®

82 PREVENTIVE CARDIOLOGY Spring 2008

Clinical StudY

The Effect of Tai Chi Exercise on Blood

Pressure: A Systematic Review
Gloria Y. Yeh, MD, MPH;1,3 Chenchen Wang, MD, MSc;2 Peter M. Wayne, PhD;3 Russell S. Phillips, MD1,3

A systematic review of the literature on the Tai chi exercise may reduce BP and serve as a
effect of tai chi exercise on blood pressure (BP) practical, nonpharmacologic adjunct to conven-
was performed. The authors searched Medline, tional hypertension management. Prev Cardiol.
CAB, Alt HealthWatch, BIOSIS previews, Science 2008;11:82–89. ©2008 Le Jacq
Citation Index, and EMBASE systems (incep-
tion through January 2007); researched Chinese
Medical, China Hospital Knowledge, China
National Knowledge Infrastructure, and China
H ypertension is the most common primary
office diagnosis in the United States, with
more than 35 million visits per year. Despite phar-
Traditional Chinese Medicine databases (incep- macologic advances and nationwide education
tion to June 2005); and performed hand searches campaigns, only one-third of adult patients with
at the medical libraries of Beijing and Nanjing hypertension in 2000 had adequate blood pres-
Universities. Clinical studies of tai chi examining sure (BP) control, far below the Healthy People
BP as an outcome published in English or Chinese 2010 goal of 50%.1 Clinical trials have consistently
were included. Studies reporting only acute exer- shown the benefits of lowering BP, with substantial
cise effects were excluded. Data were extracted in reductions in cardiovascular risk, stroke, myocar-
a standardized manner and 2 independent inves- dial infarction, heart failure, and cardiovascular-
tigators assessed methodologic quality. Twenty- related death.2 While pharmacologic therapy is
six studies examining patients with and without often emphasized, the critical importance of non-
cardiovascular conditions met inclusion criteria: pharmacologic approaches and lifestyle modifica-
9 randomized controlled trials, 13 nonrandom- tions, including physical activity and exercise, has
ized studies, and 4 observational studies. Study continued to be recognized by the most recent Joint
heterogeneity precluded formal meta-analyses. National Committee on Prevention, Detection,
Twenty-two studies (85%) reported reductions in Evaluation, and Treatment of High Blood Pressure
BP with tai chi (3–32 mm Hg systolic and 2–18 (JNC 7) report for both primary and secondary
mm Hg diastolic BP reductions). Five randomized prevention of high BP.1
controlled trials were of adequate quality (Jadad In recent years, with the popularity and preva-
score ≥3). No adverse effects were reported. lence of mind-body therapies, there has been a
growing interest in tai chi exercise for patients with
From the Division of General Medicine and Primary hypertension.3–5 Tai chi (t’ai chi or taiji) has origins
Care, Department of Medicine, Beth Israel Deaconess in ancient Chinese martial arts and combines gentle
Medical Center;1 the Division of Rheumatology, Tufts- physical activity with elements of meditation, body
New England Medical Center;2 and the Division awareness, imagery, and attention to breathing. The
for Research and Education in Complementary and scientific literature describing tai chi is varied, with
Integrative Medical Therapies, Harvard Medical School,3 studies reporting benefits in a number of health con-
Boston, MA ditions, from balance and reduction of falls in frail
Address for correspondence:
Gloria Y. Yeh, MD, MPH, Beth Israel Deaconess Medical
adults, to improvements in quality of life and symp-
Center, Harvard Medical School Osher Institute, 401 toms of rheumatoid arthritis, the human immunode-
Park Drive, Suite 22A, Boston, MA 02215 ficiency virus, cancer, and heart failure.6,7 A substan-
E-mail: [email protected] tial amount of research examines the cardiovascular
Manuscript received August 8, 2007; effects of tai chi, including cardiorespiratory fitness
revised September 28, 2007; and exercise capacity, although BP is the most com-
accepted October 9, 2007 monly evaluated effect in these studies.4,8
To date, there have been no comprehensive sys-
www.lejacq.com ID: 7565 tematic reviews examining long-term BP effects of

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

Spring 2008 PREVENTIVE CARDIOLOGY 83

tai chi, and very little is known about what has been impractical in tai chi studies, our modification gives
published in the Chinese language. Our objective 1 point for proper single-blinding of the outcome
was to conduct a systematic review of Chinese and assessors. Summary quality grading criteria for each
English language literature on the effects of tai chi of the 3 design strata are listed in Table I.
on BP and hypertension and to offer recommenda-
tions for future research. Results
We screened 829 English language and 859 Chinese
Methods language abstracts and full text articles for poten-
Literature Search tially relevant data. A total of 26 studies (11 in
We conducted electronic literature searches of English, 15 in Chinese) met the inclusion criteria
Medline (from 1966), CAB (from 1973), Alt and were analyzed. This includes 9 RCTs, 13
HealthWatch, BIOSIS previews (from 1969), NRSs, and 4 OBSs.10–35 Studies were conducted in
Science Citation Index (from 1945), and EMBASE patients with hypertension, coronary heart disease,
systems (from 1991) through January 2007 using varied cardiovascular conditions, and chronic rheu-
search terms “tai chi,” “tai chi chuan,” “ta’i chi,” matologic and dermatologic conditions, as well as
“tai ji,” and “taijiquan.” In addition, we per- in “healthy volunteers.” Within these studies, in
formed searches of the Chinese Medical, China addition to BP, other reported outcomes included
Hospital Knowledge, China National Knowledge heart rate, body mass index, exercise capacity,
Infrastructure, and China Traditional Chinese heart rate variability, lipids, pulmonary and cardiac
Medicine databases from inception to June 2005 function, functional measures, and quality of life.
and performed hand searches at the medical librar- Table II details studies examining BP in patients
ies of Beijing and Nanjing Universities in China. We with hypertension.10–17 Table III details studies
also performed hand searches of retrieved articles examining BP in other cardiovascular popula-
for additional references. tions.18–20 Table IV details studies examining BP
in noncardiovascular populations and “healthy”
Eligibility Criteria patients.21–35 Study heterogeneity precluded formal
Available clinical studies published in English and meta-analysis. No adverse events associated with tai
Chinese that used human participants and exam- chi were reported. Of the 9 RCTs, 4 received an A
ined long-term (nonacute) BP changes (>1 week) quality score and 2 received a B. Of the 13 NRSs,
were included. 11 received a B score. Of the 4 OBSs, 1 received an
A and 3 received a B score. Of the English language
Data Extraction and Synthesis studies, 5 received an A and 6 received a B. Of the
Two independent reviewers extracted data in a stan- Chinese studies, 1 received an A, 10 received a B,
dardized manner. We extracted data from Chinese and 4 received a C. The significance of each rating
language articles with direct translation to English. is provided in Table I.

Grading of Methodologic Quality Studies in Patients With Hypertension

To assess methodologic quality of studies, we Of the 8 studies that specifically evaluated patients
developed an “ABC” summary quality grading with hypertension, all reported a statistically signifi-
system adapted from methods used in evidence cant within-group reduction in mean BP after tai chi
reports of the Agency for Healthcare Research and exercise. Three RCTs, 4 NRSs, and 1 OBS with a
Quality (AHRQ) Evidence-Based Practice Centers.9 total of 524 patients with hypertension were found
Two independent investigators assessed method- (Table II). The duration of tai chi training for the
ologic quality, evaluating specific criteria for each studies ranged from 12 weeks to 3 years. The mag-
study design type (randomized controlled trials nitude of systolic BP (SBP) and diastolic BP (DBP)
[RCTs], prospective nonrandomized controlled and change in the tai chi group ranged from –7 to –32
noncontrolled studies [NRSs], and observational mm Hg and –2.4 to –18 mm Hg, respectively.
controlled and noncontrolled studies [OBSs] and The 3 RCTs specifically designed to study patients
assigning an A, B, or C grade based on the potential with hypertension varied in quality and study design.
for bias in the study. Any discrepancies between Young and colleagues10 compared a light-intensity
assessors were resolved through discussion. The tai chi program that “emphasized physical move-
summary quality grading system evaluates and rates ments rather that meditational aspects” to moderate-
studies within each of the study design strata. By intensity walking and low-impact aerobic dance. The
design, it does not attempt to assess the compara- investigators reported comparable BP reductions
tive validity of studies across different design strata. (±SD) in both groups (–7.0 [±8.8] vs –8.4 [±8.8] mm
Thus, in interpreting the methodologic quality of Hg SBP; –2.4 [±5.5] vs –3.2 [±5.5] mm Hg DBP,
a study, one should note the quality grade that it respectively); however, there were no difference
received and the study design. For RCTs, in addi- between groups. Of note, a higher compliance with
tion to the summary quality grade, we also indicate home exercise was reported in the tai chi group. Tsai
a modified Jadad score. Because double-blinding is and colleagues11 reported significant SBP and DBP

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

84 PREVENTIVE CARDIOLOGY Spring 2008

Table I. ABC Quality Grading Criteria for 3 Study Design on the State-Trait Anxiety Inventory. The study
Strataa by Shen and Su,12 conducted in China, compared
Randomized controlled trials tai chi with antihypertensive agents (unspecified)
and with no treatment. Although this study was
Adequate randomization, proper single-blinding of
poorly described and lacked detail, the investiga-
assessors, and reporting of dropouts (modification of
tors reported a significantly larger proportion of
Jadad score)
patients meeting “effective BP control” in the tai
Adequate methods used to assess physical activity chi group (73% in tai chi vs 45% in the medication
No errors or discrepancies in reporting results group, P<.01).
Clear inclusion/exclusion criteria
Sample size estimates/justification Studies in Other Cardiovascular Populations
Adequate description of tai chi intervention (eg, style, Of the 3 studies that examined patients with car-
training schedule, frequency/duration of classes, diovascular conditions, 2 studies examined patients
instructor experience) with coronary heart disease discharged after an
Adequate description of comparison groups acute hospital stay (Table III). In the only RCT
Prospective, nonrandomized studies (controlled and
in this subset, Channer and colleagues18 random-
noncontrolled) ized patients recovering from an acute myocardial
infarction to a mixed tai chi/qigong intervention,
Unbiased selection of the cohort (prospective recruitment
conventional aerobic exercise, or a cardiac support
of patients)
group. After 8 weeks, both aerobic exercise and
Sufficiently large sample size tai chi were associated with significant reductions
Adequate description of the cohort; clear inclusion/ (±SD) in SBP (–4 [±7.5] and –3 [±3.3] mm Hg,
exclusion criteria respectively, both P<.05). DBP was improved in the
Adequate methods used to assess physical activity tai chi group only (–2 [±2.7] mm Hg, P<.01). No
Adequate description of tai chi intervention (eg, style, between-group comparisons were made. This study
training schedule, frequency/duration of classes, also reported decreases in resting heart rate and
instructor experience) greater compliance with the tai chi intervention.
Adequate description of comparison groups The prospective nonrandomized study by Zheng19
Use of validated method for ascertaining clinical outcomes showed decreases in DBP in patients with coronary
Adequate follow-up period
artery disease who had a mean baseline BP that met
current JNC 7 criteria for stage I hypertension.
Completeness of follow-up
Analysis (multivariate adjustments) and reporting of results; Studies in Noncardiovascular Populations and
use of appropriate statistical analyses Healthy Patients
Observational studies (controlled and noncontrolled) Of the 15 studies in this subset, most were conducted
Valid ascertainment of cases in middle-aged “healthy” patients, although the mean
Unbiased selection of cases age ranged from 20 to 81 years (Table IV). Five RCTs,
Appropriateness of the control population (as applicable) 7 NRSs, and 3 OBSs with a total of 1157 patients
Clear inclusion/exclusion criteria were included. The magnitude of BP decreases in
these studies ranged from –4 to –18 mm Hg SBP and
Comparability of cases and controls with respect to
–2.3 to –7.5 mm Hg DBP. Most compared tai chi with
potential confounders
usual activity, other forms of exercise, or education.
Adequate methods used to assess physical activity Of note, patients in 4 of these studies had a mean
Adequate description of tai chi intervention (eg, style, baseline SBP >140 mm Hg, although a diagnosis of
training schedule, frequency/duration of classes, hypertension was not mentioned.
instructor experience)
Adequate description of comparison groups Discussion
Appropriate statistical analyses Tai chi may have beneficial effects on BP, although
ABC grades: A, Least bias; results are valid. B, Susceptible studies have varied in methodologic quality. Each
to some bias but not sufficient to invalidate the results. of the 8 studies specifically designed to evaluate
C, Significant bias that may invalidate the results. aABC patients with hypertension reported improvements
summary quality grading system adapted from methods used in BP with tai chi, although sometimes showing no
in evidence reports of the Agency for Healthcare Research difference when compared with conventional exer-
and Quality.9 cise. Another 18 studies, including other cardio-
vascular populations and “healthy” patients, offer
supportive evidence that tai chi can reduce BP and
reductions, while there were no changes with usual may play a role in primary prevention. More than
care. They also reported significant improvements in one-half of the studies were published in Chinese
lipids (total cholesterol, triglycerides, and low- and and offer data that has historically been excluded
high-density lipoprotein), as well as improvements from other reviews.

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

Spring 2008 PREVENTIVE CARDIOLOGY 85

Table II. Studies of Tai Chi in Patients With Hypertension

Modified
Jadad
Scorec
Reference (Author, BP Results/Magnitude and/
Year, Country, Study Population of BP Change (SD or ABC
Publication Study Description, Mean Intervention/ if Available) in TC Quality
Language) Design Baseline BP, Mean Age No.a Control Details groupb Score
Young et al,10 1999, RCT High normal or stage I 60 Yang-style TC (13 Decreased BP in both +4c
United States, hypertension movements) for groups (between-group A
English 140/76 mm Hg 12 wk P value not significant)
67 y Walking/aerobic SBP –7 (8.8) mm Hg
dance DBP –2 (5.5) mm Hg
Tsai et al,11 2003, RCT High normal or stage I 76 Yang-style TC (108 Decreased BP (compared +3c
Taiwan, English hypertension postures) for 12 wk with usual care) B
145/87 mm Hg Usual care SBP –16 (7.9) mm Hg
51 y DBP –9 (7.4) mm Hg
Shen and Su,12 2000, RCT Essential hypertension 60 TC/Qigong Decreased BP (compared +2
China, Chinese Baseline BP not (18 postures; with medication and C
reported unspecified no treatment)
64 y duration) SBP –30 mm Hg
Medicationd DBP –10 mm Hg
No treatment
Wang et al,13 2000, NRS Essential hypertension 54 Yang-style TC for 3 y Decreased BP B
China, Chinese or high normal SBP –14 mm Hg
(Men) 161/97 mm Hg (hypertension); –10
(hypertension) mm Hg (high normal)
140/83 mm Hg (high DBP –7 mm Hg
normal) (hypertension); –3 mm
66 y Hg (high normal)
Fang and Wang,14 NRS Stage I/II hypertension 70 Yang-style TC Decreased BP all groups B
1985, China, 168/103 mm Hg (simplified 24 compared with no
Chinese Range 40–70 y forms) for 12 wk treatment (between-
Qigong breathing group P value not
Medication (tab significant for active
hypotensor co- interventions)
captopril)d SBP –13 mm Hg
No treatment DBP –8 mm Hg
Lu et al,15 1987, NRS Hypertension 14 TC (unspecified Decreased BP B
China, Chinese 180/99 mm Hg style)/Qigong/ SBP –32 mm Hg
66 y relaxed slow DBP –18 mm Hg
running for 6 mo
Taylor-Piliae,16 2006, NRS At least 1 CV risk 38 Yang-style TC Decreased BP B
United States, factor (92% with (24-posture short SBP –19 (2.8) mm Hg
English hypertension) form) for 12 wk (at rest); –14 (4.0) mm
150/86 mm Hg Hg (after step-test)
66 y DBP –9 (1.3) mm Hg
(at rest); –17 (3.5) mm
Hg (after step-test)
Liu and Li,17 2004, OBS Chronic hypertension 113 TC (unspecified Decreased BP, B
China, Chinese 66 y style; unspecified TC vs control: SBP 159
duration) vs 167 mm Hg
No TC
aNumber of study participants included in blood pressure (BP) analyses. bAll within-group (tai chi [TC]) pre-post changes are

significant (P≤.05) unless otherwise noted. All results in controlled trials are reported in comparison with the control group(s) and
are significant (P≤.05) unless otherwise noted. cModified Jadad score for randomized controlled clinical trials (RCTs) (which gives 1
point for proper single-blinding of outcome assessors). dNo further details reported. Abbreviations: CV, cardiovascular; DBP, diastolic
BP; NRS, prospective nonrandomized studies (that include an intervention), controlled and noncontrolled; OBS, observational,
cross-sectional studies, controlled; SBP, systolic BP; SD, standard deviation.

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

86 PREVENTIVE CARDIOLOGY Spring 2008

Table III. Studies of Long-Term BP Effects in Other Cardiovascular Populations

Study Population Modified
Reference Description, Mean BP Results/ Jadadc and/or
(Author, Year, Study Baseline BP, Mean Intervention/ Magnitude of BP ABC Quality
Country) Design Age No.a Control Details Change in TC groupb Score
Channer et al,18 RCT Coronary artery 126 Wu-style TC/Qigong Decreased BP both +2
1996, United disease (post- for 8 wk exercise groups B
Kingdom, myocardial Exercise to music (between-group P
English infarction) Support group value not reported)
133/84 mm Hg BP –3 (3.3) mm Hg
56 y DBP –2 (2.7) mm Hg
19
Zheng, NRS Coronary artery 24 Yang-style TC Decreased DBP B
2004, China, disease (simplified 24 DBP –6 mm Hg
Chinese 149/88 mm Hg forms) for 3 mo
68 y
Zhou and Li,20 NRS Various CV 143 TC (unspecified Decreased BP C
1994, China, conditions (26% style) for 3 y (compared with no
Chinese with hypertension) No TC TC)d
Baseline BP not
reported
54 y
aNumber of study participants included in blood pressure (BP) analyses; bAll within-group (tai chi [TC]) pre-post changes are

significant (P≤.05) unless otherwise noted. All results in controlled trials are reported in comparison with the control group(s) and
are significant (P≤.05) unless otherwise noted. cModified Jadad score for randomized controlled clinical trials (RCTs) (which gives 1
point for proper single-blinding of outcome assessors). dNo further details reported. Abbreviations: CV, cardiovascular; DBP, diastolic
BP; NRS, prospective nonrandomized studies (that include an intervention), controlled and noncontrolled; OBS, observational,
cross-sectional studies, controlled; SBP, systolic BP.

Comparison With Nonpharmacologic Approaches modulation of autonomic tone is often cited.46–48

to Hypertension The role of stress in cardiovascular physiology and
Traditionally, the role of exercise in the manage- the relationship between stress, chronic sympathetic
ment of BP and hypertension has been discussed stimulation, and vasoconstriction is well described.49
in combination with other lifestyle modifications, Although there is a component of aerobic exercise
including weight loss, sodium restriction, the with tai chi, to what extent the meditation, imagery,
Dietary Approaches to Stop Hypertension (DASH) breathwork, general relaxation, and stress reduction
diet with increased potassium and calcium, mod- play active roles in the overall effects is unknown.
eration of alcohol, and smoking cessation.36,37 With Studies suggest that meditative and stress reduction
tai chi, we found reductions in SBP from 7 to 32 techniques (eg, progressive muscle relaxation, bio-
mm Hg. Studies of conventional physical activity feedback, or music therapy) alone can have measur-
have reported changes in SBP from 4 to 9 mm Hg able effects on BP.50
and suggest that there may be an additive effect
when used in combination with other nonpharma- Clinical Implications and Advantages of Tai Chi
cologic strategies, lowering SBP up to 10 to 15 mm Given the existing evidence, tai chi exercise may
Hg.38–40 Although it is difficult to compare magni- be a safe and effective alternative to conventional
tudes of BP reductions across studies, the available exercise programs. It may be appropriate for
data suggest that tai chi may be as effective as other patients unable or unwilling to engage in other
lifestyle approaches. forms of physical activity or as a bridge to more
rigorous activity in frail or deconditioned patients.
Comparison With Other Mind-Body Approaches Patients with borderline BPs may be reluctant to
Similar BP-lowering effects have been reported begin drug therapy and often welcome nonphar-
with other mind-body therapies, both nonexercise macologic approaches. These lifestyle interventions
and exercise-based interventions (eg, transcenden- have been recognized as important and effective
tal meditation, mindfulness-based stress reduction, strategies for both primary and secondary preven-
qigong, and yoga).41–45 In one study, investigators tion, although they often require a high level of
reported a 3.4 mm Hg SBP reduction in patients motivation from the patient.39 Patients with either
with stable coronary disease who practiced 16 weeks prehypertension or established hypertension, who
of transcendental meditation, as compared with otherwise feel well, may find it difficult to engage
health education. The mechanisms by which mind- in and maintain a regular conventional exercise
body interventions such as tai chi may contribute to regimen. Finding an appropriate, nonthreatening,
reductions in BP are not well understood, although easy-to-perform activity that patients will maintain

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

Spring 2008 PREVENTIVE CARDIOLOGY 87

Table IV. Studies of Long-Term BP Effects in Noncardiovascular and “Healthy” Patients

Modified
Jadadc and/
Reference Study Population or ABC
(Author, Year, Study Description, Mean Intervention/ BP Results/Magnitude of BP Change Quality
Country) Design Baseline BP, Mean Age No.a Control Details (SD) in TC groupb Score
Wolf et al,21 RCT Healthy 167 Yang-style TC for 15 Decreased SBP post-walk test (compared +4c
1996, United SBP post-walk 169 mm Hg wk with balance training and education) A
States, English 76 y Computerized balance SBP –13 (27.4) mm Hg
training
Education
Wolf et al,22 RCT Transitionally frail 311 Yang-style TC for 48 Decreased SBP (compared with +4†
2006, United 149/77 mm Hg wk education) A
States, English 81 y Wellness education SBP –7 mm Hg
DBP –4 mm Hg
Thomas et al,23 RCT Healthy 207 Yang-style TC (24 No change in BP (compared with +3
2005, Hong 142/72 mm Hg forms) for 12 mo resistance training and usual activity) A
Kong, English 69 y Resistance training
Usual activity
Zhang and Yue,24 RCT Healthy men 20 TC (unspecified style) Decreased BP (compared with running) +2
2004, China, 123/79 mm Hg for 8 wk SBP –9 mm Hg C
Chinese 20 y Running DBP –6 mm Hg
Zhang et al,25 RCT Healthy 60 TC (unspecified style Decreased BP (compared with no +1
1997, China, Baseline BP not reported and duration) exercise) C
Chinese Mean age not reported No exercise Magnitude of change not reported
Wu and Ho,26 NRS Chronic rheumatic and 20 Yang-style TC Increased SBP A
1996, China, dermatologic conditions (simplified 24 forms, (+7 mm Hg)
Chinese 118/75 mm Hg 108 movements) for
54 y 3 mo
Ko et al,27 2006, NRS Healthy women; hospital staff 20 TC (unspecified style) Decreased SBP B
Hong Kong, 114/69 mm Hg for 10 wk (–6 mm Hg)
English 41 y
Schaller28 1996, NRS Healthy 46 Modified TC for 10 wk No change in BP (compared with usual B
United States, Mean BP not reported Usual activity activity)
English 70 y
Thornton et al,29 NRS Healthy women; community- 34 Yang-style TC (108 Decreased BP (compared with usual B
2004, Hong dwelling) postures) for 12 wk activity)
Kong, English 122/80 mm Hg Usual activity SBP –10 mm Hg
48 y DBP –8 mm Hg
Liu et al,30 2003, NRS Healthy women 30 Yang-style TC Decreased BP B
China, Chinese 130/83 mm Hg (simplified 24 form) SBP –18 mm Hg
64 y for 6 mo DBP –11 mm Hg
Ni and Lei,31 NRS Healthy 39 TC (42 styles) for 1 y Decreased SBP B
2000, China, 131/78 mm Hg (–4 mm Hg)
Chinese 61 y
Zhang and Fu,32 NRS Healthy 90 TC (unspecified style) Decreased BP B
1991, China, 121/74 mm Hg (young men); for 12 mo SBP –7 mm Hg (young men); -11 mm
Chinese 113/73 mm Hg (young Hg (young women); –12 mm Hg (older
women); 130/86 mm Hg men); –8 mm Hg (older women)
(older men); 126/82 mm DBP –5 mm Hg (young men); –9 mm
Hg (older women) Hg (young women); –12 mm Hg (older
25 y (young) men); –9 mm Hg (older women)
41 y (old)
Wang et al,33 OBS Healthy male TC 20 Yang-style TC (108 No difference in BP A
2001, Taiwan, practitioners movements) for 11 y TC vs control
English 69 y Age and body size– SBP 141 (28.5) vs 139 (22.1) mm Hg
matched sedentary DBP 72 (6.3) vs 77 (6.3) mm Hg
control
Gao and Tan,34 OBS Healthy elders 60 TC (unspecified style)
Decreased BP B
1997, China, Range 60–70 y for 3 y
TC vs control:
Chinese No TC SBP 123 (11.3) vs 145 (13.5) mm Hg
DBP 79 (13.5) vs 88 (9.0) mm Hg
35 2001,
Liang, OBS Healthy elders 33 Long-term TC Decreased BP B
China, Chinese 65 y practitioners TC vs control:
(unspecified style; SBP 137 (4.3) vs 148 (4.9) mm Hg
unspecified duration) DBP 77 (2.5) vs 81 (3.1) mm Hg
Non-TC practitioners
aNumber of study participants included in blood pressure (BP) analyses. bAll within-group (tai chi [TC]) pre-post changes are significant (P≤.05)

unless otherwise noted. All results in controlled trials are reported in comparison with the control group(s) and are significant (P≤.05) unless otherwise
noted. cModified Jadad score for randomized controlled clinical trials (RCTs) (which gives 1 point for proper single-blinding of outcome assessors).
Abbreviations: DBP, diastolic BP; NRS, prospective nonrandomized studies (that include an intervention), controlled and noncontrolled; OBS,
observational, cross-sectional studies, controlled; SBP, systolic BP.

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

88 PREVENTIVE CARDIOLOGY Spring 2008

is critical to therapeutic success. Clinical trials have clinical use. Further research will help to define the
reported excellent compliance with tai chi interven- potential role of tai chi exercise in both primary
tions and suggest that tai chi may promote exercise and secondary prevention of hypertension together
self-efficacy. Several studies mention better adher- with other nonpharmacologic approaches, lifestyle
ence to tai chi, as compared with standard exercise. modifications, and conventional medications.
Other studies have provided qualitative data to Disclosures: Dr Yeh is supported by a K-23 Career Investigator
support meaningful changes in social functioning Award from the National Institutes of Health (NIH) National Center
and well-being with tai chi.51,52 for Complementary and Alternative Medicine (NCCAM), Bethesda,
MD (1K23-AT002624). Dr Phillips is supported by a K-24 Career
Safety Investigator Award from the NIH NCCAM (1K24-AT00589). The
Collectively, these studies suggest that tai chi is contents of this manuscript are solely the responsibility of the authors
and do not necessarily represent the official views of NCCAM.
likely to be safe for patients with hypertension. Two
trials with higher-risk patients with coronary artery
disease reported no adverse effects.18,19 Another References
1 Seventh Report of the Joint National Committee on
study appeared to be safe in frail elders.22 The Prevention, Detection, Evaluation, and Treatment
exercise intensity of tai chi can be easily modified. of High Blood Pressure: the JNC 7 report. JAMA.
Many studies have reported metabolic equivalents 2003;289:2560–2572.
of 1.5 to 4.0 (approximately low-moderate aerobic 2 Ogden LG, He J, Lydick E, et al. Long-term absolute ben-
exercise), a reasonable intensity for even the more efit of lowering blood pressure in hypertensive patients
according to the JNC VI risk stratification. Hypertension.
deconditioned cardiac patient.53–55 2000;35:539–543.
3 Wong SS, Nahin RL. National Center for Complementary
Limitations and Alternative Medicine perspectives for complementary
The quality of studies within this review varied and alternative medicine research in cardiovascular diseases.
Cardiol Rev. 2003;11:94–98.
significantly. Overall, quality was poorer in the 4 Taylor-Piliae RE. Tai Chi as an adjunct to cardiac rehabilitation
Chinese language studies although adequate qual- exercise training. J Cardiopulm Rehabil. 2003;23:90–96.
ity studies were found in both languages. Since the 5 Luskin FM, Newell KA, Griffith M. A review of mind-body
substantial majority of studies reported positive therapies in the treatment of cardiovascular diseases part
1: implications for the elderly. Altern Ther Health Med.
results, the possibility of publication bias certainly 1998;4:46–61.
exists. Also, since most of the studies reported posi- 6 Wang C, Collet JP, Lau J. The effect of Tai Chi on health
tive outcomes, it is difficult to determine whether outcomes in patients with chronic conditions: a systematic
there was an inverse relationship between study review. Arch Intern Med. 2004;164:493–501.
quality and positive findings (eg, lower quality stud- 7 Klein PJ, Adams WD. Comprehensive therapeutic ben-
efits of Taiji: a critical review. Am J Phys Med Rehabil.
ies reporting more positive results). It is, however, 2004;83:735–745.
worth noting that 2 of the 3 studies that reported 8 Verhagen AP, Immink M, van der Meulen A, et al. The effi-
no change in BP received an A quality score. cacy of Tai Chi Chuan in older adults: a systematic review.
In addition, we were unable to perform meta- Fam Pract. 2004;21:107–113.
9 EPC evidence reports. Agency for Healthcare Research and
analyses, and our synthesis of results was limited Quality Web site. http://www.ahrq.gov/clinic/epcindex.htm.
by study heterogeneity, with differences in design, Accessed February 15, 2008.
and types of control, as well as protocol imple- 10 Young DR, Appel LJ, Lee SH. The effects of aerobic exercise
mentation, intervention style, intensity, and dose/ and T’ai Chi on blood pressure in older people: results of a
duration. There were also inherent limitations in randomized trial. J Am Geriatr Soc. 1999;47:277–284.
11 Tsai JC, Wang WH, Chan P, et al. The beneficial effects of Tai
our use of the Jadad scale, given the nature of tai Chi Chuan on blood pressure and lipid profile and anxiety
chi trials and the frequent difficulty and impracti- status in a randomized controlled trial. J Altern Complement
cality of double-blinding. Despite these limitations, Med. 2003;9:747–754.
this study provides a valuable synthesis of informa- 12 Shen ZS, Su XH. The effect of Tai Chi Qigong form 18 on
hypertension. Modern Rehabil. 2000;4:33.
tion on a therapy that has the potential to benefit 13 Wang ZY, Liu DN, Kong DZ. The effect of Tai Chi on lower-
a majority of our population. Further, this review ing blood pressure. Chinese Chi Gong. 2000;10:4–5.
includes the first comprehensive synthesis of both 14 Fang Z, Wang ZY. Clinical comparison of simplified
English and Chinese language literature describing Taichiquan, breathing exercise, tab, hypotensor co, and
long-term BP effects of tai chi. simple convalescence in treatment of hypertension. J Chin
Phys. 1985;2:96–97.
15 Lu JB, Wang YF, Hu M, et al. A Preliminary observation on
Conclusions the therapeutic effect of breathing exercise and shadow box-
Further studies of tai chi exercise are needed. Future ing training on hypertension in the elderly. J WCUMSHua Xi
Yi Ke Da Xue Xue Bao. 1987;18:37–39.
research should include prospective, adequately 16 Taylor-Piliae RE, Haskell WL, Froelicher ES. Hemodynamic
powered RCTs in carefully chosen and defined responses to a community-based tai chi exercise intervention
patient populations, with SBP and DBP as primary in ethnic Chinese adults with cardiovascular disease risk fac-
outcomes. Importantly, tai chi study interventions tors. Eur J Cardiovasc Nurs. 2006;5:165–174.
need to be well characterized and validated so 17 Liu TM, Li SX. Effect of shadow boxing on the cardiovas-
cular excitability, adaptability and endurance in middle-aged
that meaningful comparisons can be made across and elderly patients with hypertension. Chinese Journal of
studies and practical inferences made regarding Clinical Rehabilitation. 2004;8:7508–7509.

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.
 ®

Spring 2008 PREVENTIVE CARDIOLOGY 89

18 Channer KS, Barrow D, Barrow R. Changes in haemody- 37 Appel LJ, Champagne CM, Harsha DW, et al. Effects of
namic parameters following Tai Chi Chuan and aerobic exer- comprehensive lifestyle modification on blood pressure
cise in patients recovering from acute myocardial infarction. control: main results of the PREMIER clinical trial. JAMA.
Postgrad Med J. 1996;72:349–351. 2003;289:20892083–2093.
19 Zheng JQ. The effect of Tai Chi on coronary heart dis- 38 Pinto A, Di Raimondo D, Tuttolomondo A, et al. Twenty-
ease rehabilitation in elderly. Chin J Rehabilitation Theory four hour ambulatory blood pressure monitoring to evaluate
Practice. 2004;10:429. effects on blood pressure of physical activity in hypertensive
20 Zhou WS, Li Z. The effect of Tai Chi Chuan training on patients. Clin J Sport Med. 2006;16:238–243.
blood pressure, ECG and microcirculation in older people. 39 Tejada T, Fornoni A, Lenz O, et al. Nonpharmacological
Chin J Sports Med. 1995;14:249. therapy for hypertension: does it really work? Curr Cardiol
21 Wolf SL, Barnhart HX, Kutner NG, et al. Selected as the Rep. 2006;8:418–424.
best paper in the 1990s: Reducing frailty and falls in older 40 O’Shaughnessy KM. Role of diet in hypertension manage-
persons: an investigation of tai chi and computerized balance ment. Curr Hypertens Rep. 2006;8:292–297.
training. J Am Geriatr Soc. 2003;51:1794–1803. 41 Cheung BM, Lo JL, Fong DY, et al. Randomised controlled
22 Wolf SL, O’Grady M, Easley KA, et al. The influence of trial of qigong in the treatment of mild hypertension. J Hum
intense tai Tai chi Chi training on physical performance and Hypertens. 2005;19:697–704.
hemodynamic outcomes in transitionally frail, older adults. J 42 Sivasankaran S, Pollard-Quintner S, Sachdeva R, et al. The
Gerontol A Biol Sci Med Sci. 2006;61A:184–189. effect of a six-week program of yoga and meditation on bra-
23 Thomas GN, Hong AWL, Tomlinson B, et al. Effects of tai chial artery reactivity: do psychosocial interventions affect
chi and resistance training on cardiovascular risk factors in vascular tone? Clin Cardiol. 2006;29:393–398.
elderly Chinese subjects: a 12-month longitudinal, random- 43 Paul-Labrador M, Polk D, Dwyer JH, et al. Effects of a
ized, controlled, intervention study. Clin Endocrinol (Oxf). randomized controlled trial of transcendental meditation on
2005;63:663–669. components of the metabolic syndrome in subjects with coro-
24 Zhang JH, Yue GF. The effect of Tai Chi and running nary heart disease. Arch Intern Med. 2006;166:1218–1224.
exercise on blood pressure. Chinese Journal of Clinical 44 Walton KG, Schneider RH, Nidich SI, et al. Psychosocial
Rehabilitation. 2004;8:2957. stress and cardiovascular disease Part 2: effectiveness of the
25 Zhang Q, Wei X, Zhao SY, et al. The effect of Tai Chi on Transcendental Meditation program in treatment and pre-
microcirculation system in elderly. J of Binzhou Medical vention. Behav Med. 2002;28:106–123.
Institute. 1997;20:11. 45 Mayer M. Qigong and hypertension: a critique of research.
26 Wu XP, Ho ZQ. Effect of Tai Chi on cardiorespiratory func- J Altern Complement Med. 1999;5:371–382.
tion among older people with chronic conditions. Massage 46 Vaananen J, Xusheng S, Wang S, et al. Taichiquan acutely
Instruct. 1996;71:9–10. increases heart rate variability. Clin Physiol Funct Imaging.
27 Ko GTC, Tsang PCC, Chan HCK. A 10-week tia Tai-Chi 2002;22:2–3.
program improved the blood pressure, lipid profile and 47 Lu WA, Kuo CD. The effect of Tai Chi Chuan on the auto-
SF-36 scores in Hong Kong Chinese women. Med Sci Monit. nomic nervous modulation in older persons. Med Sci Sports
2006;12:CR196–CR199. Exerc. 2003;35:1972–1976.
28 Schaller KJ. Tai Chi Chih: an exercise option for older 48 Audette JF, Jin YS, Newcomer R, et al. Tai Chi versus brisk
adults. J Gerontol Nurs. 1996;22:12–17. walking in elderly women. Age Ageing. 2006;35:388–393.
29 Thornton EW, Sykes KS, Tang WK. Health benefits of 49 Esch T, Stefano GB, Fricchione GL, et al. Stress in cardiovas-
Tai Chi exercise: improved balance and blood pressure in cular diseases. Med Sci Monit. 2002;8:RA93–RA101.
middle-aged women. Health Promot Int. 2004;19:33–38. 50 Vogel JH, Bolling SF, Costello RB, et al. Integrating comple-
30 Liu J, Chen PJ, Qiu PX, et al. The follow up investigation of mentary medicine into cardiovascular medicine. A report of
long time effect of Tai Chi Chuan on heart-lung function in the American College of Cardiology Foundation Task Force on
elder women. Chin J Sports Med. 2003;22:290–293. Clinical Expert Consensus Documents (Writing Committee to
31 Ni HY, Lei XS. The effect of performing 42-form Tai Chi Develop an Expert Consensus Document on Complementary and
exercise on hemodynamics and hemorheology in elderly. Integrative Medicine). J Am Coll Cardiol. 2005;46:184–221.
Chin J Sports Med. 2000;19:194–196. 51 Morris Docker S. Tai Chi and older people in the com-
32 Zhang WJ, Fu SY. The effect of performing Tai Chi on munity: a preliminary study. Complement Ther Clin Pract.
cardiovascular function. J of Beijing Institute of Traditional 2006;12:111–118.
Chinese Medicine. 1991;14:46–47. 52 Kutner NG, Barnhart H, Wolf SL, et al. Self-report benefits
33 Wang JS, Lan C, Wong MK. Tai Chi Chuan training to of Tai Chi practice by older adults. J Gerontol B Psychol Sci
enhance microcirculatory function in healthy elderly men. Soc Sci. 1997;52:242P242–P246.
Arch Phys Med Rehabil. 2001;82:1176–1180. 53 Lan C, Chou SW, Chen SY, et al. The aerobic capacity and
34 Gao LZ, Tan AL. Effect of Tai Chi on cardiorespira- ventilatory efficiency during exercise in Qigong and Tai Chi
tory function in the elderly. Bo Bu Medical School J. Chuan practitioners. Am J Chin Med. 2004;32:141–150.
1997;22:436–437. 54 Chao YF, Chen SY, Lan C, et al. The cardiorespiratory
35 Liang YW. The effect of Tai Chi on the heart-lung function response and energy expenditure of Tai-Chi-Qui-Gong. Am
in elderly. Journal of Physical Education. 2001;8:64–75. J Chin Med. 2002;30:451–461.
36 Reisin E. Nonpharmacological approaches to hypertension. 55 Fontana JA, Colella C, Wilson BR, et al. The energy
Weight, sodium, alcohol, exercise, and tobacco consider- costs of a modified form of T’ai Chi exercise. Nurs Res.
ations. Med Clin North Am. 1997;81:1289–1303. 2000;49:91–96.

Preventive Cardiology® (ISSN 1520-037X) is published quarterly (Jan., April, July, Oct.) by Le Jacq, located at Three Enterprise Drive, Suite 401, Shelton, CT 06484. Le Jacq is an imprint of Blackwell Publishing, which was acquired by John Wiley & Sons
in February 2007. Blackwell’s programme has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell. Copyright ©2008 by Le Jacq. All rights reserved. No part of this publication may be reproduced or
transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers. The opinions and ideas expressed in this
publication are those of the authors and do not necessarily reflect those of the Editors or Publisher. For copies in excess of 25 or for commercial purposes, please contact Ben Harkinson at [email protected] or 781-388-8511.