Effects of a Pulmonary Rehabilitation Program With Balance Training on Patients With COPD

PURPOSE: Patients with chronic obstructive pulmonary disease (COPD) have balance impairments. However, pulmonary rehabilitation (PR) is associated with only minor improvements in functional balance. Therefore, there is a need to explore the role of balance training within PR. This study aimed at assessing the effects of a PR program, with a specific component of balance training, on functional balance of patients with COPD. METHODS: Outpatients with COPD (N = 22, age = 68.0 ± 11.8 years; forced expiratory volume in 1 second = 72.2 ± 22.3% predicted) participated in a 12-week PR program including exercise training and psychosocial support and education. Exercise training sessions comprised endurance, strength, and a specific component of balance training. The Timed Up and Go (TUG) test was used to assess functional balance before and after the PR. Health-related quality of life (St George's Respiratory Questionnaire), quadriceps muscle strength (10 repetition maximum), and exercise tolerance (6-minute walk test) were also assessed. RESULTS: Patients demonstrated significant improvements in TUG scores after PR (mean change −1.7 ± 1.4 seconds; P = .001; effect size = 1.249). Before PR, 9 (41%) participants and after PR only 1 (4.5%) participant had a TUG performance worse than the average performance of age-matched healthy peers (P = .008). The St George's Respiratory Questionnaire symptoms score (P = .012), quadriceps muscle strength (P = .001), and exercise tolerance (P = .001) were also improved. CONCLUSIONS: Pulmonary rehabilitation with a specific component of balance training had a large effect on functional balance in patients with COPD. Findings highlight the value of including balance training in PR programs. Further research is needed to determine the optimal intervention to improve balance and its specific components among patients with COPD.

■ PURPOSE: Patients with chronic obstructive pulmonary disease (COPD) have balance impairments. However, pulmonary rehabilitation (PR) is associated with only minor improvements in functional balance. Therefore, there is a need to explore the role of balance training within PR. This study aimed at assessing the effects of a PR program, with a specific component of balance training, on functional balance of patients with COPD.
■ METHODS: Outpatients with COPD (N = 22, age = 68.0 ± 11.8 years; forced expiratory volume in 1 second = 72.2 ± 22.3% predicted) participated in a 12-week PR program including exercise training and psychosocial support and education. Exercise training sessions comprised endurance, strength, and a specific component of balance training. The Timed Up and Go (TUG) test was used to assess functional balance before and after the PR. Health-related quality of life (St George's Respiratory Questionnaire), quadriceps muscle strength (10 repetition maximum), and exercise tolerance (6-minute walk test) were also assessed.
■ CONCLUSIONS: Pulmonary rehabilitation with a specific component of balance training had a large effect on functional balance in patients with COPD. Findings highlight the value of including balance training in PR programs. Further research is needed to determine the optimal intervention to improve balance and its specific components among patients with COPD.
intervention for COPD, which includes exercise training and psychosocial support and education for patients to achieve maximal independence and function in the community. 7 However, this recommended standard of care is mainly focused on training peripheral muscles through endurance and strength training and does not include a specific component of balance training. Although exercise has been shown to improve balance and decrease fall risk in older adults, 8 when the effect of PR on balance in patients with COPD was studied, only minor improvements on balance were found. 9 The primary aim of this study was to assess the effects of a PR program, including a specific component of balance training, on functional balance of patients with COPD.

Data Collection
Data on sociodemographic, fall history, and lung function were obtained to characterize the sample. Fall history was assessed with a standardized question "Have you had any falls in the last 12 months?" Lung function was assessed with spirometry and the COPD grade was determined. 10 To assess the effects of the PR program, activities limitation resulting from breathlessness and health-related quality of life (HRQOL) were collected, followed by the assessment of quadriceps muscle strength, exercise tolerance, and functional balance. All questionnaires and tests were administered before and after the PR program.

Intervention
The PR program was conducted for 12 weeks and included 2 main components-exercise training (3 times a week for 60 minutes each session) and psychosocial support and education (once a week for 90 minutes). The exercise training component comprised the following: 1. A warm-up of 5 to 10 minutes including range of motion, stretching, low-intensity aerobic exercises, and breathing techniques, such as pursed lips breathing, body positions, diaphragmatic breathing, and airway clearance techniques. 2. Endurance training (walking) for 20 minutes at 60% to 80% of the average speed achieved during the 6-minute walk test (6MWT). Training intensity was adjusted according to patient symptoms of fatigue and dyspnea (Modifi ed Borg Scale). 7 3. Strength training (15 minutes) included 7 exercises (2 sets of 10 repetitions) of the major upper and lower limb muscle groups using elastic bands, free weights, and ankle weights. The amount of weight was between 50% and 85% of the 10 repetition maximum (10-RM). 7 The training load was increased when the patient performed 2 additional repetitions with a given load on 2 consecutive sessions. The training intensity was adjusted as described in the endurance training. 4. Balance training for 5 minutes consisted of static and dynamic exercises using mainly upright positions. Balance exercises were organized in the following 4 levels 6 : (1) postures that gradually reduced the base of support; (2) dynamic movements that perturbed the center of gravity; (3) stressing postural muscle groups; and (4) dynamic movements while performing a secondary task individually or in groups, with a progressively narrowed base of support. In this component, patients were also trained on how to lie down and get up from the fl oor. 5. A cool down of 10 minutes included similar exercises to the warm-up.
The psychosocial support and education component provided information about COPD; healthy lifestyles; falls and fall prevention strategies; emotion-management strategies; problem solving techniques, and community resources.

Functional Balance
The primary outcome measure was functional balance, defined as the ability to maintain equilibrium in dynamic situations required for daily activities. The Timed Up and Go (TUG) test was used to assess functional balance, as recent research has suggested that assessment of balance under multitask conditions may be a more sensitive indicator of balance problems and falls than assessment of balance in a single-task context. 11 The test requires the patient to rise from a standard chair, walk 3 m, turn around, walk back to the chair, and sit down. Patients were instructed to www.jcrpjournal.com to a varicose vein surgery (n = 1), and no reasons given (n = 4). Therefore, 22 participants (13 males, age 68.0 ± 11.8 years; forced expiratory volume in 1 second = 72.2 ± 22.3% predicted) completed the study. Regarding COPD severity, 9 patients had mild, 8 moderate, and 5 severe to very severe COPD. Almost half of the participants (n = 10; 45.5%) reported at least 1 fall in the preceding year. Table 1 provides the sociodemographic and clinical characteristics of the participants. A comparison of measures before and after PR is presented in Table 2 .
After the PR program, the activities limitation resulting from breathlessness was significantly reduced (mean change = − 0.5; P = .025; ES = 0.466). The SGRQ total score did not change significantly (mean change = − 4.3 units; P = .089; ES = 0.391), although the change reached clinically relevance. 13 There was a significant improvement of the SGRQ symptoms score (mean change = − 12. Regarding the effect of PR on functional balance, the TUG score showed a significant improvement ( Figure 1 ) (mean change = − 1.7 seconds; P = .001; ES = 1.249). Before PR, approximately half of the sample (n = 9; 40.9%) and after PR only 1 participant (4.5%) ( P = .008) had a TUG score worse than the average of age-matched healthy peers. 12 There were no statistically significant correlations between change in functional balance and change in SGRQ total score ( r = 0.171), SGRQ symptoms score ( r = − 0.018), quadriceps muscle strength ( r = 0.207), or 6MWT distance ( r = − 0.060). walk quickly, but as safely as possible. Two TUG tests were performed and the best performance was recorded. This test has been shown to predict falls in community-dwelling older adults, 11 and normative reference scores are available. 12

Secondary Outcome Measures
Activities limitation resulting from breathlessness was assessed by patients selecting the statement from the modified British Medical Research Council questionnaire that best described their limitation. Healthrelated quality of life was assessed using the St George's Respiratory Questionnaire (SGRQ), and a change of 4 units was considered clinically relevant. 13 Quadriceps muscle strength was assessed using the 10-RM with ankle weights. Exercise tolerance was measured using the 6MWT. Two tests were performed and the best performance was recorded.

Statistical Analysis
Descriptive statistics were used to describe the sample. Measurements collected before and after the PR program were compared using paired t tests for normally distributed data, the Wilcoxon signed-rank test for nonnormally distributed data and ordinal data, and the McNemar test for dichotomous categorical data. Correlations between functional balance changes and changes in SGRQ total score, SGRQ symptoms score, quadriceps muscle strength, and 6MWT were analyzed with the Pearson or Spearman correlation coefficient. All statistical analyzes were performed using PASW Statistics version 18.0 for Windows (SPSS Inc, Chicago, IL), and graphs were created using GraphPad Software 5.0 (La Jolla, CA).
Effect sizes (ES) were also determined to explore the degree to which the PR program was responsible for changes in activities limitation resulting from breathlessness, HRQOL, muscle strength, exercise tolerance, and functional balance. ESs were calculated as the mean difference in values before and after PR divided by the mean standard deviation (ie, mean ± SD before and after PR) using G*Power Software 3.1.1 (Kiel, Germany). The magnitudes of the ES were interpreted as ≥ 0.2 small effect, ≥ 0.5 medium effect, and ≥ 0.8 large effect. 14 The level of significance considered was 0.05.

RESULTS
A total of 34 patients enrolled in the study. However, 12 (35.3%) dropped out due to professional reasons (n = 2), relocation (n = 1), deterioration of health status due to muscular impairments (n = 2), respiratory exacerbation (n = 2), the person was submitted contribute to greater functional balance improvements than standard PR. However, this comparison must be considered with caution because the 2 studies included patients with different grades of the disease (forced expiratory volume in 1 second = 72.2 ± 22.3% predicted in the present study vs 46.3 ± 22.3% predicted in Beauchamp et al 9 ) and employed different TUG protocols. Even with the uncertainty that PR programming with balance training improves patient functional balance to a greater extent than standard PR, our findings clearly demonstrate that most participants with a TUG score worse than the average of their age-matched healthy peers at baseline 12 achieve a normal TUG performance after the program. As TUG assesses balance under multitask conditions and this is similar to the performance of activities of daily living, 11 this finding suggests a lessening of overall fall risk in these patients.
Currently, it is not possible to determine whether the TUG score improvement found in this study has clinical relevance, as TUG minimal detectable change (MDC) has not been determined for COPD. The MDC is the smallest amount of difference in individual scores between 2 points in time that represents a true statistical change. Therefore, we cannot firmly state that including a 5-minute component of balance training in PR programs is sufficient to promote clinically relevant functional balance improvements in patients with COPD. Nevertheless, in our study, balance training was implemented with the recommended frequency (3 times per week) 15 and incorporated static and dynamic exercises similar to everyday activities. Leung et al 16 assessed the effect of Short-form Sunstyle Tai Chi compared with no exercise training on balance in patients with COPD. The intervention was 12 weeks, twice a week, for 60 minutes. It was

DISCUSSION
The main finding of this study was that a PR program with a specific component of balance training had a large effect on functional balance improvements in patients with COPD.
Similar to other studies, which have assessed the effects of PR on patients with COPD, significant improvements on activities limitation resulting from breathlessness, HRQOL, muscle strength, and exercise tolerance were found. This shows that the implemented program was, overall, at least as effective as it has been previously reported. Regarding the effect of the program on patient functional balance, a change of − 1.7 ± 1.4 seconds in TUG score was found. This result is slightly higher than the value obtained by Beauchamp et al 9   observed that the Tai Chi group had a significant improvement in functional reach distance, compared with the control group (ES = 1.45). 16 The distinctly different dose trainings in our study and Leung et al 6 demonstrate the urgent need for guidelines addressing the optimal dose-response ratio of balance training.

Limitations and Future Work
The dropout rate (35.3%) found in this study may be interpreted as an attrition bias. Nevertheless, it is similar to the dropout rates reported in other PR programs, between 20% and 40%. 17 , 18 The absence of a control group can also be seen as a limitation of this study and should be considered in future studies. As the intervention consisted of a multicomponent PR program, the TUG score results cannot be interpreted as solely the result of the balance training component. Thus, while continuing to address the clinical relevance of the TUG score and the effectiveness of a multicomponent PR program, future studies should be aimed at studying the specific components of balance, such as posture and muscle strength, in reducing fall risk in this population. -Acknowledgments-