Aerobic and Strength Training in Patients with

Chronic Obstructive Pulmonary Disease

SARAH BERNARD, FRANÇOIS WHITTOM, PIERRE LEBLANC, JEAN JOBIN, ROGER BELLEAU,
CHANTAL BÉRUBÉ, GUY CARRIER, and FRANÇOIS MALTAIS
Unité de Recherche, Institut de Cardiologie et de Pneumologie de Québec, Université Laval, Ste-Foy, Québec, Canada

The purpose of this study was to evaluate whether strength training is a useful addition to aerobic
training in patients with chronic obstructive pulmonary disease (COPD). Forty-five patients with
moderate to severe COPD were randomized to 12 wk of aerobic training alone (AERO) or combined
with strength training (AERO 1 ST). The AERO regimen consisted of three weekly 30-min exercise
sessions on a calibrated ergocycle, and the ST regimen included three series of eight to 10 repetitions
of four weight lifting exercises. Measurements of peripheral muscle strength, thigh muscle cross-sec-
tional area (MCSA) by computed tomographic scanning, maximal exercise capacity, 6-min walking
distance (6MWD), and quality of life with the chronic respiratory questionnaire were obtained at
baseline and after training. Thirty-six patients completed the program and constituted the study
group. The strength of the quadriceps femoris increased significantly in both groups (p , 0.05), but
the improvement was greater in the AERO 1 ST group (20 6 12% versus 8 6 10% [mean 6 SD] in the
AERO group, p , 0.005). The thigh MCSA and strength of the pectoralis major muscle increased in
the AERO 1 ST group by 8 6 13% and 15 6 9%, respectively (p , 0.001), but not in the AERO group
(3 6 6% and 2 6 10%, respectively, p . 0.05). These changes were significantly different in the two
study groups (p , 0.01). The increase in strength of the latissimus dorsi muscle after training was
modest and of similar magnitude for both groups. The changes in peak exercise work rate, 6MWD,
and quality of life were comparable in the two groups. In conclusion, the addition of strength train-
ing to aerobic training in patients with COPD is associated with significantly greater increases in mus-
cle strength and mass, but does not provide additional improvement in exercise capacity or quality of
life. Bernard S, Whittom F, LeBlanc P, Jobin J, Belleau R, Bérubé C, Carrier G, Maltais F. Aero-
bic and strength training in patients with chronic obstructive pulmonary disease.
AM J RESPIR CRIT CARE MED 1999;159:896–901.

Although exercise training is now recognized as an essential useful addition to whole-body aerobic training in patients with
component of pulmonary rehabilitation (1, 2), there is no con- COPD. In addition to improving peripheral muscle function,
sensus about the optimal training strategy for this purpose (3). this training modality may enhance exercise tolerance in pa-
Lower-extremity aerobic training consistently improves exer- tients with COPD. A greater strength of the quadriceps femo-
cise tolerance in patients with chronic obstructive pulmonary ris muscle after training may reduce the perception of muscle
disease (COPD), but has little effect on muscle atrophy and fatigue, a common limiting symptom during exercise in pa-
weakness, two problems common in patients with COPD and tients with COPD (5, 8). In older normal subjects, strength
which can contribute to their poor exercise tolerance and qual- training has also been associated with modest but significant
ity of life (4–6). increases in skeletal muscle oxidative capacity (9), another po-
Strength training can promote muscle growth and strength- tential factor involved in exercise intolerance in patients with
ening in normal subjects (7), and may therefore represent a COPD (10).
Simpson and colleagues have reported that 8 wk of
strength training produces an improvement in muscle strength
and in submaximal exercise tolerance in patients with COPD
(6). In patients with coronary heart disease, strength and aero-
(Received in original form July 8, 1998 and in revised form September 17, 1998) bic training have complementary effects on peripheral muscle
Supported in part by the Fonds de la Recherche en Santé du Québec and by la function and exercise capacity (11). On the basis of these ob-
Fondation J. D. Bégin, Université Laval.
servations, we were interested in evaluating whether strength
Dr. Maltais is a Clinician-Scientist of the Fonds de la Recherche en Santé du
training would provide any additional benefits in COPD pa-
Québec.
tients involved in an aerobic training program. Accordingly,
Correspondence and requests for reprints should be addressed to Dr. François
Maltais, Centre de Pneumologie, Hôpital Laval, 2725 Chemin Ste-Foy, Ste-Foy,
the aim of the present study was to compare the effects of aer-
QC, G1V 4G5 Canada. E-mail: medfma@hermes.ulaval.ca obic training alone and in combination with strength training
Am J Respir Crit Care Med Vol 159. pp 896–901, 1999 on peripheral muscle mass and strength, exercise tolerance,
Internet address: www.atsjournals.org and quality of life in patients with COPD.
Bernard, Whittom, LeBlanc, et al.: Exercise Training in COPD 897

METHODS leagues (13), Goldman and Becklake (14), and Cotes and Hall (15),
respectively. Only spirometry was repeated at the second evaluation.
Subjects Computed tomography. A computed tomographic scan of the right
Forty-five patients with COPD (12 females and 33 males) were re- and left thigh, halfway between the pubic symphysis and the inferior
cruited for this study. The diagnosis of COPD was based on smoking condyle of the femur, was done with a fourth-generation scanner
history and on pulmonary function tests showing irreversible bron- (Model 900S; Toshiba Inc., Tokyo, Japan). Each image was 10 mm
chial obstruction (12–15). The patients were stable at the time of entry thick and was taken at 120 kV and 200 mA with a scanning time of 1 s
into the exercise program, and none had clinical evidence of exercise- while the subject was lying in the supine position. The thigh muscle
limiting cardiovascular or neuromuscular diseases. The research pro- cross-sectional area (MCSA) was obtained by measuring the surface
tocol for the study was approved by the ethics committee of our insti- area of the tissue with a density of 40 to 100 Hounsfield units (HU).
tution, and written consent was obtained for each participant. This range of density was chosen because it corresponds to the density
of muscle tissue (17) and eliminates the bone and fat components. All
Study Design computed tomographic images were analyzed blindly by one investi-
gator (G.C.). The bilateral thigh MCSA was calculated by adding the
In the rehabilitation program conducted at our institution, patients values for the right and left thighs, and was used for data analysis.
are recruited in blocks of 12 to 14. They are then divided into two Strength measurements. Measurement of maximal voluntary
groups in order to facilitate supervision and coaching during exercise strength of the quadriceps, femoris, pectoralis major, and latissimus
sessions. For the purpose of our study, four blocks of patients were dorsi muscles was made during dynamic contractions against hydrau-
successively recruited. Patients in the first three blocks were randomly lic resistance (HF STAR, Hydrafitness Total Power; Henley Health
divided into two groups for each block, who underwent either aerobic Care, Belton, TX), as previously reported (18). The strength of the
training alone (AERO; n 5 19) or combined with strength training three muscle groups was measured with the subject seated comfort-
(AERO 1 ST; n 5 17). Only nine patients could be recruited for the ably and performing a rapid and powerful movement of the chosen
fourth block. To facilitate their supervision during training, it was de- body segment. The strength of the lower limbs was measured during
cided to assign these nine patients to only one training modality, bilateral knee extension (mainly involving the quadriceps femoris);
AERO 1 ST. The choice of this training modality was made in a ran- that of the shoulder girdle was measured during a seated press
dom manner (toss of a coin). Accordingly, the AERO group included (mainly involving the pectoralis major muscles) and a bilateral move-
19 patients, whereas the AERO 1 ST group contained 26 patients. ment combining elbow flexion and shoulder adduction (mainly in-
volving the latissimus dorsi muscles). To ensure that the best possible
Exercise Training efforts were measured, subjects were carefully instructed to exert
The 12-wk training program included a period of aerobic training that maximal effort at high velocity for each of six resistance levels, and
was supplemented with a 45-min period of relaxation and breathing were closely supervised during the procedure. The three lowest hy-
exercises for the AERO group and a 45-min period of strength train- draulic levels were used to familiarize the subjects with the technique.
ing for the AERO 1 ST group. The relaxation was done in a semire- The same movements were repeated at the higher levels until the gen-
cumbent position, and the breathing exercises consisted of diaphrag- erated strength reached a plateau. Two sets of measurements, sepa-
matic and pursed-lip breathing and effective coughing techniques. rated by 30 min, were made for each muscle group, and the highest of
During the training sessions, each group was closely supervised by a the two values obtained was reported.
physiotherapist or an exercise physiologist. Exercise test. After an arterial cannula was placed in a radial ar-
Aerobic training. The aerobic training consisted of leg exercise on tery, subjects were seated on an electrically braked ergocycle (Quiton
a calibrated ergocycle (Monark; Monark-Crescent, Varberg, Sweden) Corival 400; A-H Robins Company, Seattle, WA) and connected to
for 30 min in each of three weekly sessions. We aimed at high-inten- the exercise circuit through a mouthpiece. The exercise circuit con-
sity training, and the work rate corresponding to 80% of the peak sisted of a pneumotachograph, O2 and CO2 analyzers, and a mixing
work rate achieved during the baseline incremental exercise test was chamber (Quinton Qplex; A-H Robins). After 5 min of rest, the pa-
selected as the target training intensity. The physiotherapist or exer- tient performed a progressive stepwise exercise test to maximum indi-
·
cise physiologist was aware of the training intensity prescribed for vidual capacity. Five-breath averages of ventilation ( V E), oxygen up-
· ·
each patient, and encouraged him or her to reach it. The intensity and take (V O2), and CO2 exhalation (V CO2) were measured at rest and
duration of training were monitored as previously described (16). during exercise. Each exercise step lasted 1 min, and stepwise incre-
Supplemental oxygen was used during training for five patients in the ments of 10 W were used. During exercise, arterial blood was sampled
AERO group and six in the AERO 1 ST group for persistent daytime at 1-min intervals for determination of the lactic acid concentration.
hypoxia (n 5 3) or exercise-induced oxygen desaturation (SaO2 , Blood samples were placed on ice until the end of the exercise test, at
90%) (n 5 8). which time they were centrifuged at room temperature. Plasma lactic
Strength training. The strength training program included different acid concentrations were measured with an enzymatic technique (Kit
exercises involving four muscle groups, which were performed with lactate; Boehringer Mannheim, Mannheim, Germany). Blood gases
the following weight-lifting procedures: (1) a seated press (mainly for were also analyzed with the patient at rest and at maximal exercise ca-
·
strengthening of the pectoralis major muscle); (2) a bilateral move- pacity. The peak values for V O2 and exercise work rate were related
ment combining elbow flexion and shoulder adduction (mainly for the to the normal values of (19).
latissimus dorsi); (3) a leg press (mainly for the gluteus maximus); and 6MWD. The 6MWD test was done in a long hall in which the pa-
(4) a bilateral knee extension (mainly for the vastus lateralis muscle). tients could walk freely around two cones separated by 20 m. They
During the first 2 wk of the program patients were asked to perform were asked to walk as far as possible for 6 min with standardized en-
two sets of eight to 10 repetitions of each exercise at a workload of couragement (20). Three trials were done, and the best of the three
60% of the one repetition maximum (1 RM). This was progressively was used for data analysis.
increased to three sets of eight to 10 repetitions at 80% of 1 RM. Quality of life. Health-related quality of life was assessed with the
Thereafter, the training workload was increased when more than 10 Chronic Respiratory Questionnaire (CRQ) (21). A recently validated
repetitions per set could be performed. French Canadian version was used (22), and was administered by a
trained interviewer.
Evaluation
Subjects were evaluated 1 or 2 wk before and at the end of the 12-wk Data and Statistical Analysis
exercise training program. Values are reported as means 6 SD. The average training intensity
Pulmonary function tests. Standard pulmonary function tests, in- for a given training week was obtained by dividing the total work per-
cluding spirometry, lung volume measurement, and measurement of formed during that week by the duration of the training during the
the single-breath diffusion capacity for carbon monoxide (DLCO), week. Physiologic parameters during exercise were compared at peak
were conducted at baseline according to previously described guide- exercise and also at an identical exercise work rate (isoexercise) that
lines (12), and were related to normal values of Knudson and col- corresponded to the greatest exercise work rate achieved at both of
898 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 159 1999

the two study evaluations. For each group, pre- and post-training for the two groups. The strength training objectives were
comparisons were made, using a repeated measures design. The mag- achieved for the four muscle exercises in the AERO 1 ST
nitude of the pre- and posttraining changes in the two study groups group, as indicated by the number of sets and repetitions per
was compared through the use of a two-way analysis of variance set performed for each muscle exercise and the training work-
(ANOVA) (group, training effect), with repeated measures for the
loads, which were close to their target values. No injury oc-
second factor (training effect). A value of p , 0.05 was considered sta-
tistically significant. curred with either training modality.
Effects of Exercise Training on
RESULTS Peripheral Muscle Function
Nine patients (four in the AERO group and five in the AERO 1 In the AERO 1 ST group, there was an 8% increase in bilat-
ST group) failed to complete the program because of exacer- eral thigh MCSA after training, from 160 6 36 cm2 to 171 6 38
bations of COPD requiring hospitalization (n 5 3), lower limb cm2 (p , 0.0001); a 20% increase in the strength of the quadri-
injury unrelated to the program (n 5 1), surgery that could not ceps, from 57 6 20 kg to 67 6 21 kg (p , 0.0001); a 15% in-
be postponed (n 5 3), amyotrophic lateral sclerosis that was crease in the strength of the pectoralis major, from 64 6 16 kg
diagnosed at the beginning of the program (n 5 1), and uncon- to 73 6 17 kg (p , 0.0001), and an 8% increase in the strength
trolled diabetes mellitus (n 5 1). of the latissimus dorsi, from 53 6 12 kg to 56 6 11 kg (p ,
Baseline characteristics of the remaining 36 patients, who 0.05). The quadriceps strength also increased significantly af-
constituted the study population, are presented in Table 1. On ter training in the AERO group, from 51 6 14 kg to 55 6 15
average, patients had a normal body mass index (BMI), mod- kg (p , 0.005), but the other muscle characteristics did not
erate to severe airflow obstruction, a slightly reduced PaO2, change significantly in this group (160 6 27 cm2 versus 164 6
and a normal resting PCO2. No significant difference between 30 cm2, 60 6 14 kg versus 61 6 14 kg, and 47 6 11 kg versus
the two study groups was found for these variables, although 48 6 9 kg for the bilateral thigh MCSA, pectoralis major
the FEV1% predicted tended to be greater in the AERO 1 strength, and latissimus dorsi strength before and after train-
ST group than in the AERO group. No change in these vari- ing, respectively; p . 0.05).
ables occurred with training (data not shown). No differences
in age, gender distribution, BMI, or pulmonary function were Effects of Exercise Training on Exercise
found in patients belonging to the first three blocks and the Capacity and Quality of Life
nine patients of the fourth block (data not shown). None of The physiologic parameters obtained at peak exercise are
the patients was employed; all were either retired or disabled shown in Table 3. Peak exercise work rate increased by 19%
because of severe respiratory impairment. and 12% in the AERO and AERO 1 ST groups, respectively,
but this reached statistical significance only
·
in the
·
AERO 1
Amount of Training ST group (p , 0.05). No changes in peak VO2, VE, heart rate,
The attendance rate at the training sessions averaged 92 6 9% or arterial lactate concentration were found in either study
·
and 94 6 4% in the AERO and AERO 1 ST groups, respec- group. In contrast, significant reductions in VE (6%, p , 0.05),
tively. The most common reasons for missing exercise sessions
were fatigue, transient worsening of dyspnea not requiring
modification of the usual bronchodilator therapy, and trans- TABLE 2
portation difficulties. The average amount of training during MEAN TRAINING INTENSITY AND DURATION ACHIEVED DURING
each training sessions is provided in Table 2. The duration of TRAINING SESSIONS FOR THE 12-wk PROGRAM*
aerobic training was comparable in the two study groups. As a Aerobic Training Aerobic 1 Strength Training
result of a greater baseline exercise capacity, the average abso- (n 5 15) (n 5 21)
lute training intensity was significantly greater in the AERO 1
Aerobic training
ST group than in the AERO group (p , 0.05). However,
Exercise session duration, min 26 6 1 28 6 1
when the training intensity was expressed as a percent of Training intensity, watts 28 6 9 38 6 13†
the baseline peak exercise capacity, no difference was found Training intensity, % W max 73 6 24 65 6 13
Strength training
Pectoralis major
No. sets — 2.8 6 0.1
TABLE 1
No. repetitions — 10 6 2
PATIENT CHARACTERISTICS* Workload, kg — 28 6 8
Workload, % 1 RM — 81 6 11
Pretraining Latissimus dorsi
Aerobic Training Aerobic 1 Strength Training No. sets — 2.8 6 0.1
(n 5 15 ) (n 5 21) No. repetitions — 11 6 1
Workload, kg — 32 6 7
Age 67 6 9 64 6 7 Workload, % 1 RM — 76 6 11
Sex, M/F 11/4 17/4 Gluteus maximus
Height, m 1.64 6 0.09 1.67 6 0.09 No. sets — 2.8 6 0.1
Weight, kg 67 6 14 72 6 18 No. repetitions — 12 6 2
Body mass index, kg/m2 25 6 4 27 6 5 Workload, kg — 100 6 27
FEV1, L 0.96 6 0.29 1.23 6 0.24 Workload, % 1 RM — 93 6 14
FEV1, % pred 39 6 12 45 6 15 Vastus lateralis
FVC, L 2.30 6 0.96 2.94 6 0.71 No. sets — 2.8 6 0.1
FVC, % pred 63 6 23 74 6 15 No. repetitions — 10 6 2
TLC, % pred 115 6 25 127 6 24 Workload, kg — 54 6 12
DLCO, % pred 75 6 15 68 6 22 Workload, % 1 RM — 80 6 8
PaO2, mm Hg 78 6 9 77 6 8
PaCO2, mm Hg 43 6 6 41 6 4 * Values are mean 6 SD and represent the amount of training, on average, at each
training session for the entire program.
* Values are mean 6 SD.
†
p , 0.05 AERO 1 ST group versus AERO group.
Bernard, Whittom, LeBlanc, et al.: Exercise Training in COPD 899

TABLE 3 TABLE 5
PHYSIOLOGIC PARAMETERS AT PEAK EXERCISE BEFORE MEAN DIFFERENCE IN HEALTH-RELATED QUALITY
AND AFTER AEROBIC TRAINING* OF LIFE BEFORE AND AFTER TRAINING*

Aerobic Training Aerobic 1 Strength Training Aerobic Training Aerobic 1 Strength Training
(n 5 15) (n 5 21) (n 5 15) (n 5 21)

Pretraining Posttraining Pretraining Posttraining Dyspnea 1.1 6 0.8† 1.0 6 0.8†

· Fatigue 0.7 6 0.6† 0.5 6 0.9†
V O2, L/min 0.91 6 0.25 0.94 6 0.25 1.13 6 0.40 1.13 6 0.41 Emotion 0.4 6 0.7† 0.4 6 0.7
·
V O2, % pred 56 6 11 59 6 20 66 6 20 66 6 22 Mastery 1.0 6 0.7† 0.4 6 0.7†
Wmax, W 43 6 18 51 6 19 60 6 23† 67 6 29‡
Wmax, % pred 36 6 12 44 6 16 45 6 16 51 6 19‡ * Values are mean 6 SD.
· †
p , 0.05.
V E, L/min 35.5 6 10.5 35.4 6 11.4 43.3 6 15.9 42.1 6 15.9
Heart rate, beats/min 124 6 17 123 6 19 132 6 21 125 6 21
PaO2, mm Hg 70 6 14 70 6 14 66 6 11 64 6 13
PaCO2, mm Hg 48 6 8 48 6 7 47 6 7 48 6 8
pH 7.35 6 0.04 7.35 6 0.03 7.35 6 0.04 7.34 6 0.05 found that the combination of aerobic and strength training
Lactate, mmol/L 3.06 6 1.32 3.20 6 1.15 3.84 6 1.23 3.62 6 1.49
was safe and well tolerated despite the severity of the underly-
* Values are mean 6 SD. ing lung disease, and was associated with greater improvement
†
p , 0.05 AERO 1 ST group versus AERO group at baseline. in peripheral muscle strength and in thigh MCSA than was
‡
p , 0.05 pre- versus posttraining within each group.
aerobic training alone. However, these changes did not trans-
late into further improvement in exercise tolerance or quality
of life as measured in this study.
heart rate (7%, p , 0.05), and arterial lactate concentration
(17%, p , 0.05) for an identical exercise work rate were ob- Effects on Strength Training Combined with
served after training in the two groups (Table 4). After train- Aerobic Training on Muscle Function
ing, the functional exercise capacity also increased in both Together with the results of a previous study showing im-
groups, as indicated by the improvement in the 6MWD, which proved skeletal muscle oxidative capacity after exercise train-
increased from 388 6 78 m to 454 6 50 m in the AERO group ing in patients with moderate to severe COPD (24), the results
and from 411 6 81 m to 499 6 68 m in the AERO 1 ST group of the present study confirm that such patients retain the ca-
(p , 0.0005). Health-related quality of life improved in both pacity for improved peripheral muscle function with an ade-
study groups (Table 5). For most dimensions of the CRQ, the quate training stimulus. Furthermore, the magnitude of im-
effect size corresponded to a small (0.5-point) to moderate provement in midthigh MCSA and peripheral muscle strength
(1.0-point) improvement (23). found in our study was similar to that reported in older normal
subjects after a period of strength training (7). As compared
Comparisons between the Two Study Groups with those of normal subjects of similar age studied in our lab-
The percent changes in bilateral thigh MCSA and in periph- oratory, peripheral muscle strength and thigh MCSA in our
eral muscle strength before and after training are shown in COPD patients were not completely corrected after training
Figure 1. As can be seen, the changes in thigh MCSA and in (18), suggesting that the training period was not of sufficient
quadriceps and pectoralis major muscle strength were signifi- duration or that factors other than chronic inactivity are in-
cantly greater in the AERO 1 ST group than in the AERO volved in explaining muscle atrophy and weakness in patients
group (p , 0.05). In contrast, the magnitude of the changes in with COPD.
exercise capacity (Table 3), physiologic parameters at isoexer- In young and older normal subjects, muscle hypertrophy
cise (Table 4), 6MWD, and quality of life (Table 5) were com- and improved neural recruitment patterns account for the in-
parable for the two groups. crease in muscle strength following muscle training (7, 25, 26).
Both mechanisms probably played a role in explaining the im-
DISCUSSION provement in muscle strength after training in our patients.
The increase in midthigh MCSA and the ability to demon-
This study sought to evaluate whether strength training is a
strate increased strength on an apparatus other than the train-
useful addition to aerobic training in patients with COPD. We
ing devices used in the study support the idea that structural
adaptation to strength training took place in our patients. On
the other hand, the occurrence of neural adaptation is sug-
TABLE 4
gested by the proportionally greater improvement in muscle
PHYSIOLOGIC PARAMETERS AT ISOEXERCISE BEFORE strength than in muscle cross-sectional area.
AND AFTER AEROBIC TRAINING*
Effects of Training on Exercise Capacity
Aerobic Training Aerobic 1 Strength Training
(n 5 15) (n 5 21) and Quality of Life
In accordance with the literature on pulmonary rehabilitation,
Pretraining Posttraining Pretraining Posttraining
both training modalities examined in the present study re-
·
V O2, L/min 0.89 6 0.24 0.86 6 0.25 1.13 6 0.39 1.05 6 0.31 sulted in modest improvements in peak exercise capacity,
Wmax, W 41 6 17 41 6 17 58 6 23 58 6 23 whereas the change in functional exercise capacity (6MWD)
·
V E, L/min 35.1 6 10.4 31.9 6 9.0† 42.8 6 15.7 39.1 6 13.7†
Heart rate, beats/min 123 6 19 116 6 20† 131 6 21 120 6 21†
was more pronounced ·
(1). Most notably, significant reduc-
PaO2, mm Hg 68 6 15 69 6 13 69 6 11 67 6 12 tions in heart rate, VE, and blood lactate concentration for a
PaCO2, mm Hg 49 6 9 47 6 8 46 6 8 46 6 7 given exercise work rate were found in both study groups,
pH 7.34 6 0.04 7.37 6 0.05 7.35 6 0.04 7.36 6 0.04 strongly suggesting that a physiologic training effect occurred.
Lactate, mmol/L 3.06 6 1.32 2.62 6 1.00† 3.66 6 1.12 3.10 6 1.43† Despite the greater improvement in midthigh MCSA and
* Values are mean 6 SD. muscle strength in the AERO 1 ST group, the changes in ex-
†
p , 0.05 pre- versus posttraining within each group. ercise capacity and in quality of life were of similar magnitude
900 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 159 1999

Figure 1. Mean 6 SD percent change in bilateral thigh MCSA and in the strength of the quadriceps, pec-
toralis major, and latissimus dorsi muscles before and after training in the AERO and AERO 1 ST groups.
The improvement in bilateral thigh MCSA and in the strength of the three muscle groups was statistically
significant in the AERO 1 ST group. Quadriceps strength also showed a significant increase in the AERO
group. As can be seen, the magnitude of the changes in thigh MCSA and in the strength of the quadri-
ceps and pectoralis major muscles was significantly greater in the AERO 1 ST group than in the AERO
group. (*p , 0.05 for pre- versus posttraining within each study group; †p , 0.05 for the AERO group ver-
sus the AERO 1 ST group.)

in the two study groups. However, it is possible that a longer strength training has several interesting features that could
duration of training would have produced a greater improve- help patients with severe COPD. We found, as did Simpson
ment in exercise capacity with the combination of strength and colleagues, that the muscle resistive exercises induced less
and aerobic training than with aerobic training alone. dyspnea than did the cycling exercise, and as a result were well
These results are remarkably similar to those previously re- tolerated by the patients (6). The implementation of muscle
ported in two studies of the effects of anabolic drugs and exer- resistive exercises also helped diversify the training sessions
cise training in patients with COPD (27, 28). In both cases the and maintain the patients’ interest and motivation during the
authors showed that the combined use of anabolic drugs and training program. In addition to its beneficial effects on pe-
aerobic training produced a greater improvement in periph- ripheral muscle function, strength training may help increase
eral muscle mass and strength than did exercise training alone. bone density (29), a potentially interesting effect for patients
However, no further gain in exercise tolerance was obtained with COPD, in whom osteoporosis is highly prevalent (30).
with the combination of anabolic drugs and exercise training. Strength training should be used cautiously in patients with
Perhaps the changes in peripheral muscle function in our musculoskeletal disorders and/or osteoporosis, in order to
study and these other studies were not of sufficient magnitude avoid injury such as bone fracture. However, when adapted
to further improve exercise capacity. Another possible expla- to the individual patient’s condition, we found that intense
nation for the dissociation between changes in muscle mass muscle exercises could be performed safely in patients with
and strength and exercise capacity is the relative task specific- COPD.
ity of any training stimulus: the greatest improvement in mus-
cle function is shown in tests that closely mimic the character- Criticisms of the Study
istics of the training movement (7). An important implication The improvement in peak exercise work rate was of similar
of this observation is that the training movements should re- magnitude for the two groups investigated in our study. How-
semble activities that are relevant to the patient’s daily activi- ever, the change in peak exercise work rate reached statistical
ties. Further studies will be needed to determine whether significance only in the AERO 1 ST group; this was probably
greater improvement in peripheral muscle function can en- due to a type II error, since fewer patients were included in
hance exercise capacity, and whether these muscle changes the AERO group. It is unlikely that including more patients in
can be translated into greater ease in performing activities of the study would have changed the conclusion that the im-
daily living. provements in peak exercise capacity and 6MWD were similar
for the two study groups. Using the baseline peak exercise
Role of Strength Training in Exercise Training work rate values, we calculated that more than 1,000 patients
Programs in Patients with COPD per group would have been required to show that the ob-
Although we could not exclude the possibility that prolonging served pre- to posttraining improvements in maximal exercise
the duration of leg cycling exercises would have had the same capacity were statistically different for the two study groups.
beneficial effects on peripheral muscle function as the combi- A study of this size would not only be difficult to conduct, but
nation of strength training and aerobic training, we found that its results would also be of no clinical significance. Because of
Bernard, Whittom, LeBlanc, et al.: Exercise Training in COPD 901

practical considerations (see METHODS), patients in the fourth 945.

block were allocated to the same training modality (AERO 1 12. American Thoracic Society. 1987. Standards for the diagnosis and care
of patients with chronic obstructive pulmonary disease (COPD) and
ST). Although this may have influenced our results, we be-
asthma. Am. Rev. Respir. Dis. 136:225–244.
lieve that any potential biases were minimized, since: (1) the 13. Knudson, R. J., R. C. Slatin, M. D. Lebowitz, and B. Burrows. 1976. The
training modality used for the patients in the fourth block was maximal expiratory flow-volume curve: normal standards, variability
selected in a random fashion; and (2) the characteristics of the and effects of age. Am. Rev. Respir. Dis. 113:587–600.
patients in the first three blocks and those in the fourth block 14. Goldman, H. I., and M. R. Becklake. 1959. Respiratory function tests:
were similar. normal values at median altitudes and the prediction of normal re-
sults. Am. Rev. Tuberc. 79:454–467.
In summary, the addition of strength training in aerobic
15. Cotes, J. E., and A. M. Hall. 1970. The transfer factors for the lung: nor-
training was safe and well tolerated in patients with severe mal values in adults. In P. Arcangeli, editor. Normal Values for Respi-
COPD, and was accompanied by greater improvement in ratory Function in Man. Panminerva Medica, Torino, Italy. 327–343.
muscle mass and strength then was aerobic training alone. 16. Maltais, F., P. Leblanc, J. Jobin, C. Bérubé, J. Bruneau, L. Carrier, M. J.
This study therefore confirms that the peripheral muscles may Breton, G. Falardeau, and R. Belleau. 1997. Intensity of training and
show structural adaptation with an appropriate training regi- physiologic adaptation in patients with chronic obstructive pulmonary
disease. Am. J. Respir. Crit. Care Med. 155:555–561.
men despite severe ventilatory impairment. This observation
17. Kelley, D. E., B. S. Slasky, and J. Janosky. 1991. Skeletal muscle density:
is important, given the potentially adverse effects of periph- effects of obesity and non-insulin-dependent diabetes mellitus. Am. J.
eral muscle dysfunction on exercise tolerance and quality of Clin. Nutr. 54:509–515.
life in patients with COPD. However, further studies are re- 18. Bernard, S., P. Leblanc, F. Whittom, G. Carrier, J. Jobin, R. Belleau, and
quired to clarify the extent to which any improved peripheral F. Maltais. 1998. Peripheral muscle weakness in patients with chronic
muscle function that could be obtained with exercise training obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 158:
629–634.
might improve exercise tolerance in patients with COPD.
19. Jones, N. L. 1988. The interpretation of stage 1 exercise test results. In
Acknowledgment : The authors thank Marthe Bélanger, Anne-Marie Bezeau, Carol Trumbold, editor. Clinical Exercise Testing, 3rd ed. W. B. Saun-
Marie-Josée Breton, Gisèle Deshaies, Anita Leblanc, Jacqueline Lepage, Ja- ders, Philadelphia. 158–185.
cynthe Rondeau, and Ginette Turbide for their assistance; Serge Simard for 20. Guyatt, G. H., S. Pugsley, M. J. Sullivan, P. J. Thompson, L. Berman,
his statistical assistance; and Drs. Yvon Cormier, Yves Lacasse, and Frédéric N. L. Jones, E. L. Fallen, and D. W. Taylor. 1984. Effect of encourage-
Sériès for helpful suggestions about the manuscript. The authors are also ment on walking test performance. Thorax 39:818–822.
grateful to Joan Bruneau, Suzanne Lemay, Agathe Tremblay, and Didier 21. Guyatt, G. H., L. B. Berman, M. Townsend, S. O. Pugsley, and L. W.
Saey for their help in supervising the training sessions. Chambers. 1987. A measure of quality of live for clinical trials in
chronic lung disease. Thorax 42:773–778.
22. Bourbeau, J., C. Guimont, W. Gibbons, F. Maltais, M. Rouleau, G. Ger-
References ardi, M. Mondor. 1998. Validation of French translated chronic respi-
1. Lacasse, Y., R. S. Goldstein, E. Wong, G. H. Guyatt, D. King, and D. J. ratory questionnaire (CRQ) in patients with chronic obstructive pul-
Cook. 1996. Meta-analysis of respiratory rehabilitation in chronic ob- monary disease (COPD) (abstract). Am. J. Respir. Crit. Care Med.
structive pulmonary disease. Lancet 348:1115–1119. 157:A762.
2. ACCP/AACVPR Pulmonary Rehabilitation Guidelines Panel. 1997. 23. Jaeschke, R., J. Singer, and G. H. Guyatt. 1989. Measurement of health
Pulmonary rehabilitation: joint ACCP/AACVPR evidence-based guide- status: ascertaining the minimal clinically important difference. Con-
lines. Chest 112:1363–1396. trolled Clin. Trials 10:407–415.
3. Celli, B. R. 1995. Pulmonary rehabilitation in patients with COPD. Am. 24. Maltais, F., P. Leblanc, C. Simard, J. Jobin, C. Bérubé, J. Bruneau, L.
J. Respir. Crit. Care Med. 152:861–864. Carrier, and R. Belleau. 1996. Skeletal muscle adaptation to endur-
4. Gosselink, R., T. Troosters, and M. Decramer. 1996. Peripheral muscle ance training in patients with chronic obstructive pulmonary disease.
weakness contributes to exercise limitation in COPD. Am. J. Respir. Am. J. Respir. Crit. Care Med. 154:442–447.
Crit. Care Med. 153:976–980. 25. Rogers, M. A., and W. J. Evans. 1993. Changes in skeletal muscle with
5. Hamilton, A. L., K. J. Killian, E. Summers, and N. L. Jones. 1995. Mus- aging: effects of exercise training. In J. O. Holloszy, editor. Exercise
cle strength, symptom intensity and exercise capacity in patients with and Sports Science Reviews. Williams and Wilkins, Philadelphia. 65–
cardiorespiratory disorders. Am. J. Respir. Crit. Care Med. 152:2021– 102.
2031. 26. Fiatarone, M. A., E. C. Marks, N. D. Ryan, C. N. Meridieth, L. A. Lip-
6. Simpson, K., K. Killian, N. McCartney, D. G. Stubbing, and N. L. Jones. sitz, and W. J. Evans. 1990. High-intensity strength training in nonage-
1992. Randomised controlled trial of weightlifting exercise in patients narians. J.A.M.A. 263:3029–3034.
with chronic airflow limitation. Thorax 47:70–75. 27. Schols, A. M. W. J., P. B. Soeters, R. Mostert, R. J. Pluymers, and E. F. M.
7. Frontera, W. R., C. N. Meredith, K. P. O’Reilly, H. G. Knuttgen, and Wouters. 1995. Physiologic effects of nutritional support and anabolic
W. J. Evans. 1988. Strength conditioning in older men: skeletal muscle steroids in patients with chronic obstructive pulmonary disease: a pla-
hypertrophy and improved function. J. Appl. Physiol. 64:1038–1044. cebo-controlled randomized trial. Am. J. Respir. Crit. Care Med. 152:
8. Killian, K. J., P. Leblanc, D. H. Martin, E. Summers, N. L. Jones, and 1268–1274.
E. J. M. Campbell. 1992. Exercise capacity and ventilatory, circula- 28. Casaburi, R., E. Carithers, J. Tosolini, J. Phillips, and S. Bhasin. 1997.
tory, and symptom limitation in patients with airflow limitation. Am. Randomized placebo controlled trial of growth hormone in severe
Rev. Respir. Dis. 146:935–940. COPD patients undergoing endurance exercise training (abstract).
9. Frontera, W. R., C. N. Meredith, K. P. O’Reilly, and W. J. Evans. 1990. Am. J. Respir. Crit. Care Med. 155:A498.
Strength training and determinants of VO2max in older men. J. Appl. 29. Heinonen, A., P. Kannus, H. Sievanen, P. Oja, M. Pasanen, M. Rinne, K.
Physiol. 68:329–333. Uusi-Rasi, and I. Vuori. 1996. Randomised controlled trial of effect of
10. Maltais, F., A. A. Simard, C. Simard, J. Jobin, P. Desgagnés, and P. Leb- high-impact exercise on selected risk factors for osteoporotic frac-
lanc. 1996. Oxidative capacity of the skeletal muscle and lactic acid ki- tures. Lancet 348:1343–1347.
netics during exercise in normal subjects and in patients with COPD. 30. McEvoy, C. E., K. E. Ensrud, E. Bender, H. K. Genant, W. Yu, J. M.
Am. J. Respir. Crit. Care Med. 153:288–293. Griffith, and D. E. Niewoehner. 1998. Association between corticoste-
11. McCartney, N., R. S. Mckelvie, D. R. S. Haslam, and N. L. Jones. 1991. roid use and vertebral fractures in older men with chronic obstructive
Usefulness of weightlifting training in improving strength and maxi- pulmonary disease. Am. J. Respir. Crit. Care Med. 157:704–709.
mal power output in coronary artery disease. Am. J. Cardiol. 67:939–