Pulmonary Rehabilitation* Joint ACCP/AACVPR Evidence-Based Clinical Practice Guidelines

Andrew L. Ries, MD, MPH, FCCP (Chair); Gerene S. Bauldoff, RN, PhD, FCCP; Brian W. Carlin, MD, FCCP; Richard Casaburi, PhD, MD, FCCP; Charles F. Emery, PhD; Donald A. Mahler, MD, FCCP; Barry Make, MD, FCCP; Carolyn L. Rochester, MD; Richard ZuWallack, MD, FCCP; Carla Herrerias, MPH


CHEST. 2007;131(5):4S-42S. 

In This Article

Outcomes of Comprehensive Pulmonary Rehabilitation Programs

As currently practiced, pulmonary rehabilitation typically includes several different components, including exercise training, education, instruction in various respiratory and chest physiotherapy techniques, and psychosocial support. For this review, comprehensive pulmonary rehabilitation was defined as an intervention that includes one or more of these components beyond just exercise training, which is considered to be an essential, mandatory component.

In addition to the clinical trials reviewed in the evidence tables in this document, several systematic reviews and metaanalyses have been published within the past decade that support the beneficial effects from comprehensive pulmonary rehabilitation programs. In a Cochrane Review published in 2006, Lacasse[30] analyzed 31 RCTs in patients with COPD and concluded that rehabilitation forms an important component of the management of COPD. They reported statistically and clinically significant improvements in important domains of quality of life (i.e., dyspnea, fatigue, emotions, and patient control over disease). Improvement in measures of exercise capacity were slightly below the threshold for clinical significance. Similarly, after a systematic review, Cambach and colleagues[31] identified 18 articles for inclusion in a metaanalysis of outcome measures of exercise capacity and HRQOL in patients with COPD. They found significant improvements for exercise measures of maximal exercise capacity, endurance time, and walking distance, and for HRQOL measures in all dimensions of the Chronic Respiratory Disease Questionnaire (CRDQ) [ie, dyspnea, fatigue, emotion, and mastery]. Improvements in maximal exercise capacity and walking distance were sustained for up to 9 months after rehabilitation.

Dyspnea. In the previous evidence-based review document[2,3] the 1997 guidelines panel concluded that the highest strength of evidence (A) supported the recommendation for including lower extremity exercise training as a key component of pulmonary rehabilitation for patients with COPD. In addition, the panel concluded that there was high-grade evidence (A) that pulmonary rehabilitation improves the symptom of dyspnea in patients with COPD. This panel concluded that the evidence presented in Table 3 in this document further strengthens those conclusions and recommendations.

1. A program of exercise training of the muscles of ambulation is recommended as a mandatory component of pulmonary rehabilitation for patients with COPD. Grade of recommendation, 1A

2. Pulmonary rehabilitation improves the symptom of dyspnea in patients with COPD: Grade of recommendation, 1A

Regarding changes in HRQOL, the previous panel concluded that there was B level strength of evidence supporting the recommendation that "pulmonary rehabilitation improves health-related quality of life in patients with COPD." Based on the current review, this panel believes that the additional published literature now available strengthens support for this conclusion and upgrades the evidence to grade A. In this document, the term HRQOL will be used interchangeably with the term health status.

In one of the larger RCTs reported (200 patients), Griffiths and colleagues[32] reported significant improvements in HRQOL 1 year after a 6-week pulmonary rehabilitation program. Troosters and colleagues[33] reported sustained improvement in HRQOL over 18 months after patients participated in a 6-month outpatient pulmonary rehabilitation program compared with the decline observed in the control group. The study reported by Green and colleagues[34] reported improvement in HRQOL after pulmonary rehabilitation and found that improvements after a 7-week intervention were greater than those after 4 weeks of pulmonary rehabilitation. Strijbos and colleagues[35] reported significant improvement in reported well-being after pulmonary rehabilitation that was maintained over 18 months in rehabilitation-treated subjects, while most patients in the control group felt unchanged or worse. Foglio and colleagues[36] reported sustained improvements in HRQOL up to 2 years after pulmonary rehabilitation. In a study of early pulmonary rehabilitation after hospital discharge for an exacerbation of COPD, Man and colleagues reported significant improvements in HRQOL measures. Finnerty and colleagues[37] reported marked improvements in HRQOL after pulmonary rehabilitation that persisted for 6 months. Similar findings were reported by Bendstrup and colleagues.[38] In the study reported by Wedzicha and colleagues,[39] which stratified patients according to baseline dyspnea, improvement in HRQOL after pulmonary rehabilitation was observed in patients with moderate dyspnea (Medical Research Council [MRC] score, 3 or 4) but not in control subjects or patients with severe baseline dyspnea (MRC score, 5). The study by Ries and colleagues[40] evaluated a maintenance program after pulmonary rehabilitation. However, observational results after pulmonary rehabilitation that had been administered to all patients before randomization demonstrated consistent improvements in several different measures of both general and disease-specific measures of HRQOL. Guell and colleagues[41] reported significant improvement in HRQOL that persisted, although diminished, for up to 2 years of follow-up after the pulmonary rehabilitation intervention.

Of the studies reported in Table 3 , only one small study by White and colleagues[42] reported only modest improvements in measured HRQOL that did not consistently reach statistically or clinically significant levels. In addition to the studies reported in Table 3 , which generally were performed in single specialized centers, two observational studies[43,44] provide strong evidence of the effectiveness of pulmonary rehabilitation as routinely practiced in clinical centers. Although neither of these studies[43,44] was an RCT, they provide important information regarding the generalizability of the practice of pulmonary rehabilitation beyond specialized centers and as currently practiced in the general medical community in the United States. A multicenter evaluation of pulmonary rehabilitation in 522 patients in nine centers throughout California[43] reported consistent improvements in symptoms of dyspnea and HRQOL after pulmonary rehabilitation. Similar findings were reported in a multicenter observational study in Connecticut.[44] In this study, significant improvement was reported in the pulmonary functional status scale in 164 patients in 10 centers and in the CRDQ in 60 patients in 3 centers. Also, in the National Emphysema Treatment Trial (NETT),[26] a randomized study that evaluated lung volume reduction surgery in 1,218 patients with severe emphysema, all subjects were required to complete a pulmonary rehabilitation program as part of the eligibility requirements before randomization. Pulmonary rehabilitation was conducted at the 17 NETT centers as well as at 539 satellite centers throughout the United States. Observational results demonstrated significant improvements in measures of exercise tolerance, dyspnea, and HRQOL after rehabilitation that were quite comparable among the specialized NETT centers and the largely community-based satellite centers.

3. Pulmonary rehabilitation improves HRQOL in patients with COPD. Grade of recommendation, 1A

Regarding changes in health-care utilization resulting from pulmonary rehabilitation, the previous panel concluded that there was B level strength of evidence supporting the recommendation that "pulmonary rehabilitation has reduced the number of hospitalizations and the number of days of hospitalization for patients with COPD."

In the current review, some additional information is available about changes in health-care utilization after pulmonary rehabilitation. In the study by Griffiths and colleagues,[32] over 1 year of follow-up the number of patients admitted to the hospital was similar in both the pulmonary rehabilitation group and the control group (40 of 99 vs 41 of 101 patients); however, the number of days spent in the hospital was significantly lower in the rehabilitation patients (10.4 vs 21.0 days, respectively). In a subsequent cost-utility economic analysis of the results in this pulmonary rehabilitation trial, Griffiths and colleagues[45] found that the cost per quality-adjusted life-years indicated that pulmonary rehabilitation was, in fact, cost-effective and would likely result in financial benefits to the health-care system (quality-adjusted life-year is a measure of effectiveness that is commonly used in cost-effectiveness analyses, reflecting survival adjusted for quality of life, or the value that individuals place on expected years of life). In the trial reported by Foglio and colleagues,[36] results indicated a significant decrease in yearly hospitalizations and exacerbations > 2 years after pulmonary rehabilitation.

Goldstein and colleagues[46] conducted a cost analysis that was associated with an RCT of a 2-month inpatient pulmonary rehabilitation program (followed by 4 months of outpatient supervision) that produced statistically and clinically significant improvements in measures of HRQOL and exercise capacity. Although the cost analysis in this study was driven largely by the inpatient phase of the program and, as such, is not applicable to the large majority of outpatients programs, the authors found cost-effectiveness ratios for the CRDQ component measures to range from $19,011 to $35,142 (in Canadian dollars) per unit difference. Even with the added costs associated with the inpatient program, these cost/benefit ratios are within a range that has been typically considered to represent reasonable cost-effectiveness for other widely advocated health-care programs.[47]

In a small randomized trial of early pulmonary rehabilitation after hospitalization for acute exacerbation, Man and colleagues[48] reported a significant reduction in emergency department visits and a trend toward reduced numbers of hospital admissions and days spent in the hospital over the 3 months after hospital discharge in the pulmonary rehabilitation group compared to the usual-care group. Also, in a multicenter randomized trial of a self-management program of patients with severe COPD, Bourbeau and colleagues[49] reported a significant reduction in the numbers of hospital admissions and days spent in the hospital in the year following the intervention compared to the usual-care control group.

In a multicenter, observational evaluation[43] of the effectiveness of pulmonary rehabilitation in centers throughout California (not included in Table 3 ), self-reported measures of health-care utilization were found to decrease substantially over 18 months of observation after the rehabilitation intervention. In the 3-month period prior to pulmonary rehabilitation, 522 patients reported 1,357 hospital days (2.4 per patient), 209 urgent care visits (0.4 per patient), 2,297 physician office visits (4.4 per patient), and 1,514 telephone calls to physicians (2.7 per patient). Over the 18 months after rehabilitation, the average per patient reported health-care utilization (in the past 3 months) was reduced approximately 60% for hospital days, 40% for urgent care visits, 25% for physician office visits, and 30% for telephone calls. It should be recognized that the results of an observational, noncontrolled study like this may be influenced by the selection of patients for pulmonary rehabilitation shortly after an exacerbation or episode of increased health-care utilization.

4. Pulmonary rehabilitation reduces the number of hospital days and other measures of health-care utilization in patients with COPD. Grade of recommendation, 2B

5. Pulmonary rehabilitation is cost-effective in patients with COPD. Grade of recommendation, 2C

The previous panel concluded that there was little evidence (strength of evidence, C) regarding survival after pulmonary rehabilitation and made the recommendation that "pulmonary rehabilitation may improve survival in patients with COPD." Only one RCT[50] of pulmonary rehabilitation was included in the previous review. In that study of 119 patients, Ries and colleagues[50] reported 11% higher survival over 6 years after comprehensive pulmonary rehabilitation (67%) compared with an education control group (56%). This difference was not statistically significant. Other evidence for improved survival was derived from nonrandomized and observational studies. This lack of evidence does not necessarily indicate that pulmonary rehabilitation has no effect on survival, but in order to be reasonably powered to detect an effect of this magnitude the sample size would have to be a magnitude larger than those found in existing studies. The timed walk distance and MRC-rated dyspnea do improve with pulmonary rehabilitation, and these variables are correlated with survival in patients with COPD.

In the current review, few additional data were found regarding the effect of pulmonary rehabilitation on survival. Similar to previous published studies, the trial reported by Griffiths and colleagues[32] that followed 200 patients over 1 year found fewer deaths in the rehabilitation group (6 of 99 patients) compared with the control group (12 of 101 patients).

6. There is insufficient evidence to determine whether pulmonary rehabilitation improves survival in patients with COPD. No recommendation is provided.

Regarding psychosocial outcomes of pulmonary rehabilitation, the previous panel concluded that "scientific evidence was lacking" (strength of evidence, C). Reviews of the research literature pertaining to psychosocial outcomes of pulmonary rehabilitation programs indicate that comprehensive pulmonary rehabilitation is generally associated with enhanced psychological well-being (ie, reduced distress) and improved quality of life.[51,52] In addition, it has been found that increased self-efficacy associated with exercise may mediate the effect of exercise rehabilitation on quality of life.[53] Other positive psychosocial outcomes of exercise rehabilitation include improved cognitive function,[54,55,56] reduced symptoms of anxiety[32,55] and depression,[32] and improved patient perceptions of positive consequences of the illness.[57]

In the current review of randomized studies, Griffiths and colleagues[32] reported reduced symptoms of anxiety and depression following a 6-week pulmonary rehabilitation program, with symptoms of depression remaining significantly reduced at the 12-month follow-up. Emery and colleagues[55] found reduced anxiety and improved cognitive function following a 10-week pulmonary rehabilitation intervention. In a study of 164 patients participating in pulmonary rehabilitation prior to being randomly assigned to a long-term follow-up intervention, Ries and colleagues[40] observed significant improvements in measures of depression and self-efficacy for walking immediately following the 8-week pulmonary rehabilitation program.

7. There are psychosocial benefits from comprehensive pulmonary rehabilitation programs in patients with COPD. Grade of recommendation, 2B

The formal component of most pulmonary rehabilitation programs is of relatively short duration, usually ranging from 6 to 12 weeks. Regarding the issue of long-term benefits following the short-term intervention, the previous panel did not specifically address this topic but recommended it as an important area for future research. Since that time, additional important studies have addressed this topic. The next section discusses the issue of the duration of pulmonary rehabilitation treatment (ie, beyond 12 weeks).

Several clinical trials of 6 to 12 weeks of comprehensive pulmonary rehabilitation that have followed patients over a longer term have found that benefits typically persist for about 12 to 18 months after the intervention but gradually wane thereafter. In many ways, this is surprising given the severity of illness for many of these patients with chronic lung disease and the complex set of behaviors incorporated into pulmonary rehabilitation (eg, exercise training, breathing control techniques, complex treatment regimens with medications, use of supplemental oxygen, and relaxation or panic control techniques). More recent clinical trials substantiate these findings ( Table 4 ).

Griffiths and colleagues[32] reported improvements in measures of exercise tolerance, HRQOL, anxiety, and depression after pulmonary rehabilitation that remained significant but declined gradually over 1 year of follow-up. The study reported by Wijkstra and colleagues[58] evaluated the effects of weekly vs monthly follow-up over the 18 months after pulmonary rehabilitation in a small sample of patients with COPD (n = 36). They reported no long-term improvement in exercise tolerance in the two experimental groups, although this was better than the decline observed in the control group. There were, however, more sustained improvements in dyspnea. Engstrom and colleagues[59] reported sustained improvement in exercise tolerance at 12 months after pulmonary rehabilitation with minimal improvements in either a general or disease-specific measure of HRQOL (although there was a trend for worsening HRQOL in the control group). Strijbos and colleagues[35] reported significant improvement in reported well-being after pulmonary rehabilitation that was maintained over 18 months (compared to most control subjects who reported being unchanged or worse). The study reported by Guell and colleagues[41] also found persistent, but diminished, benefits in measures of exercise tolerance, dyspnea, and HRQOL over the 2 years of follow-up after pulmonary rehabilitation.

The study reported by Ries and colleagues[40] examined the effects of a telephone-based maintenance program for 1 year after a short-term rehabilitation intervention. The experimental effects of the maintenance program are discussed in a subsequent section on postrehabilitation maintenance. However, as an observational study, it is notable that the control group (without postprogram maintenance) demonstrated a progressive decline in benefits over 2 years of follow-up. Another multicenter observational evaluation of the effectiveness of pulmonary rehabilitation in centers throughout California (not included in Table 3 )[43] found that improvements in symptoms of dyspnea, HRQOL, and indexes of health-care utilization declined over 18 months but still remained above baseline levels.

8. Six to twelve weeks of pulmonary rehabilitation produces benefits in several outcomes that decline gradually over 12 to 18 months. Grade of recommendation, 1A. Some benefits, such as HRQOL, remain above control levels at 12 to 18 months. Grade of recommendation, 1C


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.