ORIGINAL RESEARCH ARTICLE

Aerobic Exercise Training in Very Severe Chronic

Obstructive Pulmonary Disease

A Systematic Review and Meta-Analysis

Mara Paneroni, PT, MSc, Carla Simonelli, PT, Michele Vitacca, MD, FERS, and Nicolino Ambrosino, MD, FERS

Objective: To evaluate the effectiveness of exercise training in patients with very severe chronic obstructive pulmonary disease (COPD).
Design: We searched MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and Cumulative Index to Nursing and Allied Health
Literature databases using the following as search terms: COPD, Chronic Obstructive Pulmonary Disease, Exercise, and Pulmonary
Rehabilitation. We included randomized controlled trials (RCTs) of subjects with forced expiratory volume in the first second of less than 35%
of the predicted normal value enrolled in in-patient, outpatient, or home- or community-based training programs lasting at least 4 weeks with
respect to usual care. We included RCTs with outcome measures including the 6-minute walking test and/or health-related quality of life
assessed by the St. George's Respiratory Questionnaire (SGRQ).
Results: Of 580 articles screened, 10 were included. The programs' duration ranged from 4 to 52 weeks with 1 to 5 sessions per week lasting
15 to 40 minutes each. The intervention group improved in 6-minute walking test [weighted mean difference, 67.1 (95% confidence interval
[CI], 37.897–98.927); standardized mean difference, 3.86 (95% CI, 2.04–5.67)], and St. George's Respiratory Questionnaire [weighted mean
difference, −8.041 (95% CI, −15.273 to −0.809); standardized mean difference, −1.23 (95% CI, −2.14 to −0.31)].
Conclusions: Exercise training improves exercise tolerance and health-related quality of life in patients with very severe COPD. However, because
few studies on severely affected patients are available and the training programs are Highly heterogeneous, larger RCTs are needed.
Key Words: COPD, Chronic Obstructive Pulmonary Disease, Exercise, Pulmonary Rehabilitation
(Am J Phys Med Rehabil 2017;00:00–00)

xercise training is a core component of pulmonary reha-
EEuropean
bilitation (PR). The joint American Thoracic Society/
Respiratory Society guidelines emphasize that
PR including aerobic exercise training improves exercise capac-
ity, dyspnea, and health-related quality of life (HRQL) more so
than programs without exercise for people with chronic ob-
structive pulmonary disease (COPD).1 Advanced COPD is a
life-limiting condition that causes symptoms that dramatically
reduce HRQL. Typically, extreme breathlessness and/or pe-
ripheral muscle fatigue can occur during exercise and reduce
the benefit from PR.2
Although data and recommendations are available for
patients with mild to moderate COPD,3 current guidelines
From the Respiratory Rehabilitation Unit, Fondazione Salvatore Maugeri, IRCCS,
Institute of Lumezzane, Lumezzane (BS), Italy (MP, CS, MV); and Auxilium
Vitae, Volterra, Pisa, Italy (NA).
All correspondence and requests for reprints should be addressed to: Mara Paneroni,
PT, MSc, Fondazione Salvatore Maugeri, IRCCS, Divisione di Pneumologia
Riabilitativa, Via Giuseppe Mazzini 129, 25065 Lumezzane (BS), Italy.
AUTHORSHIP: All authors made substantial contributions to conception and
design, and in analysis and interpretation of data; took part in drafting the article
or revising it critically for important intellectual content; gave final approval of
the version to be published; and agreed to be accountable for all aspects of
the work.
Financial disclosure statements have been obtained, and no conflicts of interest have
been reported by the authors or by any individuals in control of the content of
this article.
Supplemental digital content is available for this article. Direct URL citations appear
in the printed text and are provided in the HTML and PDF versions of this article
on the journal’s Web site (www.ajpmr.com).
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.
ISSN: 0894-9115
DOI: 10.1097/PHM.0000000000000667
provide few recommendations on the modalities and effec-
tiveness of training for more severe patients, namely, those
at Global Initiative for Chronic Obstructive Lung Disease stage
IV, with or without chronic respiratory failure (CRF). A recent
study by Spruit et al.4 reported that the severity of airway flow
limitation did not seem to be related to PR outcomes and some
patients with very severe COPD were considered either as
“very good responders” or “poor responders” independently
of the extent of their airflow obstruction. Even when patients
are stable, it can be difficult to determine whether those with
advanced COPD should be offered PR including exercise
training or not, and the safe and effective intensity of exercise
training has not yet been established. Therefore, the aim of this
systematic review and meta-analysis of randomized controlled
trials (RCTs) was to evaluate, in patients with very severe but
stable COPD, the effectiveness of exercise training defined as
a change in functional capacity and HRQL.
METHODS
This study conforms to all Preferred Reporting Items for
Systematic Reviews and Meta-Analyses guidelines and reports
the required information accordingly (Supplementary Check-
list, http://links.lww.com/PHM/A364).
Data Sources and Search Strategies
We searched the following databases: MEDLINE,
EMBASE, Cochrane Central Register of Controlled Trials,
and the Cumulative Index to Nursing and Allied Health

American Journal of Physical Medicine & Rehabilitation • Volume 00, Number 00, Month 2017 www.ajpmr.com 1

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Paneroni et al.
Literature, from their inception to October 2015, with no lan-
guage restriction. We also reviewed the bibliographies of retrieved
articles for additional studies. We used as search terms COPD,
Chronic Obstructive Pulmonary Disease, Exercise, and Pul-
monary Rehabilitation. The search was limited only to RCTs.
Patients
We included RCTs involving very severe COPD, defined
as the following: (1) clinical diagnosis of COPD according to
the Global Initiative for Chronic Obstructive Lung Disease rec-
ommendations5; (2) mean forced expiratory volume in the first
second(FEV1) of less than 35% of predicted value at baseline.
This level was proposed as it represents the lowest level of the
body mass index (B), the degree of airflow obstruction (O) and
dyspnea (D), and exercise capacity (E)] score.6 Studies in pa-
tients with CRF, defined as the presence of arterial oxygen ten-
sion less than 55 mm Hg and home treatment with oxygen for
at least 6 months before the start of the study were also in-
cluded. We excluded studies involving patients with (1) an
acute exacerbation within 4 weeks before starting the interven-
tion; (2) major comorbidities such as chronic heart failure,
asthma, and sleep-related disorders.
Programs
We included studies of training programs with the follow-
ing characteristics:
- in-patient or outpatient or home- or community-based;
- minimum duration of 4 weeks;
- at least one kind of endurance training included, involving
lower and/or upper limbs.
For the purpose of this review, usual care was defined as
conventional medical care without prescription of exercise train-
ing or physical activity or any form or additional intervention
(such as formal programs of education during the study).
General recommendations on physical activity and lifestyle given
during medical visits were considered as part of usual care.
We excluded studies that included interventions in which
the exercise component was not considered to be aerobically
demanding (eg, respiratory muscle training, tai chi, neuromus-
cular electrical stimulation, vibrating platforms, etc). We also
excluded studies that compared exercise to something that
was not usual care (eg, comparison between two types of exer-
cise, or exercise using positive expiratory pressure compared to
exercise alone, etc).
Outcome Measures
We included any study assessing functional capacity by
the 6-minute walking test (6MWT) and/or HRQL by means
of the Chronic Respiratory Disease Questionnaire or St.
George's Respiratory Questionnaire (SGRQ).
Data Collection and Analysis
Two investigators (C.S., M.P.) independently conducted
the search of the bibliographic databases screening all titles
and/or abstracts for the inclusion criteria. Reviewers then re-
trieved abstracts and/or full-text papers of all potentially eligible
studies and maintained records on all studies that did not meet
the inclusion criteria, providing the rationale for their exclusion.
2 www.ajpmr.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 00, Number 00, Month 2017
The investigators put the data of potentially eligible articles
in a dedicated electronic database with the use of Microsoft
Excel software (2010 version, Microsoft, Redmond, WA). Dis-
agreement between investigators about eligibility was resolved
by discussion and consensus. If consensus could not be
reached, a third investigator (M.V.) adjudicated. All entered
data of the studies finally selected were checked for accuracy.
Missing data were requested by e-mail from the papers'
authors. The lead investigator (C.S.) separated the original
full-text papers of the studies finally selected for inclusion in
the meta-analysis from the others and inserted their data into
a dedicated electronic database. The second investigator (M.P.)
independently extracted data from the same studies.
The information included were background characteristics
of the research reports; characteristics of participants in the
study; number of participants who dropped out or withdrew
from the study; full description of the exercise training pro-
grams (setting, components, duration, and characteristics); ex-
ercise capacity and HRQL outcomes, and associated results. If
a study reported multiple group comparisons (eg, exercise ther-
apy plus urban walking circuit vs exercise therapy alone or vs
conventional community care), treatment groups considered
relevant to exercise capacity were combined into one virtual in-
tervention group, and this group was compared to the group re-
ceiving conventional care.
The investigators assessed papers for bias using the
Cochrane Collaboration's tool for assessing risk of bias in
RCTs.7 Risk of bias was assessed according to the following
domains: sequence generation; allocation concealment; blinding
of outcome assessment; incomplete outcome data; selective out-
come reporting, and other biases. Blinding of participants and
personnel was not considered as a possible bias because the de-
sign of the study (exercise vs no exercise) precluded blinding.
Statistical Analysis
Statistical analysis was performed using STATA version
11.2 software (Stata, College Station, TX). All data were ex-
trapolated from the corresponding full-text studies. For each
outcome (exercise tolerance and HRQL), we recorded mean
and standard deviation (SD) of variation from baseline to the
end of the study. When SD was not available, we used the mean
value of SD of the studies reported for that outcome. We ex-
cluded studies with data other than mean and SD.8
The standardized mean difference (SMD) and weighted
mean difference (WMD) were calculated. Heterogeneity of
studies was assessed by performing the Q-test considering
values of P < 0.01 as significant. A first step of evaluation
was conducted by fixed-effect models. If a significant heteroge-
neity among studies was found, a random effect evaluation by
the Der-Simonian and Laird method approach9 was performed.
Forest plots were used to demonstrate the results. Funnel plots
were used to detect any publication bias.10
RESULTS
We identified 580 potentially relevant articles. Of these,
289 were excluded after the abstracts were read and 291
full-texts were analyzed for inclusion criteria. Ten of the 291
satisfied the inclusion criteria and were included in the
meta-analysis (Fig. 1).
© 2017 Wolters Kluwer Health, Inc. All rights reserved.
Volume 00, Number 00, Month 2017 Aerobic Exercise Training in Very Severe COPD

FIGURE 1. Flow chart of the study.

Table 1 shows the baseline characteristics of the included test) and alternating between them every 2 minutes. A low-
subjects. They had severe COPD with at least moderate impair- intensity interval training was used in the study by Regiane
ment of functional capacity. They also had moderate dyspnea Resqueti et al.,13 ie, cycling at a maximum load of 30 W alter-
as assessed by the Medical Research Council score and had a nating with periods of rest. Pleguezuelos et al.19 began from
high impact of disease on the HRQL. Almost half [46.9% moderate intensity (50 W) and increased the load according
(33.3%)] were in CRF. to symptoms; free intensity was reported in two studies,12,17
Outcome measures in the included studies are described in whereas two others18,20 did not specify any training intensity.
Table 2. The 6MWT was used in 8 of 10 studies,11,12,15–20 and
HRQL was assessed by the SGRQ in five studies11,14,16,18,20
and by the Chronic Respiratory Disease Questionnaire in two.12,13 Duration
Table 3 shows the main characteristics of all interventions The total duration of programs ranged from 4 to 52 weeks.
carried out in the studies considered. The duration of the sessions ranged from 25 to 30 minutes in
six studies,11,12,15,16,19,20 and was less than 15 minutes in two
Setting
The settings differed, with four studies12,14,15,19 conducted
mainly in an outpatient setting, four at home,13,16,18,20 and two TABLE 1. Anthropometric and clinical characteristics of
included patients
studies11,17 both in the outpatient setting and home.
No. Studies Preintervention
Frequency Characteristics (No. Participants) Mean (SD)
There were one to five sessions per week. All 10 Age, years 10 (458) 65.6 (6.7)
studies11–20 included leg exercise training: cycling in three BMI, kg/m2 7 (355) 25.53 (3.92)
studies,11,13,19 treadmill walking in one,15 free walking in FEV1, % of predicted 10 (458) 31.52 (8.59)
two,12,16 and a combination of these in four.14,17,18,20 Only FVC, % of predicted 6 (318) 57.28 (13.27)
one study12 included structured arm training (lifting dumbbells). 6MWT, m 8 (396) 281.40 (70.50)
MRC, score 2 (63) 2.80 (0.65)
Intensity CRDQ, dyspnea score 2 (78) 3.24 (0.85)
Exercise was of high intensity ranging from 70% to 90% SGRQ, total score 5 (182) 53.10 (11.26)
of the maximum load or velocity reached during incremental CRF patients 6 (233) —
tests in three studies.14–16 Engstrom et al.11 used high-intensity BMI, body mass index; CRDQ, Chronic Respiratory Disease Questionnaire;
bilevel training consisting of cycling at two levels of intensity FVC, forced vital capacity; N, number; PaO2, arterial O2 tension.
(42% and 85% of the maximum load obtained on an incremental

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.ajpmr.com 3

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Paneroni et al.
TABLE 2. Outcome measures
Reference Spirometry 6MWT MRC CRDQ SGRQ
11
Engstrom et al X X
Singh et al12 X X X
Regiane Resqueti et al13 X X
Paz-Diaz et al14 X X
Borghi-Silva et al15 X Cardiopulmonary exercise test, electrocardiography
Fernandez et al16 X X X
Theander et al17 X X
Ghanem et al18 X X SAS
Pleguezualos et al19 X Physical activity, MIP, MEP, BODE index
De Sousa Pinto et al20 X X
BODE score, body mass index (B), the degree of airflow obstruction (O) and dyspnea (D), and exercise capacity (E) (BODE)] score; CRDQ, Chronic Respiratory
Disease Questionnaire; MEP, maximal expiratory pressure; MIP, maximal inspiratory pressure; MRC, Medical Research Council (assessment of dyspnea); SAS,
self-administered scale (alternative version of CRDQ).
other studies.13,17 In the study by Paz-Diaz et al.,14 the session
duration was 40 minutes, divided into 10 minutes of treadmill
walking, 10 minutes of cycling, and 20 minutes of arm exercises.
The study by Ghanem et al.18 did not specify exercise time.
Other Rehabilitation Program Components
Eight studies11,13–18,20 also incorporated upper and/or
lower limb strengthening exercises, seven11–14,16,18,20 included
breathing exercises (eg, pursued lip breathing and diaphrag-
matic breathing), one14 included relaxation techniques, and
four11,14,15,20 included stretching and/or flexibility/mobility
exercises. Seven studies11–14,17,19 included structured educa-
tional sessions. Secretion management12 and inspiratory mus-
cle training16 were included in one study only.
Outcomes
Functional Capacity
Figure 2 shows the forest plot for the eight studies9,10,15–20
(396 patients: 207 treatments and 189 controls) assessing func-
tional capacity via 6MWT. Owing to high heterogeneity in the
“fixed analysis model” (P = 0.0001), a random effect model
was performed. The intervention group, performing aerobic
exercise training, improved significantly more than the control
group [WMD, 67.066 (95% CI, 35.205–98.927), SMD, 3.86
(95% CI, 2.04–5.67)]. Only the studies by Borghi-Silva et al.15
and Ghanem et al.18 repeated the 6MWT twice at each time of
their evaluations with the aim of avoiding the learning effect
that can account for improvement of that parameter.
HRQL
Figure 3 shows the forest plot for the five stud-
ies11,14,16,17,20 (182 patients: 98 treatments and 84 controls)
reporting effects on HRQL measured by the SGRQ. For the
analyses, we used the total score of the questionnaire. Owing
to high heterogeneity in the statistical analysis (P = 0.0001),
a random effect model was performed. It showed that the inter-
vention group improved significantly more than the control
4 www.ajpmr.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 00, Number 00, Month 2017
Others
Cardiopulmonary exercise test, fat free mass, sickness impact profile, mood
adjective check list, hospitalization
Body mass index, 3-minute walk test
Beck Depression Inventory, State Trait Anxiety Inventory
Body mass index, handgrip, Fatigue Impact Scale, Canadian Occupational
Performance Measure, Medical Outcomes Survey Short Form SF-36
Medical Outcomes Survey Short Form SF-36
London Chest Activity of Daily Living, Sickness Impact Profile
group [WMD, −8.041 (95% CI, −15.273 to −0.809)]; SMD,
−1.23 (95% CI, −2.14 to −0.31).
Although five studies included patients with CRF and ox-
ygen supplementation during exercise,13,15,16,18,20 only one of
them16 evaluated patients with CRF alone. For this reason,
the SMD and WMD could not be calculated.
The assessment of the risk of bias by funnel plots revealed
low quality in the design of the included studies. Table 4 shows
results of the risk of bias assessment assessed by the Cochrane
collaboration tool7 modified by the Jadad Scale score.10 Most
of the studies did not report any details on randomization and
allocation modalities, and only the study by Engstrom et al.11
specified that the operator was blinded to patients' allocation.
Furthermore, the group of patients investigated was often small
or sample size analysis was not performed, and data on with-
drawals and dropouts were often missing.
DISCUSSION
This meta-analysis supports the effectiveness of exercise
training in improving functional capacity and HRQL for very
severe but stable patients with COPD whether or not in CRF.
Late-stage COPD is characterized by a fluctuating course with
frequent exacerbations and recurrent hospitalizations, a vicious
cycle of increasing dyspnea resulting in decreased exercise tol-
erance and activity, depression, and social isolation, all condi-
tions associated with poor quality of life and high risk of
mortality.21 Skeletal muscle wasting in these patients contrib-
utes to muscle fatigue during exercise, causing patients to stop
exercising even before reaching their aerobic capacity.22
One study on a large cohort reported that exercise train-
ing can be effective for patients with severe as well as mild
COPD,23 supporting the prescription of exercise training for
them. Unfortunately, current guidelines1 are mainly based
on the evidence from studies conducted in mild to moderate
COPD and do not suggest prescriptions for patients with se-
vere COPD. This current study highlights the type of training
that has been used in severe COPD and suggests that high-
intensity endurance exercise, mainly in a continuous training
© 2017 Wolters Kluwer Health, Inc. All rights reserved.
Volume 00, Number 00, Month 2017 Aerobic Exercise Training in Very Severe COPD

TABLE 3. Characteristics of the interventions

FEV1 Mean (Range), Duration, Withdrawal/

Reference % Predicted Weeks Dropouts Participants, n Intervention and Control Intervention Details
11
32 (19.3–44.3) 18 5 26 Intervention: outpatient sessions 30 min of cycling bilevel training
including breathing techniques (bouts of 2 min) at 42%–85%
and bicycle training + home of maximum load reached
program of strength training and during an incremental test,
mobility exercises + educational adjusted according to
sessions with occupational symptoms
therapist, dietician and nurse,
1–2 days/wk
24 Control: usual medical care.
12
27 (18.9–35.5) 4 Not available 20 Intervention: outpatient sessions 30 min of walking at a
including breathing techniques, submaximal speed, twice a day.
removal of secretions and
education on energy conservation
techniques, + home walking
exercise, 2 days/week.
20 Control: usual medical care.
13
29 (18.5–37.6) 9 13 19 Intervention: supervised home Interval Training (bouts of 5 min
program including educational exercise and 2 min rest) with
sessions, respiratory physiotherapy, maximum load 30 watts,
muscle training and cycling duration 5–15 min
exercise training, 5 days/wk
19 Control: educational sessions during
hospitalization, no treatments
after discharge
14
32 (21–45) 8 Not available 10 Intervention: outpatient sessions 40 min of aerobic exercises at
including disease education, 80% VO2 max: 10 min
breathing exercises, flexibility, treadmill + 10 min
aerobic exercise for lower and cycling + 20 min lifting
upper limbs, 3 days/week. dumbbells with upper limbs
14 Control: usual medical care.
15
34 (24–46) 6 6 20 Intervention: outpatient sessions 30 min of treadmill ambulation
including aerobic training with at 70% of the maximal speed
treadmill ambulation and reached during an incremental
stretching, 3 days/week test
14 Control: usual medical care.
16
35.5 (24–50) 52 9 27 Intervention: partially supervised 30 min of walking at 90% of the
all with CRF home program including respiratory maximal speed reached in
re-education, inspiratory muscle the 6MWT.
training, upper limb strength training,
lower limb strength training, aerobic
exercise, 5 days/wk.
14 Control: 3 sessions of education, and
usual medical care.
17
33.5 (22.8–42.7) 12 4 12 Intervention: outpatient and home 10–15 min of bicycle training
multidisciplinary rehabilitative (plus home free daily walks)
program including training
(aerobic bicycle training, strength
exercises and walking), plus
nutritional, occupational and
educational interventions,
2 days/wk
14 Control: usual medical care.
18
26 (15.5–42.6) 8 Not available 25 Intervention: home program Walking and cycling
including respiratory exercises,
endurance training and strength
training, 5 days/wk

(Continued on next page)

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.ajpmr.com 5

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Paneroni et al. Volume 00, Number 00, Month 2017

TABLE 3. (Continued)

FEV1 Mean (Range), Duration, Withdrawal/

Reference % Predicted Weeks Dropouts Participants, n Intervention and Control Intervention Details
14 Control: usual medical care.
19
31.5 (30.8–33.2) 12 35 54 Intervention: outpatient sessions 25 min of cycling exercise,
including aerobic cycling exercise, starting at 50 W of load and
strength training and relaxation increasing according to
(plus a walking urban circuit in a symptoms (plus walking
subgroup of 34 patients), urban circuit in 34 patients)
3 days/week.
71 Control: education on physical
activity and lifestyle, and usual
medical care
20
33.5 (25–44) 12 9 23 Intervention: supervised home 5–30 min of aerobic exercise
program including breathing including one modality
and stretching exercises, between: walking, stair
endurance training, strength climbing, cycling and
training, at least 2 days/week. treadmill walking
18 Control: a nursing counselling
session and usual care

mode by cycling or walking, can be prescribed and beneficial
for severely affected patients. Similarly as for patients with
mild severity, it seems that using 70% to 90% of the maxi-
mum loads or velocities reached during incremental testing
may be a good strategy for determining the exercise prescrip-
tion for this patient population as well.14–16
To compare the efficacy of the exercise programs, the fol-
lowing minimal clinically important differences (MCID) were
used: +33 m on 6MWT24; and −4 points on SGRQ.25 Remark-
ably, we found an increase in 6MWT of 67 m after PR in the
reviewed studies. This value was higher than the MCID previ-
ously described24 and higher than the learning effect on
6MWTs described in the literature.26 Significant increases in
6MWT have been associated with reduced risk of acute exac-
erbations, hospitalizations, and death, independently of airflow
obstruction.27 Similarly, the decrease of 8 points observed in
the WMD of SGRQ was twice that reported for MCID.2
The outcome measures described in Table 2 indicate con-
siderable heterogeneity of tests and exercise strategies. This
heterogeneity of approaches has also been reported for less se-
vere patients. It, therefore, continues to be necessary to estab-
lish optimal assessment instruments as well as exercise
strategies for patients with advanced COPD.23
Spruit et al.4 observed that patients undergoing PR may be
clustered in “very good”, “good”, “moderate,” and “poor” re-
sponder groups according to their multidimensional response
profiles. The very good responders had more severe dyspnea;
more hospitalizations; worse exercise tolerance, performance,
and life satisfaction scores; more severe anxiety and depres-
sion; worse health status; and were more likely to be in an
in-patient program compared with the other three patient cat-
egories. Although the best responders have been reported to
be the more severely affected, no covariates related to tests
on lung function were entered into their prediction analyses.

FIGURE 2. Forest plot of SMD on results of the 6MWT.

6 www.ajpmr.com © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 00, Number 00, Month 2017 Aerobic Exercise Training in Very Severe COPD

FIGURE 3. Forest plot of SMD on results of the SGRQ.

Moreover, patient population of Spruit et al. seems to have
been less severe than those included in our meta-analysis.4
Limitations
Although exercise training has been shown to be effective
in improving exercise capacity and HRQL in this meta-
analysis, the results should be interpreted with caution because
the meta-analysis included very few studies and the sample
populations were small. Some of the included studies proposed
training programs involving free walking or low-intensity in-
terval training that are not recommended by the more recent
guidelines on aerobic training in COPD. Limiting the inclusion
criteria to studies that propose only criterion-standard training
protocols could have led to better results.
Furthermore, we selected studies that reported a mean value
of FEV1 of less than 35% predicted, but this means that some
patients with a higher actual value of FEV1 could have been
included in the present meta-analysis. An analysis with stricter
inclusion criteria might have yielded more definitive results.
Another aspect to consider regards comorbidities of pa-
tients with severe COPD. Most studies so far have enrolled
patients with “pure” COPD and excluded patients with co-
morbidities (that might limit exercise performance) and pa-
tients after acute exacerbations of COPD or after surgery.
However, since it is quite rare to find patients with advanced
COPD without any comorbidities, more complex and frail pa-
tients than those included in this meta-analysis might not de-
rive the same benefits.28,29 Furthermore, we did not assess for
chronic hypercapnia. Nevertheless, Foster et al.30 have re-
ported that hypercapnic patients with COPD can also signifi-
cantly improve their exercise capacity. Also excluded were
new exercise training strategies such as interval training,
training including the use of noninvasive ventilation, helium,
etc.3 that have been reported to help improve exercise perfor-
mance in severe COPD.
Finally, the methodological level of the included studies
was low with high dropout and withdrawal rates (33%–68%)
as well as a high heterogeneity in the frequency, duration, in-
tensity, and type of exercise training. These methodological

TABLE 4. Results of the risk of bias assessment using the Cochrane collaboration tool and modified Jadad Scale score,10 of the
included studies

Random Generation Blinding of Participants

of Allocation Sequence Allocation Concealment And Personnel Incomplete Selective Modified by
Reference Generation (Selection Bias) (Selection Bias) (Performance Bias) Outcome Data Outcome Reporting Jadad Scale
11
Uncertain Uncertain High Low Low 2
12
Uncertain Uncertain High High High 1
13
Uncertain Low High Low Low 2
14
Uncertain High High Low High 2
15
Uncertain Uncertain High Low High 1
16
Low Low High Low High 3
17
Low Low Uncertain High Low 3
18
High Uncertain High High High 0
19
High High Uncertain High Low 2
20
Low Low High Low Low 3
Low means that there is plausible bias unlikely to seriously alter the results; High means that plausible bias exists that can seriously weaken confidence in the
results; Uncertain is stated if there are unclear risks of bias or if plausible bias can raise some doubt about the results.

© 2017 Wolters Kluwer Health, Inc. All rights reserved. www.ajpmr.com 7

Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Paneroni et al.
flaws strongly suggest the need for further large RCTs for
patients with very severe COPD.
CONCLUSION
Exercise training apparently improves functional capacity
and HRQL in patients with very severe COPD. The evidence is
limited, however, because of the few applicable studies and the
heterogeneity of their methodologies and outcome measures.
Large RCTs for patients with severe and very severe COPD
are needed.
ACKNOWLEDGMENTS
The authors thank Rosemary Allpress for the English
revision of the paper.
REFERENCES
1. Spruit MA, Singh SJ, Garvey C, et al: ATS/ERS Task Force on Pulmonary Rehabilitation. An
official American Thoracic Society/European Respiratory Society statement: key concepts
and advances in pulmonary rehabilitation. Am J Respir Crit Care Med 2013;188:e13–64
2. Ambrosino N, Strambi S: New strategies to improve exercise tolerance in chronic obstructive
pulmonary disease. Eur Respir J 2004;24:313–22
3. McCarthy B, Casey D, Devane D, et al: Pulmonary rehabilitation for chronic obstructive
pulmonary disease. Cochrane Database Syst Rev 2015:2:CD003793
4. Spruit MA, Augustin IM, Vanfleteren LE, et al: Differential response to pulmonary
rehabilitation in COPD: multidimensional profiling. Eur Respir J 2015;46:1625–35
5. Vestbo J, Hurd SS, Agustì AG, et al: Global strategy for the diagnosis, management, and
prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir
Crit Care Med 2013;187:347–65
6. Celli BR, Cote CG, Marin JM, et al: The body-mass index, airflow obstruction, dyspnea, and
exercise capacity index in chronic obstructive pulmonary disease. N Engl J Med
2004;350:1005–12
7. Higgins JP, Altman DG, Gøtzsche PC, et al: The Cochrane Collaboration's tool for assessing
risk of bias in randomised trials. BMJ 2011;343:d5928
8. Cochrane WG: The combination of estimates from different experiments. Biometrics
1954;10:101–29
9. Der-Simonian R, Laird NM: Meta-analysis in clinical trials. Control Clin Trials
1986;7:177–88
10. Olivo SA, Macedo LG, Gadotti IC, et al: Scales to assess the quality of randomized controlled
trials: a systematic review. Phys Ther 2008;88:156–75
11. Engström CP, Persson LO, Larsson S, et al: Long-term effects of a pulmonary rehabilitation
programme in outpatients with chronic obstructive pulmonary disease: a randomized
controlled study. Scand J Rehabil Med 1999;31:207–13
8 www.ajpmr.com
Copyright © 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Volume 00, Number 00, Month 2017
12. Singh V, Khandelwal DC, Khandelwal R, et al: Pulmonary rehabilitation in patients with
chronic obstructive pulmonary disease. Indian J Chest Dis Allied Sci 2003;45:13–7
13. Regiane Resqueti V, Gorostiza A, Gáldiz JB: [Benefits of a home-based pulmonary
rehabilitation program for patients with severe chronic obstructive pulmonary disease].
Arch Bronconeumol 2007;43:599–604
14. Paz-Díaz H, Montes de Oca M, López JM, et al: Pulmonary rehabilitation improves
depression, anxiety, dyspnea and health status in patients with COPD. Am J Phys Med Rehabil
2007;86:30–6
15. Borghi-Silva A, Arena R, Castello V, et al: Aerobic exercise training improves autonomic
nervous control in patients with COPD. Respir Med 2009;103:1503–10
16. Fernandez AM, Pascual K, Ferrando C, et al: Home-based pulmonary rehabilitation in very
severe COPD: is it safe and useful? J Cardiopulm Rehabil Prev 2009;29:325–31
17. Theander K, Jakobsson P, Jörgensen N, et al: Effects of pulmonary rehabilitation on fatigue,
functional status and health perceptions in patients with chronic obstructive pulmonary
disease: a randomized controlled trial. Clin Rehabil 2009;23:125–36
18. Ghanem M, ELaal EA, Mehany M, et al: Home-based pulmonary rehabilitation program:
effect on exercise tolerance and quality of life in chronic obstructive pulmonary disease
patients. Ann Thorac Med 2010;5:18–25
19. Pleguezuelos E, Perez ME, Guirao L, et al: Improving physical activity in patients with COPD
with urban walking circuits. Respir Med 2013;107:1948–56
20. De Sousa Pinto JM, Martin-Nogueras AM, Calvo-Arenillas JI, et al: Clinical benefits of
home-based pulmonary rehabilitation in patients with chronic obstructive pulmonary disease.
J Cardiopulm Rehabil Prev 2014;34:355–9
21. Ambrosino N, Simonds A: The clinical management in extremely severe COPD. Respir Med
2007;101:1613–24
22. Gea J, Pascual S, Casadevall C, et al: Muscle dysfunction in chronic obstructive pulmonary
disease: update on causes and biological findings. J Thorac Dis 2015;7:E418–38
23. Carone M, Patessio A, Ambrosino N, et al: Efficacy of pulmonary rehabilitation in chronic
respiratory failure (CRF) due to chronic obstructive pulmonary disease (COPD): The Maugeri
Study. Respir Med 2007;101:2447–53
24. Singh SJ, Puhan MA, Andrianopulos V, et al: An official systematic review of the European
Respiratory Society/American Thoracic Society: measurement properties of field walking
tests in chronic respiratory disease. Eur Respir J 2014;44:1447–78
25. Schünemann HJ, Griffith L, Jaeschke R, et al: Evaluation of the minimal important difference
for the feeling thermometer and the St. George's Respiratory Questionnaire in patients with
chronic airflow obstruction. J Clin Epidemiol 2003;56:1170–6
26. Hernandes NA, Wouters EF, Meijer K, et al: Reproducibility of 6-minute walking test in
patients with COPD. Eur Respir J 2011;38:261–7
27. Moy ML, Gould MK, Liu IA, et al: Physical activity assessed in routine care predicts mortality
after a COPD hospitalisation. ERJ Open Res 2016;2:00062–2015
28. Mesquita R, Vanfleteren LE, Franssen FM, et al: Patients with COPD and multiple
comorbidities can still benefit from pulmonary rehabilitation. Eur Respir J 2015;46:545–8
29. Franssen FM, Rochester CL: Comorbidities in patients with COPD and pulmonary
rehabilitation: do they matter? Eur Respir Rev 2014;23:131–41
30. Foster S, Lopez D, Thomas HM: Pulmonary rehabilitation in COPD patients with elevated
pCO2. Am Rev Respir Dis 1988;138:1519–23
© 2017 Wolters Kluwer Health, Inc. All rights reserved.