Published ahead of print on March 18, 2005, doi:10.1164/rccm.200408-1109SO
American Journal of Respiratory and Critical Care Medicine Vol 172. pp. 19-38, (2005)
© 2005 American Thoracic Society
doi: 10.1164/rccm.200408-1109SO
Pulmonary Rehabilitation in Chronic Obstructive Pulmonary Disease
Thierry Troosters,
Richard Casaburi,
Rik Gosselink and
Marc Decramer
Respiratory Rehabilitation and Respiratory Division, University Hospital; Department of Rehabilitation Sciences, Faculty of Physical Education and Physiotherapy, Katholieke Universiteit Leuven; and Respiratory Rehabilitation, Universitaire Ziekenhuizen Gasthuisberg, Leuven, Belgium; and Rehabilitation Clinical Trials Center, Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, California
Correspondence and requests for reprints should be addressed to Thierry Troosters, Ph.D., P.T., Respiratory Rehabilitation and Respiratory Division, UZ Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium. E-mail: Thierry.Troosters{at}med.kuleuven.ac.be
Key Words: chronic obstructive pulmonary disease • exercise training • multidisciplinary • muscle • pulmonary rehabilitation
 |
CONTENTS
|
|---|
Effectiveness in Terms of Clinically Important Outcomes
Exercise Performance
Health-related Quality of Life and Related Outcomes
Use of Health Care Resources
Survival
Exercise Training
Skeletal Muscle Dysfunction in COPD: An Update
Principles of Design for Exercise Training Programs
Can Training Intensity Be High Enough in Patients with COPD to Yield a Physiologic Training Effect?
Specific Strategies to Increase Training Intensity
Should Exercise Training Be Prescribed for All Patients with COPD?
Other Components of Pulmonary Rehabilitation Programs
Improving Education and Self-management
Psychosocial Support: Rationale and Practical Implication
Improving Activities of Daily Living: Role for Occupational Therapy
Nutritional Interventions
Evaluation before, during, and after Pulmonary Rehabilitation
Organizing Pulmonary Rehabilitation Programs
Where to Organize Rehabilitation Programs
Strategies to Maintain the Effects of Pulmonary Rehabilitation
Suggestions for Further Research
Optimal Combination of Program Length and Proper Maintenance in COPD
New Strategies to Improve Rehabilitation Outcome in COPD
Summary
In patients with pulmonary disease, disease severity and prognosis are determined not only by lung function impairment (1, 2). In patients with mild, moderate, or severe disease, exercise capacity, health-related quality of life, and participation in activities of daily living are often impaired out of proportion to lung function impairment (3–5). Hence, therapies that improve the patient's lung function may have relatively limited impact on the above-mentioned outcomes (6, 7). Optimal bronchodilatation can be seen as a first step in the treatment of patients with chronic obstructive pulmonary disease (COPD); greater treatment effects (e.g., improvements in exercise performance, symptoms, and health-related quality of life) are often achieved only after the addition of pulmonary rehabilitation (8). Comprehensive pulmonary rehabilitation programs aim at tackling the systemic consequences of COPD, as well as the behavioral and educational deficiencies observed in many patients (9).
Since the first controlled trials on pulmonary rehabilitation in the mid-1970s (10, 11) (for review see Casaburi and Petty [12]) and initial skepticism in the early 1980s (13), pulmonary rehabilitation has proven to result in clinically significant improvements in more than 20 methodologically well-designed randomized controlled trials (14). According to the World Health Organization's Global Initiative for Chronic Obstructive Lung Disease (GOLD) consensus document on the management of COPD (15), pulmonary rehabilitation should be considered in patients with an FEV1 below 80% of the predicted value. In addition, most national and international guidelines consider pulmonary rehabilitation an important treatment option (9, 16–18). Hence, ideal candidates show, despite optimal medical treatment, significant abnormalities in their function and their participation in everyday life, leading to impaired health-related quality of life. In addition, candidates should be motivated to engage in a rehabilitation process that is often demanding both in terms of time and in physical and psychological investment (19). At the severe end of the disease spectrum, the results of the National Institutes of Health National Emphysema Treatment Trial (NETT), investigating the effects of lung volume reduction surgery, are strong encouragement for the implementation of pulmonary rehabilitation programs for patients with COPD.
The NETT, indeed, identified lung volume reduction surgery as an expensive (20), but effective treatment option in selected patients after they had undergone pulmonary rehabilitation (21). In patients screened for lung volume reduction surgery, pulmonary rehabilitation programs were thought to be an essential preparatory step before considering surgery. When, despite pulmonary rehabilitation, patients still have severely impaired exercise capacity, and have predominantly upper lobe emphysema, they are likely to benefit substantially from lung volume reduction surgery. Moreover, because the improved lung function seen after lung volume reduction surgery (22) facilitates exercise at higher intensity, rehabilitation programs after the surgery are beneficial as well. The results of the NETT study have triggered discussions on the reimbursement of pulmonary rehabilitation programs in the United States (23).
Today the question is no longer "should patients with chronic obstructive lung disease receive pulmonary rehabilitation?" but rather "how should pulmonary rehabilitation be delivered to patients with COPD?" and "which components form the basis of the success of pulmonary rehabilitation programs?" The present review focuses on the present state of the art and science concerning the physiological rationale for exercise training; we also discuss the potential of the multidisciplinary approach generally adopted during rehabilitation programs.
|
EFFECTIVENESS IN TERMS OF CLINICALLY IMPORTANT OUTCOMES
|
|---|
The goals of pulmonary rehabilitation programs are to reduce symptoms, improve activity and daily function, and restore the highest level of independent function in patients with a respiratory disease (9, 16–19). The success of achieving these goals can be assessed by using both physiological and psychosocial outcome measures.
Rehabilitation programs are recommended to be set up as individualized and multidisciplinary interventions. The different components of the program aim at improving physical and psychological functioning of patients in interaction with his or her environment, enhancing knowledge of the disease, and improving self-management.
Exercise Performance
In a meta-analysis, the effects of pulmonary rehabilitation programs on exercise tolerance were systematically reviewed (14). Improvement of exercise tolerance has been estimated from incremental exercise tests, constant work rate tests, or free walking tests. In the following, pooled data are presented as a reasonable estimate of the anticipated effects of rehabilitation programs. Although clinically these programs differed in terms of duration, number of sessions, training intensity, and so on (factors that are discussed below), statistical heterogeneity was not significant (24): for maximal work rate Q = 4.9, p = 0.55; for peak oxygen uptake Q = 11.8, p = 0.22; for constant work rate tests Q = 3.55, p = 0.47; and for the 6-minute walking test Q = 14.5, p = 0.2.
In incremental tests, peak work rate improves on average by 18% compared with baseline (weighted mean of studies [14, 25–27], reporting maximal exercise tolerance in a parallel group design). The interquartile range of the improvement reported across studies was 13 to 24% improvement in the treatment groups, compared with the respective control groups. Peak oxygen uptake improved by 11% when the rehabilitation groups are compared with the respective controls (interquartiles of different studies, 4 to 18% improvement). The effect of pulmonary rehabilitation on whole body constant work rate exercise tolerance is much greater. As a weighted average of five randomized controlled studies (27–31), endurance exercise time improved by an average of 87%. One study showed a 14% increase in the theoretical maximal sustainable work rate, another outcome that may be relevant to everyday life (32). A combination of improved mechanical efficiency, improved muscle force (33) and oxidative capacity (34, 35), adaptations in the breathing pattern (36, 37), and consequently reduced dynamic hyperinflation (38) are likely to contribute to the improved exercise tolerance that is consistently reported.
Exertional dyspnea is consistently reported to be reduced after pulmonary rehabilitation (36). The reduction in dyspnea is partially mediated through the reduced ventilatory requirements at identical work rates and at identical oxygen consumption. In addition, patients have also reported reduced sensation of dyspnea at identical levels of ventilation (39). The latter may be the consequence of less dynamic hyperinflation at isoventilation and desensitization to dyspnea. In patients with inspiratory muscle weakness (40, 41), but not in patients with better preserved inspiratory muscle force (39, 42), improved sensation of dyspnea at isoventilation may be due to improved inspiratory muscle function after rehabilitation.
The clinical relevance of the benefit of pulmonary rehabilitation is illustrated by the improved functional capacity, as measured by the 6-minute walk test. The pooled effect size of all randomized controlled studies of the results of pulmonary rehabilitation is 49 m, with a 95% confidence interval of 26–72 m (14). The minimal clinically important difference of the 6-minute walking test has been estimated to be 54 m (43). In Figure 1 the effects of rehabilitation programs are displayed along with the program characteristics (25, 31, 33, 44–52). Although a formal test for statistical heterogeneity failed to be significant (Q = 14.5, p = 0.2), clinically relevant heterogeneity is observed across studies. From Figure 1 it is suggested that programs with more rehabilitation sessions (median across programs, 28 sessions) had somewhat more effect than programs with fewer sessions (34.5 m with 28 sessions or fewer, versus 50.3 m with more than 28 sessions). Other factors that may interfere with the overall effect of the program are the setting, the intensity, the duration of the training session, and the duration of the program. A meta-analysis attempting to investigate whether these factors would lead to statistically significant difference in results, however, revealed only a trend for longer programs (more than 6 months, 70 [41–93] m) to be superior to shorter programs (42 [10–72] m) and showed a strong trend for enhanced effects (60 [34–86] versus 18 [15–50] m) when close supervision of the patients was ensured (14).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 1. Left: Summary of published differences in the 6-minute walking distance (6MWD, expressed in meters) between groups of patients who underwent rehabilitation including exercise training and their respective control groups receiving usual care. The minimal clinically important difference is indicated by the thin vertical line. Right: The type of program (outpatient [O], inpatient [I], or home based [H]), the duration (in weeks), and the number of sessions (Sess). US = unsupervised sessions performed at home; the number within parentheses indicates that the exact number of sessions was not reported. In addition, the various components of the rehabilitation program are displayed (Ex = exercise; B = breathing exercises; Ed = education sessions; Di = dietary intervention; Occ = occupational therapy; Psy = psychotherapy; Lei = leisure activities; Nu = nurse specialist).
|
|
From a randomized controlled study conducted in the authors' center (33), we calculated that the number needed to treat, to have one patient with a clinically significant benefit, was three (95% confidence interval, 1.7–6.4). Others found similar results (52), even if sustained improvement over a 24-month follow-up period was used as a criterion. The addition of pulmonary rehabilitation to the treatment of patients with stable COPD, hence, seems to result in more significant improvements in exercise tolerance than adding an additional bronchodilator (53).
Health-related Quality of Life and Related Outcomes
A review of the published literature shows that the improvement in health-related quality of life after pulmonary rehabilitation clearly exceeds the minimal clinically important difference. When disease-specific instruments were used, the lower limit of the 95% confidence interval exceeded the minimal clinically important difference (14). Figure 2 summarizes the effects of various randomized controlled trials investigating the effect of outpatient rehabilitation (33, 48, 50–52, 54, 55), inpatient rehabilitation (49), and home rehabilitation (30, 31, 55, 56) on health-related quality of life. For comparison, the effects of several large trials of bronchodilator administration, inhaled corticosteroids, and combination products are also depicted. These pharmacologic trials have been reviewed in detail elsewhere (57). Because these studies used different health-related quality of life instruments, effects are displayed as a fraction of the minimal clinically important difference; therefore an effect size exceeding "1" is considered to be of clinical relevance. The two rehabilitation trials not showing improvement in health-related quality of life above the minimal clinically important difference are one trial of home rehabilitation in patients with severe dyspnea (55) and one trial of outpatient rehabilitation applying a low training frequency (27 training sessions over 1 year) (48). In the 11 remaining rehabilitation trials, the effects were both clinically relevant and statistically significant. In contrast, few trials with bronchodilators, inhaled steroids, or combination products yielded improvements in health-related quality of life exceeding the minimal clinically important difference.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 2. Effect of home- or community-based (open circles) (30, 31, 55, 56), outpatient (solid circles) (33, 48, 50–52, 54, 55), or inpatient (open square) (49) pulmonary rehabilitation compared with usual care on health-related quality of life (HRQoL), expressed as a fraction of the minimal clinically important difference (MCID) (10 points for the Chronic Respiratory Disease Questionnaire, 4 points for the St. George's Respiratory Questionnaire), and various medication trials with inhaled corticosteroids (+), long-acting bronchodilators (*), or combinations (open triangles). These trials are reviewed in an evidence-based review elsewhere (57). Treatment or follow-up time is given in weeks.
|
|
Improved health-related quality of life is observed even in the absence of clinically significant improvements in exercise capacity (58, 59). Long-term follow-up has shown that quality of life benefits are maintained above control levels if rehabilitation yields clinically significant effects on exercise tolerance (33, 52, 60). Whether the effects on health-related quality of life are maintained after the rehabilitation program is stopped in patients not achieving a significant physiological training effect has not been studied.
Effects of rehabilitation on psychological well-being (e.g., anxiety and depression) are less studied. Factor analysis, however, has shown that these effects cannot be simply deduced from general health-related quality of life questionnaires (61). One large randomized controlled study did not find a reduction in anxiety and depression (27). Other studies, however, support the effectiveness of these programs in ameliorating anxiety and depression in an appreciable fraction of patients (62–64). This discrepancy may be explained by the fact that effects of pulmonary rehabilitation on anxiety and depression should be expected only in the subgroup of patients who are anxious and depressed before starting pulmonary rehabilitation. This subgroup of patients has been estimated to be as large as 20 to 40% of patients referred for pulmonary rehabilitation (65–68). Exercise training in depressed, but otherwise healthy, elderly patients showed improvements in mood status comparable to antidepressant pharmacotherapy (69). If improvement in anxiety and depression is achieved with a rehabilitation program, this contributes to the improved health-related quality of life, as these variables have been shown to be partially interrelated (70).
Use of Health Care Resources
Few randomized controlled studies have examined the effectiveness of pulmonary rehabilitation programs on use of health care resources and assessed the cost-effectiveness of this intervention. To assess the cost-effectiveness of pulmonary rehabilitation, long-term follow-up is mandatory. In a well-conducted study (54), Griffiths and coworkers reported that patients with COPD spent fewer days hospitalized during a 1-year follow-up period. Smaller studies of outpatient rehabilitation showed similar trends in reducing hospital days, but lacked statistical power to confirm whether the differences in use of health care resources were statistically significant (52). Several uncontrolled studies also support the ability of pulmonary rehabilitation to reduce hospital days (71–75), which are the primary cost driver of COPD care (76). When pulmonary rehabilitation is conducted as an outpatient rehabilitation program, the cost is relatively modest. An analysis specifically posits that pulmonary rehabilitation is cost-effective (77).
It is unclear, however, whether this benefit of rehabilitation programs should be attributed to the physiological improvements, or rather to the improved knowledge of the disease and enhanced self-management. In a randomized controlled trial, hospital admissions were reduced by 40% among patients with a history of hospital admissions who followed a self-management program, and in whom a "case manager" was assigned to follow up the patients (78). In contrast, a study incorporating long-term exercise training, but not including individualized education sessions and self-management strategies, reported no decrease in hospital days, despite significant physiological effects (48). Interestingly, the latter study is one of the few randomized controlled trials in which no significant improvement was observed in health-related quality of life. In this study only a limited number of sessions (27 sessions) were spread out over a 1-year program. The effect of pulmonary rehabilitation on the occurrence of mild exacerbations has not been studied extensively. One Spanish study (52) reported significantly fewer mild exacerbations after a 6-month pulmonary rehabilitation program in which exacerbations were tracked with diary cards. Reduction in mild exacerbations may lead to a further, albeit more modest, reduction in use of health care resources. It is appealing to speculate that a reduction in mild and moderate acute exacerbations may lead in the long term to a modification of disease progression, as exacerbation frequency has been linked to decline in FEV1 (79). No study has so far confirmed this hypothesis. Nevertheless, in the aforementioned study (52), during the 24 months of follow-up more patients from the control group became oxygen dependent. Oxygen dependency can perhaps be seen as a marker of disease progression.
Survival
So far no study has convincingly shown evidence of improved survival after pulmonary rehabilitation (27, 33, 48, 52, 54, 56, 60). None of these studies, however, has been powered to detect a reduction in mortality. When all these studies are pooled, 1-year to 18-month mortality risk was 7.8% in the rehabilitation group (23 of 315 randomized patients died) and 9.9% in the control group (28 of 283 patients died). The pooled odds of dying in the rehabilitation group, as compared with the control group, was 0.69 (relative risk, 0.38–1.25; p = 0.395 [heterogeneity Q = 2.02, p = 0.91]). This suggests that our current best estimate is that rehabilitation reduces the short-term risk of dying by 31%. This risk reduction is, however, not statistically significant at all because, even when pooled, the number of patients studied is insufficient to detect this magnitude of effect. Because patients who enroll in pulmonary rehabilitation are generally in a relatively stable state, their likelihood of dying in the short term is rather low. Hence the absolute reduction in mortality is likely to be relatively modest. Studies investigating patients with higher mortality risk (e.g., after discharge from the hospital for an acute exacerbation) may be more successful in finding effects on survival (80). Indeed, in these patients inactivity has been identified as a factor independently related to hospital readmission (81) and mortality (82). Because increasing the patient's activity level is a central goal of rehabilitation, rehabilitation programs initiated after severe exacerbations may show improved survival in smaller studies. Alternatively, a large-scale multicenter randomized trial employing a prolonged follow-up period (perhaps 3 years) is required to conclusively establish whether pulmonary rehabilitation impacts the progression of disease and survival. If one assumes, on the basis of the above-described analysis of the available literature, an absolute 2% mortality difference per year, and a durable effect of rehabilitation (i.e., with proper maintenance programs; see below), up to 750 to 1,000 patients would be required in both the treatment and control groups with 3 years of follow-up.
|
EXERCISE TRAINING
|
|---|
Exercise reconditioning is the key to a successful rehabilitation program (83). The challenge is to design a training program that stimulates the cardiovascular system and the skeletal muscle so that physiological adaptations in the skeletal muscle are induced that reverse the deleterious impact of deconditioning and other systemic manifestations of COPD that impact skeletal muscle. Training programs in lung disease must accommodate the limitations of the patient with COPD. These limitations include inability to increase oxygen delivery to the peripheral muscle because of gas exchange inefficiency in the lungs, constraints on lung mechanics (dynamic hyperinflation and flow limitation) (84), and the development of pulmonary hypertension during exercise (85). Beyond these limitations, muscle dysfunction is a factor limiting exercise tolerance in a substantial fraction of patients (4, 86, 87). Exercise training programs should be adapted to the individual limitations of the patient, taking cardiovascular, pulmonary, and skeletal muscle limitations into account.
Skeletal Muscle Dysfunction in COPD: An Update
Skeletal muscle dysfunction has now been widely accepted in COPD, and has been reviewed in several excellent reviews in the Journal (88–90). Skeletal muscle atrophy or weakness is clearly a negative prognostic factor (91–93) and requires appropriate action. The potential for at least partial reversibility of this dysfunction is believed to be an important determinant of the physiological gains achieved with pulmonary rehabilitation. Skeletal muscle dysfunction is typically characterized by reduced muscle strength, reduced muscle endurance, impaired muscle oxidative capacity (as shown by reduced activity of the enzymes citrate synthase and hydroxyacetyl-CoA dehydrogenase [94, 95]), and a shift toward a glycolytic fiber type distribution (low fraction of Type I fibers) (96–100). These changes result in reduced skeletal muscle endurance (100), increased fatigability (101, 102), a lowering of the lactate threshold (94), and increased ventilatory requirements during exercise.
Although several complementary mechanisms may be the basis of the observed skeletal muscle dysfunction, the exact factors that cause skeletal muscle dysfunction in individual patients remain largely unknown. Hence, wide variation in skeletal muscle weakness among subjects is observed, ranging from normal function in some patients to severe impairment in others (103).
A first mechanism potentially contributing to skeletal muscle dysfunction is the sedentary lifestyle of patients with COPD (104–107). The accompanying deconditioning leads to skeletal muscle atrophy. Atrophy can be observed at the whole muscle level (108) or at the myocyte level (109), but can also be indirectly assessed by the loss in (extremity) fat-free mass (110). The proportional reduction in muscle mass and muscle force (108) and the preserved in vitro contractile properties of the quadriceps give further support to the importance of inactivity-induced atrophy (111). Partial reversibility of low muscle oxidative capacity (35, 112), as well as increases in muscle cross-sectional area (113), are seen with exercise conditioning programs. In addition, skeletal muscle strength (33, 39, 114, 115) and skeletal muscle fatigability (116) have also been reported to be significantly improved after exercise training in COPD. These observations give support to the hypothesis that inactivity is an important driver of the observed abnormalities in the skeletal muscles of these patients.
There are several reasons, however, to posit that inactivity, albeit an important factor, is not the only factor responsible for the observed skeletal muscle dysfunction: (1) a study employing 6 months of high-intensity exercise training failed to normalize skeletal muscle force (33); (2) studies investigating healthy subjects with an activity level similar to that of patients with COPD observed reductions in myosin heavy chain Type I expression in the skeletal muscles of patients with COPD (96), or found that the muscles of patients with COPD adapted differently to strenuous exercise stimuli (117); (3) in a subset of patients with COPD, exercise training responses of the peripheral muscle are clearly abnormal, reducing rather than improving the scavenging potential for free radicals (118). Hence, other mechanisms have been suggested to exacerbate the observed skeletal muscle abnormalities in COPD. Acute or chronic treatment with high-dose oral corticosteroids has been directly associated with the development of muscle weakness (119, 120) and poor survival (121). Relatively modest doses of steroids administered over a period of 14 days, administered to stable patients (e.g., in the context of a reversibility trial), however, seem not to exert demonstrable adverse effects on quadriceps and respiratory muscle force in the average patient (122). More recent studies have underlined the importance of systemic inflammation as a mechanism for development of muscle weakness. Systemic inflammation has been suggested to contribute to excessive muscle weakness, especially during severe exacerbations of COPD (123). Circulating levels of interleukin-8 were significantly correlated with muscle weakness (123). Weight loss and especially loss of fat-free mass (a surrogate measure of muscle mass) have been associated with systemic inflammation (124). In addition, increased levels of skeletal muscle apoptosis were observed in patients presenting with weight loss (125). Another factor related to the process of muscle wasting in COPD is oxidative stress. Patients with COPD are exposed to increased levels of oxidative stress, both when stable and (more evidently) during acute exacerbations (126, 127). During exercise, larger amounts of free radicals are formed as a result of increased mitochondrial activity. When there is insufficient scavenging potential for these free radicals, oxidative damage to lipids and proteins may result. There is still debate on the oxidant–antioxidant balance in COPD. Several authors have reported impaired systemic or skeletal muscle antioxidant capacity in patients with COPD (117, 128) and lipid peroxidation was observed, even at rest (117, 129, 130). Others, however, do not confirm that skeletal muscle antioxidant capacity is reduced at rest (118). After local (117) or whole body (131) exercise, markers of oxidative stress damage were significantly increased. Hence, it seems likely that some patients are more susceptible to oxidative damage than others. Inflammatory status, nutritional status, and steroid treatment (132) as well as activity level (133) may interact to magnify the deleterious effects of oxidative damage. Although muscle antioxidant capacity was found to be improved after exercise training in patients with heart failure (134) and in healthy subjects, the response to exercise training in patients with COPD is highly variable (118, 135).
A last factor may be related to the angiotensin-converting enzyme genotype of patients with COPD. Hopkinson and coworkers reported that the deletion (D) allele of the angiotensin-converting enzyme gene polymorphism was associated with better preserved quadriceps force (136). Interestingly, the insertion (I) allele of the angiotensin-converting enzyme polymorphism may be related to enhanced endurance capacity in healthy subjects (137), athletes (138, 139), and patients with COPD (140). In addition, the insertion allele may result in somewhat greater training effects in terms of maximal exercise capacity (140). This is surely an exciting avenue for further research, and the integration of the genotype of patients with COPD may prove to be helpful in our understanding of the variability in muscle weakness seen across patients with COPD in future research.
Irrespective of the precise mechanism leading to skeletal muscle abnormalities, it is clear that reversing this dysfunction is an important, and clinically relevant, objective of exercise training programs in COPD.
Principles of Design for Exercise Training Programs
In healthy subjects, cardiorespiratory fitness is improved when the following guidelines for exercise training intensity are observed. Exercises should be performed 3 to 5 days per week at an intensity above 40 to 85% of the oxygen uptake reserve (difference between resting and peak oxygen uptake) for more than 20 minutes (or at lower intensity, preferably for 30 minutes), continuously or in intervals (141). These guidelines were established by rather painstaking studies (142) in which matrices of programs with various characteristics were studied, with the goal of establishing the minimal parameters of an effective program. Whether these guidelines are appropriate for patients with COPD has been debated. Systematic studies to determine the minimal program duration, session duration, or number of sessions per week have not been performed. Without evidence to the contrary, it has generally been conceded that these program parameters are unlikely to differ substantially between healthy subjects and patients with COPD (9, 143). Nevertheless, controversy exists.
The duration (weeks of training) of the rehabilitative exercise training program is a much discussed but poorly investigated topic. There is substantial sentiment in the rehabilitation community that longer programs than absolutely required to elicit a physiologic training effect may be of benefit, taking into account that patients generally start off with severe muscle weakness and deconditioning. Clearly, shorter programs (6–8 weeks) are less expensive and allow more patients to experience rehabilitation. However, a key goal of pulmonary rehabilitation is to change the patient's behavior (e.g., from a sedentary toward a more active lifestyle). Although measurable physiological changes may occur within weeks, behavioral changes may require longer time periods (144). It is therefore not surprising that some pulmonary rehabilitation programs of longer duration (6 months or longer) have shown better long-term effects (33, 52) as compared with shorter interventions (27, 145, 146). On the other end of the spectrum, responses to short-term rehabilitation programs seem to be strongly influenced by the duration of the program (147–149) and the training intensity (34). Programs shorter than 6 weeks currently lack validation, as no randomized controlled trial has been performed. One trial compared 4 weeks with 7 weeks of rehabilitation (147) and concluded that 4 weeks of rehabilitation was less effective. Two other studies reported effects at 4 and 8 weeks (150) and at 6 and 12 weeks (51) and confirmed that longer rehabilitation programs yield significantly greater effects. In a large multicenter trial, greater effects were shown after 6 months compared with 3 months of rehabilitation (151).
At present, rehabilitation programs including high-intensity endurance training in which patients are engaged for a relatively long period of time have been demonstrated to reduce symptoms and increase exercise tolerance, skeletal muscle function, and health-related quality of life. A minimum of about 8 weeks seems necessary to achieve substantial effects, but longer programs generally achieve more favorable results. It is worth noting that neither patients nor health care providers are well served by programs that yield only modest benefits. The goal should be to induce as great a benefit as possible and these benefits should be long-lasting.
Can Training Intensity Be High Enough in Patients with COPD to Yield a Physiologic Training Effect?
During whole body exercise, at the peak exercise level tolerated significant metabolic reserve remains in the skeletal muscle of many patients with severe COPD due to a ventilatory limitation (152). In contrast, when small muscle groups are exercised in isolation, a higher metabolic rate can be achieved (153) because the ventilatory demand of exercising a small muscle mass is not great. It has been posited that in patients with severe COPD it may be difficult to achieve a level of training intensity high enough to result in physiological training effects due to the central ventilatory limitation (13). However, there is now a wealth of experience that indicates that physiological evidence of a training effect can be achieved, if high training targets are employed. As first pointed out by Punzal and coworkers, patients with COPD can exercise for prolonged periods at high fractions of their peak exercise tolerance (154). In patients with COPD with primarily moderate disease, exercise training conducted at about 75% of the peak work rate (60% of the difference between the lactate threshold and peak oxygen uptake) resulted in significant physiological effects (34). A similar training strategy was shown to be effective in patients with severe disease. (37) Others have confirmed that high training intensity is required to elicit physiologic training effects (155, 156). These findings probably partially explain the disappointing training effects in earlier studies (13), in which exercise intensity may have been insufficient. Maltais and coworkers showed that training at 60% of the peak work rate achieved in an incremental exercise test resulted in significantly improved exercise tolerance (157), but that few subjects were able to achieve a target training intensity of 80% of the peak work rate, despite careful supervision and encouragement. A word of caution is warranted when a fraction of the peak work rate achieved in an incremental exercise test is used to specify a target training intensity. The rate of work rate increment used during the maximal incremental exercise test is a determinant of the peak work rate obtained (158). In the study by Maltais and coworkers (157), 10-W/minute increments were used. If a lower rate of work rate increment had been used, peak work rate would have been lower and the authors would probably have concluded that patients could tolerate a higher fraction of this revised peak work rate. Using the appropriate work rate increment, Neder and coworkers confirmed that the work rate that could be sustained by patients was about 80% of the peak work rate, on average (159). When walking is used as the primary form of exercise, even higher fractions of the peak work can be sustained (154, 160). Walking engages larger muscle groups and hence results in less stress on the skeletal muscle at isoventilation. Hence, during walking, patients may reach the limits of the ventilatory system before the skeletal muscle is fatigued during exercise (161). This is also reflected in the lower lactate levels at isoventilation and identical oxygen consumption (160, 162). As long as ventilation is just below peak ventilation, patients with COPD without significant respiratory muscle weakness are able to sustain the exercise for a relatively long time (159, 160). From the foregoing, it can be speculated that exercise intensity targets in the range of 60 to 80% of the peak work rate may be practical depending on the exercise type and the incremental exercise protocol applied. The lower limit of effective training intensities (i.e., the critical training intensity) has not been identified with any certainty, although it seems likely that the higher the intensity, the better the physiological training effect. It should be stressed that these considerations yield only an initial exercise target. Rehabilitation therapists need to be given the latitude to adjust these targets up or down depending on the patient's symptomatic response. Hence close supervision of the patients is needed. Further, as the rehabilitation program proceeds, therapists should be encouraged to adjust training intensity. Weekly increments of training intensity are possible in patients with COPD (33, 115, 157, 163). Symptom scores have been posited to be useful in establishing an exercise prescription and for subsequently adjusting it. One group has proposed that on the modified Borg Rating of Perceived Exertion Scale (with a range of 0 to 10 points) a score of 4–6 generally indicates adequate training intensity (164). Alternatively, the target dyspnea score can be obtained from incremental exercise testing. These targets remain stable for a relative work rate over the course of a rehabilitation program (165). At present it remains unclear whether symptom scores are a sufficiently reliable method for establishing an exercise prescription. Nevertheless, the Borg Scale has been used in many clinical studies to guide increments in training load (27, 33, 48, 54, 55, 166).
As indicated above, it is difficult to ensure that all patients with COPD achieve and tolerate a level of exercise intensity high enough that a physiologic training effect is achieved. Strategies have been investigated to allow the patient to tolerate higher exercise intensities with the aim of achieving a greater training effect, or a more comfortable training regimen. This is especially true for the patients with very severe disease: those with severe muscle weakness or profound ventilatory limitation to exercise (this has been a particular concern in patients awaiting lung volume reduction surgery and lung transplantation).
Specific Strategies to Increase Training Intensity
Reducing work of breathing during training.
Many patients with COPD reach the boundaries of the ventilatory system during incremental exercise (167), and high-intensity constant work rate exercise (159). Hence interventions that allow higher levels of ventilation, or reduce the ventilation requirement at a given level of exercise, may allow patients to perform exercise training at higher intensities. This may lead to enhanced training results, typically in patients in GOLD Stages III and IV (FEV1 less than 50% predicted).
OXYGEN SUPPLEMENTATION.
Oxygen supplementation during exercise clearly reduces the ventilatory requirements for a given work rate and increases maximal exercise tolerance, even in patients without appreciable exercise desaturation (168). At high exercise levels, oxygen delivery to the exercising muscles is enhanced (169). At near maximal exercise levels, modest reductions in circulating lactate levels are correlated with the reduced ventilation during exercise (170, 171). These findings were confirmed in healthy subjects breathing hyperbaric, or hyperoxic, air (172, 173). At lower exercise levels, the major mechanism of benefit is reduction in carotid body drive to breathing (174). Acute administration of oxygen was also shown to reduce the exercise-induced pulmonary hypertension in moderate to severe COPD (175). Despite the consistent improvements found in exercise tolerance, most studies seeking to determine whether providing supplemental oxygen during training improved the gains in exercise tolerance elicited by rehabilitative training have failed to show an additional benefit (176–178). This led to the belief that oxygen supplementation during exercise training would be of limited benefit. In one study, however, Emtner and colleagues studied patients with COPD who did not experience clinically significant desaturation with exercise as they participated in a training program with high-intensity targets (179). Those patients randomized to respire supplemental oxygen during exercise training succeeded in training at higher training intensities than those respiring supplemental air. Consequently the improvement in exercise tolerance was greater in patients who trained with oxygen supplementation. The observation that the benefit was more pronounced in patients with a greater acute response to oxygen supplementation is of special interest to clinicians prescribing oxygen therapy. Despite the fact that this is a relatively small study, these data suggest that oxygen supplementation may be a useful tool with which to enhance training intensity in patients with COPD (180). This in turn may lead to greater physiological effects of the exercise training program. Although these results are promising, so far, no studies showed that use of oxygen supplements during training results in superior and clinically significant benefits in quality of life and activities of daily living. Neither have studies confirmed that providing oxygen during exercise training sessions results in more durable effects of the training programs. This is surely a point of interest for further research. These missing elements preclude making strong recommendations, at present, on the use of oxygen supplementation in rehabilitation practice. Nevertheless, if confirmed in longer term studies these findings may form the basis of a more rational prescription of oxygen during high-intensity exercise training programs.
NONINVASIVE MECHANICAL VENTILATION.
Noninvasive mechanical ventilation, administered in the form of continuous positive pressure ventilation or pressure support ventilation (bilevel positive airway pressure or proportional assist ventilation), unloads the respiratory muscles (181). Noninvasive mechanical ventilation has therefore found use during severe exacerbations of COPD (182) and hypoxic respiratory failure (183). It also has been shown to improve dyspnea and exercise endurance in patients with COPD (184–186), healthy subjects (187), and even patients with chronic heart failure (188). Using inspiratory support, the load on the inspiratory muscles is reduced, with a consequent reduction in the work of breathing (184). Blood gases improve (189) and patients are able to sustain lactic acid accumulation for a longer period of time (190). Hence, the application of inspiratory support, through proportional assist ventilation, or continuous positive airway pressure ventilation may potentially lead to enhanced training intensity. Two trials indeed confirmed this hypothesis (191, 192), showing somewhat greater physiologic benefits of training with "proportionally assisted ventilation" in patients with severely impaired, hypercapnic COPD. In patients with less severe disease, however, others could not confirm these findings (193). Patient selection seems crucial when clinicians apply this intervention. The improvement obtained by noninvasive mechanical ventilation was more pronounced in patients with respiratory muscle weakness (186). Given the difficulties in adjusting the mask and ventilator settings, individual supervision of patients is necessary. In addition, two of the four studies investigating the effect of inspiratory pressure support during exercise training report that some patients do not support the ventilatory assist (192, 193). It should be noted, however, that the total dropout reported in the four groups treated with noninvasive mechanical ventilation (191–194) was not significantly different from the dropout among patients not receiving this intervention. The aforementioned factors may complicate the clinical application of assisted ventilation in exercise programs. As an alternative, noninvasive positive pressure ventilation at home in combination with an outpatient rehabilitation program resulted in an additional increase in the shuttle walking distance and health-related quality of life compared with exercise training alone (195).
HELIUM–OXYGEN BREATHING.
Breathing gas mixtures with low density (e.g., 79% He, 21% O2; heliox) lowers airflow resistance and therefore decreases exercise-induced hyperinflation (196). Exercise tolerance is improved (152, 197). So far, however, no trial has shown that heliox breathing during rehabilitative exercise training yields a superior increase in exercise tolerance outcomes (194). It should be mentioned that, with current technology, the cost of respiring heliox during an entire exercise program is appreciable.
BREATHING EXERCISES.
The benefit of adding breathing exercises to pulmonary rehabilitation programs is yet unclear. Diaphragmatic breathing, long a technique taught in rehabilitation programs, seems on shaky ground, as it been shown to decrease breathing efficiency (198). The deterioration of breathing efficiency during diaphragmatic breathing may result from the recruitment of additional muscles or a deteriorated mechanical coupling of inspiratory muscles, rib cage, and abdomen, reducing the pleural pressure developed for a given muscle contraction (198). Pursed-lip breathing is often adopted by patients spontaneously (9, 199). With pursed-lip breathing, recruitment of the abdominal muscles during expiration is facilitated (200). During quiet pursed-lip breathing, pulmonary gas exchange is improved, as is illustrated by an increase in arterial oxygen saturation (200). Pursed-lip breathing has also been shown to have favorable effects on breathing pattern (201–203), increasing tidal volume, reducing the ratio of the inspiratory time to the total duration of the respiratory cycle (TI/Ttot), and reducing dyspnea. Significant reductions in end-expiratory lung volumes, and hence hyperinflation, were observed with optoelectronic plethysmography, especially in patients with more severe airflow obstruction (201). In selected patients pursed-lip breathing was also shown to reduce the oxygen cost of breathing (202). The effect of pursed-lip breathing during exercise has not been thoroughly studied. Adding 5 cm H2O of positive expiratory pressure (much like the pressure achieved during pursed-lip breathing) did not alleviate dyspnea during exercise in patients with moderate COPD (204). According to the above-mentioned observations, however, these patients may not have been the best candidates. In clinical practice, patients with severe COPD can be readily trained to employ pursed-lip breathing and the effectiveness can be assessed both symptomatically and by oxygen saturation measurements. When effective, patients generally will adopt this breathing pattern spontaneously at rest, during exercise, or during recovery from exercise.
Adjustments to the training program to deal with central limitations.
INTERVAL TRAINING. Some patients have significant problems with achieving the target training intensity for the 20- to 30-minute duration recommended by the American College of Sports Medicine (157). Other forms of whole body exercise training have therefore been developed in COPD. During interval training, the 30-minute exercise bout is divided into smaller exercise bouts with appropriate training intensity (e.g., 70–100% of the peak work rate in an incremental test). These bouts may be as long as 2 to 3 minutes (205, 206), or as short as 30 seconds (207, 208). Using 1-minute exercise bouts, Sabapathy and coworkers showed that ventilation was reduced by about 20%, compared with constant work at identical work rate. Consequently training time and total work performed were increased by 63% and 31%, respectively (209). With interval training higher work rates are achieved with lower symptom scores (207). The total exercise time of 30 minutes, however, is respected. Besides its effectiveness in patients with COPD, interval training was shown to be a suitable training modality in congestive heart failure (210) and in healthy elderly subjects (211). Although no randomized trial has shown that interval training is superior to continuous exercise training in rehabilitative exercise programs for patients with COPD (206, 207), these programs have been shown to elicit fewer symptoms during exercise training (207). The latter has been linked to less dynamic hyperinflation during this form of exercise (208, 209). Hence this form of training may be more comfortable in patients with more severe dynamic hyperinflation. It would be worthwhile to compare the effects of interval training in patients with severe airflow obstruction, who might benefit most from this training modality
STRENGTH (RESISTANCE) TRAINING.
Training small muscle groups alleviates the ventilatory burden of exercise training. Hence these muscles, not limited by the central cardiorespiratory limitations, can achieve significantly higher relative work rates (152, 153). Resistance training (weight lifting) is a way of training small muscle groups, and has been applied in several studies in COPD (29, 113, 114, 212–214). Training volume is expressed as the number of repetitions and the weight lifted (generally expressed as a fraction of the maximum weight that can be lifted once, the one-repetition maximum). A typical program might incorporate two or three sets of eight repetitions each at 70% of the one-repetition maximum (114, 115). Studies using resistance training as the only training modality consistently find improved skeletal muscle strength after at least 8 weeks of resistance training. Whether there is transfer toward improved endurance in patients suffering from muscle dysfunction is still controversial. Among healthy elderly subjects, studies showed that resistance training improved skeletal muscle oxidative capacity and endurance (215) and increased capillarization and oxygen flux through the skeletal muscle (216, 217). In COPD, increased whole body endurance, 6-minute walking distance, and  O2 peak were also reported after resistance training (29, 114, 115). However, other trials in COPD failed to show convincing effects on whole body endurance when resistance training was added to endurance training (113), or with resistance training as the only intervention (214). Studies looking more closely at muscular changes after resistance training (muscle biopsy material, or magnetic resonance spectroscopy) are not yet available in COPD. Nevertheless, the enhanced muscle strength reported in all trials that included resistance training is in itself a clinically relevant improvement in these patients, who generally suffer from muscle weakness. Resistance training is currently often used in combination with endurance training. This combination was shown in two studies to have complementary effects (i.e., effect on strength and endurance) (113, 114). Therefore, the combination of resistance and endurance exercise is advised in clinical rehabilitation. It should be acknowledged, however, that the greater improvement in skeletal muscle strength after the addition of resistance training has not been shown to translate into superior benefits in terms of health-related quality of life, or symptomatology.
NEUROMUSCULAR ELECTRICAL STIMULATION.
During neuromuscular electrical stimulation, specific muscle groups of interest (generally lower limb) are activated using low-intensity electrical currents. Two trials have evaluated the effectiveness of 6 weeks of neuromuscular electrical stimulation in COPD (218, 219). Both studies showed that in stable patients with severe muscle weakness, transcutaneous neuromuscular electrical stimulation of the lower limb muscles resulted in significant increases in skeletal muscle strength and functional exercise capacity. Peak oxygen uptake increased in one study (218). In this study increased endurance of a constant work rate task was also shown. In more debilitated patients with respiratory failure who were receiving mechanical ventilation and had a history of being bed bound for at least 30 days, Zanotti and coworkers showed a faster functional recovery of patients when neuromuscular electrical stimulation was applied (220). Although preliminary reports (in the literature a total of 24 patients have received the intervention so far) are in favor of using neuromuscular electrical stimulation as a modality of resistance training in the weakest patients, further study is needed to determine its place in conjunction with other interventions. Neuromuscular electrical stimulation, however, is apparently safe, can be conducted at home, has few reported side effects, and is relatively inexpensive.
Ergogenic drugs: growth hormone/insulin-like growth factors and anabolic steroids.
A review of the literature reveals that five studies have investigated the effects of anabolic steroids in COPD. Most studies have included only men. Studied drugs are oxandrolone (221), nandrolone decanoate (222, 223), stanozolol (224), and testosterone (214). All studies report an increase in body weight. Anabolic steroids have the benefit over nutritional interventions (222) and appetite stimulants (225) that the increase in body weight is achieved mainly through a gain in lean body mass (221, 222, 224) (in fact, fat mass generally decreases). Gain in lean body mass may be beneficial because increase in lean mass has been associated with decrease in all-cause mortality in patients with severe COPD (226). Four studies were conducted in the context of a pulmonary rehabilitation program consisting of endurance training, strength training, and/or respiratory muscle training (214, 222–224). Anabolic steroids have been found to increase muscle strength (and to enhance the effects of strength training programs) (227), but not to improve exercise endurance (or to enhance the effects of endurance programs) (228, 229). This is not surprising, because anabolic steroids induce muscle hypertrophy and do not induce the capillary increases and increases in aerobic enzymes that would be expected to increase exercise endurance (230).
So far the clinical relevance of anabolic steroid use in unselected patients with COPD admitted to rehabilitation is not clear (231), as no systematic improvements in skeletal muscle force or exercise tolerance were observed (221, 224).
Some comments should be made regarding the limitation of these studies. First, the dose of anabolic steroids that appropriately balances risks and benefits has not been established. This is made more difficult when testosterone analogs are used. These drugs yield decreased levels of circulating testosterone (through pituitary inhibition of gonadotropin secretion) so that blood assays are not generally helpful in assessing the adequacy of dosing. When testosterone is administered, levels within the physiologic range for young men or supraphysiologic levels can be targeted (232). It deserves mention that the dose–response relationship between testosterone dose and anabolic effects (muscle mass gain, strength increase) has been found to be linear well beyond the physiologic range (233). A second issue regards patient selection for this intervention. Hypogonadal men have been found to have substantial benefit of normalization of testosterone levels (234), so it has been proposed that candidates for testosterone supplementation have circulating levels in the low (or at least low normal) range. In fact, many men with severe COPD have testosterone levels in this range (214, 235, 236). Third, as many patients with COPD receive intermittent systemic corticosteroid courses and corticosteroids have been shown to induce muscle dysfunction (119) it has been proposed that anabolic steroids can be used to counteract the effects of corticosteroids (223). In fact, patients receiving oral corticosteroid treatment often have reduced testosterone levels (235). Administration of anabolic steroids may normalize androgen levels and "protect" patients from developing corticosteroid-induced myopathy (120). In a post hoc analysis, Creutzberg and coworkers showed that patients receiving oral corticosteroid treatment indeed derived more functional benefit from nandrolone treatment (223). Last, administration of anabolic steroids was shown to be more effective when combined with resistance training (237). In one study, Casaburi and coworkers indeed reported that quadriceps force and fatigability tended to improve more when hypogonadal patients undergoing resistance training received supplemental testosterone (strength, +27 versus +17% in resistance training only; fatigability, +81 versus +45%) (214). These results encourage further larger scale studies in this target population (hypogonadal COPD), to revisit the rather negative image of androgen replacement therapy. At present, however, data showing better preserved long-term results of resistance training in patients with COPD are lacking. It remains unclear whether patients should be maintained on testosterone treatment and/or resistance training to maintain the observed benefits. In healthy subjects testosterone administered for up to 20 weeks seemed to have an acceptable safety profile, but it remains unclear whether these data can be extrapolated to COPD. Ferreira and colleagues administered stanozolol for 6 months to patients with COPD without significant (short-term) side effects (224). Testosterone replacement therapy should not be administered to patients with prostate hypertrophy or a history of prostate cancer. Testosterone increases hematocrit (238), and therefore seems not a treatment option in patients with a hemoglobin concentration greater than 16 g · dl–1. Because testosterone can cause sodium and water retention, caution is also warranted in patients with renal disease and congestive heart failure.
Administration of recombinant human growth hormone is another pathway to stimulate anabolism in patients with muscle wasting. Although controversial in healthy elderly subjects (239), the rationale for the addition of growth hormone or insulin-like growth factor (the primary mediator of the anabolic action of growth hormone on muscle) to the therapy of patients with COPD can be found in the disturbed anabolic–catabolic balance (240). One study investigated the effectiveness of growth hormone replacement in the context of pulmonary rehabilitation but failed to show an effect on skeletal muscle force or exercise capacity (241). In heart failure, a trial confirmed that systemic growth hormone replacement therapy was safe, but it did not improve cardiac function or exercise capacity (242). Whether specific groups of patients (e.g., those with blunted insulin-like growth factor-I [IGF-I] response to exercise) may benefit from growth hormone administration or from more advanced strategies to stimulate the growth hormone axis (e.g., locally administered IGF-I [243], recombinant human IGF-I in combination with one of its binding proteins [IGF–binding protein-3 complex; 244], or administration of a growth hormone-releasing factor [245]) should be further investigated. For the present, however, the high cost of growth hormone, the requirement that it be administered by injection, and the lack of evidence of substantial functional benefits do not make it an attractive intervention in the context of a rehabilitation program.
Respiratory muscle training.
The respiratory muscles have been specifically targeted for training in COPD. Inspiratory muscle training programs can be conducted at home, using resistive breathing with target inspiratory pressures or target inspiratory flows or with threshold loading devices. Normocapnic hyperpnea has also been applied, albeit less frequently, in COPD (246). When training load is appropriate (controlled and more than 30–40% of maximal inspiratory pressure [PImax]), inspiratory muscle training leads consistently to reductions in dyspnea and improved measures of inspiratory muscle performance (247). Programs are relatively inexpensive, but require regular supervision. Whether inspiratory muscle training translates to increased exercise tolerance and quality of life is much less clear (247). Therefore there has been some debate as to whether inspiratory muscle training should be part of rehabilitation programs in COPD, with evidence-based guidelines concluding that it should not be a routine component (16, 18, 83). Its rationale has been questioned. Although measures of inspiratory muscle strength (indirectly assessed through the pressure these muscles can generate) are clearly reduced in the average patient with COPD, this is in part because the respiratory muscles act at a mechanical disadvantage because of hyperinflation. The respiratory muscles adapt to the chronically imposed increased work of breathing and, unlike the peripheral muscles, the diaphragm of patients with severe COPD has a greater oxidative capacity than is seen in healthy subjects (248–252). In addition, it has been shown that patients with COPD are able to sustain a level of ventilation close to the maximal voluntary ventilation during exercise (159), and contractile fatigue is rarely observed after exercise in stable COPD (253, 254). Hence respiratory muscles, unlike the peripheral muscles, seem not to suffer from "deconditioning" and a training intervention may not be warranted. On the other hand, diaphragm work is increased in COPD, especially during exercise (255, 256). The increased relative work of breathing is illustrated by the high PI/PImax ratio at which patients breathe during exercise. PI/PImax is related to the perceived dyspnea sensation, especially when the tidal volume/vital capacity ratio is also taken into account (257). It leads to the perception of "unrewarded inspiration," "inspiratory difficulty," and "shallow breathing" (257), which is clearly distinct from the complaints seen in healthy subjects during exercise. Hyperinflation places a further burden on the respiratory muscles. Therefore the mechanisms of adaptation of the respiratory muscles may still be insufficient in some patients with COPD (258); if respiratory muscle training increases PImax, it seems reasonable to expect that dyspnea on exertion will be reduced. It is clear that well-designed inspiratory muscle training does improve PImax and leads to hypertrophy of the Type II fibers in the intercostal muscles, and an increase in the proportion of Type I fibers (259). In patients with inspiratory muscle weakness it can be speculated that this may transform into functional benefits. This has indeed been confirmed, to some extent, by a systematic literature review (247). Therefore, in patients with inspiratory muscle weakness, the prescription of strictly standardized inspiratory muscle training may be justified as an adjunct to exercise training (64, 246, 260), with the aim of improving exercise-induced symptoms of dyspnea. It should be noted, however, that whole body exercise training, by itself, has improved inspiratory muscle force in some studies (33, 39). Inspiratory muscle training as a stand-alone treatment is clearly inferior to general exercise training in COPD if the goal is to improve function or health-related quality of life. As to the latter outcome, no study has shown that the addition of inspiratory muscle training to a whole body exercise training program yielded significant benefits. In addition, one study investigating the long-term effects of inspiratory muscle training suggested that the effects obtained from the inspiratory muscle training may—as would be expected from the responses to exercise training—wear off rapidly when the inspiratory muscle training is discontinued (261).
Should Exercise Training Be Prescribed for All Patients with COPD?
As a group, patients with COPD clearly benefit from exercise training programs. For a long time, patients with severe lung disease were believed not to be good candidates for exercise training programs (13). Later, lactic acid production was seen as a prerequisite to achieve an appropriate training response (34), as reductions in lactic acid were coupled to significant reductions of minute ventilation for a given work rate and oxygen consumption. However, more recent studies in which high-intensity training stimuli were applied clearly showed that patients with FEV1 below 40% of the predicted values or those with hypercapnia (GOLD Stage IV) should not be excluded from exercise training (37, 37, 157, 262, 263). In a large study by Berry and coworkers (166), patients with mild, moderate, and severe COPD showed the same proportional improvement in exercise tolerance after pulmonary rehabilitation. Further, in patients not manifesting significant lactic acid accumulation during heavy exercise, significant physiologic training effects were confirmed (37, 154). Despite the overall success of pulmonary rehabilitation across disease severity stages, a significant number of patients (roughly 30%) have no significant improvement in exercise capacity (264). When increases in peak oxygen consumption are used as the only measure to define response to exercise training, as many as 50% of the patients may be nonresponders (165). This is clearly more than is observed after carefully monitored endurance training in healthy subjects (265). The question concerning why a minority of patients is not responding adequately to an exercise training stimulus is intriguing, and no final answer can yet be given.
Clinical data suggest that patients with more pronounced muscle weakness and somewhat less impaired ventilatory reserve may be better candidates for exercise training programs (264). These data concur with others showing that patients with a poor walking distance—which is correlated with muscle weakness in COPD—and greater breathing reserve are the best candidates for exercise training (266). The variance explained by the proposed model, however, is too low to allow for appropriate patient selection. Older age (267), severe lung function impairment (166), the presence of hypercapnia (262), and current smoking (268) seem not to be valid exclusion criteria for pulmonary rehabilitation.
It should be acknowledged, however, that failure to respond to an exercise program, in terms of physiological benefits, does not mean that the rehabilitation of the patient should be regarded as unsuccessful. In one of the above-mentioned studies the so-called nonresponders showed a clinically relevant improvement in health-related quality of life (264). Identifying patients nonresponsive to exercise training should therefore not necessarily lead to a reduction of patients referred to comprehensive pulmonary rehabilitation. It may, however, lead to a better program design. Other interventions that are integral parts of rehabilitation (described below) improve individual patients in important respects outside of the context of exercise physiology.
|
OTHER COMPONENTS OF PULMONARY REHABILITATION PROGRAMS
|
|---|
A European Respiratory Society task force on pulmonary rehabilitation recommended that rehabilitation programs should be flexible, patient tailored, and include a variety of components besides exercise, depending on resources available and taking into account patient preferences and available evidence (19). The American Thoracic Society in its 1999 guideline on pulmonary rehabilitation (9) and the British Thoracic Society guidelines (18) also advocate the multidisciplinary nature of pulmonary rehabilitation programs with attention to educational, psychosocial, and behavioral interventions.
Improving Education and Self-management
When delivered outside of the context of a comprehensive rehabilitation program, the benefit of educational sessions is considered only minimal (27, 63, 83, 269, 270). Education aiming only to improve patient knowledge about the disease does not improve health status (269, 271). Educational programs improving "self-management" of patients may be more successful. Self-management is a form of education aiming at teaching skills to optimally control the disease, achieve behavioral change, and improve coping with the disease. Although a meta-analysis (272) and a trial in "low-risk" patients (273) showed no significant effects of these interventions, another trial including patients with more severe disease at risk for readmission after hospitalization was successful (78). Because the number of sessions is generally limited, these interventions were also shown to be cost-effective (274) and can be incorporated relatively easily in rehabilitation programs.
Several points can be addressed in educational interventions. Because up to 75% of patients may have difficulties in understanding how and when to take their inhalation medication (275), this may be a particularly useful target for educational sessions. In the context of a pulmonary rehabilitation program, education was shown to improve adherence to medication (276). Another important aspect in education programs concerns teaching patients to deal with exacerbations, using so-called action plans. These may be of particular benefit to patients with a history of being admitted to the hospital (78). With specific training, patients may adopt these action plans rapidly, and deal more appropriately with acute exacerbations (277). Education sessions may be offered to individual patients or small groups of patients by all members of the rehabilitation team.
Psychosocial Support: Rationale and Practical Implication
The incidence of depression in patients with COPD is about 2.5-fold higher compared with the general population (278). Pulmonary rehabilitation programs including psychological interventions improve these mood disturbances more than those consisting of exercise training only (279). Hence psychological counseling may be of benefit to those patients selected for pulmonary rehabilitation and presenting with symptoms of anxiety and depression (some 20–40% of patients [65–67]). In addition, spouses of patients with COPD were reported to suffer from depressive symptoms and stress (280–282), and may require the attention of the rehabilitation team. Psychologists are the best placed health care providers to provide treatment of patients with significant anxiety or depressive symptoms and also to optimize and discuss smoking cessation strategies with patients continuing to smoke. Smoking cessation counseling, education, and nicotine replacement therapy (or treatment with antidepressants such as bupropion and nortriptyline) maximize the chances for sustained smoking cessation (57). Some studies of rehabilitative therapy have excluded current smokers (49), but the rationale to do so is lacking. Smokers, however, are more likely to decline invitations to take part in rehabilitation programs (66, 283). Indeed, smokers are generally less adherent than ex-smokers (284). Nevertheless, a significant number of patients are current smokers at inclusion in pulmonary rehabilitation programs (27, 66, 285) that do not explicitly exclude smokers.
Improving Activities of Daily Living: Role for Occupational Therapy
Transforming the achieved physiological improvements into benefits that are relevant to patients is critical to the success of pulmonary rehabilitation. This integration may be facilitated by occupational therapists. Programs including specific occupational therapy interventions showed significant benefits on reported activities of daily living (51). The intervention by the occupational therapist may consist of functional training in activities of daily living, energy-conserving strategies, and the use of wheeled walking aids. The latter have been shown to result in increased functional autonomy in patients with COPD, especially in the patients with more severe disease (286, 287), through an increase in ventilatory capacity and walking efficiency (288). The prescription of rollators adds to the incremental cost of pulmonary rehabilitation (289). The latter authors report that the prescription of assistive devices (mainly rollators) led to an incremental cost of 133 U.S. dollars (1989 U.S. dollars). Only one, relatively small study has addressed this aspect to some extent in a prospective study (290). This study suggests that independence in activities of daily living is improved only when patients receive occupational therapy as part of their rehabilitation program. Although promising, the design of this single study (nonrandomized) and the fact that the results pertain to a short inpatient rehabilitation program do not allow us to draw firm conclusions.
Nutritional Interventions
Patients with COPD are often underweight (291). The loss of fat-free mass is most closely related to morbidity (291, 292) and mortality (293). Nutritional interventions as a treatment of weight loss are, in general, not effective in outpatients (294). However, one study showed that when patients gain more than 2 kg of body weight, survival is significantly improved (226). In addition, an increase of 1 body mass index unit was also associated with improved survival in underweight patients (293). It should be noted that the use of body mass index as an indicator of "underweight" may be misleading, as many patients with a normal body mass index have a low fat-free mass (295). The latter factor is related to muscle wasting (110), which is even more closely related to mortality risk (93). Nutritional specialists involved in pulmonary rehabilitation should balance the calorie intake of patients with the energy requirements of patients, taking into account that patients with COPD have increased resting energy expenditure compared with age-matched subjects (296). Patients nonresponsive to nutritional interventions are characterized by more systemic inflammation, older age, and low dietary intake (297). Exercise training or a more active lifestyle (106) may induce a negative protein balance because of the increased energy expenditure. In these patients, nutritional supplements may be beneficial and may even lead to a somewhat greater training effect in terms of functional status (298, 299) and body composition and health-related quality of life (299), but the data currently available remain insufficient to make strong recommendations regarding the use of nutritional supplements. At present prospective randomized trials supporting the use of nutritional interventions unequivocally are missing. Nevertheless, in selected patients with insufficient calorie (in particular, protein) intake, nutritional supplements may be helpful to meet prescribed calorie intake. Further research could focus on the long-term nutritional advice that should be given to patients. There is currently no evidence that nutritional supplements further increase the duration of benefits of exercise training. In addition, it should be recognized that nutritional supplementation may deter patients from consuming their regular meals, hence resulting in less rather than more calorie intake, jeopardizing any positive effect of nutritional supplements (300). Another new avenue for future research is the potential impact of supplementing specific nutrients and amino acids that may play a role in the oxidant–antioxidant balance, or are specific building blocks of key proteins (128, 301, 302). The clinical implications and precise prescription regimens of these so-called nutriceuticals in COPD are, however, currently unknown.
Obviously, in obese patients (some 14% of the patients admitted to the rehabilitation program in Leuven, Belgium), the focus of the dietary intervention may be weight loss rather than weight gain. The challenge here is to not lose fat-free mass.
|
EVALUATION BEFORE, DURING, AND AFTER PULMONARY REHABILITATION
|
|---|
|
TOP
CONTENTS
EFFECTIVENESS IN TERMS OF...