Content uploaded by Rainbow Tin Hung Ho
Author content
All content in this area was uploaded by Rainbow Tin Hung Ho
Content may be subject to copyright.
Review Article
The Effect of T’ai Chi Exercise on Immunity and Infections:
A Systematic Review of Controlled Trials
Rainbow T.H. Ho, PhD,
1,2
Chong-Wen Wang, MD, PhD,
1
Siu-Man Ng, PhD,
2,3
Andy H.Y. Ho, MSocSc,
1
Eric T.C. Ziea, MD, PhD,
4
Vivian Taam Wong, MD, FRCP,
4
and Cecilia L.W. Chan, PhD,
1,2
Abstract
Purpose:The aim of this review is to summarize and assess critically clinical trial evidence of the effect of t’ai chi
(TC) exercise on immunity and TC efficacy for treating infectious diseases.
Methods:Fourteen databases were searched from their respective inceptions through January 2011. No language
restrictions were imposed. Quality and validity of the included clinical trials were evaluated using standard
scales.
Results:Sixteen (16) studies, including 7 randomized controlled trials, 4 controlled clinical trials, and 5 retro-
spective case-control studies, met the inclusion criteria for this review. One (1) study examined clinical symp-
toms, 3 studies tested functional measures of immunity (antigen-induced immunity), and the other studies tested
enumerative parameters of immunity. such as lymphocytes, immunoglobulins, complements, natural-killer cells,
and myeloid dendritic cells. Overall, these studies suggested favorable effects of TC exercise.
Conclusions:TC exercise appears to improve both cell-mediated immunity and antibody response in immune
system, but it remains debatable whether or not the changes in immune parameters are sufficient to provide
protection from infections.
Introduction
T’ai Chi (TC), a complementary and alternative
modality of Traditional Chinese Medicine, combines
characteristics of physical exercise and meditative practice.
TC is popularly practiced by a large number of people in
Chinese communities to improve physical fitness and overall
well-being. The gentle movements and postures of the exer-
cise coordinated with breathing patterns and meditation are
designed to achieve a harmonious flow of energy (qi) in the
body. TC exercise is equivalent to moderate-intensity aerobic
exercise. However, TC is not only an exercise but also a mind–
body intervention. Its beneficial effects on health have been
documented in increasing number of studies.
1–3
It is hy-
pothesized that TC as a modality of mind–body intervention
with a moderate intensity of physical exercise may improve
immune functions of human body.
4,5
However, few reviews
have examined the scientific evidence of the effect of TC on
immunity. It is well-known that infections, such as influenza
and herpes zoster, are associated with human body’s immu-
nity, and that individuals who have substantial declines in
immune function are at increased risk for contracting a
number of infectious diseases. Epidemiologic studies have
suggested that moderate exercise training is associated with
reduction in the incidence of upper respiratory–tract infection
(URTI), whereas endurance athletes are at increased risk for
URTIs during periods of heavy training.
5,6
It is still unclear
whether or not TC exercise may reduce the incidence or the
severity of infectious diseases. Thus, this systematic review
aims to summarize and evaluate critically clinical trial evi-
dence of the effectiveness of TC exercise for improving im-
mune functions and its efficacy for treating infectious
diseases.
Methods
Data sources
The following electronic databases were searched from
their respective inceptions through January 2011: PubMed/
MEDLINE; CENTRAL; CINAHL; EMBASE; AMED; Qigong
and Energy Medicine Database; SPORTDiscus Database;
China Journals Full-text Database-Medicine/Hygiene Series,
China Proceedings of Conference Full-text Database, Chinese
Master Theses Full-text Database, China Doctor Dissertations
1
Centre on Behavioral Health,
2
Department of Social Work and Social Administration, and
3
Family Institute, The University of Hong
Kong, Hong Kong SAR, China.
4
Chinese Medicine Department, Hospital Authority of Hong Kong, Hong Kong SAR, China.
THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 19, Number 2, 2013, pp. 1–8
ªMary Ann Liebert, Inc.
DOI: 10.1089/acm.2011.0593
1
Full-text Database, Electronic Theses and Dissertation Sys-
tem (Taiwan), Taiwan Electronic Periodical Services, and
Index to Taiwan Periodical Literature System. The search
terms used for this review included: tai chi, taichi, tai ji, taiji,
taijichuan, shadowboxing, influenza, infection, infectious,
inflammation, inflammatory, immune, immunity, immuno-
logical, lymphocyte, and antibody. Both traditional and
simplified Chinese translations of these terms were used in
Chinese databases. Reference lists of all included studies,
existing reviews, and other archives of the located publica-
tions were hand-searched for further relevant articles.
Study selection
All controlled clinical trials (CCTs) were included if they
examined the effects of TC exercise on the parameters of im-
munity or for treatment of various infectious diseases. Given
the limited number of prospective clinical trials in the field,
retrospective case-control studies (RCSs) were also included to
provide alternative evidence, but uncontrolled observational
studies were excluded, because of their susceptibility to bias
and lack of significant evidence. Case reports and qualitative
studies were also excluded for lack of significant evidence. To
assess the effect of TC on immunity, studies on any subject
were included, but studies among athletes were excluded
because of their high intensity of physical exercise. To evalu-
ate the effect of TC on infections, any study about TC con-
cerning the incidence or severity of infectious diseases were
included. For all included studies, primary data from the
original sources were reviewed and analyzed.
To assess the effects of TC exercise on improvement of
immunity and TC’s effectiveness for treating infectious dis-
eases, such outcome measures as physical symptoms rele-
vant to infections and biomedical indicators of immunity
were considered. Generally, an individual’s immune status
can be assessed using either enumerative or functional
measures.
7
Functional measures assess how well a specific
immune process works (e.g., how effectively natural-killer
[NK] cells destroy laboratory-grown tumor cells or how
much lymphocytes divide following stimulation with a mi-
togen [a substance that induces mitosis or cell division]). In
clinical practice, one of the most valid and commonly used
functional measures of immunity is used to assess people’s
immune responses to antigens that people are highly sensi-
tive to. This technique involves placing a small piece of an-
tigen directly underneath the skin, a procedure that causes a
local inflammatory response consisting of induration (a
swollen round bump) and erythema (redness around the
bump), and measuring the magnitude of this response im-
mediately or 24–48 hours later, depending on the antigen
that is used. Enumerative measures involve counting dif-
ferent immune-system components (or biomarkers), includ-
ing white blood–cell populations (granulocytes, monocytes,
lymphocytes, NK cells, B-lymphocytes, T-lymphocytes,
helper T-lymphocytes, and suppressor/cytotoxic T-lympho-
cytes), antibody populations in the blood (immunoglobulin
[Ig]A, IgG, and IgM) and in saliva (secretory IgA), and an-
tibodies to specific pathogens. The current review focused on
the number and the percentage of white blood cells, mainly
T-lymphocytes, and levels of serum Igs and complements in
peripheral blood, because these biomarkers are commonly
used parameters in clinical practice and in the field of exer-
cise research. Psychosocial outcomes, such as quality of life
and psychologic well-being, were not considered because it
is difficult to attribute effects on such outcomes to the change
of immunity.
Data extraction and assessment
For each included study, data were extracted by 1 main
researcher and then verified by another researcher. All dis-
crepancies were resolved by discussion. The strength of the
evidence was evaluated for all the included studies using the
Oxford Centre for Evidence-based Medicine Levels of Evi-
dence.
8
These criteria are applied to grade the methodolog-
ical rigor of studies from level 1 or grade A (systematic
review of RCTs, 1a; individual RCT with narrow confidence
interval, 1b) to level 5 or grade D (expert opinion). The
quality and validity of the included RCTs were also evalu-
ated using the Jadad scale,
9
which is based on three criteria:
(1) description of randomization and allocation concealment;
(2) double-blinding; and (3) withdrawals or dropouts (the
score ranges from 0 to 5). This is a standard scale used in
systematic reviews of RCTs. Given that it was difficult to
blind patients to TC, only assessor blinding was evaluated.
The risk of bias in the included trials was assessed using the
framework for methodological quality recommended by Juni
et al.
10
According to this framework, biases fall into four
categories: (1) selection bias (biased allocation to comparison
groups); (2) performance bias (unequal provision of care
apart from intervention under evaluation); (3) detection bias
(biased assessment of outcomes); and (4) and attrition bias
(biased occurrence of loss to follow-up).
Results
Study description
The database searches identified 87 potentially relevant
articles (Fig. 1). Of them, 51 articles were excluded because
they were not clinical trials or not related to infection or
immunity. Full reports of 36 studies were acquired, and 20
were also excluded because they were uncontrolled observa-
tional studies (10), studies with unparallel controls (2), du-
plicate publications (3), studies using athletes (2), and studies
with other outcome measures (3). Sixteen studies published
between 1988 and 2010, including 7 RCTs,
11–17
4CCTs,
18–21
and 5 RCSS,
22–26
met this review’s inclusion criteria. These
studies were conducted in the United States,
13,14,16,19
Taiwan,
22
and mainland China.
11,12,15,1 7,18,20,21,23 –26
Seven (7)
studies
13,14,16,19,22,24,25
were published in English, 8 stud-
ies
11,12,15,17,18, 21,23,26
were published in Chinese, and the re-
maining 1 study
20
was a proceeding.
Of the 16 included studies, 4 used samples of young col-
lege students;
11,12,18,20
1 study used a sample of persons with
HIV infection,
16
and the other studies used samples of
middle-age or older healthy adults. Four (4) studies
11,12,17,18
focused solely on females and 2 studies
22,23
focused solely on
males. Sample sizes in the included studies ranged from 16
to 252. A sample shared by 2 studies
11,12
was considered to
be one sample. In total, these studies covered 939 subjects,
including 577 subjects in the TC exercise groups and 362
subjects in the control groups. Characteristics of the included
RCTs and non-RCTs (CCTs and RCSs) are presented in Ta-
bles 1 and 2, respectively.
2 HO ET AL.
Durations of TC intervention ranged from 5 weeks to 6
months for the 11 prospective studies (7 RCTs and 4 CCTs).
The five RCSs
22–26
reported durations of TC practice span-
ning 12 months to 12 years. The majority of the included
studies were conducted with a two-armed parallel-group
design except for 2 studies
18,20
with a 3-armed parallel
group-design and 3 studies
11,12,16
with a 4-armed parallel
group design. One (1) study
20
compared TC with qigong, and
another study
13
used group health education as a control
while the other studies used a wait-list group or a group
with routine activites as a control.
According to outcome measures, these studies could be
divided into 3 categories: (1) 1 study
18
on clinical symptoms
relevant to infections; (2) 3 studies
13,14,19
on functional mea-
sures of immunity (antigen-induced immunity); and (3) other
studies on enumerative parameters of immunity. The most-
often used outcome measure in the included studies was T-
cells
15,16,21,23–26
followed by Igs (IgA, IgE, IgG, and
IgM),
12,15,18,21
complements (C3, C4),
11,20
NK-cells,
17,23
and
myeloid dendritic cells.
22
Effects of TC intervention
One CCT
18
on clinical symptoms suggested a favorable
effect of TC exercise. Three (3) studies, including 2 RCTs
13,14
and 1 CCT,
19
examined the effects of TC on antigen-induced,
virus-specific cell-mediated immunity and antibody re-
sponse in the human immune system. The results of these
studies indicated that TC exercise could augment immune
responses to virus and influenza vaccines.
Six (6) studies, including 2 RCTs,
15,16
1 CCT,
21
and 3
RCSs,
23,25,26
examined number and/or percentage of T-
lymphocytes. One (1) RCT
15
on older adults suggested that
the number of CD4
+
and the ratio of CD4
+
/CD8
+
increased
significantly after 8 weeks of TC practice. Another RCT
16
on
persons with HIV suggested that lymphocyte proliferation
function was augmented significantly at the 6-month follow-
up visit after 10 weeks of TC exercise. One (1) CCT
21
and 3
RCSs
23,25,26
suggested that the number of lymphocytes,
mainly CD4
+
, and the ratio of CD4
+
/CD8
+
were signifi-
cantly higher in TC groups, compared to control groups. One
(1) RCS
24
examined the number and percentage of B-lym-
phocytes and suggested that the number of ZC-rosette-
forming cells (B-lymphocytes) was lower in a TC group at
resting status but increased significantly after 20 minutes of
exercise, compared to controls.
One (1) RCT
17
indicated that number of NK cells increased
significantly after 6 months of TC practice, but 1 RCS
23
suggested that the percentage of NK cells decreased signifi-
cantly in a TC group after 25 minutes of exercise. One RCS
22
examined the effect of TC on circulating myeloid dendritic
cells (the potent antigen-presenting cells linking innate and
adoptive immunity) and indicated a favorable effect of TC
exercise.
Five (5) studies, including 2 RCTs and 3 CCTs, examined
the concentration of Igs in peripheral blood. One (1) RCT
15
and 3 CCTs
18,20,21
suggested favorable effects of TC exercise
on IgG and IgA, while another RCT
12
only suggested a fa-
vorable effect on IgG. One (1) RCT
11
examined concentra-
tions of complements (C3, C4), and the results indicated that
concentrations of C3 and C4 increased significantly after 12
weeks of TC exercise. One (1) CCT
20
also suggested a fa-
vorable effect of TC exercise on C3.
Study quality
Jadad scores for the included RCTs ranged from 1 to 4,
with a value of 3 or above only for 2 studies. Levels of evi-
dence for the included studies were ranked as ‘‘A’’ for 3
studies, ‘‘B’’ for 12 studies, and ‘‘C’’ for 1 study.
Discussion
This review aimed to assess the efficacy and the effec-
tiveness of TC exercise for treating infectious diseases. Fol-
lowing a comprehensive search of existing literature, it was
found that clinical trials of TC in patients with infectious
diseases were very limited. Only one study
18
examined
clinical symptoms relevant to infections among female stu-
dents. The results indicated that the durations of URTIs
shortened significantly after 6 months of TC exercise. In-
stead, many studies in the field examined the effectiveness of
TC exercise on improvement of human immune function.
Overall, this review demonstrated that the available evi-
dence suggested favorable effects of TC exercise both for
increasing effective components of the immune system and
for improving immune function, as indicated by functional
measures of immunity.
The risk of bias for the studies examined in this review
was assessed, based on the descriptions of randomization,
allocation concealment, blinding, and withdrawals.
10
A high
risk of bias might have existed in some of the included trials,
which might have led to false–positive results. Of the 7 in-
cluded RCTs, only 3
13,14,16
described method of randomiza-
tion and allocation concealment; 2 RCTs
13,16
adopted
assessor blinding, and 4 RCTs
13–16
reported details about
dropouts and withdrawals and adopted intention-to-treat
FIG. 1. Selection process for included studies. RCTs, ran-
domized controlled trials; CCTs, clinical controlled trials;
RCSs, retrospective case-control studies.
T’AI CHI EFFECT ON IMMUNITY AND INFECTIONS 3
Table 1. Summary of RCTsofT’ai Chi Exercise on Functional Measures and Enumerative Measures of Immunity
Authors,
years
& references
Subjects
(ages) n
Interventions
(styles) & frequency Control Duration
Outcome
measures Results
Jadad
score
Level
of
evidence
Huang
et al.,
2006
11
Female
students
from a
nursing
school
(NR)
TG 1: 10
TG 2: 10
TG 3: 10
CG: 10
TC exercise (style: NR):
TG 1: 1 time per week
TG 2: 2 time per week
TG 3: 3 time per week
(45 minutes each time)
Wait-list 12 weeks Complement
3, 4 (C3, C4);
activities of
overall
complement
Concentration of serum C3 & C4 as well as
overall supplement activity increased
significantly in TG 3 at 10th & 12th week,
compared to CG & TG1 ( p<0. 05).
1 B (3b)
Huang
et al.,
2006
12
Female
students
from a
nursing
school
(NR)
TG 1: 10
TG 2: 10
TG 3: 10
CG: 10
TC exercise (style: NR):
TG 1: 1 times per week
TG 2: 2 times per week
TG 3: 3 times per week
(45 minutes each time)
Wait-list 12 weeks Serum IgG, IgM,
IgA, IgE
Only IgG increased significantly among
participants in TG 3 at post-intervention ( p
value: NR).
1 B (3b)
Irwin
et al.,
2007
13
Healthy
older
adults
(59–86)
TG: 59
CG: 53
TCC (a Westernized
Standardized version
of TC)
(40 minutes, 3 times
per week); varicella
vaccine at 16th week
& evaluated 9 weeks
later
Health
education
& group
discussion
25 weeks Levels of VZV-CMI,
indicated by
frequency of
peripheral blood
mononuclear cells
& memory T-cells)
Compared to CG, level of VZV-CMI
increased significantly in TC group at
postintervention ( p<0.05)
4 A (1b)
Irwin
et al.,
2003
14
Healthy
older
adults( ‡60)
TG: 18
CG: 18
TCC exercise
(45 minutes, 3 times
per week)
Wait-list 15 weeks Levels of VZV-CMI. VZV-specific CMI increased 50% from
baseline to 1-week postintervention in TCC
group ( p<0.05) but unchanged in CG
3 A (1b)
Liu, 2006
15
Older adults
(55–65)
TG: 10
CG: 10
24-style t’ai chi chuan
(1 hour, 4 times
per week)
Routine
activities
8 weeks 1. T-lymphocytes
2. Serum IgG,
IgM, & IgA
1. Significant increase in expression of CD4 +
(p<0.05) & CD4 +/CD8 +ratio ( p<0.01) &
decrease in expression of CD8 +(p<0.05)
in TG at postintervention; no significant
change in T-lymphocytes in CG
2. Concentrations of IgG & IgA increased
significantly ( p<0.05) in TG at
postintervention; no significant change for
these variables in CG
2 B (3b)
McCain
et al.,
2008
16
Persons with
HIV
infection
(NR)
TG 1: 62
TG 2: 65
TG 3: 68
CG: 57
TG 1: Focused TC
exercise
(style: NR)
TG 2: Cognitive–
behavioral
relaxation exercise
TG 3: Spiritual
growth
intervention
Wait-list 10 weeks T-lymphocytes,
NKC cytotoxicity,
Cytokines,
Lymphocyte
proliferation
Compared to CG, all treatment groups had
augmented lymphocyte proliferative
function or increased cellular proliferation
capacity at 6-month follow-up visit
(p<0.01)
4 A (2b)
Wang,
2003
17
Older
female
adults
(55–65)
TG: 10
CG: 6
Group TC exercise
(style: NR
(1 hour/day)
Routine
activities
6 months IL-2
NKC
Number of NK cells & concentration of IL-2
increased significantly in TG at
postintervention, compared to CG ( p
values: NR)
1 B (3b)
RCTs, randomized controlled trials; NR, not reported; TC, t’ai chi; TG, t’ai chi group; CG, control group; Ig, immunoglobulin; TCC, T’ai Chi Chih; VZV, varicella zoster virus; CMI, cell-mediated
immunity; HIV, human immunodeficiency virus; NKC, natural-killer cells, NK, natural killer; IL, interleukin.
4
Table 2. Summary of CCTs and RCSsofT’ai Chi Exercise on Functional and Enumerative Measures of Immunity &Clinical Symptoms
Studies
author, year
& reference Subjects (ages) n
Interventions
(frequency) Controls Duration
Outcome
measures Results
Level
of
evidence
Liu
et al.,
2005
18
Female students
in a college
TG 1: 30
TG 2: 30
CG: 30
24-style t’ai chi chuan
TG 1: 3 times per week
TG 2: 5 times per week
(1 hour each time)
Routine
activities
6 months Symptoms;
serum IgA,
IgM, IgG
1. Duration of URTI became shorter significantly at
postintervention in participants in TG 1 & TG 2,
compared to CG ( p<0.05)
2. Levels of IgA and IgG increased significantly at
postintervention, compared to CG ( p<0.05)
B (2b)
Yang
et al.,
2008
19
Older adults
(TG: 79.5 –1.9)
(CG: 74.5 –1.6)
TG: 27
CG: 23
Equal parts of TC & qigong
(1 hour, 3 classes
per week), influenza
vaccine during
1st week of
intervention
Routine
activities
20 weeks Anti-influenza
antibody
titer
1. Significant increase ( p<0.05) in magnitude & duration
of antibody response to influenza vaccine in
TG, compared to CG
2. Significant between-group difference at 3 & 20 weeks
after vaccine, & at 20 weeks TG had
significantly higher titers, compared to the prevaccine
timepoint ( p<0.05), whereas CG did not
B (2b)
Yan,
1989
20
College students TC: 40
CG: 30
T’ai chi exercise
(NR)
Qigong
exercise
1 month Serum IgA,
IgG, IgM,
complement
C3 & saliva
lysozyme
Compared with baseline measures, all measures in
TC group increased significantly; in qigong group,
only levels of serum complement C3 & saliva
lysozyme increased significantly
C(4)
Zhang,
2002
21
Older adults
(‡50)
TG: 12
CG: 12
T’ai chi exercise
(style: NR)
1 hour, 3 times
per week
Wait-list 5 weeks T-lymphocytes;
serum IgG,
IgM, IgA
1. Number of CD4 +and the ratio of CD4 +/CD8 +
increased significantly ( p<0.05; 0.01) & number of CD8 +
decreased significantly ( p<0.05) in TG, whereas number
of CD8 +decreased significantly in CG2.
2. Levels of IgG & IgA increased significantly in TG
(p<0.05); no significant change observed in CG
B (3b)
Chiang
et al.
2010
22
Healthy male adults,
(TG 1: 54.2 –8.4
TG 2: 53.8 –7.9
CG: 53.1 –7.1)
TG 1: 21
TG 2: 22
CG: 20
Yang style t’ai chi
TG 1: Practice
for >5 years
TG 2: Practice
for 2–5 years
Sedentary
lifestyle
Retrospective Myeloid
dendritic
cells in
peripheral
blood
1. Compared to CG,
2. Number of myeloid dendritic cells was significantly
greater in TGs ( p<0.05), whereas the quantity of myeloid
plasmacytoid dendritic cells was similar ( p>0.05).
2. Number of myeloid dendritic cells in TG 1 was
significantly more than in TG 2 ( p<0.05)
B (3b)
Liu and
Zhang,
2002
23
Older male adults
(55–67)
TG: 25
CG: 10
TC exercise
(style: NR)
for 6–12 years
Routine
activities
Retrospective T-lymphocytes,
NK cells
Compared to CG, percentage of CD3 +& CD4 +cells
& ratio of CD4 +/CD8 +increased significantly
(p<0.05), whereas percentage of CD16 +(NKC)
decreased significantly in TC group after 25 minutes
exercise ( p<0.05)
B (3b)
Sun
et al.,
1990
24
Healthy aging
subjects
(54–80)
TG: 24
CG: 24
88-style t’ai chi chuan,
Practice for
average of 7 years
Routine
activities
Retrospective Number
& percentage
of ZC-RFL
(B-lymphocytes)
Number & percentage of ZC-RFL in peripheral blood
increased significantly in TG after 20 minutes of exercise,
compared to CG ( p<0.01)
B (3b)
Sun
et al.,
1989
25
Healthy aging
subjects
(NR)
TG: 30
CG: 30
88-style t’ai chi chuan
Practice for
2–10 years
Routine
activities
Retrospective Total & active
T-lymphocytes
(E-RFL)
Total number of T-lymphocytes & number of active
T-lymphocytes increased significantly in TG,
compared to CG controls ( p<0.01)
B (3b)
Zhu
& Sun,
1998
26
Middle-age
& older healthy
adults
(NR)
TG: 24
CG: 24
88-style t’ai chi chuan
Practice for
average of 7 years
Routine
activities
Retrospective WBC; LC, LC%;
E-RFC, E-RFC%;
Y-RFC, Y-RFC%
1. The numbers and the percentages of LC and
E-RFC were higher in TG at resting status, compared
to CG ( p<0.01)
2. After 20 minutes of exercise, number & percentage
of E-RFC & Y-RFC increased significantly in TG,
compared to CG ( p<0.01)
B (3b)
CCTs, controlled clinical trials; RCSs, retrospective case-control studies; TG, t’ai chi group; CG, control group; URTI, upper–respiratory tract infection; Ig, immunoglobulin; TC, t’ai chi; NR, not
reported; ZC-RFL, ZC-rosette-forming lymphocytes; E-RFL, E-rosette-forming lymphocytes; WBC, white blood cells; LC, lymphocytes; E-RFC, E-rosette-forming cells; Y-RFC, Y-rosette-forming cells;
NK, natural killer; NKC, natural-killer cell.
5
(ITT) analyses. The other RCTs did not have descriptions of
their methods of sequence generation or allocation conceal-
ment and the details on dropouts, and were rated as ‘‘un-
clear’’ for these domains, thus, introducing the potential risk
of bias. Details on dropouts and withdrawals were also de-
scribed in 1 CCT,
19
but the ITT analysis was not adopted in
any CCT, which might have led to the exclusion of some
particular patients and might have introduced attrition bia-
ses. The 4 included CCTs and 5 RCSs were subject to a high
risk of selection bias caused by nonrandomized allocation.
Moreover, the 5 included RCSs
21–26
did not adjust the values
of baseline measures; thus, the reliability of the evidence
presented in these studies was clearly limited. One (1)
study
18
was presented at a conference on medical qigong and
had not undergone the process of peer-review, thus, intro-
ducing potential for a number of biases. In the majority of the
included prospective trials, group TC exercise training was
provided preferentially to the intervention groups, whereas
the control groups did not have a matched number of social
contact hours with coparticipants. Thus, these studies might
have been subject to potential risk of performance bias.
Furthermore, the majority of the included studies had small
samples ( <50 subjects) and the results were prone to a type
II error. Therefore, further vigorously designed, large-scale,
placebo-controlled, randomized studies are needed.
Despite methodological flaws, nearly all of the included
studies demonstrated a beneficial effect of TC exercise on one
or more parameters of immunity. Apart from the included
studies, one CCT
27
of TC exercise on mediators (interleukins,
transforming growth factor, and transcription factors) of the
Th1/Th2/T regulatory reaction also suggested a beneficial
effect of TC on improvement of T-cell helper function. All of
the uncontrolled observational studies
28–37
also reported fa-
vorable effects of TC exercise on different parameters of
immunity, but such data were highly susceptible to bias and,
hence, provided little scientific evidence of the specific effects
of TC exercise for improving immune function.
It has been suggested that the mechanisms underlying
exercise-associated immune changes are multifactorial and
include many neuroendocrine factors.
38,39
TC as a form of
moderate-intensity exercise may promote release of neu-
roendocrinologic factors, such as catecholamines (adrenaline,
noradrenaline), growth hormone, and cortisol, through the
sympathetic–adrenal medullary (SAM) axis.
39,40
These fac-
tors induce changes in cellular trafficking, lymphocyte pro-
liferation, and antibody production.
41
As a consequence of
muscular contraction and catecholamine-induced immediate
changes, for instance, cellular components of the immune
system may be mobilized to the blood.
38
Moreover, TC ex-
ercise may lead to an increased oxygen supply and alter-
ations in metabolism and metabolic factors, such as plasma
glutamine and plasma glucose, which also contribute to ex-
ercise-associated changes in immune function.
38,39
An addi-
tional possibility is that immune enhancement is mediated,
in part, by improvements of psychosocial factors that are
promoted by TC as a mind–body intervention.
42
It has been
suggested that psychologic stress can affect immunity
through the hypothalamic–pituitary–adrenal (HPA) axis.
43
Stress-induced activation of the HPA axis results in the re-
lease of neuroendocrine hormones, such as adrenocortico-
tropin, from the anterior pituitary gland. Adrenocotropin
then circulates through the bloodstream to the adrenal
glands, where the hormone induces release of glucocorti-
coids, which can bind receptors at the cell surfaces of lym-
phoid and myeloid cells.
44
TC as a form of mind–body
intervention may buffer the effects of stress on plasma glu-
cocorticoids and, thus, induce alterations in immune func-
tion.
It should be noted that the immune system is a complex
system, and both functional and enumerative measures of
immunity provide rough estimates of specific processes ra-
ther than global indications of the immune system’s capacity
to resist infectious disease.
7
First, the normal functioning
range is very broad for most immune measures, and it is still
unclear whether or not the magnitude of changes induced by
exercise are sufficient to move outside of the normal ranges.
Even if these changes were sufficient, it is not clear whether
or not the alterations would persist for a sufficient duration
of time to alter risk for infectious disease.
7
Second, most of
the included studies were conducted using healthy adults,
and the clinical implications of the changes in these param-
eters in healthy people are unclear. Changes in cell number
may just reflect changes in the dynamics of lymphocyte
migration and recirculation, or shifts in plasma volume, ra-
ther than absolute changes in total cell numbers.
45
In addi-
tion, absolute changes in cell number will not necessarily
result in a significant change in the capacity of the immune
system to make an effective response to antigenic chal-
lenge.
46
Thus, it would be inappropriate to conclude that TC-
induced changes in any specific immune parameter signal a
state of ‘‘immune enhancement’’ or resistance to infections.
7
This review may have had some limitations. Similar to any
systematic review, one major limitation was the potential
incompleteness of the evidence reviewed. The aim was to
identify all controlled trials in this area in a large number of
databases with no restrictions on publication language. The
current authors are confident that the search strategy used
for this review had located all relevant data; however, a
degree of uncertainty remains. Moreover, selective publish-
ing and reporting can be major causes of bias in the included
studies. In addition, it was not possible to perform meta-
analyses because of the heterogeneity of study designs and
outcome measures in the included studies.
Conclusions
The available evidence suggest that TC exercise may im-
prove both cell-mediated immunity and antibody response
in immune system, but it remains debatable whether or not
the exercise-induced alterations in immune function are
sufficient to alter human body defense, disease susceptibility,
and severity.
47
Because of methodological flaws in existing
studies, further vigorously designed large-scale placebo-
controlled, randomized trials are needed. Future studies
should also test the efficacy of TC exercise for reducing the
incidence or the severity of infectious disease.
Acknowledgments
This review was supported by the Hospital Authority of
Hong Kong (HA105/48 PT5).
Disclosure Statement
No competing financial interests exist.
6 HO ET AL.
References
1. Hall A, Maher C, Latimer J, Ferreira M. The effectiveness of
tai chi for chronic musculoskeletal pain conditions: A sys-
tematic review and meta-analysis. Arthritis Rheum
2009;61:717–724.
2. Wang WC, Zhang AL, Rasmussen B, et al. The effect of tai
chi on psychosocial well-being: A systematic review of ran-
domized controlled trials. J Acupunct Meridian Stud
2009;2:171–181.
3. Yeh GY, Wang C, Wayne PM, Phillips R. Tai chi exercise for
patients with cardiovascular conditions and risk factors: A
systematic review. J Cardiopulm Rehabil Prev 2009;29:152–
160.
4. Irwin M, Pike J, Oxman M. Shingles immunity and health
functioning in the elderly: Tai chi chih as a behavioral treat-
ment. Evid-Based Complement Alternat Med 2004;1:223–232.
5. Nieman DC. Exercise and resistance to infection. Can
J Physiol Pharmacol 1998;76:573–580.
6. Sim YJ, Yu S, Yoon KJ, et al. Chronic exercise reduces illness
severity, decreases viral load, and results in greater anti-
inflammatory effects than acute exercise during influenza
infection. J Infect Dis 2009;200:1434–1442.
7. Miller GE, Cohen S. Psychological interventions and the
immune system: A meta-analytic review and critique.
Health Psychol 2001;20:47–63.
8. Oxford Centre for Evidence-based Medicine. Levels of Evi-
dence. March 2009. Online document at: www.cebm.net/
index.aspx?o =1025. Accessed November 10, 2010.
9. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality
of reports of randomized clinical trials: Is blinding neces-
sary? Control Clin Trials 1996;17:1–12.
10. Ju
¨ni P, Altman DG, Egger M. Systematic reviews in health
care: Assessing the quality of controlled clinical trials. BMJ
2001;323:42–46.
11. Huang QP, Jiang GF, Wan YP, Xiong ML. The effect of
Taijiquan on enhancing overall supplement activity in serum
in female college students [in Chinese]. J Phys Educ
2006;13:69–71.
12. Huang QP, Wan YP, Dai KX, Jiang GF, Tang SY. Human
immune response of senior university female students en-
hanced by Taijiquan exercise [in Chinese]. J Wuhan Institute
Phys Educ 2006;40:54–56.
13. Irwin MR, Olmstead R, Oxman MN. Augmenting immune
responses to varicella zoster virus in older adults: A ran-
domized, controlled trial of tai chi. J Am Geriatr Soc
2007;55:511–517.
14. Irwin MR, Pike JL, Cole JC, Oxman MN. Effects of a be-
havioral intervention, tai chi chih, on varicella–zoster virus
specific immunity and health functioning in older adults.
Psychosom Med 2003;65:824–830.
15. Liu XD. Effect of 8-week tai chi on immune function of older
people [in Chinese]. Chin J Clin Rehabil 2006;10:10–12.
16. McCain NL, Gray DP, Elswick RK, et al. A randomized
clinical trial of alternative stress management interventions
in persons with HIV infection. J Consult Clin Psychol
2008;76:431–441.
17. Wang XJ. Effect of Taijiquan on IL-2 of aged people [in
Chinese]. J Shandong Institute Phys Educ 2003;19:48–50.
18. Liu YH, Nie H, Ma GD, Zhou JQ, Su CL. Research on impact
of Taijiquan on immunoglobulins and the relationship be-
tween immunoglobulins and Up-respiratory tract infections
(URTI) in the female college students [in Chinese]. J Beijing
Sport Univ 2005;28:1089–1090.
19. Yang Y, Verkuilen J, Rosengren KS, et al. Effects of a tradi-
tional taiji/qigong curriculum on older adults’ immune re-
sponse to influenza vaccine. Med Sport Sci 2008;52:64–76.
20. Yan H. Research on the relationship between immunity of
human body and physical training, taijiquan and qigong. 2nd
International Conference on Qigong, September 10–15, 1989,
Xi’an, China.
21. Zhang M. The impact of tai chi exercise on immune function
in older adults [in Chinese]. J Henan Normal Univ (Natural
Sci) 2002;30:85–88.
22. Chiang J, Chen YY, Akiko T, et al. Tai chi chuan increases
circulating myeloid dendritic cells. Immunol Invest
2010;39:863–873.
23. Liu SH, Zhang H. The research on the effect of tai chi exercise
on T-lymphocyte subgroup and NK cells [in Chinese]. China
Sport Sci Technol 2002;38:50–52.
24. Sun XS, Xu Y, Zhu R, Detection of ZC rosette-forming
lymphocytes in the healthy aged with Taichiquan. J Sports
Med Phys Fitness 1990;30:401–405.
25. Sun XS, Xu Y, Xia YJ. Determination of E-rosette-forming
lymphocytes in aged subjects with Taichiquan exercise. Int J
Sports Med 1989;10:217–219.
26. Zhu Y, Sun X. Effect of Taichiquan exercise on im-
munofunction in the aged [in Chinese]. J Shanghai Teachers
Univ 1998;27:70–74.
27. Yeh SH, Chuang H, Lin LW. Regular tai chi chuan exercise
improves T-cell helper function of patients with type 2
diabetes mellitus with an increase in T-bet transcription
factor and IL-12 production. Br J Sports Med 2009;43:
845–850.
28. Yang KD, Chang WC, Chuang H, et al. Increased comple-
ment factor H with decreased factor B determined by pro-
teomic differential displays as a biomarker of tai chi chuan
exercise. Clin Chem 2010;56:127–131.
29. Lee EO, Chae YR, Song R, et al. Feasibility and effects of a tai
chi self-help education program for Korean gastric cancer
survivors. Oncol Nurs Forum 2010;37:E1–E6.
30. Yeh SH, Chuang H, Lin LW, Hsiao CY, Eng HL. Regular tai
chi chuan exercise enhances functional mobility and CD4CD25
regulatory T-cells. Br J Sports Med 2006;40:239–243.
31. Jin H. The impact of tai chi exercise on T-lymphocyte subsets
in human peripheral blood [in Chinese]. Zhong Guo Yi Yue
Xue Bao 2000;15:16–17.
32. Liu J, Chan XF, Chan PJ. Effect of long-term tai chi exercise
on the concentration of serum IGF-I among middle-aged
and older women [in Chinese]. Chin J Sports Med 2004;23:
686, 697.
33. Liu J, Chan XF, Qiu PX. Effect of long-term tai chi exercise on
NK T-cell counts among middle-aged and older women [in
Chinese]. Chin J Sports Med 2007;26:738–739.
34. Liu J, Chan PJ, Wang Y. Effect of tai chi exercise on the
concentrations of IFN-r and IL-4 in prepheral blood of
middle-aged and older women [in Chinese]. Chin J Sports
Med 2009;28:557–558.
35. Li ZQ, Sun X. Effect of Wu-style tai chi on NK cell activities
in peripheral blood of older adults [in Chinese]. Chin J Sport
Med 1995;14:53–56.
36. Qi DY, Li XH, Wang YG, Liu WH. A study of influence of
Taijiquan exercise on immune function of diabetes II patients
[in Chinese]. J Beijing Sport Univ 2008;31:932–933,950.
37. Esch T, Duckstein J, Welke J, Braun V. Mind/body tech-
niques for physiological and psychological stress reduction:
Stress management via tai chi training—a pilot study. Med
Sci Monit 2007;13:CR488–CR497.
T’AI CHI EFFECT ON IMMUNITY AND INFECTIONS 7
38. Pedersen BK, Hoffman-Goetz L. Exercise and the immune
system: Regulation, integration, and adaptation. Physiol Rev
2000; 80:1055–1081.
39. Pedersen BK, Toft AD. Effects of exercise on lymphocytes
and cytokines. Br J Sports Med 2000;34:246–251.
40. Hoffman-Goetz L, Pedersen BK. Exercise and the immune
system: A model of the stress response? Immunol Today
1994;15:382–387.
41. Goldrosen MH, Straus SE. Complementary and alternative
medicine: Assessing the evidence for immunological bene-
fits. Nat Rev Immunol 2004;4:912–921.
42. Kohut ML, Lee W, Martin A, et al. The exercise-induced
enhancement of influenza immunity is mediated in part by
improvements in psychosocial factors in older adults. Brain
Behav Immun 2005;19:357–366.
43. Yang EV, Glaser R. Stress-associated immunomodulation
and its implications for responses to vaccination. Expert Rev
Vaccines 2002;1:453–459.
44. Eskandari F, Sternberg EM. Neural–immune interactions in
health and disease. Ann N Y Acad Sci 2002;966:20–27.
45. Burns VE. Psychological stress and immune function. In:
Gleeson M, ed. Immune Function in Sport and Exercise. Edin-
burgh & New York: Churchill Livingstone Elsevier & British
Association of Sport and Exercise Sciences, 2006:221–246.
46. Vedhara K, Fox JD, Wang ECY. The measurement of stress-
related immune dysfunction in psychoneuroimmunology.
Neurosci Biobehav Rev 1999;23:699–715.
47. Walsh NP, Gleeson M, Shephard RJ, et al. Position state-
ment: Part one. Immune function and exercise. Exerc Im-
munol Rev 2011;17:6–63.
Address correspondence to:
Chong-Wen Wang, MD, PhD
Center on Behavioral Health
The University of Hong Kong
5 Sassoon Road
Pokfulam, Hong Kong, SAR
China
E-mail: wangcw@hku.hk
8 HO ET AL.
