Heart Failure Reviews

https://doi.org/10.1007/s10741-018-9686-z

Influence of exercise on oxidative stress in patients with heart failure

Sabrina Weiss Sties 1,2,3,4,5 & Leonardo Vidal Andreato 2,3,6 & Tales de Carvalho 2,3 & Ana Inês Gonzáles 2,3,7 &
Vitor Giatte Angarten 8 & Anderson Zampier Ulbrich 9 & Lourenço Sampaio de Mara 2,3 & Almir Schmitt Netto 1,2,3 &
Edson Luiz da Silva 10,11 & Alexandro Andrade 2,12

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Reactive oxygen species play an important role in the pathophysiology of heart failure (HF). In contrast, regular physical exercise
can promote adaptations to reactive oxygen species that are beneficial for patients with HF. We completed a systematic review of
randomized controlled trials that evaluate the influence of exercise on oxidative stress in patients with HF. Articles were searched
in the PubMed, Cochrane, SciELO, and LILACS databases. The search was conducted according to the Preferred Reporting
Items for Systematic Reviews and Meta-Analyses guidelines. The quality of the included studies was assessed using the
Physiotherapy Evidence Database scale. We selected 12 studies with a total of 353 participants. The included patients had a left
ventricle ejection fraction of < 52% and New York Heart Association functional class II or III disease. A significant increase was
observed in peak oxygen consumption (between 10 and 46%) in the group that underwent training (TG). There was an
improvement in the oxidative capacity of skeletal muscles in the TG, related to the positive activity of mitochondrial cytochrome
c oxidase (between 27 and 41%). An increase in the expression of the enzymes glutathione peroxidase (41%), catalase (between
14 and 42%), and superoxide dismutase (74.5%), and a decrease in lipid peroxidation (between 28.8 and 58.5%) were observed
in the TG. Physical training positively influenced the cardiorespiratory capacity and enhanced the benefits of oxidant and
antioxidant biomarkers in patients with HF. High-intensity training promoted a 15% increase in the plasma total antioxidant
capacity, whereas moderate training had no effect.

Keywords Cardiovascular disease . Physical fitness . Antioxidants . Exercise

Background patients. HF is characterized by dyspnea, fatigue, and exercise

intolerance due to decreased cardiac output, skeletal muscle
Heart failure (HF) is the common final pathway of various heart myopathy, and endothelial dysfunction [1]. Inflammatory re-
diseases and has a significant impact on the prognosis of sponses and the production of reactive oxygen species (ROS)

* Sabrina Weiss Sties 6

Departamento de Educação Física, Centro Universitáriode Maringá
sabrinawsties@gmail.com (UNICESUMAR), Maringá, PR, Brazil
7
Departamento de Fisioterapia, Centro Universitário Estáciode Santa
1
Faculdade Avantis, Departamento de Fisioterapia, Balneário Catarina (ESTÁCIO), Florianópolis, SC, Brazil
Camboriú, SC, Brazil 8
Faculdade de Motricidade Humana, Lisbon, Portugal
2 9
Universidade do Estado de Santa Catarina (UDESC), Universidade Federal do Paraná, (UFPR), Curitiba, SC, Brazil
Florianópolis, SC, Brazil 10
Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC,
3
Núcleo de Cardiologia e Medicina do Exercício, Florianópolis, Brazil
SC, Brazil 11
Laboratório de Pesquisa em Lipídeos, Antioxidantes e Aterosclerose,
4
Centro de Ciências da Saúde e do Esporte – CEFID/UDESC, Rua Florianópolis, SC, Brazil
Pascoal Simone, 358 – Coqueiros, Florianópolis, SC 88080-350, 12
Brazil Laboratório de Psicologia do Esporte e do Exercício (LAPE),
Florianópolis, SC, Brazil
5
Cardiology and Exercise Medicine Center – bloco C,
Florianópolis, SC, Brazil

are modulated by deteriorating heart function. Moreover, low
cardiac output or sympathetic vasoconstriction causes tissue
hypoxia and may also trigger an increase in the production of
free radicals [2].
ROS are chemically active and formed as a by-
product of oxygen metabolism, reacting directly with
proteins, membrane lipids, and nucleic acids, impairing
cellular function, and causing death by apoptosis and
necrosis. Additionally, chronic release of ROS and per-
sistent inflammation can contribute to cardiac hypertro-
phy, while collagen deposition, fibrosis, and metallopro-
teinase activation are associated with progression of
end-stage heart failure [3].
ROS can function as signaling molecules to trigger
proinflammatory cytokine production. Potential sources
of ROS in myocardial cells include xanthine oxidase,
NADPH oxidase, lipoxygenase, cyclooxygenase, nitric
oxide synthase, the mitochondrial respiratory chain, per-
oxidase, and other hemoproteins. On the other hand,
cell defense systems can generate increases in enzymes
such as superoxide dismutase (SOD), catalase (CAT),
and glutathione peroxidase (GSH-PX) [3–6].
Physical training enhances cardiorespiratory capacity con-
comitant with the improvement of endothelial function [4].
The process that leads to the improvement of endothelial func-
tion is not fully elucidated; however, it seems to be related to
the attenuation of oxidative stress secondary to the balance of
reactive oxygen species (ROS) that may influence the bio-
availability of nitric oxide (NO), an important vasodilator
[7]. Accordingly, investigations about the mechanisms of
ROS generation and exercise-induced adaptations to ROS
[1, 8], as well as the role of these species in the pathophysio-
logical process of HF [9, 10], have considerably increased in
recent decades.
Inflammation and oxidative stress are involved in a self-
perpetuating cycle. Acute exercise is proinflammatory and
enhances oxidative stress [11] by contracting skeletal muscle
and generating ROS. ROS can oxidize extracellular molecules
and increase oxidative stress markers in other tissues and in
the circulation. There are many potential cell types in muscle
tissue (e.g., skeletal muscle fibers, various vascular cells, en-
dothelial cells, blood cells) that may contribute to ROS gen-
eration [12]. Nevertheless, exercise can stimulate chronic ad-
aptation, with a decrease in both inflammatory response and
oxidative stress by increasing the antioxidant level. These
exercise-mediated responses seem to be dependent on several
factors, such as exercise intensity, duration, and the muscle
mass involved [11].
However, systematic review articles that evaluate the influ-
ence of physical training on the oxidative and antioxidative
biomarkers of patients with HF are lacking. In addition, there
is a disagreement about the role of physical exercise in reduc-
ing oxidative stress in patients with HF.
Heart Fail Rev
Objectives
The objective of this study was to evaluate the influence of
exercise on oxidative stress in patients with HF, using a sys-
tematic review of randomized clinical trials.
Methods
This systematic review was performed according to the
PRISMA (Preferred Reporting Items for Systematic
Reviews and Meta-Analyses) guidelines [13] and registered
in the PROSPERO (International Prospective Register of
Systematic Reviews) database (no. CRD42016042287).
Search strategy
An electronic literature search was performed using the
PubMed, Cochrane, SciELO, and LILACS databases up to
January 29, 2017, but without restriction of the initial date.
The search strategy used a combination of keywords referring
to population (heart failure OR cardiac failure OR congestive
heart failure) AND intervention (exercise OR aerobic exercise
OR physical exercise) AND outcome (oxidative stress OR
antioxidant response elements OR antioxidants OR antioxi-
dant effect), and was not limited by language. A manual
search of the references of the included articles was also
performed.
Eligibility criteria
The study only included randomized clinical trials, available
in full texts, that considered intervention through physical
training (without restriction of exercise modality) and evalu-
ated oxidative or antioxidative biomarkers as endpoints in
subjects of both sexes, with a diagnosis of ischemic heart
failure, hypertensive or idiopathic etiology, and New York
Heart Association functional class I, II, or III according to a
decrease in left ventricular ejection fraction. Age range was
not restricted. Only interventions with ≥ 3 weeks of follow-up
were considered.
The following exclusion criteria were used: (1) non-
original study (e.g., case studies, letters to the editor, summary
publication, guidelines, recommendations, review articles,
and editorials); (2) studies in which all participants had other
pathology or disease (e.g., renal disease, liver disease, respi-
ratory disease), other than comorbidities usually present in
patients with heart failure (e.g., systemic arterial hypertension
and coronary artery disease); and (3) drug treatment using
supplements (e.g., nitric oxide, L-arginine, L-carnosine) or ex-
perimental agents (e.g., testing of new drugs), other than drugs
typically used in treatment.
Heart Fail Rev
Study selection and data extraction
The first phase in this process was the exclusion of studies
based on the title. This phase determined whether the title
clearly reported the presence of exclusion criteria (e.g., words
indicating associated diseases, use of experimental supple-
ments or drugs). The abstracts were also examined to deter-
mine the presence of preestablished exclusion criteria. The
analysis of titles and abstracts was not intended to include
studies, but only to delete searches based on the exclusion
criteria previously adopted.
The remaining articles were selected for evaluation of the
full text. At this stage, the researchers determined whether the
articles met the inclusion criteria or had any of the exclusion
criteria. At all stages of study selection, two independent re-
searchers evaluated study eligibility. In case of a difference of
opinion, a third evaluator determined eligibility.
Data extraction was then performed by one researcher and
confirmed by a second. Disagreements were solved by
consensus.
Publication bias assessment
Randomized controlled trials were classified according to
the Physiotherapy Evidence Database (PEDro) scale [14],
which is based on the Delphi list of criteria for quality
assessment of randomized clinical trials when conducting
systematic reviews. The PEDro scale includes 11 parame-
ters that evaluate methodological quality (internal validity
and statistical information). Except for the first, each pos-
itively answered parameter is assigned 1 point for the over-
all rating (0 to 10 points). Studies with values of 7 to 10 are
considered of high quality. Those that include parameters
2–9 have internal validity, and articles that contain enough
statistical information so that their results can be
interpreted represent parameters 10–11. It should be noted
that the PEDro scale score was not used as an exclusion
criterion, but rather as an indicator of the scientific evi-
dence in the included studies.
Results
Description of studies
The different phases of research and selection of studies
are shown in Fig. 1. A total of 82 studies were identified
from the electronic databases. Subsequently, five articles
were added through a manual search, and duplicate articles
and those with ineligible titles were excluded. Summaries
of potentially relevant studies were evaluated. For the sys-
tematic review, 12 studies that met the eligibility criteria
were selected.
Result of literature search
Identification
Cochrane Library (n = 11), PubMed (n = 63),
SciELO (n = 1), LILACS (n = 7),
Manual search (n = 5)
(N = 87)
Articles excluded by duplicity
(n = 9)
Potentially relevant articles
(n = 78)
Articles excluded by title (n = 47)
Screening
40 Drug treatment included
3 Other comorbidities included
4 Other diseases included
Potentially relevant articles
(n = 31)
Articles excluded by abstract (n = 15)
15 No relevant outcome measures
Eligibility
Full articles evaluated
(n = 16)
Articles excluded by full text (n = 4)
4 Non-randomized
Inclusion
Articles included
(n = 12)
Fig. 1 Flow diagram of the literature search
Evaluation of methodological quality
The studies were scored based on the complete reading of the
articles and consensus between two reviewers (Table 1). Three
studies were considered of high quality [19–21, 23].
Characteristics of the studies
Table 2 shows the characteristics of the population and the
protocols used in the included articles. In total, 353 subjects
of both sexes were included and 207 were subjected to the
intervention. The sample size of these groups varied between
9 and 18 persons, with a mean age of between 50 and 76 years
and mean left ventricular ejection fraction of 26–52%. The
patients presented with HF of New York Heart Association
functional class II and III. Only one study [20] did not use a
control (healthy) group, and two studies [7, 23] used only
healthy persons for comparison. The studies were published
between 1995 and 2015.
Cardiopulmonary tests were performed on bicycles [5,
15–18, 20, 21, 23] or treadmills [6, 7, 19, 22] for the prescrip-
tion of the target area for training and for the evaluation of
cardiorespiratory fitness. The tests were conducted progres-
sively, with a 25-W increase every 3 min [5, 15–18, 21] or a
10-W increase every 2 min [23]. The modified Bruce [6, 7, 22]
and ramp protocols [19, 20] were also used.
Table 1 Classification of the methodological quality of the articles according to the PEDro scale

Criteria Hambrecht Hambrecht Gielen et al. Gielen et al. Niebauer Linke et al. Wisloff et al. Laurent et al. Erbs et al. Tsarouhas Meirelles Aksoy et al.
et al. (1995) et al. (1997) (2003) [17] (2005) [18] et al. (2005) (2005) [5] (2007) [19] (2009) [20] (2010) [21] et al. (2011) et al. (2014) (2015) [23]
[15] [16] [7] [22] [6]

Eligibility criteria Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
were specified
Subjects were Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
randomly allocated
to groups
Allocation was No No No No No No Yes Yes Yes No Yes Yes
concealed
The groups were Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
similar at baseline
There was blinding of No No No No No No No No Yes No No Yes
all subjects
There was blinding of No No No No No No No No No No No No
all therapists
There was blinding of No No No No No No No No No No No No
all assessors
Appropriate monitoring Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Intent to treat Yes Yes No No No No Yes No No No No Yes
Comparison between Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
groups
Point and variability Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
measures
Total 6/10 6/10 5/10 5/10 5/10 5/10 7/10 6/10 7/10 5/10 6/10 8/10

PEDro Physiotherapy Evidence Database

Heart Fail Rev
 Heart Fail Rev

Table 2 Characteristics of the studies

Author (year) Sample Subjects Male sex Age (years) LVEF (%) Functional Follow-up (weeks) Training intervention
total class
N N N MD/SD MD/SD (NYHA) Frequency Duration Intensity
(weekly) (min)

Hambrecht et al. (1995) [15] 22 TG: 12 22 TG: 50 ± 12 TG: 26 ± 9 II, III 24 7 UT: 80 70% VO2max
CG: 10 CG: 52 ± 8 CG: 27 ± 10 ST: 60
Hambrecht et al. (1997) [16] 18 TG: 9 18 TG: 50 ± 12 TG: 26 ± 10 II, III 24 7 UT: 80 70% HRmax
CG: 9 CG: 52 ± 8 CG: 28 ± 10 ST: 60
Gielen et al. (2003) [17] 30 TG: 10 20 TG: 55 ± 2 TG: 26.1 ± 3 II, III 24 7 UT: 20 70% VO2 max
CG: 10 CG: 53 ± 3 CG: 24.7 ± 2 ST: 60
HG: 10 HG: 50 ± 3 HG: 72.7 ± 1
Gielen et al. (2005) [18] 20 TG: 10 20 TG: 55 ± 2 (SE) TG: 25 (SE) II, III 24 7 UT: 20 70% VO2peak
CG: 10 CG: 53 ± 3 (SE) ST: 60
Niebauer et al. (2005) [7] 27 TG: 18 27 TG: 53.69 ± 9 NR 2.2 ± 0.7 (MD/SD) 8 5 20 70–80% HRmax
HG: 9 HG: 51.3 ± 7
Linke et al. (2005) [5] 35 TG: 12 NR TG: 55 ± 2 TG: 26 ± 3 II, III 24 7 UT: 20 70% VO2peak
CG: 11 CG: 52 ± 3 CG: 27 ± 3 ST: 60
HG: 12 HG: 56 ± 4
Wisloff et al. (2007) [19] 27 MCTG: 9 20 MCTG: 74.4 ± 12 MCTG: 32.8 ± 5 NR 12 3 MCTG: 47 MCTG: 70% HRpeak
HIITG: 9 HIITG: 76.5 ± 9 HIITG: 28 ± 7 HIITG: 38 HIITG: 95% HRpeak
CG: 9 CG: 75.5 ± 13 CG: 26.2 ± 8
Laurent et al. (2009) [20] 24 CTG: 12 24 CTG: 55 ± 4 CTG: 28.9 ± 7 NR 3 5 80 60–70% HRR
CBTG: 12 CBTG: 53 ± 4 CBTG: 30.2 ± 7
Erbs et al. (2010) [21] 37 TG: 17 37 TG: 60 ± 11 TG: 24 ± 5 IIIb 12 7 UT: 20–30 60% VO2max
CG: 17 CG: 62 ± 10 CG: 25 ± 4 ST: 60
Tsarouhas et al. (2011) [22] 38 TG: 18 28 TG: 63.7 ± 12 TG: 31.3 ± 3 II, III 12 5 40 60% HRmax
CG: 10 CG: 65.0 ± 8.6 CG: 31.9 ± 3
HG: 10 HG: 60.8 ± 11
Meirelles et al. (2014) [6] 30 TG: 15 14 TG: 54 ± 3 TG: 31.2 ± 2 II, III 24 3 90 HR 5–15% ↑ VT
CG: 15 CG: 55 ± 2 CG: 31.7 ± 2
Aksoy et al. (2015) [23] 45 MITG: 15 39 MITG: 63.7 ± 88 MITG: 50.33 ± 7 II, III 10 3 35 75% VO2peak
MCTG: 15 CG: 15 MCTG: 59.6 ± 7 MCTG: 52 ± 5
GC: 57.5 ± 11.2 CG: 51.67 ± 6

LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; N, absolute number; MD, mean difference; SE, standard error mean; SD, standard deviation; CG, control group; HG, healthy
group; TG, training group; MITG, moderate interval training group; CTG, conventional training group; CBTG, combined training group; HIITG, high-intensity interval training group; MCTG, moderate
continuous training group; ST, supervised training; UT, unsupervised training; VO2max, maximum oxygen consumption; HRmax, maximum heart rate; VO2peak, peak oxygen consumption; HR, heart rate;
HRR, heart rate reserve; NR, not reported; VT, ventilatory threshold; ↑, over

To evaluate the oxidative and antioxidative biomarkers,
biopsy of the middle part of the vastus lateralis muscle [5,
15–18] or venous blood test [6, 7, 19–23] was performed.
The intervention time varied between 3 and 24 weeks, and the
weekly frequency varied between 3 and 7 days. A large variation
was also observed in the duration of training (20–90 min).
Patients who underwent unsupervised training were encouraged
to participate in one to two 60-min supervised training sessions.
Only one study [19] used high-intensity training and
moderate-intensity training, and the others used moderate-
intensity training [5–7, 15–18, 20–23]. The protocols included
treadmill and bicycle exercises, walking, gymnastics, water
aerobics, resistance exercises, and noncompetitive games.
Cardiorespiratory capacity
The outcomes of selected articles are shown in Table 3. In
most studies, a significant improvement in cardiorespiratory
capacity was observed in groups that underwent training (TG)
[6, 7, 15, 17–21], and the increase in maximum oxygen con-
sumption (VO2max) ranged from 16 to 32% and peak oxygen
consumption (VO2peak) ranged from 1 to 46%. Only one study
found no difference in this variable [23].
Oxidative capacity
In studies that measured the oxidative capacity of skeletal
muscles, there was an improvement in the TG in the total
volume density of mitochondria—TVVM (19%), surface den-
sity of the mitochondrial inner border membrane—SVMIMB
(92%), surface density of mitochondrial cristae—SVMC
(43%), and surface density of cytochrome c oxidase-positive
mitochondria—SVMOCOX+ (between 27 and 41%) [12, 13,
15]. The intervention promoted an increase in the expression
of antioxidative enzymes such as GSH-PX (41%) and CAT
(42%); however, there was no change in SOD expression [5].
The expression of inducible nitric oxide synthase (iNOS)
mRNA and the iNOS protein concentration (52 and 35%,
respectively) [17], as well as the nitrotyrosine level (between
35 and 38%) [5, 17] and lipid peroxidation (57%) [5] de-
creased in patients who performed exercise.
Other studies evaluated the markers by collecting blood
samples. Hypoxanthine concentration decreased (47.6%)
after the training protocol, whereas the levels of xanthine,
L-arginine, symmetric dimethylarginine (SDMA) [7], and
asymmetric dimethylarginine (ADMA) [7, 22] were un-
changed. One study [7] found no difference in the NO level
after training. On the other hand, another study [23] report-
ed a decrease (14.6%) in this score in subjects who per-
formed moderate continuous exercise. It was reported that
moderate-intensity training did not modify the total antiox-
idant capacity in plasma [22]; however, high-intensity
training was able to increase the total antioxidant capacity
Heart Fail Rev
in plasma (15%), which was accompanied by a decrease in
oxidized low-density lipoprotein (LDL) (9%) [19].
In the group that underwent combined training, nitrate
levels increased whereas nitrite levels did not change [20].
Lipid peroxidation decreased (between 28.8 and 58.5%) in
the TG in two studies [6, 21]; however, in another study, no
difference was observed in this outcome [22]. The TG showed
decreased activity of NOS and protein carbonylation (60%).
SOD enzyme levels increased 74.5% and catalase levels in-
creased 14%. However, the levels of cyclic guanosine
monophosphate did not show any difference [6].
Discussion
The results of this systematic review demonstrated that phys-
ical training protocols lasting 3–24 weeks improve cardiore-
spiratory capacity. Improvement in these parameters promotes
cardiac rehabilitation and reduces both cardiac and all-cause
deaths among patients with established cardiovascular dis-
ease. Moreover, physical activity can control risk factors and
mitigate the severity of heart failure [24].
Increment of VO2 can promote benefits related to oxidant
and antioxidant biomarkers in patients with HF of functional
class II and III. Both moderate-intensity and high-intensity ex-
ercises may contribute to the attenuation of oxidative stress.
Although high-intensity training is related to excessive ROS
production, morphofunctional and metabolic adjustments in-
duced by high-intensity exercise were found to be more impor-
tant than the benefits of moderate-intensity exercise [19]. Acute
responses to 30 min of moderate-intensity physical exercise
initially induced oxidative stress and inflammation. An anti-
inflammatory response and activation of the antioxidant defense
system was observed 1 h after exercise. Therefore, decreased
exercise load on the immune system and redox balance lead to
adaptive responses that result in decreased severity [11].
Recent results have shown that ROS do not merely repre-
sent random disruptors of cellular structures. At certain levels,
ROS can be beneficial to cells, actively participating in path-
ways that regulate cell growth, differentiation, and prolifera-
tion [25]. The myopathy observed in patients with HF not only
is limited to the myocardium but also is involved in the skeletal
musculature. In this context, mitochondria are the main source
of ROS and the sites where oxygen (O2) is reduced [26].
In this process, the enzyme cytochrome c oxidase plays an
important role; it controls the generation of free radicals,
preventing their excess production [27]. Hence, evaluation
of the benefits of exercise in the mitochondrial ultrastructure
in skeletal muscles is relevant. In patients with HF, physical
training promoted improvements related to the TVVM,
SVMIBM, SVMC, and SVMOCOX+ [15, 16, 18]. These find-
ings indicate, in part, the changes occurring in the genesis of
exercise intolerance in patients with HF and indicate that
Heart Fail Rev

Table 3 Results of the studies included in the review

Author (year) Variable Training Health Control

Group Result Group Result Group Result

Hambrecht et al. (1995) [15] VO2 VT (L/min) TG ↑ †† – – CG NS

VVMCOX+ (vol%) TG ↑§§
– – CG NS
VVMCOX− (vol%) TG NS – – CG NS
TVVM (vol%) TG ↑ §§ – – CG NS
Hambrecht et al. (1997) [16] SVMOCOX+ (vol%) TG ↑§§ – – CG NS
SVMC (m2/cm3) TG ↑ §§ – – CG NS
SVMIBM (m2/cm3) TG ↑§§ – – CG NS
VVM (vol%) TG ↑§§ – – CG NS
Gielen et al. (2003) [17] VO2peak (mL kg−1 min−1) TG ↑** HG NS CG NS
Local iNOS expression mRNA (relative U) TG ↓§,†† HG ↑† CG NS
Local iNOS protein concentration (%PTA) TG ↓§,§§ HG ↓§ CG NS
Local nitrotyrosine concentration (%PTA) TG ↓§,§§ HG ↑§ CG NS
Gielen et al. (2005) [18] VO2 peak (mL kg−1 min−1) TG ↑** – – CG NS
Local iNOS expression (relative U) TG ↓§,†† – – CG NR
Local iNOS protein concentration (% PTA) TG ↓ §,†† – – CG NS
Nitrotyrosine (%PTA) TG ↓§,§§ – – CG NR
COX activity (nmol O2 mg−1 min−1) TG ↑§,†† – – CG NS
COX protein TG NS – – CG NS
Niebauer et al. (2005) [7] VO2peak (mL kg−1 min−1) TG ↑† HG NS – –
Hypoxanthine (μmol/L) TG ↓† HG NS – –
Xanthine (μmol/L) TG NS HG NS – –
NO (μmol/L) TG NS HG NS – –
L-Arginine (μmol/L) TG NS HG NS – –
SDMA (μmol/L) TG NS HG NS – –
ADMA (μmol/L) TG NS HG NS – –
Linke et al. (2005) [5] Nitrotyrosine (%) TG ↓§ HG NS CG NS
SOD total activity (U/mg) TG NS HG NS CG NS
Catalase activity (U/mg) TG ↑§, §§ HG NS CG ↓§§§
SOD activity (mU/mg) TG ↑§,§§ HG NS CG ↓§§§
Lipid peroxidation (μmol/mg) TG ↓§ HG NS CG ↑§§§
Wisloff et al. (2007) [19] VO2peak (mL kg−1 min−1) MCTG ↑† – – CG NS
Total antioxidant activity (mM) MCTG NS – – GC NS
Oxidized LDL (U/L) MCTG NS – – CG NS
VO2peak (mL kg−1 min−1) HIITG ↑†,†† – – CG NS
Total antioxidant activity (mM) HIITG ↑§ – – GC NS
Oxidized LDL (U/L) HIITG ↓† – – CG NS
Laurent et al. (2009) [20] VO2peak (mL kg−1 min−1) CTG ↑§ – – – –
Nitrate (μmol/L) CTG NS – – – –
Nitrite (μmol/L) CTG NS – – – –
VO2peak (mL kg−1 min−1) CBTG ↑§ – – – –
Nitrate (μmol/L) CBTG ↑§ – – – –
Nitrite (μmol/L) CBTG NS – – – –
Erbs et al. (2010) [21] VO2max (mL min−1 kg−1) TG ↑** – – CG NS
Lipid peroxidation (pg/mL) TG ↓** – – CG NS
Tsarouhas et al. (2011) [22] Test time (s) TG NS HG NE CG NA
Total antioxidant activity (mmol/L) TG NS HG NA CG NA
TBARS (mmol/L) TG NS HG NA CG NA
 Heart Fail Rev

Table 3 (continued)

Author (year) Variable Training Health Control

Group Result Group Result Group Result

ADMA (μmol/L) TG NS HG NA CG NA
Meirelles et al. (2014) [6] VO2max (mL kg−1 min−1) TG ↑†,§ – – CG NS
Nitric oxide synthase activity (pmol/108 cells) TG ↓§ – – CG NS
cGMP levels (pmol/108 cells) TG NS – – CG NS
Protein carbonylation (nmol/mg protein) TG ↓§ – – CG NS
SOD (U/mg protein) TG ↑§ – – CG NS
Catalase (U/mg protein) TG ↑§ – – CG NS
TBARS (nmol MDA/mg protein) TG ↓§ – – CG NS
Aksoy et al. (2015) [23] VO2peak (mL kg−1 min−1) MCTG NS – – CG ↓§
NO levels (ng/mL) MCTG ↓§ – – CG ↓§
VO2peak (mL kg−1 min−1) MITG NS – – CG ↓§
NO levels (ng/mL) MITG NS – – CG ↓§

NE, note evaluated; NR, not reported; NS, not significant; SMB, skeletal muscle biopsy; BC, blood collection; CG, control group; HG, healthy group; TG,
training group; MITG, moderate interval training group; CTG, conventional training group; CBTG, combined training group; HIITG, high-intensity
interval training group; MCTG, moderate continuous training group; VO2max, maximum oxygen consumption; VO2peak, peak oxygen consumption; VT,
ventilatory threshold; VVMCOX+, volume density of cytochrome c oxidase-positive mitochondria; VVMCOX−, volume density of cytochrome c oxidase-
negative mitochondria; TVVM, total volume density of mitochondria; VVM, volume density of mitochondria; SVMOCOX+, surface density of cytochrome
c oxidase-positive mitochondria; SVMC, surface density of mitochondrial cristae; SVMIBM, surface density of the mitochondrial inner border membrane;
FVW, Fator de von Willebrand; PTA, positive tissue area; mRNA, messenger ribonucleic acid; COX, cytochrome c oxidase; TBARS, thiobarbituric acid
reactive substances; GSH-PX, glutathione peroxidase; iNOS, inducible nitric oxide synthase; NO, nitric oxide; ADMA, asymmetric dimethylarginine;
SDMA, symmetric dimethylarginine; cGMP, cyclic guanosine monophosphate; ↑, over
*Significantly different from baseline, P ≤ 0.001
**Significantly different from control, P ≤ 0.001
†
Significantly different from baseline, P ≤ 0.01
††
Significantly different from control, P ≤ 0.01
§
Significantly different from baseline, P ≤ 0.05
§§
Significantly different from control, P ≤ 0.05
§§§
Significantly different from the healthy group, P ≤ 0.05

chronic exercise can correct these changes [16]. The increase
in aerobic activity may have contributed to the improvement
of the oxidative capacity of the skeletal musculature, as they
are closely related [28].
ROS play a role in the regulation of NO, one of the most
relevant factors in the control of vascular function in endothe-
lial cells. NO is considered essential for vascular homeostasis
and has been studied because of its role in the prevention of
cardiovascular diseases [29].
From the molecular point of view, one study [19] reported
that high-intensity training seems to have a superior effect on
the improvement of endothelial function compared with mod-
erate exercise. This may be due to the increased bioavailability
of NO owing to the increased antioxidant capacity in plasma
and reduction of oxidized LDL. High-intensity exercise in-
creased antioxidative status through a decrease in the level
of ROS and higher NO production, according to the theory
that higher shear stress during intense exercise triggers greater
responses at the cellular and molecular level [19].
Moreover, according to results of the present review, mod-
erate exercise was not able to promote increased plasma anti-
oxidant activity in patients with HF [7, 22]. This may be
explained, in part, by the finding that other markers, such as
L-arginine, ADMA, and SDMA, did not show changes. These
markers may have an influence on antioxidant activity, as
ADMA and SDMA are structurally related to L-arginine,
which is a key substrate in numerous metabolic and signaling
pathways [30, 31]. The current results seem to suggest that
SDMA inhibits the cellular and renal uptake of L-arginine.
However, it also potentially leads to a decrease in the avail-
ability of NO [32].
When stimulated by cytokines and/or endotoxins, iNOS can
be expressed in various cell types as infiltrates of inflammatory
cells, endothelial cells, and cardiac myocytes [33]. This iso-
form, once synthesized, releases large amounts of NO, which
continues indefinitely until L-arginine or the cofactors required
for its synthesis are depleted, or until cell death occurs [34]. In
the studies selected for this review, the expression of the iNOS
Heart Fail Rev
mRNA in skeletal muscle and the concentration of iNOS pro-
tein decreased in patients who performed exercise [17], and
were correlated with a significant decrease in the level of local
nitrotyrosine [5, 17]. These results are relevant because
nitrotyrosine is associated with the process of atherosclerosis
[29] and, consequently, with mortality in these patients.
However, if there is no balance between the generation
of ROS and the effectiveness of the antioxidant mecha-
nisms, oxidative tissue damage may occur during aerobic
metabolism [28]. During tissue ischemia, adenosine tri-
phosphate (ATP) is converted to adenosine diphosphate
(ADP) and adenosine monophosphate (AMP) to provide
energy for muscle contraction. When the breakdown of
ATP exceeds its resynthesis, the accumulation of ADP
and AMP activates the purine nucleotide cycle, producing
ADP, AMP, hypoxanthine, and, in some tissues, xanthine
and uric acid [35]. However, hypoxanthine concentrations
can be reduced by physical training in patients with HF,
as this marker is related to oxygen consumption [7].
Temporary interruptions of calcium-dependent ATP
pumps lead to increased intracellular calcium concentra-
tions, which, during exercise, can activate the xanthine
oxidase pathway [28]. Xanthine oxidase is formed from
xanthine dehydrogenase under conditions of hypoxia and
hyperoxia [36, 37]. The conversion of hypoxanthine and
xanthine to uric acid occurs via the oxidase enzyme, with
molecular oxygen as a cofactor [28]. This process results
in the production of peroxide and hydrogen superoxide.
However, the superoxide radical can be eliminated by the
enzyme SOD, which catalyzes the dismutation of two
molecules of superoxide in oxygen (O2) and hydrogen
peroxide (H2O2). If the superoxide radical is not eliminat-
ed by the antioxidant enzymes GSH-PX and CAT, it can
generate the hydroxyl radical, leading to damages such as
lipid peroxidation [26]. In this process, physical training
may interfere with the reduction of lipid peroxidation [6,
21] in patients with HF, besides increasing the expression
of the antioxidant enzymes GSH-PX, CAT, and SOD [5,
6, 8].
Physical training is the mainstay of interventions for health
promotion and maintenance in patients with HF. Molecular
biology has helped elucidate the deficits in exercise capacity
in patients with HF, the beneficial effects of physical training,
and the progression and modulation of disease in the cardio-
vascular and musculoskeletal systems. The vascular system is
significantly impaired in patients with HF. During the last
decades, several studies that used exercise as a therapeutic
intervention have found beneficial effects on the vascular sys-
tem [4].
The present review suggests that in patients with HF, phys-
ical training plays an important role in systemic and muscular
metabolism, both in the homeostasis of and adaptive response
to ROS induced by exercise.
Conclusions
Physical training positively influences cardiorespiratory ca-
pacity and may promote benefits related to oxidant and anti-
oxidant biomarkers in patients with HF of functional class II
and III.
According to the results of the studies published to date,
moderate exercise increases the levels of antioxidant enzymes
such as catalase, SOD, and GSH-PX, decreases lipid peroxi-
dation, and increases oxidative capacity in skeletal muscles.
However, as reported in the only study that used high-
intensity exercise, solely this type of training was able to in-
crease the total plasma antioxidant activity.
Moreover, the expression of the iNOS mRNA and the con-
centration of iNOS protein decreased after exercise training and
were correlated with a decrease in the level of nitrotyrosine.
Implications for practice
This systematic review demonstrates that physical exercise
training improves cardiorespiratory capacity, which may pro-
mote benefits related to ROS biomarkers in patients with HF
of functional class II and III. These improvements are impor-
tant in patients with HF to attenuate their cardiovascular and
musculoskeletal limitations. The balance of ROS achieved
through exercise training may prevent exercise intolerance
contributing to vascular homeostasis, and improve endothelial
function.
Implications for research
There is a need for new randomized clinical trials with a more
desirable design, larger sample size, and more homogeneous
sample in terms of HF etiology, comorbidities, sex, and age.
Factors such as type, frequency, duration, and intensity of
exercise need to be similarly evaluated in subsequent studies
as a method for evaluating biomarkers. Some studies did not
present clear information about allocation concealment or
blinding of all subjects, therapists, or assessors. Moreover,
some studies did not add a control group without intervention,
which could have avoided bias and improved the methodo-
logical quality of the studies.
In addition, high-intensity training protocols should be in-
vestigated further to elucidate the benefits in terms of training
response and long-term health, verifying aspects such as ad-
herence, adverse events, and protocol time to achieve ROS
balance.
All these relevant outcomes and procedures must be report-
ed in future research to provide evidence for the effects of
physical training on oxidative stress as a preventive and ther-
apeutic intervention for HF.

Acknowledgments To the Coordination of Improvement of Higher Level
Personnel – CAPES.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
References
1. Marquez J, Suarez G, Marquez J (2013) Beneficios del ejercicio en
la insuficiencia cardíaca. Rev Chil Cardiol 32(1):58–65
2. Witman MA, McDaniel J, Fjeldstad AS, Ives SJ, Zhao J,
Nativi JN et al (2012) A differing role of oxidative stress
in the regulation of central and peripheral hemodynamics
during exercise in heart failure. Am J Physiol Heart Circ
Physiol 303(10):H1237–H1244
3. Ayoub KF1, Pothineni NVK1, Rutland J1, Ding Z1, Mehta JL
(2017) Immunity, inflammation, and oxidative stress in heart fail-
ure: emerging molecular targets. Cardiovasc Drugs Ther 31(5–6):
593–608
4. Adams V, Niebauer J (2015) Reversing heart failure-associated
pathophysiology with exercise: what actually improves and by
how much? Heart Fail Clin 11(1):17–28
5. Linke A, Adams V, Schulze PC, Erbs S, Gielen S, Fiehn E et al
(2005) Antioxidative effects of exercise training in patients with
chronic heart failure: increase in radical scavenger enzyme activity
in skeletal muscle. Circulation 111(14):1763–1770
6. Meirelles LR, Matsuura C, de Resende A, C, Salgado AA, Pereira
NR, Coscarelli PG et al (2014) Chronic exercise leads to
antiaggregant, antioxidant and anti-inflammatory effects in heart
failure patients. Eur J Prev Cardiol 21(10):1225–1232
7. Niebauer J, Clark AL, Webb-Peploe KM, Böger R, Coats AJ
(2005) Home-based exercise training modulates pro-oxidant sub-
strates in patients with chronic heart failure. Eur J Heart Fail 7(2):
183–188
8. Wiecek M, Maciejczyk M, Szymura J, Szygula Z,
Kantorowicz M (2015) Changes in non-enzymatic antioxi-
dants in the blood following anaerobic exercise in men
and women. LoS One 10(11):1–16
9. Theccanat T, Philip JL, Razzaque AM, Ludmer N, Li J, Xu X et al
(2016) Regulation of cellular oxidative stress and apoptosis by G
protein-coupled receptor kinase-2: the role of NADPH oxidase 4.
Cell Signal 28(3):190–203
10. Sánchez-Marteles M, Rubio GJ, Giménez LI (2015)
Pathophysiology of acute heart failure: a world to know. Rev
Clin Esp 216(1):38–46
11. Ribeiro-Samora GA, Rabelo LA, Ferreira ACC, Favero M, Guedes
GS, Pereira LSM, Parreira VF, Britto RR (2017) Inflammation and
oxidative stress in heart failure: effects of exercise intensity and
duration. Braz J Med Biol Res 7 50(9):e6393
12. Jackson MJ, Vasilaki A, McArdle A (2016) Cellular mechanisms
underlying oxidative stress in human exercise. 98:13–17
13. Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA
Group (2009) Preferred reporting items of systematic review
and meta-analyses: the PRISMA statement. Ann Intern Med
151(4):264–269
14. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M
(2003) Reliability of the PEDro scale for rating quality of random-
ized controlled trials. Phys Ther 83(8):713–721
15. Hambrecht R, Niebauer J, Fiehn E, Kälberer B, Offner B, Hauer K
et al (1995) Physical training in patients with stable chronic heart
Heart Fail Rev
failure: effects on cardiorespiratory fitness and ultrastructural abnor-
malities of leg muscles. J Am Coll Cardiol 25(6):1239–1249
16. Hambrecht R, Fiehn E, Yu J, Niebauer J, Weigl C, Hilbrich L et al
(1997) Effects of endurance training on mitochondrial ultrastructure
and fiber type distribution in skeletal muscle of patients with stable
chronic heart failure. J Am Coll Cardiol 29(5):1067–1073
17. Gielen S, Adams V, Möbius-Winkler S, Linke A, Erbs S, Yu J et al
(2003) Anti-inflammatory effects of exercise training in the skeletal
muscle of patients with chronic heart failure. J Am Coll Cardiol
42(5):861–868
18. Gielen S, Adams V, Linke A, Erbs S, Möbius-Winkler S, Schubert
A et al (2005) Exercise training in chronic heart failure: correlation
between reduced local inflammation and improved oxidative capac-
ity in the skeletal muscle. Eur J Cardiovasc Prev Rehabil 12(4):
393–400
19. Wisloff U, Stoylen A, Loennechen JP, Bruvold M, Rognmo O,
Haram PM et al (2007) Superior cardiovascular effect of aerobic
interval training versus moderate continuous training in heart failure
patients: a randomized study. Circulation 115(24):3086–3094
20. Laurent M, Daline T, Malika B, Fawzi O, Philippe V, Benoit D et al
(2009) Training-induced increase in nitric oxide metabolites in
chronic heart failure and coronary artery disease: an extra benefit
of water-based exercises? Eur J Cardiovasc Prev Rehabil 16(2):
215–221
21. Erbs S, Höllriegel R, Linke A, Beck EB, Adams V, Gielen S
et al (2010) Exercise training in patients with advanced
chronic heart failure (NYHA IIIb) promotes restoration of
peripheral vasomotor function, induction of endogenous re-
generation, and improvement of left ventricular function.
Circ Heart Fail 3(4):486–494
22. Tsarouhas K, Karatzaferi C, Tsitsimpikou C, Haliassos A, Kouretas
D, Pavlidis P et al (2011) Effects of walking on heart rate recovery,
endothelium modulators and quality of life in patients with heart
failure. Eur J Cardiovasc Prev Rehabil 18(4):594–600
23. Aksoy S, Findikoglu G, Ardic F, Rota S, Dursunoglu D (2015))
Effect of 10-week supervised moderate-intensity intermittent vs.
continuous aerobic exercise programs on vascular adhesion mole-
cules in patients with heart failure. Am J Phys Med Rehabil
10(sup.1):898–911
24. Xie B, Yan X, Cai X, Li J (2017) Effects of high-intensity
interval training on aerobic capacity in cardiac patients: a
systematic review with meta-analysis. Biomed Res Int
2017:1–16
25. Nunes-Silva A, Freitas-Lima LC (2015) The association between
physical exercise and reactive oxygen species (ROS) production. J
Sports Med Doping Stud 5(1):1–7
26. Dröge W (2002) Free radicals in the physiological control of cell
function. Physiol Rev 82(1):47–95
27. Barbosa KBF, Costa NMB, Alfenas RCG, De Paula SO, Minim
VPR, Bressan J (2010) Estresse oxidativo: conceito, implicações e
fatores modulatórios. Rev Nutr 23(4):629–643
28. Schneider CD, Oliveira AR (2004) Radicais livres de oxigênio e
exercício: mecanismos de formação e adaptação ao treinamento
físico. Rev Bras Med Esporte 10(4):308–313
29. Dusse LMS, Vieira LM, Carvalho MG (2003) Revisão sobre óxido
nítrico. J Bras Patol Med Lab 39(4):343–350
30. Boger RH (2007) The pharmacodynamics of L-arginine. J Nutr
137(6):1650S–1655S
31. Morris SM Jr (2007) Arginine metabolism: boundaries of our
knowledge. J Nutr 137(6):1602S1609S
32. Siegerink B, Maas R, Vossen CY, Schwedhelm E, Koenig
W, Böger R et al (2013) Asymmetric and symmetric
dimethylarginine and risk of secondary cardiovascular dis-
ease events and mortality in patients with stable coronary
heart disease: the KAROLA follow-up study. Clin Res
Cardiol 102(3):193–202
Heart Fail Rev
33. Cotton JM, Kearney MT, Shah AM (2002) Nitric oxide and myo-
cardial function in heart failure: friend or foe? Heart 88(6):564–566
34. Dusting GJ, Macdonald PS (1995) Endogenous nitric oxide in car-
diovascular disease and transplantation. Ann Med 27(3):395–406
35. Cruzat VF, Rogero MM, Borges MC, Tirapegui J (2007) Aspectos
atuais sobre estresse oxidativo, exercícios físicos e suplementação.
Rev Bras Med Esporte 13(5):336–342
36. Wiezorek J, Brown D, Kupperman D (1994) Rapid conversion to
high xanthine oxidase activity in viable Kupffer cells during hyp-
oxia. J Clin Invest 94(6):2224–2230
37. Moores HK, Beehler CJ, Hanley ME, Shanley PF, Stevens EE,
Repine JE et al (1994) Xanthine oxidase promotes neutrophil se-
questration but not injury in hyperoxic lungs. J Appl Phyisol 76(2):
941–948