Effect of Pulmonary Rehabilitation on Quadriceps Fatiguability during Exercise

Effect of Pulmonary Rehabilitation on Quadriceps Fatiguability during Exercise M. JEFFERY MADOR, THOMAS J. KUFEL, LILIBETH A. PINEDA, ANN STEINWALD, AJAY AGGARWAL, AMITA M. UPADHYAY, and MOHAMMED A. KHAN Division of Pulmonary and Critical Care Medicine, State University of New York at Buffalo, and the VA Western New York Healthcare System, Buffalo, New York We have recently shown that patients with chronic obstructive pulmonary disease (COPD) develop contractile fatigue of their quadriceps muscle following endurance exercise. Pulmonary rehabilitation can produce physiological adaptations in patients with COPD. We hypothesized that if pulmonary rehabilitation induces physiological adaptations in the exercising muscle, it should become more fatigue resistant. Twenty one patients with COPD, mean age 69.9 1.9 yr, FEV 1 45 4% predicted, participated in an 8-wk outpatient, supervised pulmonary rehabilitation exercise program. Quadriceps contractile fatigue was detected by a fall in quadriceps twitch force postexercise. Twitch force was measured during magnetic stimulation of the femoral nerve. Because potentiated twitches may be more sensitive at detecting fatigue, both unpotentiated (TwQu) and potentiated (TwQp) twitches were obtained before and 10, 30, and 60 min after constant load cycle exercise. Prerehabilitation, during constant load exercise, patients exercised at 37 4 W for 11.2 1.8 min. Prerehabilitation, TwQu fell significantly postexercise down to a minimum value of 82.5 3.1% of the baseline preexercise value (p 0.001). Similarly, prerehabilitation, TwQp fell significantly postexercise down to a minimum value of 73.9 3.9% of baseline (p 0.001)., for the same intensity and duration of exercise, TwQu was not significantly different from baseline at any time postexercise., TwQp fell significantly postexercise but the fall in TwQp with exercise was significantly less postrehabilitation compared with prerehabilitation (p 0.001). In conclusion, pulmonary rehabilitation resulted in increased fatigue resistance of the quadriceps muscle in patients with COPD. (Received in original form June 23, 2000 and in revised form October 3, 2000) Supported by the American Heart Association Western New York Affiliate. Correspondence and requests for reprints should be addressed to M. Jeffery Mador, M.D., Division of Pulmonary and Critical Care Medicine (111S), VAWNYHS, 3495 Bailey Avenue, Buffalo, NY 14215. E-mail: mador@acsu.buffalo.edu This article has an online data supplement, which is accessible from this issue s table of contents online at www.atsjournals.org Am J Respir Crit Care Med Vol 163. pp 930 935, 2001 Internet address: www.atsjournals.org Pulmonary rehabilitation of patients with chronic obstructive pulmonary disease (COPD) decreases exertional dyspnea and improves exercise endurance (1 3). The mechanistic basis for these improvements remains unclear. Psychological factors believed to be potentially important include increased patient motivation, phobic desensitization to unpleasant sensations induced by exertion (dyspnea, leg discomfort), and a decrease in the degree of dyspnea and leg discomfort at any given exercise workload. In healthy subjects, a vigorous endurance exercise program leads to physiological adaptations in the cardiovascular system and exercising muscle. It was initially believed that patients with COPD could not exercise at sufficient intensities to induce a physiological training effect (4). However, patients with COPD are often severely deconditioned. Following an intense pulmonary rehabilitation program, increases in the oxidative capacity of the exercised muscle has been demonstrated indicative of physiological adaptations within the exercising muscle (5). We have recently shown that patients with COPD develop fatigue of the quadriceps muscle following endurance exercise to the limits of tolerance (6). Fatigue was assessed by measuring quadriceps twitch force during magnetic stimulation of the femoral nerve before and after exercise (7). We hypothesized that if pulmonary rehabilitation induces physiological adaptations in the exercising muscle, it should become more fatigue resistant after pulmonary rehabilitation. Accordingly, we compared the degree of quadriceps contractile fatigue elicited by the same intensity and duration of exercise before and after pulmonary rehabilitation. We hypothesized that exercise would induce less contractile fatigue following pulmonary rehabilitation. METHODS Subjects Twenty-nine consecutive patients with COPD who entered our pulmonary rehabilitation program agreed to participate in the study. Five patients did not complete the program and were excluded. Maximal or near maximal stimulation of the femoral nerve could not be achieved in three patients. Thus, 21 patients were included in the study. There were 20 males and 1 female, aged 69.9 1.9 yr. Their height and weight were 1.75 0.02 m and 83.5 3.4 kg, respectively. The study was approved by the appropriate institutional review boards and written informed consent was obtained from all subjects. Pulmonary Function and Exercise Testing Pulmonary function was measured using standard techniques according to ATS recommendations (8). Pulmonary function measurements are shown in Table 1. Maximal incremental and constant workload cycle exercise was performed as previously described (6). Lactate levels were obtained before and after constant load exercise (6). After pulmonary rehabilitation was complete, three exercise tests were performed on separate days. On the first day, the maximal exercise test was repeated. On the second day the endurance test was repeated and the test was stopped by the investigator when the prerehabilitation endurance time was reached (for quadriceps and lactate measurements). On the final day, the endurance test was repeated but the patient was allowed to continue exercising to the limits of tolerance (to evaluate any potential improvement in exercise endurance). Six-minute walking tests were administered before (the best of three tests) and after completion of the rehabilitation program (one walk) as an additional test of functional capacity. Subjects were given standardized instructions to cover the greatest distance possible in 6 min. Standardized verbal encouragement was given each minute. Quadriceps Measurements Quadriceps twitch force was measured as previously described (6). In many of the patients, magnetic stimulation of the femoral nerve elicited a large shock artifact that obscured the compound motor action potential (M-wave). However, with careful positioning of the surface electrodes and ground, M-waves were obtained before rehabilitation in 10 patients and after rehabilitation in 8 patients. Following a vigorous voluntary contraction, the subsequent twitch is significantly increased in size (twitch potentiation) (14). Recent studies have sug-

Mador, Kufel, Pineda, et al.: Pulmonary Rehabilitation and Quadriceps Fatiguability during Exercise 931 TABLE 1. PULMONARY FUNCTION Actual % Predicted* FEV 1, L 1.48 0.12 45 4 FVC, L 2.95 0.16 68 4 FEV 1 /FVC, % 48 2 RV, L 4.55 0.33 183 13 TLC, L 7.60 0.28 109 4 DL CO, ml/min/mm Hg 14.2 1.2 55 5 MIP, cm H 2 O 64.7 8.0 62 7 MVV, L/min 56.3 4.4 46 4 Definition of abbreviations: DL CO diffusing capacity; FEV 1 forced expiratory volume in 1 s; FVC forced vital capacity; MIP maximal inspiratory pressure; MVV maximal voluntary ventilation; RV residual volume; TLC total lung capacity. * Predicted values as per references (9 13). gested that the potentiated twitch is more sensitive at detecting fatigue than the unpotentiated twitch, particularly when the amount of fatigue is small (15, 16). Accordingly, we measured both the unpotentiated and potentiated twitch before and 10, 30, and 60 min after exercise. Unpotentiated and potentiated twitches were measured as previously described (6, 18). To determine the degree to which our subjects could voluntarily activate their quadriceps muscle, twitch interpolation was employed during maximum voluntary contraction maneuvers as previously described (6, 17). An example is shown in Figure 1. To determine if exercise could elicit both high- and low-frequency fatigue, paired stimuli were obtained in 20 subjects as previously described (18). The paired stimuli were given at interstimulus intervals of 100 ms, 50 ms, and 10 ms corresponding to stimulation frequencies of 10, 20, and 100 Hz. The unpotentiated twitch was ensemble averaged and then digitally subtracted from the ensembleaveraged paired stimuli to obtain the amplitude of the second twitch, T 2. A significant decrease in T 2 postexercise would be indicative of fatigue at that particular stimulation frequency (19). A representative example is shown in Figure 2. Pulmonary Rehabilitation Program Patients underwent supervised exercise sessions three times a week for 8 wk. Patients exercised on the cycle ergometer and treadmill, performed calisthetics/stretching exercises (20) and received weekly educational sessions. Data Analysis Changes in variables over time were analyzed by repeated measures analysis of variance and paired t test with Bonferroni correction. Data are expressed as the mean SE. Changes in measurements before and after pulmonary rehabilitation were compared by analysis of variance and paired t test. Correlations between continuous variables were made using simple linear regression. An online data supplement to the METHODS section providing further details on subject characteristics, exercise, and quadriceps measurements and the pulmonary rehabilitation program is available online at www.atsjournals.org. Figure 2. Representative example to demonstrate how the T 2 amplitude is calculated. Ensemble average of the unpotentiated twitch ( ) and paired twitch at 10 Hz ( ), 20 Hz ( ), and 100 Hz ( ). The ensemble average was obtained by lining up all acceptable twitches at the same zero point and then obtaining the digital average. T 2 is obtained by digital subtraction of the ensemble averaged unpotentiated twitch from the ensemble averaged paired twitch at each stimulation frequency. T 2 is shown at 10 Hz ( ), 20 Hz ( ), and 100 Hz ( ). RESULTS Patients had moderately severe airflow obstruction with an FEV 1 of 1.48 0.12 l, 48 2% of the predicted value, air trapping (residual volume of 183 13% of predicted value) and a moderately decreased diffusing capacity, 55 5% of predicted value (Table 1). Cardiopulmonary Parameters during Exercise Maximal incremental exercise test results are shown in Table 2. Exercise capacity was severely decreased, 59 6 W, 41 4% predicted. Peak oxygen consumption ( O 2 ) averaged 47 3% predicted. End exercise heart rate reached 87 2% of the predicted maximum. Peak exercise minute ventilation ( E) averaged 81 4% of the 12 s maximal voluntary ventilation (MVV). Endurance exercise results are shown in Table 3. At baseline, patients exercised at 37 4 W for 11.2 1.8 min. Lactate levels 5 min postexercise were 4.1 0.4 meq/l. Effect of Pulmonary Rehabilitation on Cardiopulmonary Parameters Peak exercise responses during incremental exercise before and after pulmonary rehabilitation are shown in Table 2. Maximum exercise capacity increased from 59 6 at baseline to 79 7 W (34% improvement) (p 0.0001) (Table 2) postrehabilitation. Similarly, peak O 2 increased from 0.96 0.06 to 1.17 0.07 L/min (22% improvement) (p 0.0003) postrehabilitation (Table 2). Peak ventilation also increased postrehabilitation. Responses to identical levels of exercise (isowork) are also shown in Table 2. The highest workload reached by the patient during both the pre- and postrehabilitation exercise test was used for comparison. At the same workload, heart rate was significantly lower postrehabilitation (Table 2) (p 0.005). Exercise VE also tended to be lower postrehabilitation (p TABLE 2. MAXIMAL EXERCISE Prerehabilitation End Exercise Isowork Figure 1. Representative example of a superimposed twitch during a maximum voluntary contraction (MVC) maneuver. The patient did not make a perfect square wave and the superimposed twitch was not always obtained at the time of peak force development during the MVC maneuver. Thus, the degree of activation during the MVC maneuver will be slightly underestimated. Workload, W 59 6* 79 7 58 6 O 2, L/min 0.96 0.06 1.17 0.07 0.94 0.07 HR, beats/min 131 3 135 5 115 4 E, L/min 45.0 3.1 48.4 4.1 38.7 3.7 F, breath/min 35 8 35 1 28 1 VT, L 1.33 0.09 1.48 0.10 1.36 0.10 Definition of abbreviations: f respiratory rate; HR heart rate; E minute ventilation; O 2 oxygen consumption; VT tidal volume. * All values represent mean SE. Significant difference from prerehabilitation value.

932 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 163 2001 TABLE 3. ENDURANCE EXERCISE Prehabilitation End Exercise Exercise Isotime Endurance time, min 11.2 1.8* 23.7 3.0 11.2 1.8 O 2, L/min 0.92 0.08 1.05 0.06 0.98 0.07 HR, beats/min 134 4 136 3 123 4 E, L/min 42.5 3.3 42.5 3.0 39.7 2.2 F, breaths/min 34 2 35 1 32 1 VT, L 1.35 0.09 1.28 0.08 1.32 0.07 Peak lactate, meq/l 4.1 0.4 3.7 0.3 Definition of abbreviations: f respiratory rate; HR heart rate; E minute ventilation; O 2 oxygen consumption; VT tidal volume. * All values represent mean SE. Significant difference from prerehabilitation value. 0.05). The reduction in VE postrehabilitation was entirely due to a reduction in respiratory rate (p 0.001) whereas tidal volume (VT) was unchanged (Table 2). Because O 2 and carbon dioxide production (VCO 2 ) were unchanged postrehabilitation, significant reductions in E/VCO 2 (p 0.025) and E/ O 2 (p 0.005) were observed after rehabilitation. Endurance exercise responses (constant workload) before and after pulmonary rehabilitation are shown in Table 3. Endurance time increased from 11.2 1.8 at baseline to 23.7 3.0 min after pulmonary rehabilitation (p 0.0001). Despite the large increase in endurance exercise time, there were no significant changes in peak VE, breathing pattern (f, VT), or heart rate. Prerehabilitation, 11 patients stopped exercise because of shortness of breath, 9 patients stopped exercise because of leg fatigue, and 1 patient stopped exercise because of nasal irritation., 8 patients stopped exercise because of shortness of breath, 11 patients stopped exercise because of leg fatigue, and 1 patient stopped exercise because of knee pain and nasal irritation, respectively. Exercise responses at equivalent exercise duration (isotime) are also shown in Table 3. The longest duration reached by the patient during both the pre- and postrehabilitation exercise test was used for comparison. At exercise isotime, both heart rate (p 0.002) and E (p 0.02) were significantly lower postrehabilitation. Respiratory rate also tended to be lower postrehabilitation (p 0.068). Arterialized venous lactate levels obtained 5 min after exercise were not significantly different after rehabilitation for the group as a whole. A significant reduction (p 0.02) in lactate levels was seen postrehabilitation in patients with milder disease (FEV 1 40%) but not in patients with severe disease (FEV 1 40%). Six minute walk distance increased from 1069 71 at baseline to 1245 63 ft postrehabilitation (p 0.0001). The mean difference 176 ft (53.7 m) 95% confidence intervals 110 243 ft (33.5 74 m) is at the threshold (54 m, 95% confidence intervals 37 71 m) suggested as the minimal improvement in walking distance that can be reliably detected by patients as a significant improvement in functional status (21). Quadriceps Measurements TwQu before and after exercise, pre- and postrehabilitation is shown in Figure 3. At baseline, TwQu averaged 6.9 0.3 kg prerehabilitation and increased to 7.6 0.4 kg (p 0.049, 9.7% increase) postrehabilitation. Prerehabilitation, TwQu fell significantly postexercise to a minimum of 82.5 3.1% of the baseline value at 30 min postexercise (p 0.0001) and remained significantly decreased at 1 h postexercise. In contrast, postrehabilitation for the same duration and intensity of exercise, TwQu was unchanged from baseline at all times postexercise. TwQp before and after exercise, pre- and postrehabilitation is shown in Figure 4. TwQp at baseline averaged 12.7 0.7 kg prerehabilitation and increased to 13.8 0.7 kg (p 0.05, 8.8% increase) postrehabilitation. Prerehabilitation, TwQp fell significantly postexercise to a minimum of 73.9 3.9% of the baseline value at 10 min postexercise (p 0.0001) and remained significantly decreased at 1 h postexercise., TwQp also fell significantly postexercise reaching a minimum of 85 4% at 10 min postexercise (p 0.005) and remained significantly decreased at 1 h postexercise. However, the fall in TwQp postexercise was significantly larger pre- than postrehabilitation (p 0.001, ANOVA). Paired twitch data showed similar results (Figure 5). Prerehabilitation, T 2 was significantly decreased postexercise at 10 and 20 Hz down to minimum values of 76.4 4.9% and 77.5 3.6% of baseline, respectively. Prerehabilitation T 2 also fell significantly postexercise at 100 Hz but to a much smaller extent down to a minimum value of 86.8 5.0%. In contrast, Figure 3. Unpotentiated quadriceps twitch force (TwQu) expressed as a percentage of baseline before and 10, 30, and 60 min postexercise prerehabilitation (open circles) and postrehabilitation (closed circles). *Significant difference from baseline. Significant difference from the postrehabilitation value. TwQu fell significantly pre- but not postrehabilitation. Figure 4. Potentiated quadriceps twitch force (TwQp) expressed as a percentage of baseline before and 10, 30, and 60 min postexercise prerehabilitation (open circles) and postrehabilitation (closed circles). *Significant difference from baseline. Significant difference from the postrehabilitation value. The fall in TwQp postexercise was significantly less post- than prerehabilitation.

Mador, Kufel, Pineda, et al.: Pulmonary Rehabilitation and Quadriceps Fatiguability during Exercise 933 Figure 5. Twitch force elicited by the second of a paired stimuli (T 2 response) at 10, 20, and 100 Hz before (closed circles) and 10 (closed squares), 30 (open triangles), and 60 (open circles) min postexercise. Prerehabilitation results are shown in A. results are shown in B. *Significant difference from baseline. T 2 fell significantly postexercise pre- but not postrehabilitation. Prerehabilitation, the fall in T 2 postexercise was greater at 10 and 20 Hz than at 100 Hz consistent with low-frequency fatigue. postrehabilitation T 2 was not significantly different from baseline at any time postexercise at any stimulation frequency. Prerehabilitation maximum voluntary contraction (MVC) was 39.2 2.4 kg at baseline. Prerehabilitation, MVC decreased significantly postexercise reaching a minimum value of 88.1 2.7% of baseline at 30 min postexercise (p 0.005). In contrast, postrehabilitation MVC was not significantly different from baseline at any time postexercise. Prerehabilitation, twitches superimposed upon the MVC maneuver at baseline averaged 16.6 2.3% of the resting potentiated twitch indicating that patients did not fully activate their quadriceps muscle during the MVC maneuver (83.4 2.3% of full activation). The degree of activation of the quadriceps muscle was not significantly different from baseline at any time postexercise., the superimposed twitch averaged 13.3 2.5% of the resting potentiated twitch (patients achieved 86.7 2.5% of full activation). The degree of activation of the quadriceps muscle was not significantly different post- compared with prerehabilitation. Similar to prerehabilitation, the degree of activation of the quadriceps muscle was not significantly different from baseline at any time postexercise., the MVC was 14.9 5.0% (p 0.01) larger than prerehabilitation, indicating that rehabilitation elicited a significant improvement in quadriceps strength. Quadriceps M-waves were not significantly different from baseline at any time postexercise either pre- or postrehabilitation. DISCUSSION The major finding of this study was that pulmonary rehabilitation resulted in a significant improvement in quadriceps fatigability in patients with COPD. Quadriceps fatigability was measured with an effort-independent method (fall in TwQ postexercise). Thus, pulmonary rehabilitation must have resulted in changes within the exercising limb muscles that increased their fatigue resistance. Quadriceps Fatiguability In a previous study, we found that patients with COPD develop contractile fatigue of the quadriceps muscle following endurance exercise to the limits of tolerance (6). In that study, examination of individual responses revealed that 11 of 19 (58%) subjects had a 15% reduction in TwQu, indicative of significant contractile fatigue. In the present study, we measured both the unpotentiated and potentiated twitch. Recent evidence has suggested that the potentiated twitch is a more sensitive index of contractile fatigue (15, 16). In the present study 10 of 21 (48%) patients had a 15% reduction in TwQu postexercise. In contrast 17 of 21 (81%) patients had a 15% reduction in TwQp postexercise, supporting the view that TwQp is a more sensitive index of contractile fatigue. Furthermore, our TwQp results show that most patients with COPD develop contractile fatigue of the exercising limb muscles following exercise to the limits of tolerance. M-waves were unchanged postexercise in all patients in whom they could be measured (not obscured by stimulation artifact) indicating that transmission fatigue did not occur. The fatigue was long lasting (still present at 1 h). T 2 was significantly decreased from baseline postexercise at all stimulation frequencies but the amount by which T 2 fell was larger at the lower stimulation frequencies (10 and 20 Hz). Similarly, the MVC fell postexercise but to a lesser extent than the twitch. These results show that as expected, the fatigue induced by endurance exercise was predominantly low-frequency fatigue. Following pulmonary rehabilitation, the fall in measures of quadriceps contractility postexercise that occurred prerehabilitation was either abolished (TwQu, MVC, T 2 at all stimulation frequencies) or significantly ameliorated (TwQp). Thus, for the some exercise duration and intensity, the degree of exercise-induced quadriceps fatigue was clearly less postrehabilitation. What are the potential mechanisms for this increased fatigue resistance? We were somewhat surprised to find that the quadriceps MVC increased by 15 5% postrehabilitation indicating an increase in quadriceps strength. Our rehabilitation program focused on endurance exercise (bicycle and treadmill) and specific strength exercises were not performed. Thus, the endurance leg training performed in our rehabilitation program must have been sufficient to induce an increase in quadriceps strength as has also been observed in another study (22). An increase in quadriceps strength could be responsible for the increase in fatigue resistance. However, in the eight patients who did not have an increase in quadriceps MVC postrehabilitation, a significant reduction in exercise-induced quadriceps fatigue was still observed (22.3 7.3%, p 0.02 versus 23.3 8.4%). Similarly, there was no correlation (r 0.28) between the increase in quadriceps strength and the reduction in exercise-induced quadriceps fatigue. Thus, increased quadriceps strength does not appear to be responsible for the increased fatigue resistance observed postrehabilitation. Following an endurance training program above a critical minimum intensity, healthy subjects display a number of changes within the exercising muscle. These include proliferation of muscle capillaries (which increases oxygen delivery to the exercising muscle), transformation of type 2b fibers to type 2a (which have a higher oxidative capacity), increase in the size and number of mitochondria within the muscle fiber, and an increase in enzymes responsible for oxidative phosphorylation. All of these changes could increase the muscle s fatigue resistance. An increase in enzyme oxidative capacity has been demonstrated following pulmonary rehabilitation in patients with COPD, indicating that such changes are possible in this patient population (5). It would be interesting to determine whether increases in muscle oxidative capacity would correlate with improvements in fatigue resistance. It should be noted that our patients had a very favorable response to pulmonary rehabilitation with a significant increase in peak O 2 during maximal incremental exercise, a functionally significant increase in 6 min walking distance, and a marked increase in submaximal exercise endurance time. The response to pulmonary rehabilitation is not always this fa-

934 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 163 2001 vorable, probably reflecting differences in the exercise program and the patient population. It is probable that lesser improvement in overall performance might also be associated with less improvement in quadriceps fatigue resistance. In our prior study, we found that patients with severe COPD (FEV 1 40% predicted) had just as much exercise-induced quadriceps fatigue as those with milder disease despite exercising at lower intensities (6). Patients in this study behaved similarly. We interpreted this finding as suggesting that patients with more severe disease had greater muscle deconditioning and increased susceptibility to fatigue. The degree of improvement in exercise-induced quadriceps fatigue postrehabilitation, however, was similar in patients with severe COPD (21.0 10.5%, n 9) compared with those with milder disease (24.4 6.7%, n 12) demonstrating that the exercising muscle can be conditioned in patients with severe COPD as well as it can in patients with less severe disease. In our previous study, some patients (4 of 12) stopped exercise because of leg discomfort but did not demonstrate a significant reduction ( 15%) in TwQu postexercise. In the present study, we also measured TwQp, a more sensitive index of contractile fatigue. In the present study, only four patients did not show a significant reduction in TwQp postexercise prerehabilitation. All of these patients stopped exercise because of shortness of breath. However, there were seven patients who stopped exercise because of shortness of breath who displayed a significant reduction in TwQp postexercise, demonstrating that contractile fatigue of the quadriceps muscles commonly occurs postexercise in patients who stop exercise because of shortness of breath. Functional Significance of Increased Quadriceps Fatigue Resistance As mentioned above, increased quadriceps fatigue resistance is a sign that the patient underwent changes within the exercising muscle following pulmonary rehabilitation. In healthy subjects, the purpose of an exercise training program is to induce physiological adaptations that lead to improved exercise performance. In patients with COPD, there is the concern that patients will stop exercise because of shortness of breath before they can stress their exercising muscles and heart sufficiently to induce physiological adaptations, that is, exercise will be below a critical minimum intensity. In our study, we have shown that the quadriceps muscle will undergo adaptations secondary to an exercise program in the majority of patients, even those with severe COPD. Others have also provided evidence that physiological adaptations can be produced in patients with severe COPD (5, 23, 24). Conversely, however, when physiological adaptations probably did not occur (4), performance during endurance exercise still improved postrehabilitation. In our study, in the four patients who stopped exercise prerehabilitation because of shortness of breath prior to developing fatigue of the quadriceps muscle and therefore might have been the hardest to effectively train, improvements in exercise endurance time similar to the group as a whole were observed (11.1 4.0 min). In these patients, the improvement in exercise endurance might be due to other factors such as desensitization to dyspnea, improved efficiency, and increased motivation. In our study, we were unable to demonstrate a correlation between the improvement in quadriceps fatigue resistance and the improvement in exercise performance (r 0.13). The lack of correlation may be because we have compared changes over time in two relatively variable measurements (fall in TwQ postexercise, exercise endurance time), that is, the noise in the measurements may have obscured a relationship between the two variables. Improved quadriceps function may indirectly influence endurance time by reducing ventilatory demands for a given exercise workload. However, the reduction in VE at exercise isotime postrehabilitation also did not significantly correlate with improvements in exercise endurance time (r 0.34). Finally, the lack of correlation between the two variables may be because other factors are more important determinants of exercise endurance time in patients with COPD. Similar to our findings, O Donnell and colleagues found that improvements in quadriceps strength and a simple effort-dependent measure of quadriceps endurance did not correlate with improvements in exercise endurance (22). In conclusion, an exercise training regimen (at 50% of Wmax) is capable of increasing the fatigue resistance of the exercising muscle (quadriceps) in patients with COPD including those with severe disease. Acknowledgment : The authors thank Raymond Carter, LPN, and Sandra Brucato, RN, for helping to supervise the exercise training program. References 1. Ries AL, Kaplan RM, Limberg TM, Prewit LM. Effect of pulmonary rehabilitation on physiologic and psychological outcomes in patients with chronic obstructive pulmonary disease. Ann Intern Med 1995;122: 823 832. 2. Goldstein RS, Gort EH, Stubbing D, Avendano MA, Guyatt GH. Randomised controlled trial of respiratory rehabilitation. Lancet 1994; 344:1394 1397. 3. O Donnell DE, McGuire M, Samis L, Webb KA. The impact of exercise reconditioning on breathlessness in severe chronic airflow limitation. Am J Respir Crit Care Med 1995;152:2005 2013. 4. Belman MJ, Kendregan BA. Exercise training fails to increase skeletal muscle enzymes in patients with chronic obstructive pulmonary disease. Am Rev Respir Dis 1981;123:256 261. 5. Maltais F, LeBlanc P, Simard C, Jobin J, Berube C, Bruneau J, Carrier L, Belleau R. Skeletal muscle adaptation to endurance training in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1996;154:442 447. 6. Mador MJ, Kufel TJ, Pineda L. Quadriceps fatigue after cycle exercise in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2000;161:447 453. 7. Polkey MI, Kyroussis D, Hamnegard CH, Mills GH, Green M, Moxham J. Quadriceps strength and fatigue assessed by magnetic stimulation of the femoral nerve in man. Muscle Nerve 1996;19:549 555. 8. American Thoracic Society. Standardization of spirometry: 1994 update. Am J Respir Crit Care Med 1995;152:1107 1136. 9. Crapo RO, Morris AH, Gardner RM. Reference spirometric values using techniques and equipment that meet ATS recommendations. Am Rev Respir Dis 1981;123:659 664. 10. Crapo RO, Morris AH. Standardized single breath normal values for carbon monoxide diffusing capacity. Am Rev Respir Dis 1981;123: 185 189. 11. Crapo RO, Morris AH, Clayton PD, Nixon CR. Lung volumes in healthy nonsmoking adults. Bull Eur Physiopathol Respir 1982;18:419 425. 12. Kory RC, Callahan R, Borsen HG, Syner JC. The Veterans Administration-Army cooperative study of pulmonary function: I. Clinical spirometry in normal men. Am J Med 1961;30:243 258. 13. McElvaney G, Blackie S, Morrison NJ, Wilcox PG, Fairbarn MS, Pardy RL. Maximal static respiratory pressures in the normal elderly. Am Rev Respir Dis 1989;139:277 281. 14. Mador MJ, Magalang UJ, Kufel TJ. Twitch potentiation following voluntary diaphragmatic contraction. Am J Respir Crit Care Med 1994; 149:739 743. 15. Laghi F, Topeli A, Tobin MJ. Does resistive loading decrease diaphragmatic contractility before task failure. J Appl Physiol 1998;85:1103 1112. 16. Kufel TJ, Pineda LA, Mador MJ. Comparison of potentiated and unpotentiated twitches as an index of contractile fatigue [Abstract]. Am J Respir Crit Care Med 1998;157:A215. 17. Bellemare F, Bigland-Ritchie B. Central components of diaphragmatic fatigue assessed by phrenic nerve stimulation. J Appl Physiol 1987;62: 1307 1316. 18. Mador MJ, Kufel TJ, Pineda LA. Quadriceps and diaphragmatic func-

Mador, Kufel, Pineda, et al.: Pulmonary Rehabilitation and Quadriceps Fatiguability during Exercise 935 tion following exhaustive cycle exercise in the healthy elderly. Am J Respir Crit Care Med 2000;162:1760 1766. 19. Yan S, Gauthier AP, Similowski T, Faltus R, Macklem PT, Bellemare F. Force-frequency relationships of in vivo human and in vitro rat diaphragm using paired stimuli. Eur Respir J 1993;6:211 218. 20. Ellis B, Ries AL. Upper extremity exercise training in pulmonary rehabilitation. J Cardiopulmon Rehabil 1991;11:227 231. 21. Redelmeier DA, Bayoumi AM, Goldstein RS, Guyatt GH. Interpreting small differences in functional status: the six minute walk test in chronic lung disease patients. Am J Respir Crit Care Med 1997;155: 1278 1282. 22. O Donnell DE, McGuire M, Samis L, Webb KA. General exercise training improves ventilatory and peripheral muscle strength and endurance in chronic airflow limitation. Am J Respir Crit Care Med 1998; 157:1489 1497. 23. Casaburi R, Porszasz J, Burns MR, Carithers ER, Chang RSY, Cooper CB. Physiologic benefits of exercise training in rehabilitation of patients with severe chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1997;155:1541 1551. 24. Maltais F, LeBlanc P, Jobin J, Berube C, Bruneau J, Carrier L, Breton MJ, Falardeau G, Belleau R. Intensity of training and physiologic adaptations in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1997;155:555 561.