Complementary Therapies in Medicine (2013) 21, 585—594
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevierhealth.com/journals/ctim
The sustaining effects of Tai chi Qigong on physiological health for COPD patients: A randomized controlled trial夽 Aileen W.K. Chan a,∗, Albert Lee b, Diana T.F. Lee a, Lorna K.P. Suen c, Wilson W.S. Tam b, S.Y. Chair a, Peter Griffiths d a
The Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong Special Administrative Region b The Jockey Club School of Public Health and Primary Care, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong Special Administrative Region c School of Nursing, The Hong Kong Polytechnic University, Hunghom, Hong Kong Special Administrative Region d Faculty of Health Sciences, University of Southampton, Building 67, Room E4015, Highfield Campus, Southampton SO17 1BJ, United Kingdom Available online 8 October 2013
KEYWORDS COPD; Exercise tolerance; Lung functions; Six-minute walk test; Tai chi Qigong
Summary Objectives: To evaluate the sustaining effects of Tai chi Qigong in improving the physiological health for COPD patients at sixth month. Design: A randomized controlled trial. Subjects were in three randomly assigned groups: Tai chi Qigong group, exercise group, and control group. Setting: The 206 subjects were recruited from five general outpatient clinics. Interventions: Tai chi Qigong group completed a 3-month Tai chi Qigong program. Exercise group practiced breathing and walking as an exercise. Control group received usual care. Main outcome measures: Primary outcomes included six-minute walking distance and lung functions. Secondary outcomes were dyspnea and fatigue levels, number of exacerbations and hospital admissions. Results: Tai chi Qigong group showed a steady improvement in exercise capacity (P < .001) from baseline to the sixth month. The mean walking distance increased from 298 to 349 meters (+17%). No significant changes were noted in the other two groups. Tai chi Qigong group also
夽 The Health and Health Services Research Fund (HHSRF), the Food and Health Bureau, Hong Kong SAR Government is acknowledged as the funding source of this study. ∗ Corresponding author. Tel.: +852 39434290; fax: +852 26035269. E-mail address: [email protected] (A.W.K. Chan).
0965-2299/$ — see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctim.2013.09.008
586
A.W.K. Chan et al. showed improvement in lung functions (P < .001). Mean forced expiratory volume in 1 s increased from .89 to .99 l (+11%). No significant change was noted in the exercise group. Deterioration was found in the control group, with mean volume decreased from .89 to .84 l (−5.67%). Significant decreased in the number of exacerbations was observed in the Tai chi Qigong group. No changes in dyspnea and fatigue levels were noted among the three groups. Conclusions: Tai chi Qigong has sustaining effects in improving the physiological health and is a useful and appropriate exercise for COPD patients. © 2013 Elsevier Ltd. All rights reserved.
Introduction Chronic obstructive pulmonary disease (COPD) ranks as the fourth leading cause of death worldwide; it has a rising trend in morbidity and mortality.1 Because of the chronic and progressive nature of the disease, most of the lung function impairments are permanent and cannot be reversed.2 Aiding this population to develop disease management skills is therefore necessary to promote their physical health. Living with COPD imposes a certain degree of physiological challenge to the individuals, such as decline of lung functions and reduced exercise capacity.3,4 These challenges may lead to poor adjustment to their life. Symptoms of COPD included dyspnea and/or fatigue that limit exercise tolerance in most COPD patients. Poor motivation is also associated with exercise intolerance.5 Inactivity can lead to physical deconditioning that further limits exercise tolerance. Performing regular physical exercise can therefore be considered as a possible means to enhance physiological health among COPD patients. As Tai chi Qigong (TCQ) has characteristics that are particularly suitable for older people with chronic illnesses, the implementation of a TCQ program for the COPD patients could be of value. TCQ is a whole-body exercise that integrates breathing with harmonious movement of body training, which can benefit ventilation and improve physical capacity.6 However, limited existing systematic reviews on the effectiveness of Tai chi in enhancement of physical health in COPD patients was found.7 In view of the specific health needs among the COPD patients, our team has conducted a randomized controlled trial (RCT) to investigate the therapeutic effect of TCQ on COPD patients aimed to bridge the present knowledge gap.8 The TCQ forms were tailor-made for COPD patients. Participants were required to coordinate their breathing with the movement motions. After the 3-month intervention period, improvements in lung functions and exercise tolerance were noted in the TCQ group when compared to the exercise group and the control group. The improved exercise capacity was clinically significant,9 but the lung functions did not reached the clinically significant level.10 Since TCQ is a complex exercise routine that requires considerable practice to attain proficiency, it is possible that the improvements would increase over a longer study period. At the same time, a longer follow-up on the COPD subjects would be more informative, especially because deterioration in lung functions was identified as early as at 3 months in control group. This may become more marked over a longer period, whilst some improvement or maintenance was exhibited by the TCQ group. In addition, majority of the TCQ subjects (92%) showed interest in continuing the TCQ exercises. This study
therefore looked at longer period of follow-up assessment at the sixth month. It would be beneficial to assess clients’ long-term compliance and to monitor the effectiveness of TCQ practice in the progress of the disease.
Methods Study design and setting This was a randomized controlled study. Ethical approval for the 6-month follow-up evaluation was obtained from Joint the Chinese University of Hong Kong-New Territories East Cluster Clinical Research Ethics Committee. Informed consent was also obtained from every eligible participant. The study was carried out at five general outpatient clinics.
Eligibility criteria for participants Selection criteria were subjects who had clinically diagnosed with COPD and were ambulatory. Subjects were excluded if they suffered from severe cognitive impairment, symptomatic ischemic heart disease, or had practiced TCQ within a year.
Sample size and randomization Sample size calculation was based on previous findings of TCQ exercise on exercise capacity using six-minute walk test.11 According to Cohen,12 for ˛ = .05, a medium effect size of repeated measures analysis of variance test with three arms, 52 subjects per group can achieve a power of .80. To cover for the potential attrition rate of 25%,13 206 subjects were recruited and were randomly assigned to one of the three groups, TCQ group (n = 70), exercise group (n = 69), or control group (n = 67). Random allocation was decided by a computer-generated randomizer.14 The randomization avoids the shortcoming of yielding a highly disparate sample size in the study groups. The strength of this method is its ability to generate unpredictable sample allocation sequence. Stratification randomization was not used in this study.
Intervention TCQ program The experimental intervention was a TCQ program aimed to help COPD patients to develop the skills of TCQ exercise and to incorporate such exercise into their activity daily livings (ADLs). The program was conducted twice a week,
The sustaining effects of Tai chi Qigong
587
and each session lasted for 1 h. The TCQ program consisted of a 13-movement of TCQ, which were selected and modified from the 18 movements of Taiji Qigong.15 This modified 13-form TCQ was designed to allow patients to master the skills in a shorter period. The 13-form TCQ was tailor-made for COPD patients, it emphasized on elements of breathing, range of motion on upper limbs for enhancing lung expansion, and overall coordination. Participants were required to coordinate their breathing with the movement motions. An audiovisual DVD and TCQ pictures were given to each participant to facilitate daily self-practice at home. A selfrecording diary was given to the participants to record the frequency of their self-practicing TCQ at home.
Secondary outcome measures Secondary outcome measures included the Borg scale of dyspnea and fatigue levels, and the cutaneous oxygen saturation (SaO2 ) measured at rest and post 6MWT. The Borg scale is a self-rated scale which has a range of 0—10, where ‘‘0’’ indicates no dyspnea and no fatigue, and ‘‘10’’ indicates the maximal dyspnea and the maximal fatigue.19 SaO2 was measured by a noninvasive pulse oximetry to estimate the arterial oxyhemoglobin saturation. The number of exacerbations and number of hospital admissions in the past 3 months were also recorded. Acute exacerbation of COPD is defined as sustained worsening of symptoms and is acute in onset and often necessitates a change in medication.20
Breathing and walking exercise The goal of breathing exercise was to teach participants to relax the accessory muscles of breathing. Subjects were taught breathing techniques coordinated with walking exercise. The use of pursed-lip breathing (PLB) and diaphragmatic breathing (DB) techniques were taught by a qualified nurse. PLB can slow exhalatory flow and enable alveolar emptying, thereby creating mechanical advantage for the subsequent inhalation.16 DB is breathing in assisted by contracting the diaphragm and breathing out slowly while the diaphragm is relaxed. This maneuver increases lung capacity by lowering the diaphragm and allows air to reach the bottom of the lungs. The reduced respiratory rate leads to increased tidal volume and may improve gas exchange in the alveoli.17 Return demonstration of breathing techniques was performed to ensure proper skills practice of the participants. Participants were advised to self-practice breathing exercise and coordinated with self-paced walking exercise for one hour every day. Leaflets with pictures and instructions were given to the subjects to facilitate daily self-practice. A diary was also given to the participants for recording the frequency of their self-practice.
Data collection and statistical analysis Data collections were performed at 6-month (T3), in addition to baseline (T1) and post intervention at 3-month (T2). Data analyses were conducted using SPSS version 20.0. Repeated measures analyses of covariance (ANCOVA) were used to examine the outcome measures. A P-value of .05 was used as the level of statistical significance. An intention-totreat analysis was applied to carry forward last observation for the missing data.
Control group Participants in the control group continued to receive their usual medical treatment. They were advised to maintain their routine activities. No extra exercise was recommended. During the study period, all subjects continued their prescribed medical treatments. In order to enhance the internal validity of the study findings, participants in the exercise and control groups were arranged to join non-health-related social activities weekly. The purpose of maintaining regular gatherings for all participants was to balance the emotional effect of weekly gatherings of the TCQ group during the process of TCQ training. Primary outcome measures Primary outcome measures included exercise capacity and lung functions. Exercise capacity was assessed by six-minute walk test (6MWT). Subjects were encouraged to walk as far as possible for 6 min,18 and the total distance was recorded. Lung functions were measured via Spirometry. It required the patient to forcefully expel air from a point of maximal inspiration to a point of maximal expiration. Spirometry measures the forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1 ).
Results Two hundred and six eligible subjects were recruited in the study. The participants had an age range of 55—88 years, with a mean age (standard deviation) of 72.9 (7.7). The mean duration of COPD was 11.1 (9.6) years. The spirometry result indicated that the mean FEV1 percent predicted normal value was 53.8% (SD, 23.7). The most common coexisting disease was hypertension, followed by heart disease and diabetes mellitus. The number of medications taking among the three groups was comparable, ranged from 0 to 12, with a mean of 4.9 (2.1), which included medications for respiratory, cardiovascular, endocrine disorders. Of the 206 subjects, 187 (90.8%) were males and 19 (9.2%) were females. The obvious difference in gender may due to the fact that COPD is traditionally more common in men.21 The confounding effect of gender was therefore controlled as covariate in data analyses. A total of 158 subjects completed the RCT at the third month, and 128 subjects (TCQ group = 50, exercise group = 46, and control group = 32) received the follow-up assessment at the sixth month (Fig. 1). The self-reported adherence rate of daily practice in the TCQ group was 74% and in the exercise group was 69%. There was no significant difference in the baseline characteristics among the three study groups who completed the study (F (2, 128) = .118, P = .89) (Table 1).
Primary outcome measures Exercise capacity Results of 6MWT showed a significant time by group effect (F (5.10, 515.29) = 7.87, P < .001), indicating that the changes in walking distance over time among the three study groups were significantly different (Table 2 and Fig. 2). The mean distance of 6MWT in the TCQ group increased from 298 m to 331 m (+33 m, +11.0%) at the third month (T1 to T2),
588
A.W.K. Chan et al.
Figure 1 CONSORT (2005) flowchart to track participants through randomized controlled trail. ITT, intention to-treat; PP, per protocol, Reference: Chan et al.8
a further increased walking distance by 19 m (+5.6%) was observed at the sixth month (T2 to T3). No significant changes were noted in the exercise and control groups during the study period from T1 to T3. The between-subject effects also showed significant differences (F (2, 202) = 3.22, P = .042). Post hoc pairwise comparisons with Bonferroni adjustment indicated that significant differences between the TCQ and exercise groups (P = .03), and between the TCQ and control groups (P = .03) existed. No differences were noted between the exercise and control groups (P = .97). Lung functions Results reviewed that lung functions in the TCQ group showed a steady improvement from the T1 to T2 and from
T2 to T3 (Table 2). For FVC, a significant interaction effect of time by group (F (5.39, 544.82) = 6.33, P < .001) was noted. FVC increased by .19 l (+9.5%) in the TCQ group and by .11 l (+5.9%) in the exercise group, but decreased by .13 l (−7.2%) in the control group. Greater improvement in the TCQ group was found in the intervention period (T1 to T2, +6.6%) and a continuous improving trend was noted in the follow-up period (T2 to T3, +2.9%) (Fig. 3). For FEV1 , significant interaction effect of time by group (F (4.59, 464.00) = 6.36, P < .001) was also observed. FEV1 increased by .10 l (+11.00%) in the TCQ group and by .03 l (+3.27%) in the exercise group, whereas it decreased by .05 l (−5.67%) in the control group (T1 to T3). Again, the obvious improvement was seen in the first time segment (T1 to T2,
The sustaining effects of Tai chi Qigong Table 1
589
Demographic data and baseline characteristics of subjects by group allocation. TCQ (n = 70) Mean (SD)
Exercise (n = 69) Mean (SD)
Control (n = 67) Mean (SD)
P-value
Age (years)
71.7 (8.2)
73.6 (7.5)
73.6 (7.4)
.235
Gender Male (%) Female (%)
69 (99) 1 (1)
61 (88) 8 (12)
58 (87) 9 (13)
.021*
Body weight (kg) Body height (m) BMI (kg/m2 )
57.9 (10.4) 1.64 (.06) 21.5 (3.6)
56.5 (11.1) 1.61 (.07) 21.7 (4.0)
57.9 (12.3) 1.62 (.09) 21.9 (4.1)
.714 .061 .790
Years of COPD
10.3 (9.3)
10.6 (8.8)
12.4 (10.6)
.318
Smoking status Current smoker (%) Ex-smoker (%) Second hand smoker (%) Never smoke (%)
12 (17.1) 55 (78.6) 1 (1.4) 2 (2.9)
16 (23.2) 47 (68.1) 2 (2.9) 4 (5.8)
15 (22.4) 46 (68.7) 3 (4.5) 3 (4.5)
.813
Co-existing diseases Heart disease (%) HT (%) DM (%) No co-existing disease (%)
9 (13) 28 (40) 7 (10) 16 (23)
10 (15) 36 (52) 6 (9) 16 (23)
6 (9) 28 (42) 7 (10) 20 (30)
.581 .211 .947 .132
No. of drugs taking
4.9 (2.1)
5.1 (2.3)
4.5 (2.0)
.203
6MWD
297 (69)
285 (79)
290 (73)
.589
FEV1 FVC FEV1 % predicted
.89 (.38) 1.97 (.62) 50.1 (21.8)
.91 (.39) 1.84 (.52) 56.4 (25.6)
.89 (.39) 1.82 (.58) 55.1 (23.3)
.821 .238 .248
BMI, body mass index; 6MWD, six-minute walking distance; FEV1 , forced expiratory volume in 1 s; FVC, forced volume capacity; FEV1 % predicted, FEV1 percent predicted normal values. * P < .05.
+7.9%) and a smaller improvement trend was noted in second time segment (T2 to T3, +3.1%) (Fig. 4). Secondary outcome measures The only significant difference in the secondary outcome measures was the number of exacerbation attacks. The
Figure 2 baseline.
Changes in the six-minute walk test (6MWT, m) from
TCQ group showed a significant decrease in the number of exacerbations. The percentage of ‘‘no exacerbation’’ attack in the TCQ group increased from 65.7% (T1) to 85.7% (T2), and further increased to 92.9% in the sixth month
Figure 3 baseline.
Changes in the forced vital capacity (FVC, l) from
590
Table 2
Comparison of lung functions Outcome variables at baseline, 3rd month and 6th month. TCQ (n = 70) Mean ± SD (95%CI)
6MWD (m) Baseline (T1) 3-Month (T2) 6-Month (T3)
297.91 ± 68.53 (281.57, 314.25) 330.74 ± 61.86 (315.99, 345.49) 349.41 ± 70.69 (332.56, 366.27)
FVC (l) Baseline (T1) 3-Month (T2) 6-Month (T3)
1.97 ± .62 (1.82, 2.12) 2.10 ± .62 (1.95, 2.25) 2.16 ± .63 (2.01, 2.31)
FEV1 (l) Baseline (T1) 3-Month (T2) 6-Month (T3)
.89 ± .38 (.80, .98) .96 ± .39 (.87, 1.05) .99 ± .42 (.89, 1.09)
Borg scale post-6MWT Dyspnea Baseline (T1) 3-Month (T2) 6-Month (T3)
1.98 ± 1.21 (1.69, 2.27) 1.86 ± 1.25 (1.56, 2.16) 1.81 ± 1.20 (1.52, 2.09)
Fatigue Baseline (T1) 3-Month (T2) 6-Month (T3)
1.49 ± 1.46 (1.15, 1.84) 1.56 ± 1.39 (1.23, 1.89) 1.53 ± 1.32 (1.21, 1.84)
SaO2 % Baseline (T1) 3-Month (T2) 6-Month (T3)
94.60 ± 5.52 (93.28, 95.92) 94.03 ± 5.46 (92.73, 95.33) 94.10 ± 4.55 (93.02, 95.18)
Exercise (n = 69) Change from last measurement Mean (%)
32.83 (11.02) 18.67 (5.64)
Mean ± SD (95%CI)
284.64 ± 79.11 (265.63, 303.64) 290.04 ± 80.09 (270.80, 309.28) 298.07 ± 87.74 (277.00, 319.15)
.13 (6.60) .06 (2.86)
1.84 ± .52 (1.72, 1.97) 1.92 ± .63 (1.77, 2.08) 1.95 ± .62 (1.80, 2.09)
.07 (7.87) .03 (3.13)
.91 ± .39 (.82, 1.00) .92 ± .38 (.82, 1.01) .94 ± .42 (.84, 1.04)
−.12 (−6.06) −.05 (−2.69)
1.38 ± 1.74 (1.06, 1.71) 1.70 ± 1.38 (1.36, 2.03) 1.57 ± 1.28 (1.27, 1.88)
.07 (4.70) −.03 (−1.92)
1.38 ± 1.42 (1.04, 1.72) 1.42 ± 1.32 (1.10, 1.74) 1.52 ± 1.34 (1.20, 1.84)
−.57 (−.60) .07 (.07)
94.33 ± 4.80 (93.18, 95.49) 94.25 ± 5.07 (93.03, 95.46) 94.20 ± 5.75 (92.82, 95.58)
Control (n = 67) Change from last measurement Mean (%)
5.40 (1.90) 8.03 (2.77)
Mean ± SD (95%CI)
289.75 ± 72.97 (271.95, 307.55) 294.57 ± 78.05 (275.53, 313.60) 297.09 ± 84.25 (276.54, 317.64)
.08 (4.35) .03 (1.56)
1.82 ± .58 (1.68, 1.96) 1.74 ± .58 (1.60, 1.88) 1.69 ± .48 (1.57, 1.81)
.01 (1.10) .02 (2.17)
.89 ± .39 (.79, .99) .85 ± .35 (.76, .93) .84 ± .31 (.76, .91)
Change from last measurement Mean (%)
RANCOVA P-value .922
Partial eta squared
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevierhealth.com/journals/ctim
The sustaining effects of Tai chi Qigong on physiological health for COPD patients: A randomized controlled trial夽 Aileen W.K. Chan a,∗, Albert Lee b, Diana T.F. Lee a, Lorna K.P. Suen c, Wilson W.S. Tam b, S.Y. Chair a, Peter Griffiths d a
The Nethersole School of Nursing, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong Special Administrative Region b The Jockey Club School of Public Health and Primary Care, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong Special Administrative Region c School of Nursing, The Hong Kong Polytechnic University, Hunghom, Hong Kong Special Administrative Region d Faculty of Health Sciences, University of Southampton, Building 67, Room E4015, Highfield Campus, Southampton SO17 1BJ, United Kingdom Available online 8 October 2013
KEYWORDS COPD; Exercise tolerance; Lung functions; Six-minute walk test; Tai chi Qigong
Summary Objectives: To evaluate the sustaining effects of Tai chi Qigong in improving the physiological health for COPD patients at sixth month. Design: A randomized controlled trial. Subjects were in three randomly assigned groups: Tai chi Qigong group, exercise group, and control group. Setting: The 206 subjects were recruited from five general outpatient clinics. Interventions: Tai chi Qigong group completed a 3-month Tai chi Qigong program. Exercise group practiced breathing and walking as an exercise. Control group received usual care. Main outcome measures: Primary outcomes included six-minute walking distance and lung functions. Secondary outcomes were dyspnea and fatigue levels, number of exacerbations and hospital admissions. Results: Tai chi Qigong group showed a steady improvement in exercise capacity (P < .001) from baseline to the sixth month. The mean walking distance increased from 298 to 349 meters (+17%). No significant changes were noted in the other two groups. Tai chi Qigong group also
夽 The Health and Health Services Research Fund (HHSRF), the Food and Health Bureau, Hong Kong SAR Government is acknowledged as the funding source of this study. ∗ Corresponding author. Tel.: +852 39434290; fax: +852 26035269. E-mail address: [email protected] (A.W.K. Chan).
0965-2299/$ — see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctim.2013.09.008
586
A.W.K. Chan et al. showed improvement in lung functions (P < .001). Mean forced expiratory volume in 1 s increased from .89 to .99 l (+11%). No significant change was noted in the exercise group. Deterioration was found in the control group, with mean volume decreased from .89 to .84 l (−5.67%). Significant decreased in the number of exacerbations was observed in the Tai chi Qigong group. No changes in dyspnea and fatigue levels were noted among the three groups. Conclusions: Tai chi Qigong has sustaining effects in improving the physiological health and is a useful and appropriate exercise for COPD patients. © 2013 Elsevier Ltd. All rights reserved.
Introduction Chronic obstructive pulmonary disease (COPD) ranks as the fourth leading cause of death worldwide; it has a rising trend in morbidity and mortality.1 Because of the chronic and progressive nature of the disease, most of the lung function impairments are permanent and cannot be reversed.2 Aiding this population to develop disease management skills is therefore necessary to promote their physical health. Living with COPD imposes a certain degree of physiological challenge to the individuals, such as decline of lung functions and reduced exercise capacity.3,4 These challenges may lead to poor adjustment to their life. Symptoms of COPD included dyspnea and/or fatigue that limit exercise tolerance in most COPD patients. Poor motivation is also associated with exercise intolerance.5 Inactivity can lead to physical deconditioning that further limits exercise tolerance. Performing regular physical exercise can therefore be considered as a possible means to enhance physiological health among COPD patients. As Tai chi Qigong (TCQ) has characteristics that are particularly suitable for older people with chronic illnesses, the implementation of a TCQ program for the COPD patients could be of value. TCQ is a whole-body exercise that integrates breathing with harmonious movement of body training, which can benefit ventilation and improve physical capacity.6 However, limited existing systematic reviews on the effectiveness of Tai chi in enhancement of physical health in COPD patients was found.7 In view of the specific health needs among the COPD patients, our team has conducted a randomized controlled trial (RCT) to investigate the therapeutic effect of TCQ on COPD patients aimed to bridge the present knowledge gap.8 The TCQ forms were tailor-made for COPD patients. Participants were required to coordinate their breathing with the movement motions. After the 3-month intervention period, improvements in lung functions and exercise tolerance were noted in the TCQ group when compared to the exercise group and the control group. The improved exercise capacity was clinically significant,9 but the lung functions did not reached the clinically significant level.10 Since TCQ is a complex exercise routine that requires considerable practice to attain proficiency, it is possible that the improvements would increase over a longer study period. At the same time, a longer follow-up on the COPD subjects would be more informative, especially because deterioration in lung functions was identified as early as at 3 months in control group. This may become more marked over a longer period, whilst some improvement or maintenance was exhibited by the TCQ group. In addition, majority of the TCQ subjects (92%) showed interest in continuing the TCQ exercises. This study
therefore looked at longer period of follow-up assessment at the sixth month. It would be beneficial to assess clients’ long-term compliance and to monitor the effectiveness of TCQ practice in the progress of the disease.
Methods Study design and setting This was a randomized controlled study. Ethical approval for the 6-month follow-up evaluation was obtained from Joint the Chinese University of Hong Kong-New Territories East Cluster Clinical Research Ethics Committee. Informed consent was also obtained from every eligible participant. The study was carried out at five general outpatient clinics.
Eligibility criteria for participants Selection criteria were subjects who had clinically diagnosed with COPD and were ambulatory. Subjects were excluded if they suffered from severe cognitive impairment, symptomatic ischemic heart disease, or had practiced TCQ within a year.
Sample size and randomization Sample size calculation was based on previous findings of TCQ exercise on exercise capacity using six-minute walk test.11 According to Cohen,12 for ˛ = .05, a medium effect size of repeated measures analysis of variance test with three arms, 52 subjects per group can achieve a power of .80. To cover for the potential attrition rate of 25%,13 206 subjects were recruited and were randomly assigned to one of the three groups, TCQ group (n = 70), exercise group (n = 69), or control group (n = 67). Random allocation was decided by a computer-generated randomizer.14 The randomization avoids the shortcoming of yielding a highly disparate sample size in the study groups. The strength of this method is its ability to generate unpredictable sample allocation sequence. Stratification randomization was not used in this study.
Intervention TCQ program The experimental intervention was a TCQ program aimed to help COPD patients to develop the skills of TCQ exercise and to incorporate such exercise into their activity daily livings (ADLs). The program was conducted twice a week,
The sustaining effects of Tai chi Qigong
587
and each session lasted for 1 h. The TCQ program consisted of a 13-movement of TCQ, which were selected and modified from the 18 movements of Taiji Qigong.15 This modified 13-form TCQ was designed to allow patients to master the skills in a shorter period. The 13-form TCQ was tailor-made for COPD patients, it emphasized on elements of breathing, range of motion on upper limbs for enhancing lung expansion, and overall coordination. Participants were required to coordinate their breathing with the movement motions. An audiovisual DVD and TCQ pictures were given to each participant to facilitate daily self-practice at home. A selfrecording diary was given to the participants to record the frequency of their self-practicing TCQ at home.
Secondary outcome measures Secondary outcome measures included the Borg scale of dyspnea and fatigue levels, and the cutaneous oxygen saturation (SaO2 ) measured at rest and post 6MWT. The Borg scale is a self-rated scale which has a range of 0—10, where ‘‘0’’ indicates no dyspnea and no fatigue, and ‘‘10’’ indicates the maximal dyspnea and the maximal fatigue.19 SaO2 was measured by a noninvasive pulse oximetry to estimate the arterial oxyhemoglobin saturation. The number of exacerbations and number of hospital admissions in the past 3 months were also recorded. Acute exacerbation of COPD is defined as sustained worsening of symptoms and is acute in onset and often necessitates a change in medication.20
Breathing and walking exercise The goal of breathing exercise was to teach participants to relax the accessory muscles of breathing. Subjects were taught breathing techniques coordinated with walking exercise. The use of pursed-lip breathing (PLB) and diaphragmatic breathing (DB) techniques were taught by a qualified nurse. PLB can slow exhalatory flow and enable alveolar emptying, thereby creating mechanical advantage for the subsequent inhalation.16 DB is breathing in assisted by contracting the diaphragm and breathing out slowly while the diaphragm is relaxed. This maneuver increases lung capacity by lowering the diaphragm and allows air to reach the bottom of the lungs. The reduced respiratory rate leads to increased tidal volume and may improve gas exchange in the alveoli.17 Return demonstration of breathing techniques was performed to ensure proper skills practice of the participants. Participants were advised to self-practice breathing exercise and coordinated with self-paced walking exercise for one hour every day. Leaflets with pictures and instructions were given to the subjects to facilitate daily self-practice. A diary was also given to the participants for recording the frequency of their self-practice.
Data collection and statistical analysis Data collections were performed at 6-month (T3), in addition to baseline (T1) and post intervention at 3-month (T2). Data analyses were conducted using SPSS version 20.0. Repeated measures analyses of covariance (ANCOVA) were used to examine the outcome measures. A P-value of .05 was used as the level of statistical significance. An intention-totreat analysis was applied to carry forward last observation for the missing data.
Control group Participants in the control group continued to receive their usual medical treatment. They were advised to maintain their routine activities. No extra exercise was recommended. During the study period, all subjects continued their prescribed medical treatments. In order to enhance the internal validity of the study findings, participants in the exercise and control groups were arranged to join non-health-related social activities weekly. The purpose of maintaining regular gatherings for all participants was to balance the emotional effect of weekly gatherings of the TCQ group during the process of TCQ training. Primary outcome measures Primary outcome measures included exercise capacity and lung functions. Exercise capacity was assessed by six-minute walk test (6MWT). Subjects were encouraged to walk as far as possible for 6 min,18 and the total distance was recorded. Lung functions were measured via Spirometry. It required the patient to forcefully expel air from a point of maximal inspiration to a point of maximal expiration. Spirometry measures the forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1 ).
Results Two hundred and six eligible subjects were recruited in the study. The participants had an age range of 55—88 years, with a mean age (standard deviation) of 72.9 (7.7). The mean duration of COPD was 11.1 (9.6) years. The spirometry result indicated that the mean FEV1 percent predicted normal value was 53.8% (SD, 23.7). The most common coexisting disease was hypertension, followed by heart disease and diabetes mellitus. The number of medications taking among the three groups was comparable, ranged from 0 to 12, with a mean of 4.9 (2.1), which included medications for respiratory, cardiovascular, endocrine disorders. Of the 206 subjects, 187 (90.8%) were males and 19 (9.2%) were females. The obvious difference in gender may due to the fact that COPD is traditionally more common in men.21 The confounding effect of gender was therefore controlled as covariate in data analyses. A total of 158 subjects completed the RCT at the third month, and 128 subjects (TCQ group = 50, exercise group = 46, and control group = 32) received the follow-up assessment at the sixth month (Fig. 1). The self-reported adherence rate of daily practice in the TCQ group was 74% and in the exercise group was 69%. There was no significant difference in the baseline characteristics among the three study groups who completed the study (F (2, 128) = .118, P = .89) (Table 1).
Primary outcome measures Exercise capacity Results of 6MWT showed a significant time by group effect (F (5.10, 515.29) = 7.87, P < .001), indicating that the changes in walking distance over time among the three study groups were significantly different (Table 2 and Fig. 2). The mean distance of 6MWT in the TCQ group increased from 298 m to 331 m (+33 m, +11.0%) at the third month (T1 to T2),
588
A.W.K. Chan et al.
Figure 1 CONSORT (2005) flowchart to track participants through randomized controlled trail. ITT, intention to-treat; PP, per protocol, Reference: Chan et al.8
a further increased walking distance by 19 m (+5.6%) was observed at the sixth month (T2 to T3). No significant changes were noted in the exercise and control groups during the study period from T1 to T3. The between-subject effects also showed significant differences (F (2, 202) = 3.22, P = .042). Post hoc pairwise comparisons with Bonferroni adjustment indicated that significant differences between the TCQ and exercise groups (P = .03), and between the TCQ and control groups (P = .03) existed. No differences were noted between the exercise and control groups (P = .97). Lung functions Results reviewed that lung functions in the TCQ group showed a steady improvement from the T1 to T2 and from
T2 to T3 (Table 2). For FVC, a significant interaction effect of time by group (F (5.39, 544.82) = 6.33, P < .001) was noted. FVC increased by .19 l (+9.5%) in the TCQ group and by .11 l (+5.9%) in the exercise group, but decreased by .13 l (−7.2%) in the control group. Greater improvement in the TCQ group was found in the intervention period (T1 to T2, +6.6%) and a continuous improving trend was noted in the follow-up period (T2 to T3, +2.9%) (Fig. 3). For FEV1 , significant interaction effect of time by group (F (4.59, 464.00) = 6.36, P < .001) was also observed. FEV1 increased by .10 l (+11.00%) in the TCQ group and by .03 l (+3.27%) in the exercise group, whereas it decreased by .05 l (−5.67%) in the control group (T1 to T3). Again, the obvious improvement was seen in the first time segment (T1 to T2,
The sustaining effects of Tai chi Qigong Table 1
589
Demographic data and baseline characteristics of subjects by group allocation. TCQ (n = 70) Mean (SD)
Exercise (n = 69) Mean (SD)
Control (n = 67) Mean (SD)
P-value
Age (years)
71.7 (8.2)
73.6 (7.5)
73.6 (7.4)
.235
Gender Male (%) Female (%)
69 (99) 1 (1)
61 (88) 8 (12)
58 (87) 9 (13)
.021*
Body weight (kg) Body height (m) BMI (kg/m2 )
57.9 (10.4) 1.64 (.06) 21.5 (3.6)
56.5 (11.1) 1.61 (.07) 21.7 (4.0)
57.9 (12.3) 1.62 (.09) 21.9 (4.1)
.714 .061 .790
Years of COPD
10.3 (9.3)
10.6 (8.8)
12.4 (10.6)
.318
Smoking status Current smoker (%) Ex-smoker (%) Second hand smoker (%) Never smoke (%)
12 (17.1) 55 (78.6) 1 (1.4) 2 (2.9)
16 (23.2) 47 (68.1) 2 (2.9) 4 (5.8)
15 (22.4) 46 (68.7) 3 (4.5) 3 (4.5)
.813
Co-existing diseases Heart disease (%) HT (%) DM (%) No co-existing disease (%)
9 (13) 28 (40) 7 (10) 16 (23)
10 (15) 36 (52) 6 (9) 16 (23)
6 (9) 28 (42) 7 (10) 20 (30)
.581 .211 .947 .132
No. of drugs taking
4.9 (2.1)
5.1 (2.3)
4.5 (2.0)
.203
6MWD
297 (69)
285 (79)
290 (73)
.589
FEV1 FVC FEV1 % predicted
.89 (.38) 1.97 (.62) 50.1 (21.8)
.91 (.39) 1.84 (.52) 56.4 (25.6)
.89 (.39) 1.82 (.58) 55.1 (23.3)
.821 .238 .248
BMI, body mass index; 6MWD, six-minute walking distance; FEV1 , forced expiratory volume in 1 s; FVC, forced volume capacity; FEV1 % predicted, FEV1 percent predicted normal values. * P < .05.
+7.9%) and a smaller improvement trend was noted in second time segment (T2 to T3, +3.1%) (Fig. 4). Secondary outcome measures The only significant difference in the secondary outcome measures was the number of exacerbation attacks. The
Figure 2 baseline.
Changes in the six-minute walk test (6MWT, m) from
TCQ group showed a significant decrease in the number of exacerbations. The percentage of ‘‘no exacerbation’’ attack in the TCQ group increased from 65.7% (T1) to 85.7% (T2), and further increased to 92.9% in the sixth month
Figure 3 baseline.
Changes in the forced vital capacity (FVC, l) from
590
Table 2
Comparison of lung functions Outcome variables at baseline, 3rd month and 6th month. TCQ (n = 70) Mean ± SD (95%CI)
6MWD (m) Baseline (T1) 3-Month (T2) 6-Month (T3)
297.91 ± 68.53 (281.57, 314.25) 330.74 ± 61.86 (315.99, 345.49) 349.41 ± 70.69 (332.56, 366.27)
FVC (l) Baseline (T1) 3-Month (T2) 6-Month (T3)
1.97 ± .62 (1.82, 2.12) 2.10 ± .62 (1.95, 2.25) 2.16 ± .63 (2.01, 2.31)
FEV1 (l) Baseline (T1) 3-Month (T2) 6-Month (T3)
.89 ± .38 (.80, .98) .96 ± .39 (.87, 1.05) .99 ± .42 (.89, 1.09)
Borg scale post-6MWT Dyspnea Baseline (T1) 3-Month (T2) 6-Month (T3)
1.98 ± 1.21 (1.69, 2.27) 1.86 ± 1.25 (1.56, 2.16) 1.81 ± 1.20 (1.52, 2.09)
Fatigue Baseline (T1) 3-Month (T2) 6-Month (T3)
1.49 ± 1.46 (1.15, 1.84) 1.56 ± 1.39 (1.23, 1.89) 1.53 ± 1.32 (1.21, 1.84)
SaO2 % Baseline (T1) 3-Month (T2) 6-Month (T3)
94.60 ± 5.52 (93.28, 95.92) 94.03 ± 5.46 (92.73, 95.33) 94.10 ± 4.55 (93.02, 95.18)
Exercise (n = 69) Change from last measurement Mean (%)
32.83 (11.02) 18.67 (5.64)
Mean ± SD (95%CI)
284.64 ± 79.11 (265.63, 303.64) 290.04 ± 80.09 (270.80, 309.28) 298.07 ± 87.74 (277.00, 319.15)
.13 (6.60) .06 (2.86)
1.84 ± .52 (1.72, 1.97) 1.92 ± .63 (1.77, 2.08) 1.95 ± .62 (1.80, 2.09)
.07 (7.87) .03 (3.13)
.91 ± .39 (.82, 1.00) .92 ± .38 (.82, 1.01) .94 ± .42 (.84, 1.04)
−.12 (−6.06) −.05 (−2.69)
1.38 ± 1.74 (1.06, 1.71) 1.70 ± 1.38 (1.36, 2.03) 1.57 ± 1.28 (1.27, 1.88)
.07 (4.70) −.03 (−1.92)
1.38 ± 1.42 (1.04, 1.72) 1.42 ± 1.32 (1.10, 1.74) 1.52 ± 1.34 (1.20, 1.84)
−.57 (−.60) .07 (.07)
94.33 ± 4.80 (93.18, 95.49) 94.25 ± 5.07 (93.03, 95.46) 94.20 ± 5.75 (92.82, 95.58)
Control (n = 67) Change from last measurement Mean (%)
5.40 (1.90) 8.03 (2.77)
Mean ± SD (95%CI)
289.75 ± 72.97 (271.95, 307.55) 294.57 ± 78.05 (275.53, 313.60) 297.09 ± 84.25 (276.54, 317.64)
.08 (4.35) .03 (1.56)
1.82 ± .58 (1.68, 1.96) 1.74 ± .58 (1.60, 1.88) 1.69 ± .48 (1.57, 1.81)
.01 (1.10) .02 (2.17)
.89 ± .39 (.79, .99) .85 ± .35 (.76, .93) .84 ± .31 (.76, .91)
Change from last measurement Mean (%)
RANCOVA P-value .922
Partial eta squared
