Pulmonary rehabilitation programs improve functional

Short- and Long-term Effects of Outpatient Rehabilitation in Patients with Chronic Obstructive Pulmonary Disease: A Randomized Trial Thierry Troosters, PhD, Rik Gosselink, PhD, Marc Decramer, PhD PURPOSE: Pulmonary rehabilitation programs are effective in patients with severe chronic obstructive pulmonary disease (COPD) in the short term, but their long-term effects are not known. We investigated the short- and long-term effects of a 6-month outpatient rehabilitation program in patients with severe COPD. SUBJECTS AND METHODS: One hundred patients were randomly assigned to receive either an exercise training program that included cycling, walking, and strength training (n 50) or usual medical care (n 50). Thirty-four patients in the training group were evaluated after 6 months (end of training), and 26 were evaluated after 18 months of follow-up. In the control group, 28 patients were evaluated at 6 months and 23 after 18 months. We measured pulmonary function, 6-minute walking distance, maximal exercise capacity, peripheral and respiratory muscle strength, and quality of life (on a 20 to 140 point scale), and estimated the cost-effectiveness of the program. RESULTS: At 6 months, the training group showed improvement in 6-minute walking distance [mean difference (training control) of 52 m; 95% confidence interval (CI), 15 to 89 m], maximal work load (12 W; 95% CI, 6 to 19 W), maximal oxygen uptake (0.26 liters/min; 95% CI, 0.07 to 0.45 liters/min), quadriceps force (18 Nm; 95% CI, 7 to 29 Nm), inspiratory muscle force (11 cm H 2 O; 95% CI, 3 to 20 cm H 2 O), and quality of life (14 points; 95% CI, 6 to 21 points; all P 0.05). At 18 months all these differences persisted (P 0.05), except for inspiratory muscle strength. For 6-minute walking distance and quality of life, the differences between the training group and controls at 18 months exceeded the minimal clinically-important difference. CONCLUSION: Among patients who completed the 6-month program, outpatient training resulted in significant and clinically relevant changes in 6-minute walking distance, maximal exercise performance, peripheral and respiratory muscle strength, and quality of life. Most of these effects persisted 18 months after starting the program. Am J Med. 2000;109: 207 212. 2000 by Excerpta Medica, Inc. Pulmonary rehabilitation programs improve functional exercise capacity and quality of life in patients with severe chronic obstructive pulmonary disease (COPD) (1). However, the effects of outpatient training programs have not been studied comprehensively in a randomized controlled trial. Well-conducted studies have previously used only limited outcome measures that focused on functional exercise performance or quality of life (2,3). In addition, the long-term maintenance of these benefits was not addressed in most of the randomized trials (4 7). In two trials that had long-term follow-up (3,8), there were no clear effects 18 months after starting the program. This may have been because of From the Respiratory Rehabilitation and Respiratory Division, University Hospitals Katholieke Universiteit Leuven, Leuven, Belgium, and Faculty of Physical Education and Physiotherapy, Katholieke Universiteit Leuven, Leuven, Belgium. Supported by the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen (G0189.97 and G0175.99), Levenslijn Grant 7.0002.94, and Onderzoeksfonds KU Leuven, Grant 27/98. Requests for reprints should be addressed to Rik Gosselink, PhD, PT, Division of Respiratory Rehabilitation, University Hospital Gasthuisberg, Herestraat 49, 3000 Leuven, Belgium. Manuscript submitted June 14, 1999, and accepted in revised form May 10, 2000. the short duration of the interventions in those studies (9). We performed a randomized trial to investigate the short- and long-term effects of a 6-month outpatient training program as compared with usual care. The effectiveness of the program was assessed in terms of pulmonary function, functional and maximal exercise capacity, peripheral and respiratory muscle strength, and quality of life. In addition, we estimated the cost-effectiveness of the training program. MATERIAL AND METHODS Patients were sampled from consecutive patients referred to our outpatient clinic between 1995 and 1997. They were randomly assigned to the training or control group using sealed envelopes that had been prepared before the start of the study. Patients were eligible for inclusion if they were younger than 75 years of age, had a forced expiratory volume in 1 second (FEV 1 ) that was less than 65% of the predicted value, and if their clinical condition was stable at inclusion, with no infection or COPD exacerbation in the previous 4 weeks. We excluded patients 2000 by Excerpta Medica, Inc. 0002-9343/00/$ see front matter 207 All rights reserved. PII S0002-9343(00)00472-1

with other severe medical problems, such as heart failure, myocardial infarction, cerebrovascular disease, cancer, or orthopedic disorders. The study was approved by the local ethics committee. Measurements were made at enrollment, and at 6 months and 18 months after the start of the study. At all visits, patients underwent spirometry and whole body plethysmography (Body Box 1085, Medical Graphics, Inc., St. Paul, Minnesota). FEV 1 and forced vital capacity were measured according to the European Respiratory Society guidelines for pulmonary function testing (10). In addition, the diffusing capacity for carbon monoxide was measured by the single breath method (Sensor Medics 6200, Bilthoven, The Netherlands) at the initial visit. Results were expressed as a percentage of the predicted normal values (10). Isometric quadriceps strength was measured using a Cybex II dynamometer (Lumex, Bay Shore, New York). Peak extension torque was evaluated at 60 degrees of knee flexion. Reference values for quadriceps force were developed in our laboratory (11). Tests were performed at least three times, and the best of two reproducible tests was used for further analysis. Inspiratory and expiratory muscle strength was measured using standard techniques (12). At least five attempts were made to measure expiratory muscle strength from total lung capacity and to measure inspiratory muscle strength from residual volume. Both were determined as the pressure that could be sustained for at least 1 second. Tests were repeated until the variability among the three best attempts was less then 5%. The highest value was expressed as a percent of the predicted value (13). Functional exercise performance was measured by a 6-minute corridor-walking test. Encouragement was standardized (14). A 54-m improvement was considered clinically important (15). To avoid learning effects, the best of two tests was used and expressed as percent of the predicted value (16). Maximal exercise capacity was assessed by maximal cycle ergometry (Partn air 5400; Medisoft, Dinant, Belgium). Patients cycled at an incremental workload ( 10 W per minute) until exhaustion. Oxygen consumption, carbon dioxide output, and ventilation were measured breath by breath. Heart rate was monitored constantly. Maximal oxygen consumption was compared with normal values (17). At 6 months, data obtained from a maximal exercise test were analyzed at 60% of the initial workload. The Chronic Respiratory Disease Questionnaire (18) was used to assess health-related quality of life. This 20- item questionnaire, which scores quality of life into four domains (dyspnea, mastery, emotional functioning, and fatigue) has been validated in the Dutch language (19). Intervention Patients assigned to the training program were invited to attend the outpatient sessions three times a week in the first 3 months; during the subsequent 3 months, training frequency was reduced to twice weekly. Each session had a duration of 1.5 hours. Training items were cycling, treadmill walking, stair climbing, and peripheral muscle training. Patients started the program at 60% of the initial maximal workload on the cycle ergometer and at 60% of their maximal walking speed during the 6-minute walking test. During the first 3 months, this workload was increased up to 80% of the maximal workload and maximal walking speed. In each session, patients also performed arm cranking and stair climbing in 2-minute blocks (1 to 3 repetitions). Peripheral muscle strength training on a multi-gym device was performed in three series of 10 repetitions at 60% of the one repetition maximum for each muscle group (triceps, latissimus dorsi, pectoralis, and quadriceps muscle). Physiotherapists ensured close supervision and continuous encouragement of the patients. Oxygen saturation and heart rate were measured during the training sessions. Supplemental oxygen was given to maintain oxygen saturation above 90%. Cost-effectiveness The program had a cost of $57 per session per patient, based on the charges as reimbursed by the National Health Insurance, which covers the salary for 1.5 hours of treatment from a physical therapist. The total cost of the program per patient was determined by multiplying this cost by the number of sessions. Statistical Analysis Baseline characteristics of the two groups were compared using unpaired t tests. The effects of treatment (training or control) were analyzed with a repeated measures analysis of variance. Unpaired t tests were used to evaluate the effects of treatment at the end of training (6 months) and follow-up (18 months). Statistical significance was set at P 0.05. All tests were performed with the Statistical Analysis System (SAS Institute, Cary, North Carolina). RESULTS One hundred consecutive eligible patients fulfilled inclusion criteria and were randomly assigned to either the training group (n 50) or the control group (n 50). Three patients (1 in the training group and 2 in the control group) refused initial testing, leaving 97 patients who underwent initial tests. Of these, 12 patients refused to participate in the training sessions, and 15 patients in the control group refused further follow-up. Except that they were older [mean ( SD) age of 66 7 years vs 61 9 years, P 0.004], these patients did not differ from the patients with follow-up. In addition, 3 of the remaining 208 August 15, 2000 THE AMERICAN JOURNAL OF MEDICINE Volume 109

Table 1. Characteristics of Subjects in the Control and Training s* Characteristic Control (n 33) Training (n 37) Mean SD or Number (Percent) Age (years) 63 7 60 9 Male sex 30 (90) 31 (81) Body mass index (kg/m 2 ) 25 5 24 4 Forced expiratory volume in 43 12 41 16 1 second (percent Forced vital capacity 84 18 81 19 (percent Diffusing capacity for 64 25 60 28 carbon dioxide (percent Maximal inspiratory 66 23 72 24 pressure (percent Maximal expiratory pressure 74 29 75 27 (percent Quadriceps force (percent 80 24 79 25 6-minute walking distance 61 18 60 19 (percent Maximal work load (watts) 86 40 76 36 Maximal oxygen uptake 72 39 74 35 (percent Disease-specific quality of life (points) 84 22 77 17 * Only patients who participated in the training sessions or follow-up are included. As assessed using the Chronic Respiratory Disease Questionnaire (range, 20 to 140). 37 patients in the training group (2 of whom died) and 5 of the remaining patients in the control group (3 of whom died) were not able to comply with short-term follow-up (6 months). Long-term (18-month) follow-up could not be performed in 11 of the 37 patients in the training group (9 of whom died) and in 10 of the 33 patients (7 of whom died) in the control group. There were no differences in 18-month mortality between the two groups (P 0.79). All patients had severe COPD with moderate peripheral and respiratory muscle weakness and with impaired functional and maximal exercise capacity. There were no significant differences in the characteristics of the 70 patients who participated in the training and control groups (Table 1). Patients who died during the trial (n 16) had a significantly lower FEV 1 (32% 12% vs 45% 13% of predicted, P 0.001), diffusing capacity for carbon monoxide (41% 19% vs 64% 25% of predicted, P 0.001), walking distance (357 117 vs 429 124 m, P Figure 1. Upper panel: training intensity (mean SD) for treadmill walking (open circles) and cycle exercise (closed circles). Workload during cycle exercise (in watts) is expressed relative to the maximal workload achieved during the maximal incremental exercise test. Workload during treadmill walking is expressed as percent of the speed achieved during the initial 6-minute walking test. Lower panel: total training time (mean SD) during training sessions, by week of training. 0.05), and maximal exercise capacity (58 30 vs 79 34 W, P 0.02) than those who survived. Rehabilitation Program Patients in the training group participated in a mean of 46 11 sessions (attendance rate 77% 19%). Training intensity for walking and cycling reached a plateau, on average, after 3 months (Figure 1). Heart rate during the training session averaged 92% 9% of the maximal achieved heart rate during the incremental maximal exercise test and remained relatively constant throughout the 6 months of training. Effects of Training Compared with usual care, there were no significant effects of the training program on measures of pulmonary function. The training program did, however, improve 6-minute walking distance, maximal work load, maximal oxygen uptake, quadriceps force, and quality-of-life score August 15, 2000 THE AMERICAN JOURNAL OF MEDICINE Volume 109 209

Table 2. Effects of a Pulmonary Rehabilitation Training Program at 6 Months, Compared with Control* Measurement Absolute Values Control (n 28) Training (n 34) Difference between s in Change from Baseline (95% Confidence Interval) P Value Forced expiratory volume in 1 1.3 0.4 1.2 0.5 0.04 ( 0.09 0.12) 0.89 second (L) Quadriceps force (Nm) 121 52 144 45 18 (7 30) 0.004 Maximal inspiratory pressure 68 23 83 28 11 (3 20) 0.04 (cm H 2 O) 6-minute walking distance (m) 438 104 468 125 52 (15 89) 0.01 Maximal work load (watts) 85 37 88 36 12 (6 19) 0.003 Maximal oxygen uptake 1.54 0.63 1.63 0.64 0.256 (0.07 0.45) 0.02 (L/min) Disease-specific quality of life (points) 82 25 90 18 14 (6 21) 0.002 * Results are presented both as absolute values at 6 months and as the difference (change in treatment group change in control group) from baseline. For the comparison of difference in change from baseline. (Table 2, Figure 2). After 6 months, ventilation at isoworkload (60% of the initial maximal work load) decreased significantly ( 2.3 4.9 liters) compared with control ( 1.2 4.5 liters, P 0.05), while oxygen uptake was similar in the two groups. Changes in carbon dioxide production were not significantly different between the training and the control group. Mean change in heart rate was somewhat greater in the training group compared with the control group ( 8 11 vs 1 13, P 0.06) at this submaximal workload. After 18 months, the improvements in 6-minute walking distance (difference between mean change from baseline in the treatment group mean change from baseline in the control group of 90 m; 95% confidence interval [CI], 41 to 149 m; P 0.01), maximal work load (13 W; 95%CI,4to22W;P 0.05), maximal oxygen uptake (0.27 liters/min; 95% CI, 0.02 to 0.52 liters/min; P 0.05), quadriceps strength (15 Nm; 95% CI, 1 to 30 Nm; P 0.05), and quality of life (17 points; 95% CI, 8 to 26 points; P 0.001) were maintained in the training group compared with the control group. The improvements in quadriceps strength were correlated with the improvements in walking distance at both 6 months (r 0.38, P 0.05) and 18 months (r 0.47, P 0.05). The mean cost of the program per patient was $2,615 $625 DISCUSSION We found that a 6-month outpatient rehabilitation program that involved moderate-to-high training intensity did not alter pulmonary function, but did improve functional and maximal exercise performance, peripheral and respiratory muscle strength, and quality of life when compared with usual care in patients with severe COPD. Improvements in functional and maximal exercise performance and quality of life were clinically relevant (15) and were maintained 18 months after the onset of training. However, 31% of patients dropped out of the study by 6 months, and 36% by 18 months. Goldstein et al (2) reported similar dropout and refusal rates. In our study, the main reasons for dropping out were unwillingness to participate in the rehabilitation sessions, refusal to participate in further assessments, or death. Patients who died during the trial had severe disease, as manifested by low values for FEV 1 and diffusing capacity for carbon monoxide, and a low exercise capacity (20 22). As reported by others, the pulmonary rehabilitation program did not prevent short-term mortality (3). The training regimen that we used differed somewhat from other trials. Patients trained at a slightly greater intensity than reported by Maltais et al (23), but the training time for each session was substantially shorter in our study. The mean attendance rate in our study (77% 19%) was similar to that in the study of Bendstrup et al (6), although less than that reported by others (23). The magnitude of the training effect (approximately 20% improvement) in this study is similar to other trials, as measured by improvement in functional exercise capacity (5,24 26) and maximal exercise capacity (8,24,27,28). Although some trials have not demonstrated improvements in maximal oxygen uptake (4,24,28,29), the increase (12% 29%) that we achieved 210 August 15, 2000 THE AMERICAN JOURNAL OF MEDICINE Volume 109

Figure 2. Training effect for 6-minute walking distance (upper), expressed as mean ( SD) percent change compared with baseline. Mean SD training effect for the change in the dyspnea dimension of the quality-of-life questionnaire (lower) is expressed as change (in points) compared with baseline. The dashed line represents the minimal clinically-important difference. *P 0.05, **P 0.01 compared with controls. resembles that reported in studies that used strenuous, individually-tailored cycle exercise (3,23,30,31). The improvements that we observed in all four dimensions of quality of life were similar to those reported in a recent meta-analysis (1). Most of the benefits of training in this study persisted for 12 months after the training program had been completed. Swerts et al (9) reported that the improvements seen after an 8-week program declined gradually during 1 year of follow-up, whereas patients who continued the training program for 20 weeks (60 sessions) maintained improvements in walking distance. Two other studies that had shorter training periods were unable to show long-term benefits (3,8). Whether this apparent discrepancy is the result of the duration of the training program is not clear. In our study, the maintained improvement at 18 months not only reached statistical significance, but also the difference with controls exceeded the minimal clinically-important difference for improvements in quality of life and functional exercise capacity. For the latter test, it was estimated to be 54 m (19). In our study, after the rehabilitation period had been completed, the average improvement which was approximately 90 m in walking distance persisted in those subjects who returned for the 1-year follow-up. Similar and persistent improvements in quality of life were also seen, whereas there was a significant and clinically-important deterioration of mean quality of life in the control group. Our outpatient-based program had a mean cost per patient of approximately $2,600 to achieve a mean improvement of 52 m in 6-minute walking distance at 6 months. A previous study reported that a hospital-based program cost approximately $7,800 per patient and resulted in a 37m improvement in 6-minute walking distance (2). Thus, outpatient rehabilitation appears to be more cost-effective than an inpatient program. Compared with home rehabilitation, which was reported to cost approximately $660 per patient and resulted in approximately a 39m improvement in 6-minute walking distance (8), the costs of our rehabilitation program were great. However, the patients in our study had more severe disease than those in the study of the home-care program and might need more careful monitoring during training. Indeed, prudent training strategies in a home setting were not effective in one study that enrolled patients with severe COPD (7). We conclude that a 6-month outpatient rehabilitation program with high-intensity training resulted in clinically-relevant improvements in functional and maximal exercise capacity, peripheral and respiratory muscle strength, and quality of life in patients with severe COPD who were willing to participate in the program. Most of the effects were maintained among patients who completed the 18-month follow-up. ACKNOWLEDGMENT The authors thank Edgard Kellens, Anne Cattaert, Hilde Witvrouwen, Katleen Schrever, staff members involved in the training sessions. We also acknowledge the excellent statistical aid of Renild Schepers. REFERENCES 1. Lacasse Y, Wong E, Guyatt GH, et al. Meta-analysis of respiratory rehabilitation in chronic obstructive pulmonary disease. Lancet. 1996;348:1115 1119. 2. Goldstein RS, Gort EH, Stubbing D, et al. Randomised controlled trial of respiratory rehabilitation. Lancet. 1994;344:1394 1397. 3. Ries AL, Kaplan RM, Limberg TM, Prewitt LM. Effects of pulmonary rehabilitation on physiologic and psychosocial outcomes in patients with chronic obstructive pulmonary disease. Ann Intern Med. 1995;122:823 832. 4. Reardon J, Awad E, Normandin E, et al. The effect of comprehen- August 15, 2000 THE AMERICAN JOURNAL OF MEDICINE Volume 109 211

sive outpatient pulmonary rehabilitation on dyspnea. Chest. 1994; 105:1046 1052. 5. Weiner P, Azgad Y, Ganam R. Inspiratory muscle training combined with general exercise reconditioning in patients with COPD. Chest. 1992;102:1351 1356. 6. Bendstrup J, Ingemann Jensen J, Holm S, Bengtsson B. Out-patient rehabilitation improves activities of daily living, quality of life and exercise tolerance in chronic obstructive pulmonary disease. Eur Respir J. 1998;10:2801 2806. 7. Wedzicha JA, Bestall JC, Garrod R, et al. Randomized controlled trial of pulmonary rehabilitation in severe chronic obstructive pulmonary disease patients, stratified with the MRC dyspnea scale. Eur Respir J. 1998;12:363 369. 8. Strijbos JH, Postma DS, Van Altena R, et al. A comparison between outpatient hospital-based pulmonary rehabilitation program and a home-care pulmonary rehabilitation program in patients with COPD. A follow-up of 18 months. Chest. 1996;109:366 372. 9. Swerts PMJ, Kretzers LMJ, Terpstra-Lindeman E, et al. Exercise reconditioning in the rehabilitation of patients with chronic obstructive pulmonary disease: a short- and long-term analysis. Arch Phys Med Rehabil. 1990;71:550 573. 10. Quanjer PH, Tammeling GJ, Pedersen OF, et al. Lung volumes and forced expiratory flows. Eur Respir J. 1993;6(suppl 16):5 40. 11. Decramer M, Lacquet LM, Fagard R, Rogiers P. Corticosteroids contribute to muscle weakness in chronic airflow obstruction. Am J Respir Crit Care Med. 1994;150:11 16. 12. Black LF, Hyatt RE. Maximal respiratory pressures: normal values and relationship to age and sex. Am Rev Respir Dis. 1969;99:696 702. 13. Rochester D, Arora NS. Respiratory muscle failure. Med Clin North Am. 1983;67:573 598. 14. Guyatt GH, Pugsley SO, Sullivan MJ, et al. Effect of encouragement on walking test performance. Thorax. 1984;39:818 822. 15. Redelmeier DA, Bayoumi AM, Goldstein R, Guyatt G. 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. 16. Troosters T, Gosselink R, Decramer M. Six minute walking distance in healthy elderly subjects. Eur Resp J. 1999;14:270 274. 17. Jones NL, Makrides L, Hitchcock C, et al. Normal standards for an incremental progressive cycle ergometer test. Am Rev Respir Dis. 1985;131:700 708. 18. Guyatt GH, Berman LB, Townsend M, et al. A measure of quality of life for clinical trials in chronic lung disease. Thorax. 1987;42:773 778. 19. Gosselink HAAM, Wagenaar RC, Keimpema VA, Chadwick- Straver RVM. The effects of a rehabilitation program in patients with COPD and asthma. Ned Tijdschr Fysioth. 1990;100:193 199. 20. Dubois P, Machiels J, Smeets F, et al. CO transfer capacity as a determining factor of survival for severe hypoxemic patients under long-term oxygen therapy. Eur Respir J. 1990;3:1042 1047. 21. Ebi-Kryston KL. Predicting 15 year chronic bronchitis mortality in the Whitehall study. J Epidemiol Community Health. 1989;43:168 172. 22. Gerardi DA, Lovett L, Benoit-Connors ML, ZuWallack RL. Variables related to increased mortality following outpatient pulmonary rehabilitation. Eur Respir J. 1996;9:431 435. 23. Maltais F, Leblanc P, Jobin J, et al. Intensity of training and physiologic adaptation in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 1997;155:555 561. 24. O Donnell DE, McGuire MA, 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. 25. ZuWallack RL, Patel K, Reardon JZ, et al. Predictors of improvement in the 12-minute walking distance following a six-week outpatient pulmonary rehabilitation program. Chest. 1991;99:805 808. 26. White RJ, Rudkin ST, Ashley J, et al. Outpatient pulmonary rehabilitation in severe chronic obstructive pulmonary disease. J Roy Coll Phys London. 1997;31:541 545. 27. Wanke T, Formanek D, Lahrmann H, et al. The effects of combined inspiratory muscle and cycle ergometer training on exercise performance in patients with COPD. Eur Respir J. 1994;7: 2205 2211. 28. Casaburi R, Porszasz J, Burns MR, et al. 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. 29. Clark CJ, Cochrane JE, Mackay E. Low intensity peripheral muscle conditioning improves exercise tolerance and breathlessness in COPD. Eur Respir J. 1996;9:2590 2596. 30. Vallet G, Ahmaidi S, Serres I, et al. Comparison of two training programmes in chronic airway limitation patients: standardized versus individualized protocols. Eur Respir J. 1997;10:114 122. 31. Wijkstra PJ, Van der Mark TW, Kraan J, et al. Effects of home rehabilitation on physical performance in patients with chronic obstructive pulmonary disease (COPD). Eur Respir J. 1996;9:104 110. 212 August 15, 2000 THE AMERICAN JOURNAL OF MEDICINE Volume 109