Eighteen sedentary males with COPD volunteered to participate in this study. The diagnosis of COPD was based on spirometry showing moderate to severe irreversible airflow obstruction (FEV₁ of less than 60% predicted and FEV₁/FVC of less than 70%) (13). Subjects were stable at the time of the study, and none suffered from cardiovascular, neurologic, skeletal muscle, or any other condition that could alter their capacity to perform or influence an exercise test. The research protocol was approved by the institutional ethics committee, and a signed, informed consent was obtained for each subject.

After reviewing their medical history and familiarization with procedures, subjects filled out a physical activity questionnaire. Anthropometric measurements, pulmonary function testing, computed tomography of the thigh, and a symptom-limited incremental cycle exercise test were then obtained. On two subsequent visits, which were separated by two days, subjects performed constant work-rate cycling exercise to exhaustion. These tests were preceded by nebulization of 500 μg of IB (IB exercise) or placebo (placebo exercise), which were administered in a crossover, randomized, and double-blind fashion. Quadriceps force measurements were obtained at rest and at 10, 20, and 30 minutes after each exercise. Twitch force was measured during magnetic stimulation of the femoral nerve and maximum voluntary contractions (MVCs) were obtained. The study design is schematically depicted in Figure 1

Figure 1. Schematic representation of the experimental protocol. Placebo (P) and ipratropium bromide (IB) were administered in a random order. Measurement of quadriceps strength includes femoral nerve stimulation and maximal voluntary contraction (MVC).

Download Figure | Download Powerpoint

[More] [Minimize]

Standard pulmonary function tests, including spirometry, lung volumes, and diffusion capacity were obtained at the initial evaluation in all subjects according to previously described guidelines (14). Results were related to normal values of Knudson and colleagues, Goldman and Becklake, and Cotes and Hall, respectively (15–17). Routine spirometry was also performed before and 45 minutes after placebo or IB nebulization on the subsequent exercise days.

The level of physical activity in daily living was assessed by using the activity questionnaire described by Baecke and colleagues (18), adapted for older individuals (19), and used in patients with COPD (20). The questionnaire attributes a score for household, sport, and other leisure-time physical activities together resulting in a global physical activity score. A score of 9 to 16 indicates a moderate level of daily physical activity, whereas a score of less than nine is seen in sedentary subjects.

A computed tomography of the dominant thigh halfway between the pubic symphysis and the inferior condyle of the femur was performed using a fourth-generation Toshiba Scanner 900S. The midthigh muscle cross-sectional area was computed as previously described (8).

During the initial evaluation, a symptom-limited incremental cycle exercise test was performed to determine peak exercise capacity. Additional details about this measurement can be found in the online supplement.

Patients were asked to withdraw from short-acting β2-agonists (6 hours), long-acting β2-agonists (12 hours), anticholinergics (6 hours), and theophillines (24 hours) before each submaximal constant work-rate exercise test. One hour after nebulization of IB or placebo, subjects were seated on an electrically braked ergocycle as described previously here. After 5 minutes of rest and 1 minute of warm-up, a constant work-rate submaximal cycling exercise test was performed until exhaustion at a working intensity corresponding to 80% of the peak work rate achieved during the incremental exercise test. The perception of dyspnea and leg fatigue was assessed at 2-minute intervals during exercise using a modified Borg scale (21).

Before and after submaximal constant work-rate cycling exercise, the force of the dominant quadriceps was evaluated during a single magnetic stimulation of the femoral nerve and during a maximum voluntary contraction in a standardized supine position. Additional details on this method and a schematic representation of the experimental protocol (Figure E1) are available in the online supplement.

The TWq/power output relationship was obtained to evaluate whether the magnetic stimulation was supramaximal . The within-subject reproducibility of TWq was also evaluated by obtaining this measurements on two separate occasions in 17 subjects. Additional information and related Figure E2 on these measurements are provided in the online supplement.

Results are mean ± SD. A statistical level of significance of 0.05 was used for all analyses. The duration of constant work-rate exercise was defined as endurance time. Isotime measurements were obtained at the same time during both constant work-rate exercises. The effects of placebo and IB on resting lung function, endurance time, MVC, and TWq were compared with paired t test. An unpaired t test was used to compare age, body mass index, lung function, endurance time, MVC, and midthigh muscle cross-sectional area between fatiguers (a TWq fall of more than 15% at 10 minutes after placebo exercise) and nonfatiguers. Possible correlations between the changes in FEV₁ with IB, TWq, and endurance time with IB were evaluated using a Pearson correlation as was the within-subject reproducibility of TWq.

Subject's characteristics are presented in Table 1

The physiologic response to incremental exercise testing is shown in Table 2

The data obtained with validation experiments indicate that the femoral nerve stimulation at 100% power output was supramaximal. The within-subject between-trial coefficient of variation (6 ± 5%, range 0 to 16%) was similar to that obtained in a prior work of ours (11). Therefore, a decrease in TWq of more than 15% was considered as a true physiologic signal indicating contractile fatigue. The validation data are presented in greater details in the online supplement (Figure E2).

Prenebulization FEV₁ was similar for both endurance exercise tests, but FEV₁ was significantly greater 1 hour after IB nebulization than after placebo nebulization (1.07 ± 0.42 L vs. 0.93 ± 0.36 L, respectively, p < 0.05) (Table 3)

Results of quadriceps force at rest and after placebo and IB exercises are shown in Table 3. Compared with resting values, MVC at 10 and 30 minutes was stable after placebo exercise (96 ± 14% and 98 ± 13% of resting values) (Figure 2)

Figure 2. Quadriceps twitch force (TWq, A) and MVC (B) at rest and at 10, 20, and 30 minutes after placebo (closed circle) and IB exercise (open circle) (n = 18, mean ± SD). The dashed line indicates the fatigue threshold. *p < 0.05 versus resting values; ^†p < 0.05 versus placebo exercise.

Download Figure | Download Powerpoint

[More] [Minimize]

Nine patients demonstrated a more than 15% fall in TWq 10 minutes after placebo exercise and were considered as fatiguers. Fatiguers and nonfatiguers could not be differentiated on the basis of their age, body mass index, resting lung function, MVC, midthigh muscle cross-sectional area, and level of physical activity (Table 1). Apart from a tendency toward a lower endurance time during constant work-rate exercise in the nonfatiguers, the response to incremental and constant work-rate exercise did not differ significantly between fatiguers and nonfatiguers (Table 2). The dyspnea score was higher at peak exercise in the nonfatiguers compared with fatiguers, whereas the perception of leg fatigue was similar between the two groups. The reasons for stopping exercise were similar between the two groups (fatiguers: leg fatigue, n = 2; dyspnea, n = 4; both, n = 3; nonfatiguers: leg fatigue, n = 0; dyspnea, n = 6; both, n = 3).

In fatiguers, there was no significant change in endurance time with IB (394 ± 220 seconds after placebo versus 400 ± 119 seconds after IB) despite an 11 ± 18% improvement in FEV₁. For the nine subjects who did not develop contractile muscle fatigue after placebo exercise, a similar gain in FEV₁ with IB (13 ± 18%) was associated with a large improvement in endurance time (249 ± 124 seconds after placebo vs. 479 ± 298 seconds after IB_, p < 0.05). There was a significant correlation between changes in endurance time with IB compared with placebo (Δ endurance time) and TWq 10 minutes after placebo exercise (expressed in percentage of TWq resting value, r = 0.59, p = 0.01); that is, a large fall in TWq after placebo exercise predicted a small improvement in endurance time with IB (Figure 3)

Figure 3. Relationship between change in endurance time with IB compared with placebo (Δ endurance time) and the fall in TWq at 10 minutes after placebo exercise expressed in percentage resting value. There was a significant correlation between these two variables (r = 0.59, p = 0.01); that is, a large fall in TWq after placebo exercise predicted a small improvement in endurance time with IB.

Download Figure | Download Powerpoint

[More] [Minimize]

To evaluate the possible role of exercise duration in the development of contractile fatigue, the changes in TWq in relationship to exercise duration were evaluated in fatiguers and nonfatiguers (Figure 4)

Figure 4. TWq at rest and 10 minutes after placebo exercise (open symbols) and after IB exercise (closed symbols) in fatiguers (circle) and nonfatiguers (square). The dashed line indicates the fatigue threshold. The fall in TWq after IB exercise was still greater in fatiguers compared with nonfatiguers despite similar exercise duration. *p < 0.05 versus fatiguers.

Download Figure | Download Powerpoint

[More] [Minimize]

The validity of muscle strength measurements obtained through MVC is critically dependent on subjects making a maximal effort, and it is recognized that submaximal activation is common during volitional muscle testing (22). A strength of this investigation is that the force of the quadriceps muscle was assessed in an effort-independent manner. In our experience and other's experience, supramaximal stimulation of the femoral nerve using magnetic stimulation is a painless method for assessing quadriceps strength (23), which can be successfully employed to detect contractile fatigue of the quadriceps (10, 23). There are, however, potential pitfalls to this method. For the results to be valid, a supramaximal muscle twitch should be obtained. This cannot always be ascertained using the M wave, whose detection is often blurred by large artifacts during magnetic stimulation. This problem was addressed by obtaining a power output/twitch force relationship and showing that increasing the power output beyond 90–95% power output did not elicit further significant increases in the resulting TWq (see online supplement).

A second factor to deal with is the fact that twitch force resulting from magnetic stimulation is affected by prior muscle activation (potentiated twitch) and changes in leg position. Because the potentiated twitch may be a more sensitive and reproducible index of contractile fatigue, especially in mild contractile fatigue (24, 25), this method was employed in this study, and a standardized posture was used during stimulation. In accordance with our previous studies, a fall in TWq of more than 15% was used to define contractile fatigue. The use of this threshold in this study was further justified by our validation study showing that the variability of the TWq measurement was less than 15% in most circumstances.

The improvement in endurance time with IB in this study almost reached statistical significance, and its magnitude was comparable to that observed in a prior study (3). It is important to realize that the primary objective of this study was not to evaluate the effects of IB on exercise endurance; consequently, it was not powered to evaluate this outcome. Our results nevertheless indicate that the impact of a given bronchodilator response on exercise endurance is not uniform in patients with COPD and that contractile leg fatigue occurs after exercise in a significant proportion of patients with COPD.

The role of premature skeletal muscle fatigue in limiting exercise in COPD has been indirectly inferred from studies showing that this symptom is commonly perceived at peak exercise in these patients (4). In this investigation, we assessed directly how contractile fatigue of the quadriceps may influence exercise tolerance in COPD. There was a significant correlation between the fall in TWq after placebo exercise and the change in endurance time (Figure 3). In other words, the occurrence of contractile fatigue after placebo exercise predicted a small improvement in exercise tolerance with IB. Moreover, there was a nonsignificant change in endurance time with IB in fatiguers as opposed to nonfatiguers, despite similar improvement in FEV₁ with IB. These data provide further support to the contention that early peripheral muscle fatigue is involved in exercise limitation in some patients with COPD.

This study was not intended to investigate the cause of muscle fatigue in COPD. Although muscle fatigue is a complex phenomenon, changes in muscle energy metabolism such as increased lactate accumulation and early muscle acidosis reported in COPD (26, 27) may be involved. Increased production of reactive oxygen species is among other potential candidates (28, 29). Whether these alterations in muscle metabolism are related to an inadequate supply of energy to meet demands or to intrinsic muscle changes is debatable. Although whole-body cardiac output during exercise is usually preserved in COPD (30), the increased O₂ cost of breathing seen in this disease (31) may trigger a phenomenon of blood flow redistribution between lower limb and ventilatory muscles that could deprive the lower limb muscles from adequate O₂ delivery. In line with this concept, Harms and colleagues elegantly demonstrated that a reduction in work of breathing with noninvasive ventilatory support improved blood flow availability for the lower limb muscles and thus exercise tolerance in athletes (32, 33). However, in this study, the reduction in work of breathing with bronchodilation in patients with largely irreversible disease was probably of insufficient magnitude to affect significantly the blood flow redistribution between lower limb and ventilatory muscles, which would explain why IB did not increase exercise tolerance in patients who developed contractile fatigue after exercise.

The duration of exercise likely contributes to muscle fatigue, as suggested by the positive correlation between endurance time and the fall in TWq after exercise. Patients who become rapidly intolerant to exercise because of early dyspnea or limitation in ventilation do not sufficiently activate their peripheral muscles to develop muscle fatigue. Other patients with greater capacity to exercise will be able to challenge their peripheral muscle to the point of fatigue. However, the exercise duration was not the sole factor in explaining the occurrence of leg fatigue in our subjects because the fall in TWq after IB exercise was still greater in fatiguers compared with nonfatiguers despite similar exercise duration (Figure 4). This observation is consistent with a role of intrinsic muscle changes in the occurrence of muscle fatigue after exercise in COPD.

In the past decade, numerous studies have provided a long list of structural and functional changes occurring in peripheral muscles of patients with COPD (5–7). An attractive hypothesis is that these intrinsic and important muscle changes may be responsible in part for the exercise limitation observed in patients with COPD by increasing the muscle's susceptibility to contractile fatigue. The contention that origin of early muscle fatigue may lie in the skeletal muscle itself is supported by previous studies showing that peripheral O₂ delivery is usually preserved in normoxic patients with COPD (27) and that changes in muscle metabolism do not correlate with O₂ delivery in these individuals. It is very likely that the mechanisms underlying muscle fatigue may vary among patients according to their cardiorespiratory function, level of oxygenation, and the status of their peripheral muscles.

Optimal bronchodilation is the cornerstone in the treatment of patients with COPD whose objectives are not only to improve lung function but also to ameliorate shortness of breath, exercise tolerance, and ultimately quality of life. Previous investigations have reported that bronchodilators may fail to improve exercise tolerance despite significant improvement in lung function (3, 34). Our results indicate that the presence of leg fatigue may prevent optimal bronchodilation from translating into appreciable gains in exercise tolerance. These results thus confirm the multifactorial origin of exercise intolerance in COPD. A therapeutic approach, including both pharmacologic (bronchodilation) and nonpharmacologic (muscle training) interventions, is likely to provide patients with COPD the best chance to reach an optimal functional status.

Because of potential therapeutic implications, it could be useful to be able to identify with simple clinical tools patients who develop leg fatigue during exercise. This study was not designed nor powered to make detailed physiologic comparisons between fatiguers and nonfatiguers. However, these two groups of patients could not be differentiated on the basis of age, body mass index, resting lung function, physical activity score, and muscle mass. Furthermore, the perception of symptoms at peak exercise and the reasons for exercise limitation did not enable us to discriminate accurately between fatiguers and nonfatiguers. The absence of a close relationship between symptom's perception and physiologic function is commonly observed in patients with COPD (10) and other chronic diseases. Further studies are needed to help differentiate fatiguers from nonfatiguers.

We conclude that contractile fatigue occurs during exercise in some patients with COPD and may be a factor limiting exercise tolerance. Although, the mechanisms underlying muscle fatigue were not explored, the response to bronchodilation in terms of exercise tolerance was clearly different between fatiguers and non fatiguers, giving additional support to the thesis that peripheral muscle fatigue is a potential contributor to exercise intolerance in COPD.