THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE

Volume 13, Number 10, 2007, pp. 1107–1113

© Mary Ann Liebert, Inc.
DOI: 10.1089/acm.2007.0615

Effects of Yoga and the Addition of Tui Na in

Patients with Fibromyalgia.

GERSON D. DA SILVA, M.S.,1 GERALDO LORENZI-FILHO, M.D., Ph.D.,2

and LAIS V. LAGE, M.D., Ph.D.1

ABSTRACT

Objectives: This study aimed to verify whether techniques of yoga with and without the addition of Tui Na
might improve pain and the negative impact of fibromyalgia (FMS) on patients’ daily life.
Design: Forty (40) FMS women were randomized into two groups, Relaxing Yoga (RY) and Relaxing Yoga
plus Touch (RYT), for eight weekly sessions of stretching, breathing, and relaxing yogic techniques. RYT pa-
tients were further submitted to manipulative techniques of Tui Na.
Outcome measure: Outcome measures comprised the Fibromyalgia Impact Questionnaire (FIQ), pain thresh-
old at the 18 FMS tender points, and a verbal graduation of pain assessed before treatment and on the follow-
up. The visual analog scale (VAS) for pain was assessed before and after each session and on the follow-up.
Results: Seventeen (17) RYT and 16 RY patients completed the study. Both RY and RYT groups showed
improvement in the FIQ and VAS scores, which decreased on all sessions. The RYT group showed lower VAS
and verbal scores for pain on the eighth session, but this difference was not maintained on the follow-up. Con-
versely, RY VAS and verbal scores were significantly lower just on the follow-up.
Conclusions: These study results showed that yogic techniques are valid therapeutic methods for FMS. Touch
addition yielded greater improvement during the treatment. Over time, however, RY patients reported less pain
than RYT. These results suggest that a passive therapy may possibly decrease control over FMS symptoms.

INTRODUCTION tary and Alternative Medicine (CAM) techniques have been

widely used by patients with FMS.5,6 Some studies have

F ibromyalgia (FMS) is a syndrome of unknown etiology

characterized by widespread musculoskeletal pain and
multiple tender points.1 In addition to pain, patients also pre-
sent generalized fatigue, sleep disturbances, stiffness, and
mood disturbances.2 There are many problems affecting the
psychosocial life of these patients and it’s commonly ac-
cepted that they have difficulty managing their symptoms.1
Most specialists agree that medication therapy alone is
not very effective in the treatment of FMS patients. Other
therapeutic approaches, such as physical, psychologic, and
educational measures, have been proposed.3,4 Complemen-
shown that strategies that included stretching, relaxation,
and massage may help FMS patients to overcome their
symptoms.*,†,7–10 Yogic sessions with stretching postures,
breathing control, and relaxation have been already proven
useful in other chronic pain conditions.11,12 There is some
evidence that massage is also a useful treatment for FMS
patients9 and general chronic pain conditions,13 and it is a
therapeutic option that deserves more investigation.14
Our aim was to test the effects of a relaxing yogic prac-
tice consisting of physical postures, breathing exercises, and
relaxation techniques on the treatment of FMS patients in

1Rheumatology Division and 2Pulmonary Division, Faculty of Medicine, University of São Paulo, São Paulo, Brazil.
*Lenard RV. Educação somática no tratamento da fibromialgia: Estudo clínico controlado [dissertation]. Faculty of Medicine, Uni-
versity of São Paulo, 2002.
†Souza LPM. A utilização de técnicas de relaxamento no tratamento de pacientes com diagnóstico de fibromialgia [dissertation].
FMUSP São Paulo University, 2001.

1107
1108
an 8-week intervention. In a separate intervention, we also
aimed to verify the impact of the addition of a massage tech-
nique to the yoga intervention, according to the fundamen-
tals of traditional Eastern medicine.
MATERIALS AND METHODS
Patients
The study included 40 consecutive women diagnosed
with FMS recruited from the rheumatology out-patient clinic
of the Hospital das Clínicas, Universidade de São Paulo (São
Paulo, Brazil). The diagnosis of FMS was made by the at-
tending rheumatologist, according to the American College
of Rheumatology (ACR) 1990 criteria for the classification
of FMS.15 Selected subjects had widespread pain in at least
three of four quadrants of the body for more than 3 months,
and a minimum of 11 tender points among the 18 points
tested. The patients did not present other conditions that
could justify chronic pain, such as neurologic or other
rheumatic diseases. Inclusion criteria also included an age
range between 25 and 60 years, normal cognitive function,
and the ability to understand instruction.
Experimental programs
Patients were randomized into two intervention groups,
Relaxing Yoga (RY) and Relaxing Yoga plus Touch (RYT).
The therapy program consisted of eight weekly sessions,
with a median duration of 50 minutes each. Patients who
did not attend one or two sessions were given replacement
classes in the following week to complete the eight-session
program. All intervention sessions were individually ad-
ministered, and each subject received the full attention of
the therapist. The environment was prepared for each ses-
sion with low lighting levels and relaxing background mu-
sic. The format of the practice sessions consisted of adapted
yogic classes, according to Gharote’s methodology,16 start-
ing with simple postures of stretching, involving all single
movements of vertebral column: lateroflexion, rotation, ex-
tension, and flexion. These postures were done beginning
with the neck only (lateroflexions more “Brahma Mudrá”)
and then with the entire body, adopting the traditional pos-
tures named according to Gharote “trikonásana-iv,”
“vakrásana-ii,” “kapotásana-i,” and “uttánásana-ii.”17 If
necessary, adaptations were applied to keep any posture sta-
ble, comfortable, and relaxing, as stated in Patanjali Yoga
Sutras.18 After the postures, patients were asked to lay down
on a cot, where they were taught to practice 7 minutes of
diaphragmatic yogic breathing that involved regulating the
respiratory cycle, so that the duration of the exhale was twice
as long as that of the inhale, with a regular and comfortable
respiration pattern. After a controlled regulation of breath-
ing, a 15-minute relaxation technique followed, focusing the
subject’s attention sequentially on each major body segment.
DA SILVA ET AL.
At the end of each session, a text describing aspects and
principles of yogic philosophy was read by the therapist in
order to facilitate the understanding of the purpose of the
treatment.19–21 Patients assigned to the RYT group were
submitted to touch while they were in the relaxation phase
of the treatment session. The touch techniques used were
derived from the Tui Na technique and consisted of sliding
(“tui fa”) and pressuring (“na fa”) the trunk and limbs in the
supine position. Sliding maneuvers in the trunk were mul-
tidirectional (“tui fa” and “mo fa”), but in the limbs, “qi”
flow was followed. Patients were asked to focus their at-
tention on each touched part, while maintaining full con-
sciousness of the entire process. Maneuvers were applied
over the clothes.
Evaluation
One (1) week before the first session and 4–6 weeks af-
ter the last session, the myalgic score based on pain thresh-
old at the 18 tender points and the verbal score for pain were
assessed and the Fibromyalgia Impact Questionnaire (FIQ)
was administered. Three Visual Analog Scales (VAS) for
pain were assessed on each visit during treatment and two
were applied in the follow-up. VAS were assessed to mea-
sure how much pain patients felt before and after each ses-
sion, and also during each week between two sessions. The
VAS was not evaluated after the treatment session in the
follow-up visit. The myalgic score was measured by using
the apparatus known as Fisher’s algometer on the 18 tender
points, according to the description of Wolfe et al.15 Pain
thresholds at each point were measured with the algometer
in kg/cm2, expressing how many kilograms of pressure over
an area of 1 cm2 would be needed for generating a pain sen-
sation, so that higher values express less sensibility. Our pro-
cedure took into consideration the total sum of the 18 re-
sults for each patient.
The FIQ is a 10-item questionnaire that measures physi-
cal functioning, job difficulty, pain, fatigue, morning tired-
ness, stiffness, overall well-being, depression, and anxiety.
The instrument has been validated in English and Portuguese
versions22,23 and has a score that may vary from 0 to 100,
where higher values indicate worse conditions. The VAS is
a 100-mm line on which subjects indicate the degree of pain
perceived at that given moment and the weekly average. The
left extremity of the scale (0 mm) corresponds to “no pain”
and the right one (100 mm) to “unbearable pain,” so that
higher values indicate worse conditions.25 The verbal grad-
uation of pain was indicated by the subjects by choosing a
number from 0 to 10, with higher numbers corresponding
to higher pain levels. This verbal graduation was multiplied
by 10 for proper comparison with VAS and FIQ scores.
Statistical analysis
The comparison between groups was made using the Stu-
dent’s t-test and the Mann-Whitney test when the supposi-
YOGA AND TUI NA FOR FIBROMYALGIA 1109

tion of normality was rejected. For the comparison between FIQ results
different moments on each group, the Wilcoxon test was 80
used. The level of statistical significance was p  0.05. 70
60
50

scores
RESULTS RY
40
RYT
30
Of the 40 patients initially included, 7 were excluded be-
20
cause they did not return to treatment sessions (n  5) or
were unable to follow the prescribed recommendations (n  10
2). We, therefore, analyzed data from 17 patients random- 0
ized to the RY group and from 16 patients in the RYT group. before treatment follow-up
The demographic characteristics of the two groups were sim- moments
ilar in respect to age, race, educational, and economical lev- FIG. 1. Comparison of Fibromyalgia Impact Questionnaire
els and are described in Table 1. The majority were Cau- (FIQ) scores before treatment and in the follow-up. Relaxing Yoga
casian, middle-age, and in a low-income bracket. The groups (RY) group (p  0.009); Relaxing Yoga plus Touch (RYT) group
also did not differ regarding medication use for FMS and (p  0.007).
initial values for FIQ and pain threshold.
We observed significant decreases in FIQ scores for both
groups after the treatment. Based on the Wilcoxon test, the Based on the Wilcoxon test, we verified a very significant
RY group was better at the follow-up, compared with the decrease of pain intensity in the comparison of VAS values
initial visit (p  0.009), and the same was true for the RYT before and after each session. In all sessions, both groups
group (p  0.007). There were no significant differences be- showed significant immediate changes, where the results in 6
tween the groups in FIQ results (Fig. 1). of the 8 RY sessions and in all 8 of the RYT sessions were
The increase in pain threshold, according to dolorimetry very marked (p  0.001) (Fig. 2A and 2B). The tendency of
after the treatment, was not significant for either of the greater VAS changes in the RYT group, compared to the RY
groups. Using the Wilcoxon test, we obtained small, but sta- group, was significant just in sessions 2 (p  0.008) and 6
tistically insignificant, changes in the results of the RY (p  (p  0.033), according to Mann-Whitney test.
0.868) and RYT groups (p  0.877). Variance analysis with Also, using the Wilcoxon test, we obtained significant re-
repetitive measures revealed no difference between groups. sults expressing different tendencies of each group when a
comparison was made among the values of momentary
VAS, median-week VAS, and verbal graduation for pain at
TABLE 1. SUMMARY OF DEMOGRAPHIC three time points: before treatment, at the last session of the
VARIABLES ON BOTH GROUPS treatment, and at the follow-up visit. The RY group showed
RY RYT a significant decrease of pain in all evaluations only in the
Group (n  17) (n  16) comparison of the initial results with the follow-up (mo-
mentary VAS, p  0.020; median-week VAS, p  0.026;
Age, year 46.3  8.9 44.4  11.0 verbal graduation, p  0.001), but not in the comparison of
FIQ score 60  15 62.9  16.5 the initial results with the last session, which indicates a con-
Pain threshold on 18 TPs 39.6  16.4 34.1  7.8
(kg/cm2) tinued improvement even after the end of the treatment. The
Race, n (%) RYT group, on other hand, showed a significant decrease
Caucasian 12 (71) 11 (69) of pain in all evaluations only when the initial results were
Mixed 3 (17) 3 (19) compared with the last session (momentary VAS, p 
Black 1 (5.9) 1 (6) 0.007; median-week VAS, p  0.004; verbal graduation,
Asian 1 (5.9) 1 (6)
Years of study, n (%) p  0.001), but not when these initial results were compared
Less than 8 3 (18) 6 (38) with the follow-up, showing an improvement during the
From 8 to 10 6 (35) 3 (19) treatment but a worsening after it. Momentary VAS, weekly
More than 10 5 (29) 4 (25) VAS, and verbal graduation for pain in three moments are
University degree 3 (18) 3 (19) shown in Figures 3A and 3B and 4.
Family income in US$, n (%)
Up to $300 3 (18) 3 (19)
From $300 to $600 9 (53) 7 (44)
From $600 to $900 5 (29) 4 (25) DISCUSSION
More than $900 0 (0) 2 (13)

Note: With regard to demographic variables, there were no Techniques that include stretching, relaxation, and mas-
significant differences between the groups. sage have been used with good results in FMS pa-
1110 DA SILVA ET AL.

A yoga as well as meditation may play an important role in

YR momentary VAS values before and after each session
reducing this hyperactivity.28,29 We may also consider a
causal relationship between pain and muscle tension,30 pro-
80
viding a rationale for the effects of stretching and relaxing
70
in reducing tension-related musculoskeletal pain.
60 The addition of touch seems to help immediate pain de-
VAS scores

50 crease and may be, in part, explained by its social and emo-
before
40 tional comforting effects.31 One alternative explanation is
after
30 based on the gate control theory, which considers tactile and
20
proprioceptive stimulations as pain-modulating factors.32,33
The RYT group perceived more significant immediate pre-
10
to postsession decreases in VAS pain intensity than the RY
0 group in just two of the eight sessions. This limited benefi-
1 2 3 4 5 6 7 8
cial effect may be explained by the fact that the RY group
sessions
alone had already significant results.

B
YRT momentary VAS values before and after each session A
80 momentary visual analogic scale
70 80
VAS median values

60 70
50 60
before
40 50
after
scores

30 RY
40
RYT
20 30
10 20
0 10
1 2 3 4 5 6 7 8
0
sessions
session 1 session 8 follow-up
FIG. 2. A. The Relaxing Yoga (RY) group visual analog scale moments
(VAS) values before and after each session (p  0.05 in all ses-
sions; p  0.001 in sessions 1, 2, 3, 4, 6, and 8). B. The Relaxing
Yoga plus Touch (RYT) group VAS values before and after each B
session (p  0.001 in all sessions). weekly visual analogic scale
80
70
tients.*,†,7–10 The results of our treatment support these ear- 60
lier reports, showing significant decrease in pain intensity 50
scores

and impact of FMS on daily life. Most of these previous RY

40
studies, however, measured pain intensity just before and RYT
30
after the treatment, and not during treatment. We observed
that each session can be very effective in producing an im- 20
mediate perception of analgesic effects, even when long- 10
term effects are not significant. Therefore, the immediate 0
short-term usefulness of stretching and relaxation must be session 1 session 8 follow-up
considered as an option for a patient that is suffering from moments
a high-intensity pain. Analgesic effectiveness of yogic tech- FIG. 3. A. Momentary visual analog scale (VAS) values for both
niques may be related to their relaxing effects, since relax- groups in three moments: before sessions 1, 8, and in the follow-
ation techniques are considered a very effective method for up: Relaxing Yoga (RY) group session 1  follow-up (p  0.02);
dealing with a painful stressor.25 Relaxing Yoga plus Touch (RYT) group sessions 1  8 (p 
Our study was not designed to determine the exact mech- 0.007). Other comparisons were without significant changes. B.
Weekly VAS values in three moments: before sessions 1, 8, and
anism by which the techniques employed are effective. It is in the follow-up: RY group session 1  follow-up (p  0.026).
already known that FMS is associated with a sympathetic RYT group sessions 1  8 (p  0.004). Other comparisons were
hyperactivity that may be related to its pathogenesis,26,27 and without significant changes.
 YOGA AND TUI NA FOR FIBROMYALGIA 1111

Verbal graduation for pain The continuous improvement of pain in group RY, even
9 weeks after the end of treatment, may be explained by the
8 recommendation made to the patients to maintain regular
7 yoga practice. Most of the RY subjects kept the practice
6
of yogic techniques at least up to the follow-up visit, sug-
scores

YR gesting a greater degree of independence from the thera-

5
YRT
4 posite tendency was observed in the RYT group after
3 treatment. The addition of Tui Na massage showed very
2 good results up to the last session, but RYT patients did-
1 n’t maintain this improvement, having follow-up scores
session 1 session 8 follow-up
moments
FIG. 4. Verbal graduation for pain in three moments: before ses-
sions 1, 8, and in the follow-up: Relaxing Yoga (RY) group ses-
sions 1  8 (p  0.031); session 1  follow-up (p  0.001). Re-
laxing Yoga plus Touch (RYT) group sessions 1  8 (p  0.001);
session 8  follow-up (p  0.007). Other comparisons were with-
out significant changes.
The impact of fibromyalgia on daily life, measured by
the FIQ, was changed by both strategies—with and without
touch—but with no statistically significant difference in im-
provements between them. Therefore, both strategies ap-
peared to be equally effective in changing quality of life at
4–6 weeks after the end of the treatment. It is possible, how-
ever, that the groups would be in different conditions at the
last session, when the FIQ was not administered.
The myalgic scores measured by Fishers’ dolorimeter
presented very little changes in both groups, suggesting that
our techniques were not effective to significantly elevate
pain threshold. Other therapeutic intervention studies, how-
ever, have also shown that FMS subjects that improved on
other symptoms did not show an improvement in an over-
all count of 18 tender point (TP) thresholds.34–37 There is
not a strong correlation between several FMS symptoms and
dolorimetry score,38 so that it is also not surprising that the
patients were observed to have improvements in other as-
pects evaluated in our study, but not in myalgic score.
We considered that the most interesting results of our
study are the comparisons of pain intensity at three differ-
ent time points, namely beginning, end, and 4–6 weeks af-
ter the end of treatment. We evaluated pain in three differ-
ent ways. The momentary VAS was applied to show the
pain intensity at the moment of the visit. Considering that
FMS symptoms may change abruptly, we also applied a
mean-week VAS to show how much pain patients feel dur-
ing each week, so that we may expect that these values might
show smaller oscillations. The verbal graduation of pain was
used as an option to reinforce other acquired data, because
many Brazilian patients seem to have problems in under-
standing analog scales. These three different ways of pain
quantification showed similar results, clearly indicating dif-
ferent tendencies in each group.
pist in dealing with their symptoms. Surprisingly, an op-
for pain similar to those obtained before treatment. Other
studies with chronic pain sufferers treated with massage
also show they get worse months after the end of thera-
peutic interventions,9,39,40 but our study showed the same
worsening in just a few weeks after treatment, perhaps be-
cause each session included just around 15 minutes of
massage. It is also important to remember that limited ef-
fects of massage in this study may have been owing to the
applied maneuvers: Just two categories of relaxing ma-
neuvers were selected among the several used in Tui Na.41
Other results could have been obtained if other techniques
had been applied.
It is possible that the additional involvement inherent in
the application of massage may enhance a patient’s depen-
dence, and if this approach is incorporated into the patient’s
chronic pain behaviors, it may become disabling, because
the patient will expect things to be done to him or her and
not by him or her.42 Some touched patients (i.e., the RYT
group) expressed dissatisfaction with the end of treatment,
perhaps because of the therapist’s dependency, and such
feelings were not observed in RY group.
Yogic techniques in this study were used as a whole, so
it is not possible to determine in this study if significant re-
sults could have been achieved just with postures, breathing
control, relaxation, or even just with individualized atten-
tion. Considering such limitations of this design, conclusions
may be considered as probable tendencies that deserve more
investigations.
CONCLUSIONS
The results from this study point to a possibility that touch
intervention may be, in the short term, a very helpful re-
source. On the other hand, for long-term effects, it may be
better to teach FMS patients to fully participate in thera-
peutic programs, changing their self-defeating beliefs, and,
for this purpose, it is possible that an intervention with yo-
gic techniques alone may be a better option. Considering
that yoga and Eastern massage are widely used by FMS in-
dividuals,5,6 it is strongly recommended that more studies
should be done, comparing yoga and/or massage with con-
trols or other active groups, so that more precise conclusions
may be achieved.
1112
ACKNOWLEDGMENTS
We are grateful to Creuza dal Bo for the statistical analy-
sis and especially to Dr. Sat Bir S. Khalsa, Ph.D., for the
text review.
REFERENCES
1. Hudson JI, Pope HG. Fibromyalgia and psychopathology: Is
fibromyalgia a form of “affective spectrum disorder?” J
Rheumatol 1989;16(suppl. 19):15–22.
2. Scotton AS, Fraga RO. Fibromialgia. Rev Bras Med 2000;57:
45–50.
3. Lautenschlager J. Present state of medication therapy in fi-
bromyalgia syndrome. Scand J Rheumatol Suppl 2000;113:
32–36.
4. Forseth KO, Gran JT. Management of fibromyalgia—what are
the best treatment choices? Drugs 2002;62:577–592.
5. Holdcraft LC, Assefi N, Buckwald D. Complementary and al-
ternative medicine in fibromyalgia and related syndromes.
Best Pract Res Clin Rheumatol 2003;17:667–683.
6. Barbour C. Use of complementary and alternative treatments
by individuals with fibromyalgia syndrome. J Am Acad Nurse
Pract 2000;12:311–316.
7. Han SS. Effects of self-help stretching exercise on symptom
reduction in patients with fibromyalgia. Taehan Kanho
1998;37:78–80.
8. Creamer P, Singh BB, Hochberg MC, Berman BM. Sustained
improvement by nonpharmacologic intervention in fibromyal-
gia: Results of a pilot study. Arthritis Care Res 2000;13:
198–204.
9. Brattberg G. Conective tissue massage in the treatment of fi-
bromyalgia. Eur J Pain 1999;3:235–244.
10. Kevin WC, Afton LH, Faxiang H, et al. A pilot study of ex-
ternal qigong therapy for patients with fibromyalgia. J Altern
Complement Med 2006;12:851–856.
11. Kolasinski SL, Garfinkel M, Tsai AG, et al. Iyengar yoga for
treating symptoms of Osteoarthritis of the knees: A pilot study.
J Alt Complement Med 2005;11:689–693.
12. Garfinkel MS, Schumacher HR, Jr. Yoga. Rheum Dis Clin of
North Am 2000;26:125–32.
13. Walach H, Güthlin C, König M. Efficacy of massage therapy
in chronic pain: A pragmatic randomized trial. J Altern Com-
plement Med 2003;9:837–846.
14. Fitzcharles MA. Is it time for rheumatologists to rethink the
use of manual therapies? J Rheumatol 2002;29:1117–1119.
15. Wolfe F, Smythe HA, Yunus MB, et al. The American Col-
lege of Rheumatology 1990 Criteria for the Classification of
Fibromyalgia. Arthritis Rheum 1990;33:160–172.
16. Gharote ML. Yoga Aplicada—da Teoria à Prática. [Applied
Yoga]. São Paulo, Brazil: Phorte, 1996.
17. Gharote ML, Jha VK, Devnath P, Sakhalkar SB. Encyclopae-
dia of Traditional Asanas. Lonavla, India: Lonavla Yoga In-
stitute, 2006.
18. Prabhavananda S, Isherwood C. How to Know God: The Yoga
Aphorisms of Patanjali. Los Angeles: Vedanta Press, 1981.
DA SILVA ET AL.
19. Galbiati H. O Evangelho de Jesus [Portuguese translation of
Gospel]. Caxias do Sul, Brazil: Paulinas, 1971:115 (Mt,c6,
v25-34).
20. Rohden H. Tao Te King de Lao Tse [Portuguese translation
of Tao Te King]. São Paulo, Brazil: Alvorada, 1988:101 (c36).
21. Coelho P. Histórias Para Pais, Filhos e Netos. [Histories for
Fathers, Sons and Grandsons]. São Paulo, Brazil: Globo, 2001.
22. Burckhardt CS, Clark SR, Benett RM. The fibromyalgia im-
pact questionnaire: Development and validation. J Rheumatol
1991;18:728–733.
23. Marques AM, Santos AMB, Assumpção A, et al. Validation
of the Brazilian version of the Fibromyalgia Impact Ques-
tionnaire (FIQ). Rev Bras Reumatol 2006;46(1):24–31.
24. Huskisson EC. Measurement of Pain. Lancet 1974;November
9:2(7889):1127–1131.
25. Bobey MJ, Davidson PO. Psychological factors affecting pain
tolerance. J Psychosom Res 1970;14:371–376.
26. Martínez-Lavín M, Hermosillo AG, Mendoza C, et al. Ortho-
static sympathetic derrangement in subjects with fibromyal-
gia. J Rheumatol 1997;24:714–718.
27. Bengtsson A, Bengtsson M. Regional sympathetic blockade in
primary fibromyalgia. Pain 1998;33:161–167.
28. Wallace RK, Benson H, Wilson AF. A wakeful hypometabolic
physiologic state. Am J Physiol 1971;221:795–799.
29. Vempati RP, Telles S. Yoga-based guided relaxation reduces
sympathetic activity judged from baseline levels. Psychol Rep
2002;90:487–494.
30. Elert J, Kendall AS, Larsson B, et al. Chronic pain and diffi-
culty in relaxing postural muscles in patients with fibromyal-
gia and chronic whiplash-associated disorders. J Rheumatol
2001;28:1361–1368.
31. Chang SO. The conceptual structure of physical touch in car-
ing. J Adv Nurs 2001;33:820–827.
32. Doehring KM. Relieving pain through touch. Adv Clin Care
1889;4:32–33.
33. Kalkigi R, Watanabe S. Pain relief by various kinds of inter-
ference stimulation applied to the peripheral skin in humans:
Pain-related brain potentials following CO2 laser stimulation.
J Peripher Nerv Syst 1996;1:189–198.
34. Meiworm L, Jakob E, Walker UA, et al. Patients with fi-
bromyalgia benefit from aerobic endurance exercise. Clin
Rheumatol 2000;19:253–257.
35. Ammer K, Melnizky P. Medicinal baths for treatment of gener-
alized fibromyalgia. Forsch Komplementarmed 1999;6:80–85.
36. O’Malley PG, Balden E, Tomkins G, et al. Treatment of fi-
bromyalgia with antidepressants a meta-analysis. J Gen Intern
Med 2000;15:659–666.
37. Burckhardt CS, Mannerkorpi K, Hedenberg L, Bjelle A. A ran-
domized, controlled clinical trial of education and physical
training for women with fibromyalgia. J Rheumatol 1994;
21:714–720.
38. Wolfe F. The relation between tender points and fi-
bromyalgia symptom variables: Evidence that fibromyalgia
is not a discrete disorder in the clinic. Ann Rheum Dis 1997;
56:268–271.
39. Hasson D, Arnetz B, Jelveus L, Edelstam B. A randomized,
clinical trial of the treatment effects of massage compared to
relaxation tape recording on diffuse long-term pain. Psy-
chother Psychosom 2004;73:17–24.
YOGA AND TUI NA FOR FIBROMYALGIA 1113

40. Cherkin DC, Eisenberg D, Sherman KJ, et al. Randomized trial Address reprint requests to:
comparing traditional chinese medical acupuncture, therapeu- Gerson D. da Silva, M.S.
tic massage, and self-care education for chronic low back pain. Rua São Daniel
Arch Intern Med 2001;161:1081–1088. 345 São Paulo
41. Chengnan S, Zhiyi W, Qinglin W, et al. Massage for Adults.
Sao Paulo 04288110
In: Chinese Massage Therapy. Beijing, China: Shandong S &
Brazil
T Press, 1990:80–153.
42. King JC, Goddard MJ. Pain rehabilitation—II: Chronic pain
syndrome and myofascial pain. Arch Phys Med Rehabil
1994;75:9–14. E-mail: ge.daddio@ig.com.br