoL) in chronic heart failure (CHF) patients, due to coronary heart dis-
corded breath-by-breath via open-circuit spirometry (Vmax 229, SensorMedics, Califor-
ease or dilated cardiomyopathy, under optimal therapy. nia, USA), at both examinations. The Minnesota Living with Heart Failure Questionnaire
(MLHFQ) [21] and Zung's Depression Rating Scale (ZDRS) translated into Greek were
2. Methods used to evaluate Patient's quality of life (QoL) and depression status respectively [22]. De-
tailed echocardiography assessment was performed in all patients, using a Hewlett
2.1. Participants Packard 5500 Sonos with a multifrequency transducer (2.5–4 MHz), equipped with tissue
Doppler imaging (TDI) technology. Images were acquired with patients in left lateral
Chronic heart failure (CHF) patients due to left ventricular dysfunction (NYHA classes decubitus position and subjects with left ventricular ejection fraction b 50% defined by
II–IV, ejection fraction ≤ 50%), that were visiting the Heart Failure Unit of our Hospital's the Simpson's method were enrolled. From the apical four-chamber view, a 10 mm3 sam-
Cardiology department, between September of 2010 and September of 2011 were eligible ple volume was placed at the lateral mitral and tricuspid annuli, and spectral TDI was re-
for participation. Among them, those on a stable heart failure stage (I to III) due to ischemic corded, with the motion of annuli parallel to the TDI cursor. Pulse wave TDI was
or dilated heart failure and without severe valve diseases were included in the study. Pa- characterized by the systolic wave (Smv and Stv) and two diastolic waves (Emv and
tients that would have substantial changes in their medication, including neurohormonal Amv and Etv and Atv, respectively). Furthermore, using pulse wave Doppler at the tips
drugs and invasive treatment during the study period, were excluded from the final of mitral leaflets, early and late velocity waves (E and A) were recorded. Heart rate (HR)
analysis. was defined as beats per minute. Left atrial volumes were measured at end-diastole and
The study's flow diagram, according to the CONSORT guidelines [17] has been pre- systole and left atrial ejection fraction was calculated. Left ventricular outflow area was
sented elsewhere [18]. calculated from the left parasternal long axis view, while stroke volume index (SVI) was
Based on an a priori statistical power calculation (using East 3, 2003, Cytel Software calculated by the formula π/4 × (LVOT)2 × VTILVOT / BSA, where VTILVOT is the time-
Corporation, USA), the number of studied patients (n = 30 per group) was adequate to velocity integral of the left ventricular outflow tract, as detected by pulse wave Doppler,
evaluate standardized differences of the investigated parameters between the groups of from the apical four chamber view. All measurements were averaged on three to five mea-
the study, greater than 0.5 two-tailed, since a power of 85% at a significance level of 0.05 surements obtained during end-expiration. Furthermore, using the color M-mode by plac-
was achieved. ing the M-mode cursor aligned parallel to the left ventricular inflow extending from the
apex to the tips of the mitral valve, we measured the propagation velocity (Vp) [23]. Ad-
2.2. Randomization ditionally, arterial elastance (Ea) was calculated from the formula 0.9 × SBP / SV [24]. Fur-
thermore, a 24-hour Holter monitoring of heart rate was applied in all patients before and
Eligible patients after giving their consent were randomized to exercise and no after intervention.
exercise-control groups. The Biostatistics Unit of the First Cardiology Department of
our Institution performed the randomization of the participants using a block-
2.7. Bioethics
randomization design (by age group, sex, NYHA class, years of known CHF, and ischemic
CHF); each CHF patient was allocated to one or other group from a randomization list
The Internal Reviewing Board (IRB) of our Institution approved the protocol of the
that was created for this purpose.
study. All participants were informed about the aims, procedures, benefits and potential
risks of the trial, and agreed to participate and signed an informed consent form.
2.3. Peak exercise tolerance

The cardiopulmonary stress test consisted of an incremental exercise protocol on an 2.8. Data analysis
electromagnetically braked cycle ergometer (Ergoline 800, SensorMedics, California,
USA) to the limit of tolerance (WRpeak). After a 3-min baseline measurement, followed Continuous variables are presented as mean values ± standard deviation, while cate-
by a 3-min unloaded pedaling, the work rate was increased every minute by 10 or 20 W gorical variables are presented as frequencies. A univariate analysis was initially applied in
to the limit of tolerance, while subjects maintained a pedaling of 60 revolutions per mi- order to compare baseline characteristics of the patients between the two study groups.
nutes. During the test, gas exchange and flow rate variables were recorded breath-by- Associations between categorical variables were tested using contingency tables and the
breath (Vmax 229, SensorMedics, California, USA). The same test was performed at the be- calculation of Pearson's chi-squared test, while comparisons of normally distributed con-
ginning and the end of the training-intervention period. Criteria used to end the stress-test tinuous variables were performed by the calculation of the independent samples Student's
were: symptoms of fatigue/exhaustion, dyspnea, or leg fatigue/pain. t-test, after testing for equality of variances (homoscedacity), or the non-parametric
Mann–Whitney test. For within group comparisons at baseline and at follow-up examina-
2.4. Training intervention tions, the paired samples t-test, the Wilcoxon test or the McNemar chi-square test was ap-
plied. Correlations between normally distributed continuous variables were evaluated by
The intervention group followed a high-intensity intermittent aerobic training. the calculation of Pearson's r-coefficient and correlations between skewed continuous or
Patients were instructed to exercise on electromagnetically braked cycle ergometers discrete variables were evaluated by the use of Spearman's rho-coefficient. The
(Cateye Ergociser, ECl600; Cat Eye; Osaka, Japan) at an intensity equivalent to 80% Intention-to-Treat (ITT) procedure was applied in the current analysis of the data. The re-
WRpeak with a pedaling rate of 50 to 60 rpm and progressively to 100% of WRpeak for search hypothesis was evaluated using generalized estimating equations (GEE), with the
30 s, alternated with 30 s of rest for a accumulative period of 45 min/day, 3 days/week linear as the link function. The explanatory variables entered in the model were: the
for 12 consecutive weeks. Resistance training was performed with a fitness equipment. group of study (i.e., intervention or control) and those variables that showed a significant
Each session was consisting of 4 exercises i.e. knee extension, seated chest press, peck association with the outcome, i.e., QoL and ZDRS, in the univariate analysis (at a 0.05 sig-
deck and lateral pull-down. Training was initiated at an intensity of 30% of the one repeti- nificance level). Normality tests were applied using the Kolmogorov–Smirnov criterion.
tion maximum (1RM, performed at baseline) with a training amount of 3 sets of 8–10 rep- Assumptions of linearity for continuous independent variables and constant variance of
etitions for the first 2 weeks, followed by 3 weeks of training at an intensity of 50% of the the standardized residuals were assessed through plotting the residuals against the fitted
baseline 1RM. For the remaining 7 weeks the intensity was set up at an intensity of 90% of values. All reported p-values are based on two-sided tests and compared to a significance
the 1RM. Patients in the control group were managed as usual by the admitting physician level of 0.05. However, due to multiple significance tests we used the Bonferroni correc-
in the Heart Failure Unit, and no advice for any specific exercise protocol was given. tion (since the number of comparisons was less than ten) in order to account for the in-
crease in Type I error. SPSS 18.0 software (SPSS Inc. 2010, Chicago, IL, USA) was used for
all statistical calculations.
2.5. End points

The primary end points of this trial were quality of life (QoL), left ventricular diastolic
3. Results
function, and aortic elastic properties. Other secondary end point was the depression sta-
tus of the patients, and physical activity status.
In Table 1 the baseline characteristics of the CHF patients are pre-
2.6. Measurements sented. As seen, patients in the intervention group were older; however,
other characteristics were similar in both groups.
Information regarding demographic characteristics, family status and education
Exploratory analysis showed that evidence of a worse quality of
level was gathered for all patients at baseline. Moreover, overall lifestyle habits
were defined using standard procedures. Body-surface area was calculated [BSA = life at baseline (i.e., higher MLHFQ score) was positively associated
(height)0.725 ∗ (weight)0.425 ∗ 0.007184] and the New York Heart Association (NYHA) with age (rho = 0.268, p = 0.03), higher likelihood of being divorced
functional classification system was used to categorize all CHF patients [19]. In all statisti- or widowed (p = 0.03), higher NYHA classification (rho = 0.411,
cal analysis, continuous measurements were used for blood pressure and biochemical pa- p = 0.001), as well as higher likelihood of having a history of hyperten-
rameters. The translated short version (9 items) of the validated International Physical
Activity Questionnaire (IPAQ), suitable for assessing population levels of self reported
sion (p = 0.04); no significant associations were noted between
physical activity was used in all patients at baseline examination [20]. Furthermore, the MLHFQ score and medical history (i.e., years of known CHF, p = 0.18,
distance that patients were able to walk over a total of 6 min on a hard, flat surface (six- history of diabetes, p = 0.98, hypercholesterolemia, p = 0.38), the
 C. Chrysohoou et al. / International Journal of Cardiology 179 (2015) 269–274 271

Table 1 intervention group sub-tripled their MLHFQ score after 12 weeks of in-
Baseline characteristics of patients randomized to usual care or exercise intervention. tervention, whereas patients in the control group had similar scores (p
Control group Intervention group p for between groups comparisons b 0.001). No differences between
(n = 39) (n = 33) groups were observed regarding patient depression status (Table 2);
Age (years) 56 ± 11 63 ± 9 0.01 however, it should be noted that the ZDRS score was significantly
Men (%) 72 88 0.36 lower after intervention in the exercise group (p = 0.005), while re-
Years of school 11.5 ± 3.9 10.8 ± 4.0 0.49 mains without change in the control group (p = 0.19). Analysis was
Married (%) 51 82 0.07
subsequently focused on those CHF patients having an implantable
Body mass index (kg/m2) 31.3 ± 7 28.85 ± 4.2 0.08
Obesity (%) 49 42 0.60 cardioverter-defibrillator; patients in the exercise intervention group
Waist circumference (cm) 105.2 ± 11.5 101.6 ± 9.7 0.18 reduced by 1.5-times their MLHFQ score (p b 0.001), whereas no
Hip circumference (cm) 104.8 ± 10.4 101.8 ± 6.9 0.19 change was observed in the control group (p = 0.54). Similarly, exer-
Smoking (ever) (%) 87 82 0.56 cise intervention CHF patients with implantable cardioverter-
Physical activity status 0.61
Physically inactive (%) 31 39
defibrillator improved the distance walked during the 6MWT, achieved
Minimally active (%) 41 31 higher VO2max, VCO2max and WRpeak, as compared with the control
HEPA (%) 28 30 group (all p's b 0.05). Obesity status did not affect the aforementioned
MedDietScore (0–55) 35 ± 6 37 ± 7 0.03 results, since patients in the exercise intervention group had a lower
Ischemic CHF (%) 70 70 0.99
MLHFQ score as compared with the control group, irrespective of their
Ejection fraction (%) 32 31 0.91
Implantable 33 35 0.85 obesity status (pobese = 0.02 and pnon-obese = 0.04). A significant age-
cardioverter-defibrillator (%) by-group interaction was observed (p = 0.02) for quality of life; thus,
Pacemaker (%) 24 28 0.71 the analysis was stratified to patients under and above 60 years old. Pa-
Years of known CHF 3.9 ± 4.8 4.2 ± 4.6 0.78 tients in the exercise-intervention group that were under the age of
NYHA class 0.46
60 yrs, sub-tripled the MLHFQ score following 12 weeks of rehabilita-
I (%) 39 52
II (%) 51 36 tion, whereas patients' score in the control group remained in similar
III (%) 9 12 values (p for between groups comparisons b 0.001); however, no sig-
History of hypertension (%) 25 42 0.15 nificant change in the MLHFQ score was observed between groups
History of hypercholesterolemia (%) 61 63 0.83
when only patients over 60 yrs were included in the analysis (p =
History of diabetes (%) 36 33 0.45
B-blockers (%) 89 100 0.22
0.48). No sex-by-intervention group interaction was observed (p =
A- and B-blockers (%) 3 0 0.32 0.83), no NYHA class-by-group and no baseline physical activity
Angiotensin converting 43 61 0.15 status-by-group interactions on patients' quality of life were observed
enzyme inhibitors (%) (p = 0.35, p = 0.28, respectively).
Angiotensin 1 receptor inhibitors (%) 51 33 0.13
Aldosterone antagonists (%) 57 37 0.09
Loop diuretics (%) 71 49 0.05 4. Discussion
Coumadin anticoagulants (%) 17 21 0.67
Statins (%) 74 82 0.46 In this randomized clinical trial the effect of high-intensity intermit-
Insulin (%) 11 6 0.44
tent exercise training was studied among CHF patients in parameters
Amiodarone (%) 3 12 0.16
Oral hypoglycemic (%) 20 21 0.90
evaluating cardiovascular performance and quality of life. The influence
of high intensive exercise was beneficial on the diastolic function of the
Values are mean ± SD or percentages. p-Values derived using the independent samples
Student's t-test (age, BMI), Mann–Whitney test (waist, hip, MedDietScore, years of
left ventricle, on elastic properties of the aorta, and on the quality of life
known CHF) and chi-square test (sex, marital status, medical history, NYHA class, physical of heart failure patients, even among CHF patients having an implant-
activity). able cardioverter-defibrillator or being in higher NYHA class.
Heart failure is the final clinical condition of all cardiovascular dis-
eases. As the cardiovascular system involves beyond the heart and the
BMI level (p = 0.16), physical activity status (p = 0.51) or education whole arterial tree, any alterations in the function of the cardiac muscle
status (p = 0.18). pose an impact on arterial properties, and vice versa. Thus, the mea-
All patients adopted the exercise intervention program well; more sured central aortic pressure waveform represents a summation of the
specifically, no serious adverse events, such as arrhythmias, muscle in- forward arterial wave, which is generated by the left ventricle during
jury, hip fracture, or hospitalization related to exercise were observed, ejection and a backward moving wave caused by the reflection of the
suggesting that exercise training was well tolerated and safe. Further- forward moving wave within the arterial system. The forward wave is
more, no substantial changes were observed on their medications or affected by the systolic function of the left ventricle, the elastic proper-
need for invasive treatment during the study period in those who ties of the entire arterial tree and the interaction between the ventricle
were finally analyzed (Fig. 1). and aorta, described as aortic–ventricular coupling [25,26]. Normally,
Results from the intervention trial are presented in Table 2. Follow- the reflected wave arrives in diastole and aids in coronary artery
ing rehabilitation, there was a significant improvement in the interven- perfusion. Aging, thaw, and other condition (atherosclerosis,
tion group concerning exercise tolerance, as indicated by the greater hypertension, diabetes mellitus, obesity, and chronic renal disease)
distance walked during the 6MWT (by 13%, p b 0.05) and the greater create arterial stiffness, where the forward aortic velocity is increased
cycle ergometry WRpeak (by 25%, p b 0.01), higher VO2max (by 31%, and the reflected waves arrive earlier to the heart, causing an
p b 0.001) and VCO2max (by 28%, p b 0.001), and lower VE/VCO2, where- augmentation of systolic pressure which increases ventricular work.
as patients in the control group had no changes in the aforementioned Additionally, this unfavorable alteration of ventricular/vascular
indices. Also, in the intervention group pulse wave velocity was de- coupling is associated with increased arterial stiffness that has been
creased by 9% (p = 0.03); Emv/Vp by 14% (p = 0.06); E to A ratio by linked directly to the subsequent development of adverse
24% (p = 0.004) and E to Emv ratio by 8% (p = 0.05); while they in- cardiovascular outcomes including chronic heart failure and death.
creased augmentation index by 26%; and VTI by 4% (p = 0.05); Those Over a century ago, Fleming and Lewis [27,28] introduced the
parameters were not significantly changed in the control group (all morphological changes in the pressure wave contour in systolic heart
p N 0.05). At the end of the training intervention the MLHFQ score failure; while recent data, including the Systolic Hypertension in the
was lower among CHF patients in the training group as compared Elderly (SHEP) and Survival and Ventricular Enlargement (SAVE)
with the control group; in particular, patients in the exercise studies, illustrate that increased conduit artery stiffness, as assessed
272 C. Chrysohoou et al. / International Journal of Cardiology 179 (2015) 269–274

Assessed for eligibility (n=150)

Excluded (n=50)
Not meeting inclusion criteria (n=10)
Declined to participate (n=40)
Other reasons (n=0)

Randomized (n=100)

Allocated to intervention (n=50) Allocated to non-intervention (n=50)

Received allocated intervention (n=50) Received allocated intervention (n=50)
Did not receive allocated intervention (n=0) Did not receive allocated intervention (n=0)

Lost to follow-up (n=0) Lost to follow-up (n=0)

Discontinued intervention (10: denied, 7: changed phone Discontinued intervention (5: denied, 4: lack of time, 1:
number) (n=17) atrial flutter, 1: ventricular tachycardia) (n=11)

Analysed (n=33) Analysed (n=39)

(63±9 years, 88% men, 70% ischemic heart failure, (56±11 years, 82% men, 70% ischemic heart
4.2±4.6 years with known heart failure) failure, 3.9±4.8 years with known heart failure)

Excluded from analysis (n=0) Excluded from analysis (n=0)

Fig. 1. The flow diagram of the study.

non-invasively by a pulse pressure measurement in the brachial artery, force to overcome the late systolic augmented pressure. Thus, augment-
may contribute to increased adverse outcomes in patients with left ven- ed pressure and systolic pressure are therefore reduced in parallel with
tricular dysfunction [3]. It seems that as left ventricle dilates and loses the decrease in ejection duration. In those patients, wave reflection does
compensation, the heart muscle cannot generate the necessary extra not boost systolic pressure because the heart is incapable of responding,

Table 2
Changes between diastolic function measurements, aortic stiffness indices and clinical characteristics of patients randomized to usual care or exercise intervention.

Control group (n = 39) Intervention group (n = 33) p

Baseline 12-weeks after Baseline 12-weeks after

MLHFQ score (0–105) 19 ± 12 21 ± 13 21 ± 7 7 ± 9‡ 0.003

ZDRS score (0–80) 37 ± 8 41 ± 10 37 ± 8 30 ± 6‡ 0.54
ZDRS score N 50 (%) 18 5‡ 24 3‡ 0.51
6-minute-walk (m) 406 ± 64 423 ± 65 422 ± 77 476 ± 82‡ 0.05
VO2max (ml/kg/min) 17 ± 6 18 ± 4 16 ± 6 21 ± 5‡ 0.001
VCO2max (l/min) 1.57 ± 0.6 1.59 ± 0.6 1.49 ± 0.7 1.92 ± 0.6‡ 0.001
VE/VCO2 32.7 ± 3.4 30 ± 2.5 33.6 ± 5.7 31 ± 4.2 0.05
WRpeak (Watt) 88 ± 42 85 ± 29 84 ± 39 105 ± 37‡ 0.005
Mean heart rate 75 ± 12 78 ± 13 72 ± 16 69 ± 8 0.04
E/A 1.77 ± 1.45 1.45 ± 1.17 1.04 ± 0.79 0.78 ± 0.38 0.015
E/E′ 9.36 ± 4.41 13 ± 7.1 9.2 ± 4.2 9 ± 4.3 0.05
E/vp 4.9 ± 2.9 2.7 ± 0.9 3.4 ± 1.4 3 ± 1.9 0.06
Left atrium ejection fraction (%) 49 40 47 49 0.3
Stroke volume index (ml/m2) 28.3 ± 8 26.4 ± 6 30.6 ± 11.5 31.4 ± 9 0.3
TDI S tricuspid annulus 12.3 ± 2.7 11.3 ± 3 11 ± 3 14.2 ± 2 0.05
Pulse velocity (m/s) 8.8 ± 1.3 9.1 ± 1.9 9.5 ± 2.5 8.7 ± 2.7 0.05
Augmentation index 23.6 ± 10 22 ± 8.7 21 ± 8 27 ± 6 0.013
Arterial elastance (Ea) 1.9 ± 0.6 1.9 ± 0.5 1.9 ± 0.9 1.8 ± 0.6 0.82

Values are mean ± SD or percentages. Between groups p-values derived using GEE, after adjusting for age, sex, BMI, and MedDietScore.
‡
p b 0.05, for the within groups comparisons.
 C. Chrysohoou et al. / International Journal of Cardiology 179 (2015) 269–274 273

so that systole is terminated prematurely, and wave reflection poses a improve ventricular stroke volume, left ventricular diastolic indices, ar-
negative influence on flow [29,30]. It could be reasonable to hypothesize terial function and ventricular–aortic association, offering positive ef-
that any intervention that can increase contractility capacity of the left fects on quality of life.
ventricle could also have beneficial effect on arterial function, which
would be evident by an increase in augmentation index reflecting im- Conflict of interest
proved ventricular–aortic coupling. In this work, the exercise training
program in patients with systolic heart failure, offered a significant im- The authors report no relationships that could be construed as a con-
provement in diastolic and systolic function indices, implying a benefi- flict of interest.
cial effect on ventricular work. This also improved arterial function, by
decreasing PWV and increasing augmentation index. Since both param-
eters are related to vascular performance, those findings could be Funding
misinterpreted as a pathophysiology contradiction. However, the de-
crease in PWV represents reduced central arterial stiffness, which al- None.
lows in the case of weak pumping force of the failing heart to create a
forward flow able to generate a stronger reflected wave. Thus, the im- References
provement in heart and vessel performance by exercise, allows the for-
[1] W.W. Nichols, M.F. O'Rourke, McDonald's Blood Flow in Arteries: Theoretic, Exper-
mation of an integrated arterial pulse force, capable of overcoming the imental and Clinical Principles, 5th ed. Edward Arnold, London, 2005.
severe heart failure tissue hypo-perfusion and generating a functional [2] M.J. Domanski, G.F. Mitchell, J.E. Norman, D.V. Exner, B. Pitt, M.A. Pfeffer, Indepen-
arterio-ventricular system. As a consequence, a benefit was also evident dent prognostic information provided by sphygmomanometrically determined
pulse pressure and mean arterial pressure in patients with left ventricular dysfunc-
in exercise capacity, as demonstrated by a 6-minute walking test, car- tion, J. Am. Coll. Cardiol. 33 (1999) 951–958.
diorespiratory performance and in the quality of life of those patients. [3] J.B. Kostis, J. Lawrence-Nelson, R. Ranjan, A.C. Wilson, W.J. Kostis, C.R. Lacy, Associa-
The link between quality of life and exercise tolerance has been attribut- tion of increased pulse pressure with the development of heart failure in SHEP. Sys-
tolic Hypertension in the Elderly (SHEP) Cooperative Research Group, Am. J.
ed to improvements in neurohormonal activation and peripheral vascu- Hypertens. 14 (2001) 798–803.
lar reactivity as assessed by flow mediated dilatation, ventilatory and [4] G.F. Mitchell, L.A. Moye', E. Braunwald, et al., Sphygmomanometrically determined
diastolic function, reduced brain natriuretic peptide levels and arrhyth- pulse pressure is a powerful independent predictor of recurrent events after myo-
cardial infarction in patients with impaired left ventricular function. SAVE investiga-
mia genesis, significant increase in muscle aerobic capacity, and in myo-
tors Survival and Ventricular Enlargement, Circulation 96 (1997) 4254–4260.
fibril cross-sectional area and mitochondrial density [31,32]. Also, [5] S.F. Mohammed, B.A. Borlaug, S. McNulty, et al., Resting ventricular–vascular func-
different exercise types have been shown to exhibit different effects tion and exercise capacity in heart failure with preserved ejection fraction: a
RELAX trial ancillary study, Circ. Heart Fail. 7 (2014) 580–589.
on muscular capacity, with the combination of moderate and high in-
[6] J.M. Tartie're, D. Logeart, M.E. Safar, A. Cohen-Solal, Interaction between pulse wave
tensity resistance training to illustrate improvement in muscular velocity, augmentation index, pulse pressure and left ventricular function in chronic
strength and endurance by 25–100% [33,34]. heart failure, J. Hum. Hypertens. 20 (2006) 213–219.
The most well-known large scale exercise intervention study is the [7] S.J. Denardo, R. Nandyala, G.L. Freeman, G.L. Pierce, W.W. Nichols, Pulse wave anal-
ysis of the aortic pressure waveform in severe left ventricular systolic dysfunction,
HF-ACTION (Heart Failure–A Controlled Trial Investigating Outcomes Circ. Heart Fail. 3 (2010) 149–156.
of Exercise TraiNing) study which evaluated the benefits from exercise [8] J. Myers, Exercise and cardiovascular health, Circulation 107 (2003) e2–e5.
training among 2331 patients with heart failure revealed an 11% lower [9] I. Vogiatzis, S.G. Zakynthinos, The physiological basis of rehabilitation in chronic
heart and lung disease, J. Appl. Physiol. 115 (1985) (2013) 16–21.
risk of all-cause death or hospitalization and a 15% lower risk of [10] McMurray JJ, Adamopoulos S, Anker SD, et al; Task Force for the Diagnosis and
cardiovascular-related mortality and morbidity, and scored significantly Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Car-
better on a quality-of-life questionnaire [35]. In the present trial, the ex- diology, Bax JJ, Baumgartner H, Ceconi C, et al; ESC Committee for Practice Guide-
lines. ESC guidelines for the diagnosis and treatment of acute and chronic heart
ercise program was more intense than the one used in the HF-ACTION failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic
trial. A recent meta-analysis suggested that combined (strength and in- Heart Failure 2012 of the European Society of Cardiology. Developed in collabora-
termittent) exercise training elicits larger benefits in peak VO2 than in- tion with the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail. 2012;14:
803-69.
termittent exercise training alone, although intermittent exercise
[11] V. Anagnostakou, K. Chatzimichail, S. Dimopoylos, et al., Effects of interval cycle
training induced larger benefits in peak VO2 and VE/VCO2 than continu- training with or without strength training on vascular reactivity in heart failure pa-
ous exercise training in patients with moderate to severe heart failure tients, J. Card. Fail. 17 (2011) 585–591.
[12] S. Adamopoulos, A.J. Coats, F. Brunotte, et al., Physical training improves skeletal
[36]. This beneficial effect of intermittent exercise may be due to a great-
muscle metabolism in patients with chronic heart failure, J. Am. Coll. Cardiol. 21
er generation of large shear stress forces within the endothelium lead- (1993) 1101–1106.
ing to improvement in endothelial function, increasing muscle mass, [13] R. Hambrecht, S. Gielen, A. Linke, et al., Effects of exercise training on left ventricular
hence improving oxygen metabolism, diastolic function, left ventricular function and peripheral resistance in patients with chronic heart failure: a random-
ized trial, JAMA 283 (2000) 3095–3101.
work and aortoventricular coupling. All of these mechanisms confer to [14] B. Nilsson, A. Westheim, M. Risberg, Effects of group-based high intensity aerobic in-
regain quality of life, which is crucial in heart failure patients, as it is a terval training in patients with chronic heart failure, Am. J. Cardiol. 102 (2008)
powerful predictor of morbidity and mortality. 1361–1365.
[15] C. Freyssin, C. Verkindt, F. Prieur, P. Benaich, S. Maunier, P. Blanc, Cardiac rehabilita-
tion in chronic heart failure: effect of an 8-week, high-intensity interval training ver-
4.1. Strengths and limitations sus continuous training, Arch. Phys. Med. Rehabil. 93 (2012) 1359–1364.
[16] R. Arena, J. Myers, D. Forman, C. Lavie, M. Guazzi, Should high-intensity-aerobic in-
terval training become the clinical standard in heart failure? Heart Fail. Rev. 18
Due to the wide heterogeneity of clinical characteristics and origin of (2013) 95–105.
heart failure, only patients suffering from ischemic or dilated cardiomy- [17] K.F. Schulz, D.G. Altman, D. Moher, CONSORT Group, CONSORT 2010 statement: up-
opathy were enrolled (even with implanted defibrillator or cardiac dated guidelines for reporting parallel group randomized trials, BMJ 340 (2010)
c332.
resynchronization system); while few patients with severe chronic ob-
[18] C. Chrysohoou, G. Tsitsinakis, I. Vogiatzis, et al., High intensity, interval exercise im-
structive pulmonary disease and sleep apnea were included. proves quality of life of patients with chronic heart failure: a randomized controlled
trial, QJM 107 (2014) 25–32.
[19] L.V. Athanasopoulos, A. Dritsas, H.A. Doll, D.V. Cokkinos, Comparative value of NYHA
4.2. Conclusions and public health implications
functional class and quality-of-life questionnaire scores in assessing heart failure, J.
Cardiopulm. Rehabil. Prev. 30 (2010) 101–105.
Application of a high intensity interval exercise program in patients [20] C.L. Craig, A.L. Marshall, M. Sjostrom, et al., International physical activity question-
with heart failure of ischemic or idiopathic origin was feasible and of- naire: 12-country reliability and validity, Med. Sci. Sports Exerc. 35 (2003)
1381–1395.
fered substantial beneficial effects on several hemodynamic and clinical [21] T. Hak, D. Willems, G. van der Wal, F. Visser, A qualitative validation of the Minne-
factors. This combined aerobic with strength exercise program seems to sota Living with Heart Failure Questionnaire, Qual. Life Res. 13 (2004) 417–426.
274 C. Chrysohoou et al. / International Journal of Cardiology 179 (2015) 269–274

[22] K.N. Fountoulakis, A. Lacovides, S. Samolis, et al., Reliability, validity and psychomet- [29] M.F. O'Rourke, The arterial pulse in health and disease, Am. Heart J. 82 (1971) 687–702.
ric properties of the Greek translation of the Zung Depression Rating Scale, BMC Psy- [30] N. Westerhof, M.F. O'Rourke, Haemodynamic basis for the development of left ven-
chiatry 1 (2001) 6. tricular failure in systolic hypertension and for its logical therapy, J. Hypertens. 13
[23] G. Ronan, M.J. Wolk, S.R. Bailey, et al., ACCF/AHA/ASE/ASNC/HFSA/HRS/SCAI/SCCT/ (1995) 943–952.
SCMR/STS 2013 multimodality appropriate use criteria for the detection and risk as- [31] A.I. Larsen, S. Lindal, P. Aukrust, I. Toft, T. Aarsland, K. Dickstein, Effect of exercise
sessment of stable ischemic heart disease: a report of the American College of Car- training on skeletal muscle fibre characteristics in men with chronic heart failure.
diology Foundation Appropriate Use Criteria Task Force, American Heart Correlation between skeletal muscle alterations, cytokines and exercise capacity,
Association, American Society of Echocardiography, American Society of Nuclear Int. J. Cardiol. 83 (2002) 25–32.
Cardiology, Heart Failure Society of America, Heart Rhythm Society, Society for Car- [32] J.V. Papathanasiou, E. Ilieva, B. Marinov, Six-minute walk test: an effective and nec-
diovascular Angiography and Interventions, Society of Cardiovascular Computed To- essary tool in modern cardiac rehabilitation, Hell. J. Cardiol. 54 (2) (2013) 126–130.
mography, Society for Cardiovascular Magnetic Resonance, and Society of Thoracic [33] U. Wisløff, A. Støylen, J.P. Loennechen, et al., Superior cardiovascular effect of aerobic
Surgeons, J. Nucl. Cardiol. 21 (1) (2014) 192–220. interval training versus moderate continuous training in heart failure patients. A
[24] H.M. Cheng, W.C. Yu, S.H. Sung, K.L. Wang, S.Y. Chuang, C.H. Chen, Usefulness of sys- randomized study, Circulation 115 (2007) 3086–3094.
tolic time intervals in the identification of abnormal ventriculo-arterial coupling in [34] K. Meyer, L. Samek, M. Schwaibold, et al., Interval training in patients with severe
stable heart failure patients, Eur. J. Heart Fail. 10 (12) (2008) 1192–1200. chronic heart failure: analysis and recommendations for exercise procedures,
[25] A.A. Voors, C.J. Petrie, M.C. Petrie, et al., Low pulse pressure is independently related Med. Sci. Sports Exerc. 29 (1997) 306–312.
to elevated natriuretic peptides and increased mortality in advanced chronic heart [35] R.J. Mentz, P.J. Schulte, J.L. Fleg, et al., Clinical characteristics, response to exercise
failure, Eur. Heart J. 26 (2005) 1759–1764. training, and outcomes in patients with heart failure and chronic obstructive pul-
[26] N. Westerhof, P. Sipkema, G.C. van den Bos, G. Elzinga, Forward and backward monary disease: findings from Heart Failure and A Controlled Trial Investigating
waves in the arterial system, Cardiovasc. Res. 6 (1972) 648–656. Outcomes of Exercise TraiNing (HF-ACTION), Am. Heart J. 165 (2013) 193–199.
[27] W.J. Flemming, Pulse dicrotism, J. Anat. Physiol. 15 (1881) 278–291. [36] N.A. Smart, G. Dieberg, F. Giallauria, Intermittent versus continuous exercise training
[28] T. Lewis, The factors influencing the prominence of the dicrotic wave, J. Physiol. 34 in chronic heart failure: a meta-analysis, Int. J. Cardiol. 166 (2013) 352–358.
(1906) 414–429.