General Exercise Training Improves Ventilatory
and Peripheral Muscle Strength and Endurance
in Chronic Airflow Limitation
DENIS E. O’DONNELL, MAUREEN MCGUIRE, LORELEI SAMIS, and KATHERINE A. WEBB
Respiratory Investigation Unit, Department of Medicine, Queen’s University and Department of Physiotherapy,
St. Mary’s of the Lake Hospital, Kingston, Ontario, Canada
We studied the impact of a 6-wk supervised, multimodality endurance exercise training program
(EXT) on strength and endurance of ventilatory and peripheral muscles in patients with chronic airflow limitation (CAL), and determined whether potential improvements contributed to relief of exertional breathlessness (B) and perceived leg effort/discomfort (LE), respectively. Twenty breathless
patients with stable CAL (FEV1 5 41 6 3% predicted; mean 6 SEM) were tested at 6-wk intervals at
baseline, after a nonintervention control period (pre-EXT), and post-EXT. Measurements included:
pulmonary function tests (PFTs), maximal inspiratory/expiratory pressures (MIP, MEP), inspiratory
muscle endurance (VLIM), quadriceps strength and endurance, exercise endurance, and submaximal
cycle exercise with cardioventilatory and symptom responses. Measurements at baseline and pre-EXT
were identical. Post-EXT, PFTs did not change; exercise endurance measured on the treadmill, cycle
ergometer, arm ergometer, and by 6-min walk distance increased 40 6 8%, 43 6 10%, 12 6 5%, and
34 6 9%, respectively (p , 0.05); quadriceps strength increased 21 6 5% (p , 0.01); MIP and MEP
increased 29 6 11% and 27 6 11%, respectively (p , 0.05); VLIM increased almost threefold (p ,
.
0.05). At isotime near end-exercise, B, LE, carbon dioxide production ( V CO2), oxygen consumption
.
( V O2), ventilation, and breathing frequency (F) all fell after EXT (p , 0.05): DB correlated with DF (r 5
0.58, p , 0.01). Increased MIP and VLIM did not correlate with improved breathlessness or exercise
endurance. Similarly, changes in quadriceps strength and endurance did not correlate with changes
in LE or exercise endurance. In conclusion, general nonspecific EXT improved ventilatory and peripheral muscle function in severe CAL, but such improvements did not appear to contribute significantly
to reduced exertional symptoms and enhanced exercise performance. O’Donnell DE, McGuire M,
Samis L, Webb KA. General exercise training improves ventilatory and peripheral muscle
AM J RESPIR CRIT CARE MED 1998;157:1489–1497.
strength and endurance in chronic airflow limitation.
In clinically stable patients with severe chronic airflow limitation (CAL) who are on optimal pharmacotherapy, supervised
endurance exercise training (EXT) has been shown to confer
substantial additional benefits that include alleviation of exertional symptoms and improvement of exercise endurance (1–
4). We have previously shown that improved exercise tolerance following EXT in advanced CAL occurs primarily as a
result of reduced exertional dyspnea and in some instances because of reduced sense of leg effort (4). Although improvement in dyspnea is multifactorial, it is related in part to reduced ventilatory demand (4). However, we also found that
(Received in original form August 4, 1997 and in revised form December 30, 1997)
Supported by Ontario Thoracic Society/Canadian Physiotherapy Cardio-Respiratory Society.
Denis O’Donnell holds a career scientist award from the Ontario Ministry of
Health.
Presented at the ALA/ATS International Meeting, San Francisco, May 1997.
Correspondence and requests for reprints should be addressed to Dr. Denis
O’Donnell, Richardson House, 102 Stuart Street, Kingston General Hospital,
Kingston, ON, K7L 2V7 Canada.
Am J Respir Crit Care Med Vol 157. pp 1489–1497, 1998
dyspnea intensity was reduced at any given level of ventilation
after EXT and postulated that improved ventilatory muscle
function (increased strength and endurance) may have contributed to symptom relief in addition to the reduced ventilation (4).
It is unclear whether general EXT can lead to clinically important and consistent increases in respiratory muscle strength
and endurance in severe CAL. Small improvements have been
reported in normal subjects (5, 6) and in patients with cystic fibrosis (7), but responses in CAL have been minimal or nonexistent (8, 9). Nevertheless, we postulated that a training stimulus that ensured high levels of ventilation, if sustained for
sufficient duration, would result in functional adaptation in
the ventilatory muscles, given their increased intrinsic mechanical loading, i.e., elastic/threshold and resistive loads. A
second question is whether improved ventilatory muscle function leads to decreased symptoms in CAL. Studies that have
examined the relationship between increased inspiratory muscle strength and improved symptoms and exercise capacity in
CAL have yielded conflicting results (10, 11) despite there being a strong theoretical basis for such an association (12). The
first objective of this study was therefore to determine whether
exercise training targeted at high Borg dyspnea ratings could
1490
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
result in significantly improved ventilatory muscle function in
severe CAL and whether this, in turn, contributed to dyspnea
amelioration.
The effect of improved ventilatory muscle function on
breathing pattern, ventilation, and operational lung volumes
has not previously been systematically studied during exercise.
Casaburi and associates recently postulated that alteration in
breathing pattern toward a deeper, slower pattern after EXT
might have its basis in improved ventilatory muscle endurance
causing reduced dynamic hyperinflation with increased tidal
volume (13). Reduced submaximal ventilation in this study
was attributed to reduced physiological dead space and consequently more efficient CO2 elimination. Alternatively, reduced submaximal ventilation after EXT could be explained
by improved efficiency and metabolic alterations (4). Therefore, the second objective of this study was to explore mechanisms of reduced ventilation and to determine the relative importance of altered metabolic factors, dynamic ventilatory
mechanics, and breathing pattern. In this regard, we examined
the relationship between changes in dyspnea, ventilation,
breathing pattern, operational lung volumes, and metabolic
measurements following EXT.
Finally, there is increasing evidence that the perception of
increased leg discomfort contributes importantly to exercise
intolerance in patients with severe CAL (4) and that alleviation of this symptom is responsible for improving exercise endurance in some patients (4). This study’s third objective was
therefore to determine if the relief of leg discomfort was attributable to increased peripheral muscle strength following
EXT, and whether this contributed to increased endurance in
some patients.
METHODS
Subjects
Subjects for the study included 20 consecutive patients with stable, severe CAL who were entering an outpatient pulmonary rehabilitation
program and were both eligible and consenting. Subjects satisfied the
following selection criteria: (1) moderate to severe CAL (FEV1 , 60%
predicted) with a clinical course consistent with chronic bronchitis and/
or emphysema and a long history of cigarette smoking; (2) moderate to
severe chronic breathlessness (Modified Baseline Dyspnea Index < 6)
(14); (3) on optimal bronchodilator therapy and clinically stable, defined
by no change in medication dosage or frequency of administration
with no exacerbations or hospital admissions in the preceding 4 wk;
(4) no oxygen desaturation to less than 80% during exercise testing on
room air; and (5) absence of other significant diseases that could contribute to breathlessness or exercise limitation. Subjects were well motivated to participate in the program and did not currently smoke.
Study Design
This was a single-center, two-period, controlled study in which subjects completed a 6-wk nonintervention control period before entering a 6-wk pulmonary rehabilitation program consisting predominantly of EXT. After hospital/university research ethics approval was
obtained, subjects gave informed consent and entered the study on a
staggered basis over a 2-yr period. In an initial screening visit, subjects
were tested for pulmonary function and gas exchange, were familiarized with exercise testing procedures and the various scales for rating
symptom intensity, and completed an incremental symptom-limited
cycle exercise test. Three experimental visits were held at 6-wk intervals immediately before the control period, after the control period
(this visit doubling as the pre-EXT visit), and post-EXT; therefore,
subjects acted as their own controls. At each visit, subjects completed
symptom-related questionnaires and underwent testing for pulmonary
function, ventilatory muscle strength/endurance, peripheral muscle
strength/endurance, and physiologic responses to exercise. Subjects
avoided caffeine and heavy meals at least 4 h prior to testing and
avoided alcohol and major physical exertion entirely on the day of
VOL 157
1998
each visit. All visits were conducted at the same time of day for each
subject.
Pulmonary Function Testing
Routine spirometry (6200 Autobox DL; SensorMedics, Yorba Linda,
CA) was performed before exercise testing in accordance with recommended techniques (15). Functional residual capacity (FRC) and specific airway resistance (SRaw) were determined by constant-volume
body plethysmography (6200 Autobox DL). Single-breath diffusing
capacity for carbon monoxide (DLCO) was also measured (6200 Autobox DL). Predicted normal values for spirometry, lung volumes, DLCO
and SRaw were those of Morris and associates (16), Goldman and
Becklake (17), Gaensler and Wright (18), and Briscoe and Dubois
(19), respectively.
Chronic Breathlessness
The modified Baseline Dyspnea Index (BDI) incorporates magnitude
of task, magnitude of effort, and functional impairment and was used
to generate an overall focal score of chronic activity-related breathlessness (14). The Transition Dyspnea Index (TDI) was used to detect
changes in symptom intensity over time in each of these three categories (20). BDI and TDI were assessed by an unbiased observer with no
specific knowledge of the subjects’ pulmonary function or other measures of study outcome. The same observer was kept for each subject
throughout the study period. Subjects also rated the magnitude of
their own chronic activity-related breathlessness using a 100-mm oxygen cost diagram.
Exercise Testing
During initial screening, an incremental cycle exercise test was performed to a symptom-limited maximum as described in a previous
publication (4). In experimental visits, endurance cycle exercise tests
were performed to a symptom-limited endpoint in a similar manner,
with each test being conducted at the same constant work rate equal
to approximately 75% of the maximum work rate achieved during
screening. Changes in operational lung volumes during exercise were
evaluated from measurements of inspiratory capacity (IC) as previously described (21). Exercise responses were compared with the predicted normal values of Jones (22) and ventilation was compared with
the predicted maximum ventilatory capacity (MVC 5 28.09 3 FEV1 1
18.4) of Dillard and coworkers (23).
Breathlessness was defined as “the uncomfortable sensation of labored or difficult breathing” and leg effort/discomfort as “the level of
difficulty experienced during pedalling.” Before exercise testing, subjects were familiarized with the Borg Scale (24) and its endpoints were
anchored such that zero represented “no breathlessness (leg effort or
discomfort)” and “10” was “the most severe breathlessness (leg effort
or discomfort) that they had ever experienced or could imagine experiencing.” By pointing to the Borg Scale, subjects rated the magnitude
of their perceived breathlessness (B) and leg effort/discomfort (LE) at
baseline and every minute throughout exercise. At exercise cessation,
subjects were also asked to verbalize their main reason for stopping
exercise.
Six-minute Walk
Six-minute walking tests were administered at entry (the best of two
reproducible tests) and at each testing period as additional tests of
functional capacity. Subjects were given standardized instructions to
cover the greatest distance possible in 6 min and verbal encouragement was avoided during the 6 min period. Heart rate (HR), oxygen
saturation (SaO2), and Borg symptom ratings were recorded at rest
and immediately after walking cessation.
Peripheral Muscle Strength and Endurance
Peripheral muscle endurance was assessed using constant-load exercise tests at approximately 75% of the maximum work rate achieved
upon study entry. Lower limb muscle endurance was evaluated with a
cycle ergometer (Ergoline 800S; SensorMedics Corporation, Anaheim, CA) (see above) and a treadmill (Q55; Quinton Instruments
Company, Seattle, WA). Upper limb muscle endurance was evaluated
with an arm ergometer (UBE; Cybex, Ronkonkoma, NY) at a crank-
O’Donnell, McGuire, Samis, et al.: Impact of Exercise Training on Muscle Function in CAL
TABLE 1
SUBJECT CHARACTERISTICS*
Male:Female
Age, yr
Height, cm
Weight, kg
Body mass index, kg/m2
Modified Baseline Dyspnea Index
Pulmonary function
FEV1, L
FVC, L
FEV1/FVC,%
TLC, L
FRC, L
MIP, cm H2O
MEP, cm H2O
SRaw, cm H2O · s
DLco, ml/min/mm Hg
12:8
69 6 2
166 6 2
67 6 3
24.2 6 5.0
4.7 6 0.3
“severe”
0.98 6 0.08
2.33 6 0.18
43 6 2
6.40 6 0.37
4.66 6 0.17
63 6 6
82 6 7
18.9 6 1.7
10.6 6 0.8
(41)
(67)
(61)
(119)
(149)
(70)
(48)
(444)
(54)
* All values are mean 6 SEM (% predicted).
ing rate of approximately 60 rpm. Work rate, endurance time, Borg
symptom ratings, HR, and SaO2 were monitored during each symptom-limited test.
To assess quadriceps (i.e., knee extension) strength and endurance, subjects were seated with the lower leg dependent, the knee
flexed at 908 and the pelvis secured by an adjustable belt. A strap
around the leg just proximal to the malleoli led back, parallel to the
thigh, to a fixed surface by cable extension. Static contractions were
measured by cable tensiometry (Jamar Back, Leg and Chest Dynamometer; Therapeutic Equipment Corporation, Clifton, NJ), the
maximum voluntary contraction was recorded as the best of three
contractions for each leg and compared with the predicted values of
Hamilton and associates (25). Isometric endurance was measured on
the dominant leg as the time each subject could maintain a contraction at 50% of maximum. Breath-holding maneuvers were avoided.
Static handgrip strength and endurance were evaluated using a
similar protocol as that for the quadriceps muscles. Subject were
seated erect with the shoulder adducted and neutrally rotated, the elbow flexed at 908, and the forearm and wrist resting in a neutral position. Handgrip strength and endurance were measured by squeezing
an appropriately sized rubber bulb attached to a pressure manometer
(Martin Vigorimeter; ELMED Inc., Addison, IL).
1491
tive HR measurements were used as a supplementary method of monitoring training intensity. For each exercise modality, the duration of
activity included a 3-min warm-up, exercise at target intensity for the
longest tolerable time (gradually increasing from a minimum of 5 min
to a maximum of 20 min as the program progressed and as participants became more accustomed to exercise), and a 3-min cool-down.
Secondary aspects of the program included instruction on pursed lip
breathing, thoracic mobility, coughing technique, relaxation, and unstructured general patient education.
Statistical Analysis
Results are reported as means 6 SEM. The conventional level of statistical significance of 0.05 was used for all analyses. Using standardized Borg ratings of exertional breathlessness as the primary endpoint, a relevant difference of one Borg unit, a common standard
deviation established from our own laboratory values, a 5 0.05, b 5
0.20, and a two-tailed test of statistical significance, it was estimated
that a sample size of 20 was required for each study arm. Because each
subject acted as their own control, we aimed to study 20 subjects.
Nonparametric ratings of chronic and exertional breathlessness
were compared using Wilcoxon tests. Frequency statistics examining
the patient’s reasons for stopping exercise were analyzed using
Fisher’s exact test. Summary statistics for each experimental visit were
compared using analysis of variance (ANOVA) for repeated measures and the appropriate post hoc analysis. Changes that occurred
during the EXT and control periods were compared using paired
t tests. Exercise response slopes were studied using linear regression
analysis of individual data sets: general responses were expressed rela·
tive to time or oxygen consumption (V O2), and breathing pattern
·
slopes were expressed relative to ventilation (V E). Responses were
also examined at a standardized time near end-exercise (isotime was
equal to the time of the shortest of the three constant-load exercise
tests rounded off to the nearest minute).
Pearson’s correlations were performed to examine associations between outcome measures after EXT and possible contributing factors.
The dependent variable was the change in exercise endurance expressed
as the change in total cumulative work (DWorkTOT), and independent
variables included: respiratory muscle strength and endurance (MIP,
MEP, VLIM), peripheral muscle strength and endurance (quadriceps,
hand grip), resting pulmonary function (FEV1, FVC, IC), and exercise
response slopes of cardiovascular function (HR, oxygen pulse), gas
Ventilatory Muscle Strength and Endurance
Maximum inspiratory mouth pressures (MIP) from RV and maximum
expiratory mouth pressures (MEP) from TLC were measured with a
standard mouthpiece and a direct-reading dial pressure gauge (Magnehelic; Dwyer Instruments, Inc., Michigan City, IN). Measured values were compared with the predicted normal values of Black and
Hyatt (26). Inspiratory muscle endurance at a load equal to approximately 50% of the maximum achieved in the precontrol visit was assessed using a threshold breathing device (27); this load remained constant for all tests. Mouth pressure (Pm), breathing frequency (F), HR,
SaO2, and Borg symptom ratings (breathlessness, inspiratory difficulty)
were monitored during testing to a symptom-limited endpoint (VLIM).
Endurance Exercise Training (EXT)
Outpatient pulmonary rehabilitation consisted primarily of EXT and
was conducted at a chronic care and rehabilitation hospital. The graduated program was comprised of three sessions per wk over a period
of 6 wk, with up to 12 patients per session being supervised by 2 to 3
physiotherapists. Each training session spanned a 2.5-h period and incorporated multimodality upper and lower limb exercises such as
walking, stair climbing, arm ergometry, cycle ergometry, treadmill exercise, and breathing exercises with appropriate rest intervals between
activities.
Training was targeted so that patients exercised at, or just below,
the Borg breathlessness rating corresponding to their symptom-limited maximum rating: based on Borg ratings, subjects trained at the
highest attainable work rate for the longest tolerable duration. Objec-
·
·
Figure 1. The test-retest repeatability of slopes of V E and V O2 over
HR over the duration of the study permitted quantification of
training intensity with respect to these physiologic parameters.
·
Shown here, pre- and post-EXT plots of V E/HR indicate that a
training HR of 108 6 8 beats/min (mean 6 1 SD) corresponded to
·
a training V E of 27 6 7 L/min throughout the training process
(shaded area).
1492
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
VOL 157
1998
TABLE 2
ACTIVITY-RELATED SYMPTOMS
Chronic breathlessness
Baseline Dyspnea Index
Transition Dyspnea Index
Oxygen cost diagram, mm
Acute symptomatology
B-time slopes, Borg/min
·
B- VO2 slopes, Borg/(ml/kg/min)
LE-time slopes, Borg/min
·
LE- Vo2 slopes, Borg/(ml/kg/min)
Reason for stopping exercise (% of subjects)
Breathlessness
Leg discomfort
Both
Other§
Precontrol
Postcontrol
(Pre-EXT)
Post-EXT
4.7 6 0.3
—
55 6 3
4.6 6 0.3
0.0 6 0.0
54 6 3
5.5 6 0.3*‡
3.2 6 0.3*
62 6 4†‡
0.89 6 0.07
0.59 6 0.07
1.01 6 0.10
0.68 6 0.09
0.98 6 0.11
0.69 6 0.09
0.92 6 0.12
0.65 6 0.09
0.59 6 0.09*‡
0.51 6 0.07†
0.54 6 0.09*‡
0.49 6 0.08†‡
45%
30%
20%
5%
40%
25%
25%
10%
45%
15%
15%
25%
* p , 0.001, †p , 0.01, significant difference from respective pre-intervention test; ‡p , 0.01, significant difference from precontrol
baseline.
§
Other reasons for stopping exercise included general fatigue and mouthpiece-related reasons.
·
·
·
·
exchange and efficiency (SaO2, V O2, V CO2, V CO2/ V O2), ventilation
·
·
( V E, V E /MVC), breathing pattern (F, VT, VT/TE, VT/TI, respiratory
duty cycle [TI /Ttot]), operational lung volumes (IC, inspiratory reserve volume [IRV], VT/IC) and symptom intensity (B, LE). Stepwise
multiple regression analysis was then carried out to establish the best
predictive equation for improvement; resultant models were reestimated with significant predictors only. Predictors of changes in exertional symptoms in response to EXT were evaluated similarly.
RESULTS
Subjects
Subject characteristics are provided in Table 1. Medications
being taken during the time of the study included b2-agonists
(all patients), ipratropium bromide (n 5 16), inhaled corticosteroids (n 5 17), and oral theophyllines (n 5 3). Three patients used supplemental oxygen consistently throughout the
exercise training program, however, they completed all testing
procedures while breathing room air and without desaturating
below 80% in doing so.
Training Stimulus
By targeting the training intensity at the highest tolerable Borg
breathlessness rating, patients consistently reached a training
HR of 108 6 2 beats/min during the entire EXT program. A
high degree of test-retest repeatability was shown in the
cycle
·
E and
tests
between
HR
and
concurrent
measurements
of
V
·
·
·
V O2: slopes of V O2/HR and V E/HR did not change throughout the study (Figure 1). This allowed us to quantify the physiological training stimulus to which the group was exposed,
thereby
confirming that training was carried out at high levels
·
·
of V E· (27 6 1 L/min, 61 6 3% MVC, or 89 6 5% of peak V E)
57 6 4% predicted
maximum, or
and V O2 (11 6 3 ml/kg/min,
·
·
96 6 7% of peak V O2). At this training V E, the coinciding values of VT/TI (1.0 6 0.3 L/s) and F (23 6 5 breaths/min) represented 76 6 20% and 83 6 14% of their maximum values, respectively. Therefore, the use of Borg breathlessness ratings
and HR measurements served as reliable options for the monitoring of training intensity in this population.
Progression within the EXT program (i.e., upward adjustment of exercise intensity and/or duration) was dependent on
each individual’s functional capacity and rate of symptom
improvement. Endurance exercise (i.e., walking, cycling, and
arm cranking) of the prescribed intensity increased from a
mean total duration of 24 6 1 min per session (early program)
to 37 6 1 min per session (late program), not including the
warm-up and cool-down periods. The MET (metabolic equivalents or multiples of the resting oxygen consumption) level of
activities performed during training sessions also increased
over the course of the program: the mean conditioning intensity increased during each of walking (2.5 6 0.1 to 3.7 6 0.3
METS), cycling (4.6 6 0.3 to 5.1 6 0.3 METS), and arm cranking (4.1 6 0.3 to 4.3 6 0.3 METS) from early to late program.
Symptom Limitation
Figure 2. During constant-load cycle exercise, slopes of exertional
breathlessness and leg discomfort over time fell significantly (*p ,
0.0005) after EXT (circles, solid line) compared with control
(dashed lines). Endurance time also increased significantly (p ,
0.05) after EXT. C1 5 precontrol (squares), C2 5 postcontrol/preEXT (triangles).
Upon entry into the study, exercise was limited primarily by
breathlessness and secondarily by leg discomfort
(Table 2).
·
The strongest correlates of the attained peak V O2 (%predicted)
as %predicted (r 5 0.73, p , ·0.0005),
were the FEV1 expressed
·
and the slopes of B/ V O2 (r 5 20.64, p 5 0.002) and LE/ V O2 (r 5
20.68, p 5 0.001). After accounting
for FEV1 in the multiple
·
regression analysis, peak V O2 correlated best with quadriceps
strength
(p 5 0.005), body mass index (p 5 0.012), and LE/
·
V O2 (p 5 0.021); each of these independent variables were interrelated and suggestive of reduced peripheral muscle function.
Chronic activity-related breathlessness was significantly improved after EXT but not the control period (Table 2): in re-
O’Donnell, McGuire, Samis, et al.: Impact of Exercise Training on Muscle Function in CAL
1493
Figure 4. Respiratory and peripheral muscle strength improved after EXT (solid bars) but not control (hatched bars). Values are
mean 6 SEM. *p , 0.05, **p , 0.01, significant difference from
pre-intervention baseline.
tory muscle endurance increased 2.8-fold by 1.6 6 0.4 min
(p , 0.01) in conjunction with significant reductions in the intensity of both breathlessness and inspiratory difficulty (Figure 3). Each test using the endurance breathing device was
performed with a similar breathing pattern strategy for each
subject and a mean peak inspiratory Pm of 24 6 3 cm H2O or
46 6 6% of the precontrol MIP. MIP, MEP, and inspiratory
muscle endurance did not change during the control period.
Peripheral Muscle Strength and Endurance
Figure 3. Exercise endurance, and standardized Borg ratings of
exertional breathlessness and perceived muscle discomfort improved significantly in response to EXT (solid bars) but not control
(hatched bars). Values are mean 6 SEM. *p , 0.05, **p , 0.01,
significant difference from pre-intervention baseline.
sponse to EXT, BDI and oxygen cost diagram measurements
improved by 23 6 5% (p , 0.001) and 18 6 6% (p , 0.05), respectively. In the acute setting, exertional symptom intensity
was also decreased after EXT ·but not control: slopes of both B
and LE relative to time or V O2 fell significantly after EXT
(Table 2, Figure 2), and exertional B and LE at isotime were
significantly reduced during various modes of exercise after
EXT (Figure 3). Although symptom
intensity was decreased
·
after EXT at any given time, V O2 or work rate, the reasons for
stopping exercise did not change significantly (Table 2).
Ventilatory Muscle Strength and Endurance
Static inspiratory and expiratory muscle strength increased in
response to EXT by 22 6 5 cm H2O (p 5 0.05) and 23 6 4 cm
H2O (p , 0.01), respectively (Figure 4). After EXT, inspira-
Prior to EXT, isometric strength of the quadriceps muscles
was significantly reduced at 51 6 5% of predicted normal. After EXT, quadriceps strength increased significantly (p ,
0.001) (Figure 4) with modest improvements in isometric endurance (p 5 0.068). Handgrip strength did not change in response to EXT (Figure 4); however, there were modest improvements in handgrip endurance after EXT (p 5 0.049).
There were no changes in muscle strength or endurance during the control period.
Significant improvements in dynamic exercise endurance
were found in response to EXT (Figure 3): endurance time on
the cycle ergometer increased by 2.4 6 0.6 min or 43 6 10%
(p 5 0.001); 6-min walking distance improved by 72 6 14 m or
34 6 9% (p , 0.0005); treadmill walking endurance improved
by 3.2 6 0.5 min or 40 6 8% (p , 0.0005); and upper limb endurance during arm ergometry increased by 0.8 6 0.4 min or
12 6 5% (p , 0.05). Each of these improvements in exercise
time occurred in association with reductions in ratings of perceived breathlessness and peripheral muscle discomfort (Figure 3). In contrast, cycling, walking or upper limb exercise performance or the associated ratings of exertional symptom
intensity did not change after the control period (Figure 3).
Post-EXT improvement in cycle exercise tolerance was predominantly related to improved efficiency and reduced venti·
with D V O2/
latory demand: DWorkTOT correlated significantly
·
time slopes (r 5 20.66,· p 5 0.002), D V CO2/time slopes (r 5
20.54, p 5 0.014), and V E/time slopes (r 5 20.47, p 5 0.036).
Physiologic Training Responses
Resting spirometry (FEV1, FVC, IC) did not change significantly after EXT or the control period. Assuming that TLC
1494
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
VOL 157
1998
TABLE 3
SYMPTOM-LIMITED PEAK EXERCISE RESPONSES DURING CONSTANT-LOAD
CYCLE EXERCISE AT 30 6 3 WATTS
Endurance time, min
Breathlessness, Borg
Leg discomfort, Borg
SaO2, %
HR, beats/min (%pred max)
·
VO2, L/min (%pred max)
·
VE, L/min (%MVC)
· ·
VE/ VCO2
VT, L (%IC)
F, breaths/min
IC, L
DH, L
EILV, L (%TLC)
Precontrol
Postcontrol
(Pre-EXT)
Post-EXT
5.7 6 0.4
5.3 6 0.3
5.4 6 0.4
93.7 6 0.6
1156 3 (70)
0.87 6 0.06 (68)
32.4 6 1.9 (68)
43.1 6 1.5
1.26 6 0.08 (79)
26.9 6 1.8
1.72 6 0.15
0.26 6 0.08
6.09 6 0.30 (93)
5.4 6 0.4
5.3 6 0.3
5.3 6 0.4
94.4 6 0.3
112 6 3 (68)
0.86 6 0.06 (66)
31.8 6 2.2 (67)
44.3 6 1.4
1.15 6 0.07 (76)
28.1 6 1.7
1.58 6 0.14
0.30 6 0.08
5.98 6 0.32 (93)
7.8 6 0.9*
4.3 6 0.4†
3.5 60.4†
93.8 6 0.5
113 6 3 (69)
0.86 6 0.07 (66)
31.4 6 2.3 (67)
43.3 6 1.4
1.18 6 0.08 (80)
27.3 6 1.8
1.52 6 0.12
0.29 6 0.08
6.08 6 0.32 (92)
Definition of abbreviations: % pred max 5 percent predicted maximum; DH 5 dynamic hyperinflation 5 change in end-expiratory lung
volume (EELV) from rest; EILV 5 end-inspiratory lung volume.
*p , 0.001, †p , 0.05, significant difference from previous study phase.
did not change, the absence of change in IC reflects that FRC
did not change during either EXT or control. Cardiorespiratory and breathing pattern parameters at rest were unaltered
after both EXT and control.
Despite improvements in endurance time, maximum values
for physiologic measurements at end-exercise did not change
in response to EXT (i.e., peak exercise responses in Table ·3).
There were no significant changes in exercise slopes of V E,
·
·
·
V E/MVC, IC, HR, SaO2, and V CO2 relative to V O2 or in
breathing pattern slopes
of F, VT, VT/IC, VT/TI, VT/TE, and
·
relative
to
E after EXT or control; however, slopes
TI/Ttot
V
·
·
·
of V O2, V CO2, V E, and F relative to time fell significantly (p ,
0.05) in response to EXT (Figure 5). Changes in physiologic
responses at an isotime during exercise
are summarized in Ta·
ble 4: the 3.4 6 1.1 L/min fall in V E was accomplished primarily by decreasing F by 2.5 6 0.9 breaths/min (r 5 0.50, p 5
·
Figure 5. Ventilatory responses to constant-load cycle exercise are shown. Slopes of V CO2, V E, and F over
time fell significantly (*p , 0.05) after EXT (circles, solid line) compared with control (dashed lines). C1 5
precontrol (squares), C2 5 postcontrol/pre-EXT (triangles).
O’Donnell, McGuire, Samis, et al.: Impact of Exercise Training on Muscle Function in CAL
·
TABLE 4
EXERCISE RESPONSES AT ISOTIME DURING ENDURANCE EXERCISE*
Breathlessness, Borg
Leg discomfort, Borg
SaO2, %
HR, beats/min
·
VO2, L/min
·
VCO2, L/min
·
VE,L/min
·
VE/VCO2
VT, L
VT/IC, %
F, breaths/min
TI/Ttot
IC, L
DH, L
IRV, L
1495
Precontrol
Postcontrol
(Pre-EXT)
Post-EXT
3.9 6 0.3
4.2 6 0.4
94.6 6 0.5
110 6 3
0.77 6 0.06
0.67 6 0.05
29.1 6 1.8
45.0 6 1.5
1.21 6 0.08
75 6 3
25.1 6 1.7
0.39 6 0.01
1.73 6 0.14
0.24 6 0.08
0.48 6 0.08
4.5 6 0.4
3.9 60.3
94.3 6 0.3
110 6 3
0.82 6 0.06
0.69 6 0.05
29.8 6 1.9
44.9 6 1.4
1.17 6 0.08
75 6 3
26.2 6 1.8
0.38 6 0.01
1.61 6 013
0.27 6 0.09
0.44 6 0.08
2.6 6 0.4*†
2.1 6 0.3†
94.4 6 0.4
107 6 3†
0.74 6 0.06†
0.62 6 0.05†
26.4 6 1.9†
44.3 6 1.6
1.14 6 0.07
73 6 3
23.7 6 1.6†
0.38 6 0.01
1.63 6 0.13
0.18 6 0.08
0.49 6 0.09
Definition of abbreviations as in Table 3.
* Isotime 5 4.6 6 0.3 min at 30 6 3 watts.
†
p , 0.05, significant difference from previous study phase.
·
0.026), in strong association
with reductions in both V CO2 (r 5
·
0.91, p , 0.005) and V O2 (r 5 0.88, p , 0.005). The behavior
of operational lung volumes during exercise was unaltered in
response to EXT (Tables 3 and 4).
Correlates of Symptom Improvement
The relief of activity-related breathlessness (both acute and
chronic) after EXT did not correlate significantly with any index of increased respiratory muscle strength and endurance:
improved TDI did not correlate with increased MIP (p 5
0.24), MEP (p 5 0.29), or inspiratory muscle endurance (p 5
0.97), and reduced B/time slopes did not correlate with increased MIP (p 5 0.80), MEP (p 5 0.44), or inspiratory muscle endurance (p 5 0.93). The only significant correlate of relief of exertional breathlessness (Borg at isotime) was the
concurrent reduction in breathing frequency (r 5 0.58, p ,
0.01). In this group, improved cycle endurance (DWorkTOT)
was not significantly related to reduced B/time slopes (p 5
0.17), increased MIP (p 5 0.31), or increased inspiratory muscle endurance (p 5 0.70).
Likewise, the reduction of LE/time slopes in response to
EXT did not correlate with increased quadriceps strength (p 5
0.46), increased isometric endurance (p 5 0.35), or with any
other measured parameter. However, exercise slopes of perceived leg discomfort fell in direct proportion with the reduction
in breathlessness slopes (DB/time 5 0.79[DLE/time] 2 0.10; r 5
0.78, p , 0.0005). Improved cycle endurance (DWorkTOT) was
not related to decreased LE/time slopes (p 5 0.98), increased
quadriceps strength (p 5 0.50), or increased isometric quadriceps endurance (p 5 0.82).
DISCUSSION
The novel findings of this study are as follows. First, general
exercise training, when targeted at high levels of ventilation,
significantly increased the strength and endurance of ventilatory muscles in severe chronic obstructive pulmonary disease
(COPD). Second, the improvement in exertional dyspnea and
exercise endurance after EXT was explained primarily by reduced ventilatory requirements, and not directly by mechanical changes or improved muscle strength. Third, the reduction
in ventilation during exercise after EXT correlated signifi·
cantly with the improvement in efficiency (i.e., reduced V CO2
and V O2 for a given work rate) and not with changes in dynamic ventilatory mechanics. Finally, EXT resulted in modest,
but consistent, increases in peripheral muscle strength as well
as reduced exertional leg discomfort.
Ventilatory Muscle Function
General exercise training improved inspiratory muscle strength
and endurance and increased expiratory muscle strength in
these patients with severe CAL. The magnitude of improvement was comparable to that achieved previously in studies
using specific inspiratory muscle training in CAL, but exceeded that previously reported following general EXT in
normal subjects (6), in patients with cystic fibrosis (7), and in
CAL (8, 9, 28). Differences in results among the various studies may reflect differences in the level of the training stimulus
administered and in the baseline patient characteristics.
The use of targeted Borg dyspnea ratings (29) in conjunction with HR monitoring, proved reliable in ensuring that· patients trained at the highest tolerable ventilation (and V O2)
for the longest possible duration (Figure 1). With physiologic
data obtained from the cycle exercise tests, we calculated that
using· this Borg-targeted approach, patients trained
at an aver·
age V O2 of 57% predicted maximum and at a ·V E of 27 L/min
or approximately 89% of the measured peak V E. This level of
ventilation corresponded to near maximal values of breathing
frequency and VT/TI, and an end-inspiratory lung volume
(EILV) approaching TLC (i.e., 93% TLC). Training duration
at the target intensity increased from an initial average of 24
min up to 37 min over the 6-wk training period.
Our graduated training program incorporated elements of
both endurance training and strength training. The finding
that static strength measurements consistently increased suggests that ventilatory muscles, during the course of training,
were sufficiently overloaded (beyond a critical level) to induce
important structural and functional adaptation (30). Static expiratory muscle strength increased in parallel with increased
inspiratory muscle strength after EXT. Recent studies have
provided evidence that expiratory muscle recruitment, in conjunction with accessory muscle recruitment during inspiration,
often occurs in exercising CAL patients (31, 32). It is further
postulated that this strategy serves to unload the overburdened, mechanically compromised diaphragm in these hyperinflated patients (31, 32). Although the expiratory muscles
utilize a smaller fraction of their maximal force-generating capacity than the inspiratory muscles do, our results suggest that
repetitive expiratory muscle activity during exercise can increase the strength of these muscles over a 6-wk training period.
Although the measurements of ventilatory muscle strength
and endurance that we employed are, to some extent, motivationally dependent, we believe the changes observed represent real functional improvement rather than motivational effects. First, strength and endurance measurements have been
shown to be reproducible: Black and Hyatt have shown highly
reproducible MIP measurements even after up to 3 d of practice (26). In addition, previous controlled studies that used
strength and endurance measurements as outcome measures
showed little or no change during the control period (10), as
was the situation in our group before and after the nonintervention period. We believe that the duration between testing
(i.e., 6-wk) was too long to result in learning effects or improved skill of performance that could occur if tests were done
repeatedly at short intervals. Technical factors are unlikely to
have accounted for our changes in respiratory muscle function
measurements: improvement in inspiratory muscle endurance
was not due to breathing pattern alterations during testing,
1496
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
and improvement in MIP was not the result of altered residual
lung volume after EXT.
We could not demonstrate any relationship between improved muscle function, improved symptoms, and improved
exercise performance. The lack of association between improved inspiratory muscle function and reduced dyspnea was
unexpected. Theoretically, improved inspiratory muscle strength
could lead to reduced dyspnea: improvement in the inspiratory muscle force reserve means that after EXT, inspiratory
muscles would utilize a lesser fraction of their maximal forcegenerating capacity than before (12). Therefore, less neural
activation (or effort) would be required for a given force generation by the muscle. In the sensory domain, this could lead
to reduced perceived breathing discomfort. However, in previous clinical studies using specific inspiratory muscle training,
the relationship between improved static inspiratory muscle
strength and improved symptoms has been weak (10, 11).
Moreover, the true prevalence of inspiratory muscle weakness
in the CAL population is not precisely known and there are
recent reports of relatively well preserved inspiratory muscle
function even in severely hyperinflated patients with advanced CAL (33). In the majority of our study subjects, MIP
measurements were not significantly reduced at baseline (i.e.,
63 6 6 cm H2O or 70% predicted maximum). Therefore, small
increases in strength in such patients may not lead to any appreciable symptomatic benefit. It is noteworthy that in a previous study by Harver and coworkers (10) that showed improvement in chronic dyspnea after targeted inspiratory muscle
training, baseline MIP measurements were substantially reduced (i.e., 47 cm H2O) compared with our group. In fact,
measurements of MIP in their group after training only
matched those of our group at study entry. It is conceivable
that improvement in inspiratory muscle strength after training
(either EXT or specific training) will lead to exertional dyspnea reduction only in those patients whose baseline static inspiratory muscle strength is critically diminished. Alternatively,
the lack of correlation between improved muscle function and
dyspnea relief after EXT in this study may simply reflect an insufficient sample size. However, the very low statistical probabilities of association that we found between these parameters
indicate the unlikelihood of establishing significant interrelationships among these measurements, even if the sample size
had been increased.
Mechanisms of Reduced Ventilatory Demand
Improvement in cycle exercise performance after EXT · was
explained
by improved efficiency (i.e., reduced slopes of V CO2
·
and V O2 over time) and reduced ventilation. The reduction in
ventilation after EXT primarily occurred as a result of reduced breathing frequency which was, in turn, the most important determinant of reduced exertional dyspnea. Modest,
but consistent reductions in submaximal ventilation have previously been reported following exercise training and have
variously been attributed to improved oxidative capacity (34,
35), improved efficiency (4), and, more recently, to altered
breathing pattern (13). We cannot exclude the possibility of
improved oxidative metabolism as a contributing factor in the
absence of muscle biochemistry data and blood lactate
con·
·
centrations. However, the absence of a change in V CO2/ V O2
slopes during exercise after EXT does not support this contention.
Reduction in ventilation occurred in association with an
unaltered VT. The lack of change in VT was expected, because
VT is relatively fixed in advanced COPD given the severe mechanical constraints on VT expansion. Moreover, resting IC,
which represents the respiratory system’s operating limits for
VOL 157
1998
VT, and the extent of dynamic lung hyperinflation during exercise were identical before and after EXT. Our results suggest that the main explanation for reduced ventilation after
EXT is reduced metabolic loading (Figure 5) rather than increased VT, reduced physiologic dead space, and enhanced
CO2 elimination as has been recently postulated (13). Modest
reductions in ventilation following EXT in mechanically compromised patients would likely have a salutary effect on dyspnea as a result of reduced central motor command output
from the respiratory center.
Peripheral Muscle Function
Reduced quadriceps strength was previously shown by Gosselink and coworkers (36) to correlate with reduced 6-min
walking distance (r 5 0.63, p , 0.005) and reduced maximal
oxygen consumption (r 5 0.55, p , 0.005) in a series of 41 patients with severe COPD. We confirmed these associations
and also established an association between reduced peripheral muscle strength and increased intensity of exertional leg
discomfort (effort) in COPD. However, modest increases in
lower limb static strength in our study were not shown to be
associated with reduced sense of leg discomfort or improved
exercise endurance after EXT. This finding could mean, as we
presumed to be the case with ventilatory muscles, that baseline peripheral muscle strength in our group was above a critical value associated with increased sense of leg discomfort or
that the magnitude of change in strength after EXT was insufficient to alter symptom perception. Alternatively, more sophisticated tests of peripheral muscle function and a larger
study sample may be required to uncover a true relationship
between improved function and symptom alleviation.
The corollary of our findings is that much of the mechanistic basis for the major benefits of EXT (i.e., reduced dyspnea
and improved exercise tolerance) remains unexplained. While
reduced ventilatory demand, as a result of improved metabolic factors undoubtedly contributes, such factors account for
only a small percentage of the variance in improved Borg ratings and exercise endurance times. We can only conclude that
other “nonphysiological” factors, such as the development of
increased tolerance to dyspneogenic stimuli or altered perceptual response to evoked sensations, may also importantly contribute to symptom alleviation. The fact that perceived intensity of dyspnea and leg discomfort decreased in parallel and to
a similar extent points to possible global nonspecific effects of
supervised exercise.
In summary, whole body exercise training, targeted at the
highest tolerable intensity, leads to consistent improvements
in ventilatory and peripheral muscle strength and endurance
in severe CAL. In this study, the magnitude of ventilatory
muscle functional improvement was comparable to that achieved
by specific inspiratory muscle training. Although it is likely
that improvements in ventilatory muscle strength and endurance are clinically advantageous in this population (e.g., during infective exacerbations), they were not shown to contribute to the relief of exertional dyspnea or the improvement in
exercise endurance following exercise reconditioning. Increased exercise endurance and reduced dyspnea were explained primarily as a result of a combination of induced metabolic and ventilatory alterations.
References
1. Ries, A. L., R. M. Kaplan, T. M. Limberg, and L. M. Prewit. 1995. Effect
of pulmonary rehabilitation on physiologic and psychological outcomes in patients with chronic obstructive pulmonary disease. Ann.
Intern. Med. 122:823–832.
2. Wijkstra, P. J., R. Van Altena, J. Kraan, V. Otten, D. S. Postma, and
O’Donnell, McGuire, Samis, et al.: Impact of Exercise Training on Muscle Function in CAL
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
G. H. Koeter. 1994. Quality of life in patients with chronic obstructive
pulmonary disease improves after rehabilitation at home. Eur. Respir.
J. 7:269–273.
Goldstein, R. S., E. H. Gort, D. Stubbing, M. A. Avendano, and G. H.
Guyatt. 1994. Randomized controlled trial of respiratory rehabilitation. Lancet 344:1394–1397.
O’Donnell, D. E., M. McGuire, L. Samis, and K. A. Webb. 1995. The impact of exercise reconditioning on breathlessness in severe chronic airflow limitation. Am. J. Respir. Crit. Care Med. 152:2005–2013.
Robinson, E. P., and J. M. Kjeldgaard. 1982. Improvement in ventilatory
muscle function with running. J. Appl. Physiol. 52:1400–1406.
Clanton, T. L., G. F. Dixon, J. Drake, and J. E. Gadek. 1987. Effects of
swim training on lung volumes and inspiratory muscle conditioning. J.
Appl. Physiol. 62:39–46.
Orenstein, D. M., B. A. Franklin, C. F. Doershuk, H. K. Hellerstein,
K. J. Germann, J. G. Horowitz, and R. C. Stern. 1981. Exercise conditioning and cardiopulmonary fitness in cystic fibrosis: the effect of a
three month supervised running program. Chest 80:392–398.
Belman, M. J., and B. A. Kendregan. 1982. Physical training fails to improve ventilatory muscle endurance in patients with chronic obstructive pulmonary disease. Chest 81:440–443.
Ries, A. L., B. Ellis, and R. W. Howkind. 1988. Upper extremity exercise
training in chronic obstructive lung disease. Chest 93:688–692.
Harver, A., D. A. Mahler, and J. A. Daubenspeck. 1989. Targeted inspiratory muscle training improves respiratory muscle function and
reduces dyspnea in patients with chronic obstructive pulmonary disease. Ann. Intern. Med. 111:117–124.
Lisboa, C., V. Munoz, T. Beroiza, A. Leiva, and E. Cruz. 1994. Inspiratory muscle training in chronic airflow limitation: comparison of two
different training loads with a threshold device. Eur. Respir. J. 7:1266–
1274.
Jones, N. L., and K. J. Killian. 1991. Limitation of exercise in chronic airway obstruction. In N. S. Cherniack, editor. Chronic Obstructive Pulmonary Disease. W. B. Saunders, Philadelphia. 196–206.
Casaburi, R., J. Porszasz, M. R. Burns, E. R. Carithers, R. S. Y. Change,
and C. B. Cooper. 1997. Physiologic benefits of exercise training in rehabilitation of patients with severe chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 155:1541–1551.
Stoller, J. K., R. Ferranti, and A. R. Feinstein. 1986. Further specification and evaluation of a new clinical index for dyspnea. Am. Rev. Respir. Dis. 135:1129–1134.
American Thoracic Society. 1995. Standardization of spirometry: 1994
update. Am. J. Respir. Crit. Care Med. 152:1107–1136.
Morris, J. F., A. Koski, and L. C. Johnson. 1971. Spirometric standards
for healthy non-smoking adults. Am. Rev. Respir. Dis. 103:57–67.
Goldman, H. I., and M. R. Becklake. 1959. Respiratory function tests:
normal values at median altitudes and the prediction of normal results. Am. Rev. Tuberc. Pulm. Dis. 79:457–467.
Gaensler, E. A., and G. W. Wright. 1966. Evaluation of respiratory impairment. Arch. Environ. Health 12:146–189.
Briscoe, W. A., and A. G. Dubois. 1959. The relationship between airway resistance, airway conductance, and lung volumes in subjects of
different age and body size. J. Clin. Invest. 37:1279–1285.
1497
20. Mahler, D. A., D. A. Weinberg, C. K. Wells, and A. R. Feinstein. 1984.
The measurement of dyspnea: contents, interobserver agreement, and
physiologic correlates of two new clinical indexes. Chest 85:751–758.
21. O’Donnell, D. E., and K. A. Webb. 1993. Exertional breathlessness in
patients with chronic airflow limitation: the role of lung hyperinflation. Am. Rev. Respir. Dis. 148:1351–1357.
22. Jones, N. L. 1988. Clinical Exercise Testing, 3rd ed. W. B. Saunders, Philadelphia.
23. Dillard, T. A., S. Piantadosi, and K. R. Rajagopal. 1985. Prediction of
ventilation at maximal exercise in chronic airflow obstruction. Am. Rev.
Respir. Dis. 132:230–235.
24. Borg, G. A. V. 1982. Psychophysical basis of perceived exertion. Med.
Sci. Sports Exerc. 14:377–381.
25. Hamilton, A. L., K. J. Killian, E. Summers, and N. L. Jones. 1995. Muscle strength, symptom intensity, and exercise capacity in patients with
cardiorespiratory disorders. Am. J. Respir. Crit. Care Med. 152:2021–
2031.
26. Black, L. F., and R. E. Hyatt. 1969. Maximal respiratory pressure: normal values and relationship to age and sex. Am. Rev. Respir. Dis. 99:
696–702.
27. Nickerson, B. G., and T. G. Keens. 1982. Measuring ventilatory muscle
endurance in humans as sustainable inspiratory pressure. J. Appl. Physiol. 52:768–772.
28. Berry, M. J., N. E. Adair, K. S. Sevensky, A. Quinby, and H. M. Lever.
1996. Inspiratory muscle training and whole-body reconditioning in
chronic obstructive pulmonary disease. Am. Rev. Respir. Crit. Care Med.
153:1812–1816.
29. Horowitz, M. B., B. Littenberg, and D. A. Mahler. 1996. Dyspnea ratings
for prescribing exercise intensity in patients with COPD. Chest 109:
1169–1175.
30. Belman, M. J., W. C. Botnick, S. D. Nathan, and K. H. Chon. 1994. Ventilatory load characteristics during ventilatory muscle training. Am. J.
Respir. Crit. Care Med. 149:925–929.
31. Martinez, F., J. Couser, and B. Celli. 1990. Factors influencing ventilatory muscle recruitment in patients with chronic airflow obstruction.
Am. Rev. Respir. Dis. 142:276–282.
32. Ninane, V., F. Rypens, J. C. Yernault, and A. DeTroyer. 1992. Abdominal muscle use during breathing in patients with chronic airflow obstruction. Am. Rev. Respir. Dis. 146:16–21.
33. Similowski, T., S. Yan, A. P. Gauthier, P. T. Macklem, and F. Bellemare.
1991. Contractile properties of the human diaphragm during chronic
hyperinflation. N. Engl. J. Med. 325:917–923.
34. Casaburi, R., A. Patessio, F. Ioli, S. Zanaboni, C. F. Donner, and K.
Wasserman. 1991. Reductions in exercise lactic acidosis and ventilation as a result of exercise training in patients with obstructive lung
disease. Am. Rev. Respir. Dis. 143:9–18.
35. Maltais, F., P. LeBlanc, C. Simard, J. Jobin, C. Berube, J. Bruneau, L.
Carrier, and R. Belleau. 1996. Skeletal muscle adaptation to endurance training in patients with chronic obstructive pulmonary disease.
Am. J. Respir. Crit. Care Med. 154:442–447.
36. Gosselink, R., T. Troosters, and M. Decramer. 1996. Peripheral muscle
weakness contributes to exercise limitation in COPD. Am. J. Respir.
Crit. Care Med. 153:976–980.