Abstract
BACKGROUND: The aim of this study was to clarify the effectiveness of pulmonary rehabilitation in patients after exacerbations of COPD and to explore the initiation timing of pulmonary rehabilitation.
METHODS: Systematic review and meta-analysis were performed to assess the effects of pulmonary rehabilitation in subjects with exacerbations of COPD on mortality and readmission compared with usual care. We searched for studies published up to October 2020 in MEDLINE, Embase, Cochrane Library, and other sources. Risk of bias was assessed for the randomization process, deviations from intended interventions, missing outcome data, outcome measurements, and selection of the reported result using the Risk of Bias 2 tool. We pooled mortality and readmission data and performed comparisons between pulmonary rehabilitation and usual care. The subgroup analysis compared pulmonary rehabilitation at different start times (early: ≤ 1 week from admission; and late: > 1 week from admission).
RESULTS: We identified 10 randomized trials (1,056 participants). Our meta-analysis showed a clinically relevant reduction in readmission up to 3–6 months after pulmonary rehabilitation in both early group (4 trials, 190 subjects; risk ratio [RR] 0.58, [95% CI 0.34–0.99]) and late group (3 trials, 281 subjects; RR 0.48, [95% CI 0.32–0.71]). However, pulmonary rehabilitation had no significant effect on mortality 1 y later compared with usual care (4 trials, 765 subjects; RR 1.27, [95% CI 0.91–1.79]).
CONCLUSIONS: Pulmonary rehabilitation showed short-term effects for subjects with exacerbations of COPD even if initiated within 1 week; however, further study is required to determine its long-term effects.
Introduction
Pulmonary rehabilitation is important for managing disease in patients with COPD because it reduces symptoms, increases exercise capacity, improves survival, and reduces hospitalization rates in patients with COPD.1-3 However, many previous studies have included only subjects with stable COPD, and the effects of pulmonary rehabilitation initiated early after hospital admission for patients with COPD remain unknown. After exacerbations of COPD, meta-analysis has shown that rehabilitation initiation within 4 weeks of hospitalization is effective4; but there are no clear results for earlier initiation, particularly during the hospitalization period. In recent years, a large randomized controlled trial (RCT) has shown that starting early rehabilitation within 48 h of hospitalization is associated with higher mortality after 1 y.5 Even the global guidelines regarding the effects of the timing of pulmonary rehabilitation initiation differ. The 2011 guidelines of the National Institute for Health and Care Excellence recommend that rehabilitation be initiated for patients hospitalized with COPD exacerbations.6 However, as per the 2017 guidelines of the European Respiratory Society and the American Thoracic Society, the result of the RCT5 was considered important; and thus, they suggested that pulmonary rehabilitation should not be initiated for patients hospitalized with COPD exacerbations.7 As such, there is no clear recommendation on when to start rehabilitation. In particular, a history of exacerbations has been reported to increase the risk of hospitalization, which is an important prognostic factor for decreased survival.8 COPD exacerbation has also been reported to affect the progression of emphysema.9 As such, prevention of exacerbations of COPD is an important treatment strategy; and thus, whether pulmonary rehabilitation is associated with readmission and mortality should be examined. In the meta-analysis conducted by Puhan et al,10 while rehabilitation for subjects with COPD after exacerbation showed a moderate evidence of reducing readmissions, the results obtained were heterogeneous. This meta-analysis did not consider differences in the initiation timing, and the impact of initiation time remained unclear.
Therefore, the current systematic review and meta-analysis were conducted to evaluate the effectiveness of pulmonary rehabilitation in patients after exacerbations of COPD. Then, a subgroup analysis was performed to explore the initiation timing of pulmonary rehabilitation. This study evaluated the effects of rehabilitation on mortality and readmission in patients with COPD.
Methods
Search Methods for Identifying Studies
We performed this study according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Statement for meta-analyses and systematic reviews.11 We searched for studies published up to October 2020 in the following electronic databases: Cochrane Central Register of Controlled Trials, MEDLINE, Embase, CINAHL, PubMed, and OvidSP. By carrying out discussions with an academic librarian, we finalized the search strategy and modified it appropriately for each database. The search strategy comprised a combination of free-text words; words used in titles; and/or abstracts and medical subject headings, including “lung diseases obstructive OR chronic obstructive lung disease OR chronic obstructive pulmonary disease,” “rehabilitation OR exercise OR physical endurance OR muscle training OR resistance training OR exercise therapy OR aerobic training OR endurance training,” and “randomized, controlled trial OR randomized” (Appendix A, see related supplemental material at http://www.rcjournal.com). The search strategy was limited to the inclusion of only peer-reviewed publications that involved human participants. In order to ensure that the search strategy covered all the topic-related studies, we included different terms and spellings that are used worldwide. In addition, language restrictions were not applied.
Inclusion and Exclusion Criteria
In this review, we included studies that evaluated the effectiveness of pulmonary rehabilitation on patient prognosis after exacerbation of COPD. The primary outcome was readmission for up to 3–6 months. First, we evaluated readmissions at 3 months; and if 3-month data were not available, we added readmissions at 6 months. The secondary outcomes included readmission within 1 y and mortality 1 y later.
The studies that met all of the following criteria were considered eligible: (1) RCT study design, (2) study involved subjects after experiencing exacerbations of COPD, (3) study compared pulmonary rehabilitation and usual care, (4) study reported on at least one of the abovementioned outcome measures, (5) pulmonary rehabilitation was initiated during admission for an exacerbation or within 4 weeks of hospital discharge after exacerbation, and (6) pulmonary rehabilitation was provided for more than 1 month. All published and unpublished RCTs were included. Animal and in vitro studies were excluded.
Data Extraction
To determine relevance, titles and abstracts that contained key terms were independently screened by 2 investigators (MS and SK) based on the above inclusion criteria. If a decision could not be made based on the title and abstract alone, it was left for screening. The corresponding full-text articles were subsequently retrieved to assess the eligibility of the articles remaining in the screening, which then determined their inclusion in the study. Disagreements, if any, were resolved by consulting and discussing with a third author (SY). The following data were recorded from each study: the intervention details and comparison method, subjects (number) that were randomly assigned to each group as well as attrition, outcome data, first author’s name, publication year, blinding, subject age, intervention period and follow-up duration, hospital length of stay, pulmonary rehabilitation program completion rate, and adverse events. When data were insufficient, the authors were contacted and unpublished data were requested. Wherever possible, data were collected on an intention-to-treat basis.
Assessment of Risk of Bias in the Included Studies
The quality of each study was independently assessed by MS and SK using the Risk of Bias 2 tool as per the Cochrane Handbook for Systematic Reviews of Interventions.12 In cases of differing assessments, the third author (SY) was consulted for a final decision on the assessment.
Meta-Analysis and Subgroup Analysis
We divided subjects into 2 groups: the pulmonary rehabilitation group and the usual-care group. To explore the initiation timing of pulmonary rehabilitation, subgroup analysis was planned for the early (≤ 1 week from admission) and late (> 1 week from admission) groups based on the initiation timing of pulmonary rehabilitation. Data syntheses and analyses were performed using Review Manager version 5.3. For outcomes, pooled risk ratios (RRs) were calculated.
The Tau2, I2, and chi-square statistics were used to determine the statistical heterogeneity in each meta-analysis as per the Cochrane Handbook for Systematic Reviews of Interventions.13 Heterogeneity was regarded as substantial when I2 was > 50% and the P value for heterogeneity was < .05. To increase the validity of the test results, a sensitivity analysis was performed to determine whether a high risk of bias of some of the included studies affected the study results. When ≥ 10 studies were included in a meta-analysis, a funnel plot was created, and its asymmetry was visually examined to explore any publication bias. A high risk of bias was defined as follows: (1) insufficient random sequence generation, (2) insufficient allocation concealment, or (3) many high-risk domains.12
Results
According to our search strategy, 2,390 records were identified in the initial search.11 In addition, 2 other studies were found via hand searching.14,15 A total of 29 duplicate studies, 254 reviews, and 15 animal or in vitro studies were excluded. The full texts of the remaining 19 studies were retrieved, and finally, 10 RCTs met our inclusion criteria.5,14-22 Figure 1 shows the screening process and the reasons for excluding studies.
Quality and Bias Assessment in the Included Studies
Table 1 summarizes the risk of bias for each study. The risk of bias was high in 3 (30%) studies with deviations from intended interventions, one (10%) study with missing outcome data, and 3 (30%) studies with overall risk of bias.
Study Characteristics
The characteristics of the included RCTs and participant information are presented in Table 2. Five RCTs5,15,16,18,19 had started pulmonary rehabilitation ≤ 1 week from admission, whereas 5 RCTs14,17,20-22 had started pulmonary rehabilitation > 1 week from admission. A total of 1,056 participants were included in the 10 RCTs. In one RCT, the intervention groups received 4–8 weeks of pulmonary rehabilitation; in 2 RCTs, the intervention groups received 6 weeks of pulmonary rehabilitation; in one RCT, the intervention group received 7 weeks of pulmonary rehabilitation; and in the remaining RCTs, the intervention groups received 8 or more weeks of pulmonary rehabilitation.
Readmissions for up to 3–6 Months
Seven RCTs with 471 subjects were included in the analysis of readmission for up to 3–6 months. Participants who underwent pulmonary rehabilitation had a significantly lower risk of readmission for up to 3–6 months following the initial admission (RR 0.51, [95% CI 0.37–0.70]; participants = 471; studies = 7; Fig. 2). There was no significant heterogeneity (chi-square = 3.97, P = .68, I2 = 0%). In subjects who had received pulmonary rehabilitation, significant decreases in readmission for up to 3–6 months were found in both the early (RR 0.58, [95% CI 0.34–0.99]; participants = 190; studies = 4; I2 = 0%) and late subgroups (RR 0.48, [95% CI 0.32–0.71]; participants = 281; studies = 3; I2 = 0%).
Readmissions within 1 Year
Four RCTs with 765 subjects were included in the analysis of readmission within 1 y. Participants who underwent pulmonary rehabilitation had a significantly lower risk of readmission 1 y following the initial admission (RR 0.89, [95% CI 0.78–1.00]; participants = 765; studies = 4; I2 = 83%; Fig. 3). In subjects who had received pulmonary rehabilitation, significant decreases in readmissions within 1 y were found in the late subgroup (RR 0.70, [95% CI 0.58–0.85]; participants = 376; studies = 3; I2 = 66%). The early subgroup included only one RCT, and no significant difference was observed between the participants who received pulmonary rehabilitation and those who received usual care (RR 1.08, [95% CI 0.92–1.27]; participants = 389; study = 1).
Mortality 1 Year Later
Four RCTs with 765 subjects were included in the analysis of mortality 1 y later. The overall results showed that no significant difference was observed between the patients who received pulmonary rehabilitation and those who received usual care (RR 1.27, [95% CI 0.91–1.79]; participants = 765; studies = 4; I2 = 17%; Fig. 4). The substantial heterogeneity was explained by the subgroup analysis. There was a significant increase in the early subgroup (RR 1.56, [95% CI 1.04–2.33]; participants = 389; study = 1) but no significant difference in the late subgroup (RR 0.77, [95% CI 0.39–1.51]; participants = 376; studies = 3; I2 = 0%). Heterogeneity was lower after the subgroup analysis compared with the pre-subgroup analysis.
Sensitivity Analysis
The reliability of results was assessed using a sensitivity analysis. The meta-analysis was repeated without the inclusion of studies with many high-risk domains15 or few events,16,18,20,22 respectively. In the analysis of readmissions for up to 3–6 months, this omission changed the results such that there was no significant difference between pulmonary rehabilitation and usual care in the early subgroup (Figures B.1–B.3, see related supplemental material at http://www.rcjournal.com.).
Adverse Events
Adverse events in included studies were described in 3 articles, and no intervention-related adverse events were observed.
Discussion
The present study indicated that there was a significant decrease in readmission for up to 3–6 months and within 1 y in the participants who received pulmonary rehabilitation compared with those who received usual care. Furthermore, significant decreases in readmission for up to 3–6 months in pulmonary rehabilitation were found for both the early and late subgroups. The results for mortality 1 y later indicated that there were no significant differences between pulmonary rehabilitation and usual care. Based on the results of the present study, pulmonary rehabilitation had a short-term (3–6 months) beneficial effect in subjects with exacerbations of COPD even if it was initiated within 1 week of admission, similar to after 1 week from admission.
Short-Term Effects
The Global Initiative for Chronic Obstructive Lung Disease (GOLD) reports a high level of evidence indicating that pulmonary rehabilitation improves dyspnea, health status, and exercise tolerance in stable patients.1,23 Additionally, pulmonary rehabilitation after exacerbations of COPD has been noted, and the initiation of pulmonary rehabilitation within 3 months has been shown to be significantly associated with a reduced risk of death at 1 y.24 In a recent systematic review, although heterogeneity existed, pulmonary rehabilitation initiated within 3 weeks after discharge reduced the rate of readmission and mortality.10 Studies reported in GOLD also reported that pulmonary rehabilitation within 4 weeks of prior hospitalization reduced hospitalization among participants who experienced a recent exacerbation.1,4
The GOLD also states that the effect of pulmonary rehabilitation before discharge remains unclear. In the present meta-analysis, pulmonary rehabilitation has a short-term (3–6 months) effect on participants with exacerbations of COPD even if it is initiated within 1 week of admission. The meta-analysis about early mobilization among critically ill subjects reported that there were no adverse effects even if pulmonary rehabilitation was initiated within 1 week of ICU admission.25 This study newly indicated that short-term effects for subjects with exacerbations of COPD were observed in addition to no adverse effects, similar to the previous study.
The findings also indicated that there was no difference in the timing of rehabilitation initiation and that there were no short-term adverse effects from starting early. Several studies have compared differences in the timing of rehabilitation initiation in participants with COPD after exacerbation. Revitt et al26 compared a group that received rehabilitation within 4 weeks of discharge with a group that received rehabilitation starting at 7 weeks and found that the subjects who received pulmonary rehabilitation sooner after exacerbations of COPD showed better improvements than the delayed group. However, no firm conclusions could be drawn because of the significantly small sample number. In a study that initiated rehabilitation earlier, no significant differences were found by Puhan et al27 between early pulmonary rehabilitation, which was started within 2 weeks, and late pulmonary rehabilitation, which was started 6 months after randomization and in a stable state. However, their trial indicated that early rehabilitation may lead to a faster recovery of health-related quality of life after exacerbations compared with late rehabilitation. Furthermore, pulmonary rehabilitation within 1 week after discharge led to a faster improvement in physical performance than rehabilitation initiated later in the stable phase; however, it did not decrease the mortality rate or increase the time to hospital readmission.28 Thus, early rehabilitation has been shown to have a potential for faster recovery of physical function and quality of life, but its effectiveness on mortality and readmission is unclear.28,29 The reason for the difficulty in presenting clear results in these studies may be the small number of subjects. The integration of the results of multiple studies in the present meta-analysis indicates that pulmonary rehabilitation has a short-term (3–6 months) effect for subjects with exacerbations of COPD even if it is initiated within 1 week of admission.
Long-Term Effects
However, similar to the guidelines, no significant results were obtained for the long-term (1 y) effects of pulmonary rehabilitation. This result is due to the fact that there are few papers with a long-term follow-up period of 1 y. Although early pulmonary rehabilitation was shown to increase mortality at 1 y in one large-scale RCT,5 no other large RCTs have been reported on this topic. In the present meta-analysis, because few studies were included, especially in the early group, and only one RCT was included, the long-term effects remain unclear. In the meta-analysis on nonpharmacologic therapies in subjects with exacerbation of COPD, risk of readmission was similar in rehabilitation and control groups, but only one study was included in the analysis.30 RCTs in various populations are necessary to clarify the long-term effects of pulmonary rehabilitation for patients with exacerbations of COPD.
Limitations
This study has several limitations. First, the definition of the intervention was clinically heterogeneous. The training content was not uniform and included articles in which the exercising was unsupervised and for which the exercise content may not have been fully controlled. Second, the program completion rate for early pulmonary rehabilitation after COPD exacerbation has been reported to be associated with the risk of hospital readmissions,31 but we did not examine the program completion rate in this study. Third, in the sensitivity analysis, different results were shown when omissions were made, so care should be taken in interpreting the results. Fourth, the impact of different initiation timings is a secondary comparison. Since the present results suggest that starting rehabilitation early is not a bad choice, future RCTs comparing the early and late groups will be necessary. These limitations should be considered while interpreting the study results.
Conclusions
For participants with exacerbations of COPD, both early (within 1 week of admission) and late (after 1 week from admission) initiation of pulmonary rehabilitation reduced hospital readmission for up to 3–6 months. Further studies are necessary to confirm the long-term effects of pulmonary rehabilitation, especially to compare the effects of different initiation timings of pulmonary rehabilitation.
Acknowledgments
The authors thank the rehabilitation staff for their support in this study.
Footnotes
- Correspondence: Manaka Shibuya PT PhD, Department of Rehabilitation, Kitasato University Hospital, 1–15-1, Kitasato, Minami-ku, Sagamihara, Kanagawa 252–0375, Japan. E-mail: manaka.s{at}kitasato-u.ac.jp
The authors have declared no conflicts of interest.
The study was performed at Department of Rehabilitation, Kitasato University Hospital, 1–15-1, Kitasato, Minami-ku, Sagamihara, Kanagawa, Japan.
Supplementary material related to this paper is available at http://www.rcjournal.com.
- Copyright © 2022 by Daedalus Enterprises