THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE

Volume 11, Number 2, 2005, pp. 275–284

© Mary Ann Liebert, Inc.

The Effects of Meditation and Visual Imagery on an

Immune System Disorder: Dermatomyositis

MICHAEL P. COLLINS, M.D.,1 and LUCIA F. DUNN, Ph.D.2

ABSTRACT

Objective: To analyze the relationship between a patient’s “spontaneous recovery” from dermatomyositis
and her practice of transcendental meditation and visual imagery without confounding effects of conventional
therapies.
Design: Study of time-varying relationships between (1) measures of arm strength and skin condition (rash
and pain) and (2) mind–body interventions—controlling for psychologic stress—in a patient with dermato-
myositis, using regression analysis to determine half-lives of treatments and stress.
Setting: Institutional referral center.
Intervention: Transcendental meditation and visual imagery (no drugs).
Outcome measures: Daily measurements of arm strength and skin condition over 294 days. Events pro-
ducing psychologic stress were also rated using a numerical scale.
Results: The patient recovered, which is a low-probability event without conventional therapy. Regression
analysis of time dependence between measures of arm strength, rash, and pain and application of mind-body
treatments revealed statistically significant relationships for both meditation (p values 0.02 to 0.001) and visual
imagery (p values 0.02 to 0.002). Stress had a significant negative impact on skin symptoms but not arm strength.
Beneficial effects of meditation had half-lives of 48–59 days for skin condition and no detectable decay for
arm strength. Benefits of visual imagery were more transient (half-lives 4–18 days). The effects of stress had
half-lives of only 1–3 days.
Conclusions: The results demonstrate a statistically significant relationship between mind-body therapies
and the patient’s recovery from dermatomyositis, possibly mediated by influences on the humoral immune sys-
tem. A key factor in the recovery was the slower decay rate of meditation and visual imagery compared to
stress. As dermatomyositis is a humorally mediated immune microvasculopathy, the benefits of meditation and
imagery in our patient comport with a growing body of evidence showing that these techniques influence im-
mune system function.

INTRODUCTION Jacobs, 2002; Fontanarosa, 2000). This paper focuses on two

of the most widely practiced of these alternative therapies—

I nterest in mind–body medicine has grown substantially

in the last decade (Astin et al., 2003; Fontanarosa, 2000).
There is still, however, little statistically verifiable evidence
that treatments relying on the mind–body connection alone,
without concomitant and confounding drug/surgical therapy,
have brought recovery from a serious illness (Barrows and
transcendental meditation (Benson, 1975; Kabat-Zinn,
1994) and visual imagery (Achterberg and Lawlis, 1978;
Simonton, 1980)—and examines them in relation to the ef-
fects of stress on the progression of a potentially fatal au-
toimmune disease.
The patient in this study had dermatomyositis. She de-

1Department of Neurosciences, Marshfield Clinic, Marshfield, WI.

2Department of Economics, The Ohio State University, Columbus, OH.

275
276
clined conventional treatments and elected instead to try
meditation and meditation augmented by visual imagery.
This research thus resulted from an unusual opportunity to
address the question of whether these practices alone (with-
out concomitant medical therapy) could significantly influ-
ence the disease. The patient followed a transcendental med-
itation program outlined by Vivekananda (1982). Her
clinical course was monitored over an approximate 10-
month period by daily measurements of arm strength and
skin condition, which formed the dependent variables in this
study. Explanatory variables included duration of periods of
meditation, occurrences of visual imagery, and the intensity
of psychologically stressful events, together with the exact
times of these observations. Disease progression was ar-
rested after initiation of therapy, and recovery ensued.
Clearly, a single case report of “anything” means little in
and of itself because “spontaneous” recoveries are docu-
mented for most serious illnesses, and dermatomyositis is
no exception (Lewy, 1893; Schuermann, 1939; Unverricht,
1891). However, the important, unanswered question is by
what mechanisms such “untreated” patients do recover. The
term “spontaneous remission” implies not that a cure was
effected without cause, but that recovery resulted from
poorly understood means, attesting to the ability of humans
to sometimes heal themselves (Krippner and Achterberg,
2000). The primary purpose of this study was to determine
whether our patient’s “spontaneous recovery” could be sta-
tistically linked to her two mind–body interventions. A sec-
ondary goal was to review the literature on immune system
effects of these techniques.
Therapeutic studies typically compare outcomes for a
treatment group and a control group based on preinterven-
tion and postintervention measurements of a relevant end-
point variable. However, given the relatively high mortality
rate of untreated dermatomyositis, it would be unethical to
include a meditation-only limb in a randomized controlled
trial. The current study had a unique design, applying re-
gression analysis to the time-dependent relationship between
clinical outcome and mind–body treatments using a large
series of observations on a single patient over a 294-day pe-
riod. Each administration of treatment, as well as each stress-
ful event, was assumed to produce an initial effect that de-
cayed over time. Analytic techniques similar to those used
in pharmacokinetics (Gibaldi, 1982) were chosen to capture
this phenomenon so that half-lives of the mind–body treat-
ments and of stress could be computed and compared.
Clinical background on the patient
The patient was a 54-year-old woman who developed
periorbital edema and a severe, erythematous, pruritic rash
in early 1999. On her dorsal fingers, the rash was painful
and tender. Cutaneous manifestations were accompanied by
mild proximal weakness, affecting the patient’s ability to
climb stairs, stand up from the floor, and lift heavy objects.
COLLINS AND DUNN
Neurologic and general review of systems was otherwise
notable for Raynaud’s phenomenon and photosensitivity.
The patient’s only other illness was hypothyroidism. There
was no history of substance abuse and no clinical evidence
of depression, anxiety, or other psychologic disorder before,
during, or after the intervention. Family history revealed
multiple relatives with cancers, but no neuromuscular or der-
matologic disorders. Medications were levothyroxine and
multivitamins.
The patient was diagnosed with dermatomyositis by a der-
matologist and rheumatologist and referred for neuromuscu-
lar evaluation in June 1999. Her general examination was
normal except for an erythematous-to-violaceous macu-
lopapular rash involving the forehead, eyelids (heliotrope
pattern), cheeks, posterior neck, anterior chest, lateral arms,
dorsal forearms, elbows, dorsal interphalangeal and metacar-
pophalangeal joints, anterior thighs, and knees. Gottron’s
papules and bilateral periungual erythema with dilated cap-
illary loops were identified (Fig. 1A). Neurologic examina-
tion revealed mild proximally accentuated weakness. The pa-
tient could perform only two deep knee bends.
Laboratory work-up revealed positive antinuclear anti-
bodies at titers ranging from 1:320 to 1:2560. Serum crea-
tine kinases were 50–58 U/L (190 upper limit of normal).
Normal or negative studies included complete blood count,
metabolic panel, urinalysis, thyroid functions, sedimentation
rate, rheumatoid factor, complement, chest x-ray, electro-
cardiogram, and antibodies to dsDNA, Ro, La, Sm, RNP,
PM1, and Jo1 antigens. Malignancy-directed screening tests
were unremarkable. Nerve conduction studies were normal.
Needle electromyography (EMG) examination revealed
findings consistent with a mildly irritable myopathy (Table
1). Left deltoid muscle biopsy demonstrated four small foci
of mononuclear inflammatory-cell infiltration in the per-
imysium and several myofibers with punched-out areas of
focal myofibrillar loss (Fig. 2A and 2B). There were also a
few regenerating fibers, but no vacuolated fibers or amy-
loidogenic deposits. No non-necrotic fibers were invaded by
mononuclear cells. Sections stained with antibodies against
complement membrane attack complex (MAC) revealed oc-
casional MAC deposits in the walls of perimysial arteries
and the endomysial microvasculature (Fig. 2C and 2D).
Although neuromuscular manifestations were relatively
mild, dermatomyositis was confirmed by the dermatologic
findings (including Gottron’s papules), myopathic EMG,
and diagnostic muscle biopsy features (Carpenter and
Karpati, 2001). Conventional treatments such as prednisone
and intravenous immunoglobulins were discussed, but the
patient was hesitant due to potentially serious side effects,
turning instead to mind–body medicine (transcendental
meditation and meditation augmented by visual imagery) as
a therapeutic regimen. The patient had not previously prac-
ticed meditation.
The rash stabilized in June 1999 and had improved by
the following August. In March 2000, cutaneous findings
 A

FIG. 1. A. Gottron’s papules and periungual erythema (rash 60% improved from worst condition, rated at 15 on the 25-point scale).
B. Resolution of the patient’s cutaneous dermatomyositis manifestations.
278 COLLINS AND DUNN

TABLE 1. RESULTS OF NEEDLE EMG EXAMINATIONS

Spontaneous activity MUAP Analysis Recruitment

Muscle June 1999 December 2000 June 1999 December 2000 June 1999 December 2000

R. deltoid 1  FPs N Many Few short- Very early N

and short- duration
PSWs duration
R. biceps N N Few short- N Early N
brachii duration
R. triceps N N Few short- N Early N
duration
R. extensor — N — Few short- — N
indicis duration
R. flexor — N — Few short- — N
digitorum duration
profundus
R. iliopsoas — N — Few short- — Early
duration (focal)
R. vastus N N Many N Very early N
lateralis short-
duration
R. anterior — N — N — N
tibialis

EMG, electromyography; FPs, fibrillation potentials; MUAP, motor unit action potential; N, normal; PSWs, positive sharp waves.

were limited to bilateral periungual and right elbow ery-
thema. By June 2000, skin examination was normal (Fig.
1B). The patient began to feel stronger in August 1999. Ob-
jective improvements in motor examination were noted in
early 2000. By December 2000, strengths were almost nor-
mal, and the patient was able to perform 10 deep knee bends.
Needle EMG examination in December 2000 was much im-
proved from the 1999 study (Table 1); there were a few per-
sisting myopathic features, but no signs of active denerva-
tion. When last seen in May 2002, the patient had a normal
neurologic and cutaneous examination.
MATERIAL AND METHODS
Outcome measures
The most obvious manifestations of the disease were cho-
sen as the primary outcome measures: (1) weakness in the
proximal muscles and (2) rash and pain in the nailbeds and
at the sites of Gottron’s papules. These indicators were eval-
uated frequently from August 30, 1999 to June 16, 2000.
Arm strength was measured by having the patient, while sit-
ting, apply maximal force to a spring-loaded scale placed
approximately 6 inches above her umbilicus, yielding a to-
tal of 273 scale readings for each arm. Pain and rash in the
hands were evaluated by the patient using a standard 25-
point unipolar rating scale (Andrews, 1984; Krosnick and
Fabrigar, 2002) (25
normal preillness condition; 1

most severe manifestations experienced prior to initiating
mind–body treatments). The full 1–25 range occurred, al-
lowing this scale to be treated as a continuous variable. Eval-
uations of rash and pain were obtained 423 times. The time
and date of each observation were recorded.
Independent variables
Meditation and visual imagery. Minutes per meditation
session was an independent variable. Yoga-based transcen-
dental meditation aimed at attaining a blank state of mind,
was used. This differs from variants of meditation some-
times referred to as “mindfulness” (Bishop, 2002). A total
of 259 periods of mediation were recorded. The mean length
of a meditation session was 15 minutes but some were as
long as 65 minutes. At times, the patient augmented the med-
itative state with visual imagery focused on states of well-
ness in affected areas and internal organs where tumors may
be associated with dermatomyositis. These latter episodes
were entered into the analysis as a discrete variable taking
the value one when present and zero otherwise. The time
and date of the meditation and visual imagery were recorded
for each episode.
Psychologic stress. Psychologically stressful events were
evaluated by the patient on a 25-point unipolar scale and
used as a continuous variable. For example, the highest
stress event was the death of a parent, with a score of 25; a
noninjury automobile accident had a score of 15. The time
and date of each stressful event were recorded.
Statistical analysis
We estimated a regression equation where measured arm
strength (in pounds of force) depended on the cumulative ef-
A B

C D

FIG. 2. A. Perimysial vessel with a mononuclear, perivascular, inflammatory cell infiltrate and several adjacent muscle fibers with punched-out areas of decreased or absent staining, corresponding
to focal myofibrillar loss (modified Gomori trichrome; original magnification 16). B. Mononuclear cell infiltration in the epimysial connective tissue overlying a section of normal muscle (mod-
ified Gomori trichrome; original magnification 16). C. Stained for complement membrane attack complex (MAC) using an indirect immunofluorescent technique demonstrating MAC deposits in
four adjacent endomysial capillaries (original magnification 45). D. Stained for complement membrane attack complex using an indirect immunofluorescent technique demonstrating MAC de-
posits in two larger, perimysial blood vessels (original magnification 16).
280
fects of past meditation and stressful events. Anticipated di-
minishing effects of these independent variables over time
were represented by exponential decay factors of the form
exp(
t) (analogous to pharmacokinetic analysis), where t
is the time interval between the observation on arm strength
and the relevant prior treatment (such as a period of medita-
tion). The exponential factors
were estimated from the data.
An alternative model with a power-law decay factor did not
fit the data as well as exponential decay. The half-life  of
the impact of the explanatory event (i.e., time for impact to
decay to one-half its initial value) is related to the exponen-
tial factor
by 
(ln 2)/
. The cumulative effect of med-
itation associated with arm strength observation i is then
Mi(
m)

j mj exp(
m[ta,i  tm,j])
where ta,i is the time of the ith observation of arm strength,
tm,j is the time of the jth meditation event, mj is the length
of the jth meditation event, and the sum is over events j that
occurred prior to the time of the observation (i.e., tm,j  ta,i).
Analogous variables Vi(
v) and Si(
s) were constructed to
represent the cumulative effects of visual imagery and stress
events. In terms of these constructed variables, the regres-
sion equation for arm strength a is then
ai
  mMi(
m)  vVi(
v)  sSi(
s)  i
where i is a regression error term. This equation was ap-
plied to left and right arm strength separately and to two dif-
ferent measures of hand condition—rash and pain. Estima-
tion was by nonlinear least squares, with the decay
parameters
restricted to non-negative values. SAS soft-
ware (SAS, Cary, NC) was used.
RESULTS
The measurements of arm strength and skin condition
fluctuated over the course of the study, but as the patient’s
TABLE 2. CONDITION MEASURES FOR RASH
Mean values
Condition measure August 30, 1999–September 12, 1999
Hand rash 12.1 (6.0)
Hand pain 13.5 (6.8)
Mean values
Condition measure October 10, 1999–October 23, 1999
Right arm strength 22.9 lbs. (2.8)
Left arm strength 19.5 lbs. (3.2)
aRash and pain values measured on 25-point scale; value of 25 indicates best condition (i.e., preillness). Arm strength measured by scale
reading in pounds avoirdupois. Standard deviations are in parentheses; all beginning and ending differences are significant at the 1% level.
COLLINS AND DUNN
treatments continued, the mean values of these measures
steadily improved, consistent with the patient’s clinical re-
covery (Table 2).
In interpreting regression results, the constant term is
taken as a base level reading of condition—in our case, a
reading that would be obtained in the absence of meditation,
visual imagery, and stress (Tables 3 and 4).
Meditation
Meditation had a highly statistically significant, positive
impact on arm strength, rash, and pain. Using the regression
constant as a base level, each 10 minutes of meditation was
found initially to improve upper extremity strength by 0.05%
(p
0.0001) in right and 0.07% (p
0.0001) in left arm
muscles. In the arm strength regressions, the least-squares
estimate of the exponential decay factor
for the effects of
meditation (subject to
 0) was

0, corresponding to
a half-life of infinity (i.e., the effect of meditation was cu-
mulative, with no detectable decay over the period of the
observations [Table 3]). For hand condition, 10 minutes of
meditation resulted in an improvement in the rash by 1.1%
(p
0.0001) and in pain by 0.5% (p
0.02), with corre-
sponding half-lives of 47.6 days for rash and 58.7 days for
pain (Table 4). When a time trend is included in the esti-
mation, the effects of mediation (and visual imagery) be-
come more significant for pain and rash in the hands but
less significant for arm muscle strength.
Visual imagery
When a period of meditation included visual imagery,
there was an additional 0.4% (p
0.008) improvement in
right-arm strength and 1.4% (p
0.0002) improvement in
left-arm strength. Visual imagery produced an additional
6.3% (p
0.002) improvement in rash and 4.6% (p
0.02)
improvement in pain for the hands. However, visual imagery
was shorter-lived than meditation, with half-lives (in days)
as follows: right arm strength 17.7, left arm strength 8.6,
rash 5.3, and pain 3.5 (Tables 3 and 4).
AND PAIN IN HANDS AND ARM STRENGTHa
Mean values
June 3, 2000–June 16, 2000
21.0 (2.4)
21.6 (2.3)
Mean values
June 3, 2000–June 16, 2000
29.9 lbs. (0.8)
29.2 lbs. (1.2)
EFFECTS OF MEDITATION AND VISUAL IMAGERY 281

TABLE 3. REGRESSION RESULTS: EFFECT OF MEDITATION, VISUAL IMAGERY, AND STRESS ON ARM MUSCLE STRENGTHa

Half-life decay factors (in days)

With approximate
Regression coefficients 95% confidence intervalsb

Dependent Visual Psychologic Visual

variable Constant Meditation imagery stress Meditation imagery Stress

Right arm 24.19 0.0013 0.091 0.023  17.7 c

strength p
0.0001 p
0.0001 p
0.08 p
0.32 (63.0, ) (7.2, )

Left arm 22.09 0.0016 0.32 0.043  8.6 c

strength p
0.0001 p
0.0001 p
0.0002 p
0.07 (47.7, ) (5.1, 26.0)

aTwo strength readings for each arm were taken at each observation point and averaged to improve statistical accuracy. The patient

was right-handed. n
273 for right arm; n
273 for left arm.
bConfidence intervals for the exponential decay factor
, restricted to positive values of
, were converted to intervals for the half-

life 
ln 2/
(and are therefore not symmetric). A point estimate of  for the half-life  is reported when the corresponding unre-
stricted estimate of the exponential decay factor
was negative or zero.
cThe estimate of  is not meaningful when the corresponding regression coefficient is not significantly different from zero.

Psychologic stress effectiveness of this type of treatment than many cases in

The effect of psychologic stress on arm strength was not
significant. Stress did have a significant negative impact on
the rash (p
0.0001) and pain (p
0.0002) in the hands.
For example, a mild, level-9 stress event (late for appoint-
ment) produced an initial 2.2% increase in pain and 9.0%
worsening of rash. A severe, level-25 stress event (death of
father) was associated with a 6.1% increase in pain and 25%
increase in rash. The effect of stress, however, dissipated
rapidly, with a half-life of only 1–3 days.
DISCUSSION
The patient described in this report recovered from der-
matomyositis without conventional treatment, a relatively
uncommon event in the era of corticosteroids where virtu-
ally all patients are treated. Instead, the disorder was suc-
cessfully treated with two forms of mind–body intervention
alone. Therefore, the case was a purer demonstration of the
the literature where conventional treatments and several
mind–body therapies have been administered simultane-
ously for complex conditions (Barrows and Jacobs, 2002;
Fontanarosa, 2000).
This study identified a positive, statistically significant
relationship between both meditation and visual imagery and
improvement of dermatomyositis as measured by arm
strength and rash and pain in the hands. Stressful events had
a significant negative impact on hand condition but no ef-
fect on arm strength. A key factor in the recovery was the
long half-lives of the effects of meditation and imagery rel-
ative to those of stress. For the hand condition, meditation
effects outlasted stress by factors of 43 for rash and 19 for
pain, and visual imagery produced benefits lasting up to 5
times longer than the detrimental effects of stress. For arm
muscle strength, responses to meditation showed no statis-
tically detectable decay whatsoever (which has long been
the claim of its practitioners [Da, 1991; Vivekananda,
1982]), and visual imagery effects lasted many days. Ulti-
mately, the cumulative beneficial influences of meditation

TABLE 4. REGRESSION RESULTS: EFFECT OF MEDITATION, VISUAL IMAGERY, AND STRESS ON HAND CONDITIONa

Half-life decay factors (in days)

With approximate
Regression coefficients 95% confidence intervalsb

Dependent Visual Psychologic Visual

variable Constant Meditation imagery stress Meditation imagery Stress

Rash 8.99 0.010 0.57 0.09 47.6 5.3 1.1

p
0.0001 p
0.0001 p
0.002 p
0.0001 (33.5, 82.3) (3.1, 19.5) (0.8, 2.0)
Pain 12.21 0.006 0.56 0.03 58.7 3.5 3.1
p
0.0001 p
0.02 p
0.02 p
0.0002 (32.7, 283) (1.7, ) (1.7, 14.2)
an
423 for rash; n
421 for pain.
bConfidence intervals for  were computed as in Table 3.
282
and visual imagery outweighed the negative impact of stress-
ful events, contributing to the remission.
This study might be criticized for its failure to account
for other time-varying factors related to the recovery, but to
our knowledge, no non-pharmaceutical interventions have
been demonstrated to influence the course of dermato-
myositis (Engel and Hohlfeld, 2004). It might also be ar-
gued that the patient in this study was destined to recover
without the use of alternative therapies, because “sponta-
neous remissions” of dermatomyositis have been reported.
However, our statistical results show that the estimated ef-
fects of the alternative therapies on the recovery were highly
statistically significant (p value range 0.02–0.0001). In ad-
dition, a review of the literature reveals the improbability of
spontaneous remission in older persons. The initial cases of
dermatomyositis in the German literature were fatal (Hepp,
1887; Unverricht, 1887; Wagner, 1887), but recoveries were
soon observed (Lewy, 1893; Unverricht, 1891). In the
largest review in the precorticosteroid era (153 cases with
adequate follow-up), Schuermann (1939) determined a case-
mortality rate of 61%. Most surviving patients had perma-
nent sequelae (e.g., muscle atrophy, joint contractures, and
calcinosis). In more recent studies of noncorticosteroid-
treated patients, the mortality rate improved to 37%, and
33% of patients had good recoveries (Everett and Curtis,
1957; O’Leary and Waisman, 1940; Selander, 1950; Sheard,
1951; Walton and Adams, 1958; Wedgwood et al., 1953;
Winkelmann et al., 1968). However, the prognosis of der-
matomyositis is significantly worse in older patients. In
Schuermann’s review, no patient older than 53 years re-
covered. Combined data from more recent studies yields a
60% mortality and 15% recovery rate in noncorticosteroid-
treated patients older than 50 years (Logan et al., 1966;
O’Leary and Waisman, 1940; Sheard, 1951; Walton and
Adams, 1958). Hence, recovery without immunosuppressive
treatment, although possible, is distinctly uncommon in our
patient’s age group. Nevertheless, the unique finding of this
study was not the spontaneous remission itself, but the sta-
tistical evidence that our patient’s improvement was strongly
linked to her mind–body interventions, raising the possibil-
ity that psychologic influences may be active in other such
cases as well.
Potential mechanisms of mind–body therapy in dermato-
myositis deserve consideration. Dermatomyositis is an au-
toimmune disease where muscle and skin incur ischemic
damage from an immune-mediated microvasculopathy
(Greenberg and Amato, 2004). Complement activation is an
early pathogenic step, leading to the formation of cytolytic
C5b-9 MAC, which deposits in the walls of endomysial cap-
illaries and sometimes in larger perimysial blood vessels
(Kissel et al., 1986). These complexes produce osmotic ly-
sis of endothelial cells, necrosis of capillaries, and eventual
muscle ischemia and microinfarcts. Complement activation
also induces proinflammatory cytokines and chemotactic
factors, which recruit inflammatory cells and upregulate cel-
COLLINS AND DUNN
lular adhesion molecules on leukocytes and endothelial cells
in the affected region (Engel and Hohlfeld, 2004). Inflam-
matory infiltrates in muscle are composed primarily of B
cells, CD4 (helper/inducer) T cells, and macrophages (Ara-
hata and Engel, 1984). These observations argue for the pri-
macy of humoral immune mechanisms in the pathogenesis
of dermatomyositis. Vascular damage probably ensues from
a specific antibody attack against an endothelial antigen or
deposition of circulating immune complexes.
Against this pathogenic backdrop, how might psycho-
logic interventions such as meditation and imagery influ-
ence the course of dermatomyositis? One would have to pos-
tulate an effect on the humoral immune response or a
nonspecific anti-inflammatory action. Few investigations
have considered the effect of meditation or imagery on the
latter systems. The influence of these techniques on mea-
sures of immunity has been analyzed, but often with flawed
methodology and conflicting results. Multiple psychologic
modalities have commonly been used simultaneously (Carl-
son et al., 2003; Miller and Cohen, 2001; Richardson et al.,
1997). Few meditation-specific studies have been reported.
In these studies, meditation was shown to significantly (1)
inhibit delayed-type hypersensitivity and blastogenic lym-
phocyte responses to varicella-zoster virus antigen (Smith
et al., 1985); (2) limit the normal postexercise increment in
CD8 T-cell count (Solberg et al., 1995); (3) increase T- and
natural-killer (NK)–cell counts of human immunodeficiency
virus (HIV)-positive subjects (Taylor, 1995; Robinson et al.,
2003); (4) induce cytokine sifts favoring cell-mediated im-
munity (Jones, 2001); and (5) increase antibody responses
to influenza vaccination (Davidson et al., 2003). Although
limited in scope and somewhat conflicting, these data sup-
port the premise that meditation can influence the cellular
and humoral immune systems, probably mediated by neu-
rophysiological and neurochemical effects on the brain
(Austin, 1999; Newberg and Iversen, 2003).
Guided visual imagery is based on the principle that every
thought, even those without external correlates, triggers a
physiologic response and that recurring images produce en-
during, learned physiologic changes appropriate to the con-
tent of the imagery (Achterberg and Lawlis, 1978; Simon-
ton et al., 1980). Three of four small studies comparing the
effect of relaxation/hypnosis alone to relaxation/hypnosis
combined with guided imagery on salivary IgA concentra-
tion revealed greater increases of IgA concentration in the
group practicing imagery (Gregerson et al., 1996; Jasnoski
et al., 1987; Olness et al., 1989; Rider et al., 1990). Another
investigation showed no significant effect of imagery added
to relaxation on monocyte chemotaxis, lymphoproliferative
responses to T-cell mitogens, and natural killer (NK) cell
activity (Zachariae et al., 1994). A randomized, controlled
trial of imagery versus group support and standard care in
patients with breast cancer demonstrated no significant dif-
ferences between the groups in NK cell activity and levels
of several cytokines relevant to cell-mediated immune re-
EFFECTS OF MEDITATION AND VISUAL IMAGERY
sponses (Richardson et al., 1997). An uncontrolled study
showed no sustained effect of imagery on NK cell counts in
patients with breast cancer (Bakke et al., 2002). Therefore,
imagery has been shown to influence humoral but not cel-
lular immunity, consistent with our study’s finding of a sig-
nificant effect of imagery on manifestations of a humorally-
mediated autoimmune disease.
Many studies, typically enrolling small numbers of pa-
tients, have analyzed the effects of relaxation/hypnosis on
immune system function. These are relevant to our case, as
meditation and imagery both involve some degree of focused
relaxation (Benson, 1975; Kabat-Zinn, 1994; Simonton et al.,
1980). In a meta-analysis of 10 relaxation studies, of 5 ana-
lyzed immune variables, only increases in salivary im-
munoglobulin A (IgA) concentration were significant and
consistent for direction of change (Rood et al., 1993). Less
consistent effects have included increased NK cell cytotox-
icity (Fawzy et al., 1990; Gruber et al., 1988, 1993), enhanced
mitogen-induced and mixed lymphocyte responsiveness
(Gruber et al., 1988, 1993; Kiecolt-Glaser et al., 2001), in-
creased peripheral blood CD4 T cells and interleukin-1 pro-
duction (Kiecolt-Glaser et al., 2001), increased serum im-
munoglobulin levels (Green et al., 1988; Gruber et al., 1988,
1993), and decreased herpes virus antibody titers (Gruber et
al., 1988). Results from multiple studies suggest hypnosis
can influence both IgE-mediated and delayed-type hyper-
sensitivity reactions (Zachariae, 2001). A controlled trial of
relaxation and imagery in asthmatic children revealed sig-
nificant changes indicative of TH2 (humoral) response inhi-
bition and TH1 (cellular) response activation (Castes et al.,
1999). A similar shift might be predicted to improve a hu-
morally-mediated autoimmune disease like dermatomyositis.
Thus, while some investigations offer support for the notion
that the humoral immune system may be amenable to psy-
chologic interventions, more study is required.
This study demonstrated a highly significant positive rela-
tionship between the analyzed mind–body therapies and this
patient’s recovery from dermatomyositis. Whether these re-
sults can be generalized to all patients with dermatomyositis
remains unestablished, but deserves investigation. In addition,
our finding of slower decay for meditation and visual imagery
relative to stress bears confirmation in controlled clinical and
immunologic studies. Finally, the success of safe, low-cost
mind–body therapy in this case suggests that larger scale, con-
trolled trials of this treatment modality for a variety of im-
mune disorders may be warranted.
ACKNOWLEDGMENTS
We are indebted to Eva Chapin, M.D. and Sandra
Pinkham, M.D., who worked with the patient in regard to
the documentation of her condition and recovery and thus
contributed significantly to this study. We also thank M. Is-
283
abel Periquet, M.D. and John T. Kissel, M.D. for their help-
ful suggestions and Elizabeth Stasney, Ph.D. and Stephen
Cosslett, Ph.D. (Department of Statistics and Department of
Economics, respectively, Ohio State University) for their
statistical advice.
REFERENCES
Achterberg J, Lawlis G. Imagery of Disease. Champaign, IL: In-
stitute for Personality and Ability Testing, 1987.
Andrews FM. Construct validity and error components of survey
measures: A structural modelling approach. Pub Op Q 1984;48:
409–442.
Arahata K, Engel AG. Monoclonal antibody analysis of mononu-
clear cells in myopathies. I: Quantitation of subsets according
to diagnosis and sites of accumulation and counts of muscle
fibers invaded by T cells. Ann Neurol 1984;1:193–208.
Astin JA, Shapiro SL, Eisenberg DM, Forys KL. Mind–body med-
icine: State of the science, implications for practice. J Am Board
Fam Pract 2003;16:131–147.
Austin JH. Zen and the Brain. Cambridge, MA: MIT Press, 1999.
Bakke AC, Purtzer MZ, Newton P. The effect of hypnotic guided
imagery on psychological well-being and immune function in pa-
tients with prior breast cancer. J Psychosom Res 2002;53:1131–
1137.
Barrows KA, Jacobs BP. Mind–body medicine. An introduction
and review of the literature. Med Clin North Am 2002;86:11–31.
Benson H. The Relaxation Response. New York: Morrow, 1975.
Bishop S. What do we really know about mindfulness-based stress
reduction? Psychosom Med 2002;64:71–84.
Carlson LE, Speca M, Patel KD, Goodey E. Mindfulness-based
stress reduction in relation to quality of life, mood, symptoms
of stress, and immune parameters in breast and prostate cancer
outpatients. Psychosom Med 2003;65:571–581.
Carpenter S, Karpati G. Pathology of Skeletal Muscle, 2nd ed. New
York: Oxford University Press, 2001:550–568.
Castes M, Hagel I, Palenque M, Canelones P, Corao A, Lynch NR.
Immunological changes associated with clinical improvement of
asthmatic children subjected to psychosocial intervention. Brain
Behav Immunol 1999;13:1–13.
Da L. Taoist Health Exercise Book. New York: Paragon House, 1991.
Davidson RJ, Kabat-Zinn J, Schumacher J, Rosenkranz M, Muller
D, Santorelli SF, Urbanowski F, Harrington A, Bonus K, Sheri-
dan JF. Alterations in brain and immune system function produced
by mindfulness meditation. Psychosom Med 2003;65:564–570.
Engel AG, Hohlfeld R. The polymyositis and dermatomyositis
syndromes. In: Engel AE, Franzini-Armstrong C, eds. Myology,
3rd ed. New York: McGraw-Hill, 2004:1321–1366.
Everett MA, Curtis AC. Dermatomyositis. A review of nineteen cases
in adolescents and children. Arch Intern Med 1957;100:70–76.
Fawzy FI, Kemeny ME, Fawzy NW, Elashoff R, Morton D,
Cousins N, Fahey JL. A structured psychiatric intervention for
cancer patients. II. Changes over time in immunological mea-
sures. Arch Gen Psychiatry 1990;47:729–735.
Fontanarosa PB, ed. Alternative Medicine: An Objective Assess-
ment. Chicago: American Medical Association, 2000.
Gibaldi M, Perrier D. Pharmacokinetics, 2nd ed. New York: Mar-
cel Dekker, 1982.
284
Green ML, Green RG, Santoro W. Daily relaxation modifies serum
and salivary immunoglobulin and psychophysiologic symptom
severity. Biofeedback Self Regul 1988;13:187–199.
Greenberg SA, Amato AA. Uncertainties in the pathogenesis of
adult dermatomyositis. Curr Opin Neurol 2004;17:359–364.
Gregerson MB, Roberts IM, Amiri MM. Absorption and imagery
locate immune responses in the body. Biofeedback Self Regul
1996;21:149–165.
Gruber BL, Hall NR, Hersh SP, Dubois P. Immune system and
psychological changes in metastatic cancer patients using re-
laxation and guided imagery: A pilot study. Scand J Behav Ther
1988;17:25–46.
Gruber BL, Hersh SP, Hall NR. Immunological responses of breast
cancer patients to behavioral interventions. Biofeedback Self
Regul 1993;18:1–22.
Hepp P. Ueber Pseudotrichinose, eine besondere Form von acuter
parenchymatoser Polymyositis. Berlin Klin Wochenschr 1887;
24:297–299, 322–326.
Jasnoski ML, Kugler J. Relaxation, imagery, and neuroim-
munomodulation. Ann NY Acad Sci 1987;496:722–730.
Jones BM. Changes in cytokine production in healthy subjects
practicing Guolin Qigong: A pilot study. BMC Complement Al-
tern Med 2001;1:8.
Kabat-Zinn J. Wherever You Go, There You Are: Mindfulness
Meditation in Everyday Life. New York: Hyperion, 1994.
Kiecolt-Glaser JK, Marucha PT, Atkinson C, Glaser R. Hypnosis
as a modulator of cellular immune dysregulation during acute
stress. J Consult Clin Psychol 2001;69:674–682.
Kissel JT, Mendell JR, Rammohan KW. Microvascular deposition
of complement membrane attack complex in dermatomyositis.
N Engl J Med 1986;314:329–334.
Krippner S, Achterberg J. Anomalous healing experiences. In: Car-
dena E, Lynn SJ, Krippner S, eds. Varieties of Anomalous Ex-
perience: Examining the Scientific Evidence. Washington, DC:
American Psychological Association, 2000:353–395.
Krosnick JA, Fabrigar LR. Designing Questionnaires to Measure
Attitudes. New York: Oxford University Press, 2002.
Lewy B. Zur Lehre von der primaren acuten Polymyositis. Berlin
Klin Wochenschr 1893;30:475–477.
Logan RG, Bandera JM, Mikkelsen WM, Duff IF. Polymyositis:
A clinical study. Ann Intern Med 1966;65:996–1007.
Miller GE, Cohen S. Psychological interventions and the immune
system: A meta-analytic review and critique. Health Psychol
2001;20:47–63.
Newberg AB, Iversen J. The neural basis of the complex mental
task of meditation: Neurotransmitter and neurochemical consid-
erations. Med Hypotheses 2003; 61:282–291.
O’Leary PA, Waisman M. Dermatomyositis. A study of forty
cases. Arch Dermatol Syph 1940;41:1001–1019.
Olness K, Culbert T, Uden D. Self-regulation of salivary im-
munoglobulin A by children. Pediatrics 1989;83:66–71.
Richardson MA, Post-White J, Grimm EA, Moye LA, Singletary
SE, Justice B. Coping, life attitudes, and immune responses to
imagery and group support after breast cancer treatment. Altern
Ther Health Med 1997;3:62–70.
Rider MS, Achterberg J, Lawlis GF, Groven A, Toledo R, Butler
JR. Effect of immune system imagery on secretory IgA. Biofeed-
back Self Regul 1990;15:317–333.
COLLINS AND DUNN
Robinson FP, Mathews HL, Witek-Janusek L. Psycho-endocrine-
immune response to mindfulness-based stress reduction in indi-
viduals infected with the human immunodeficiency virus: A
quasiexperimental study. J Altern Complement Med 2003;9:
683–694.
Rood YR, Bogaards M, Goulmy E, van Houwelingen HC. The ef-
fects of stress and relaxation on the in vitro immune response
in man: a meta-analytic study. J Behav Med 1993;16:163–181.
Schuermann H. Zur Klinik und Pathogenese der Dermatomyositis
(Polymyositis). Arch Dermat Syph (Wein) 1939;178:414–468.
Selander P. Dermatomyositis in early childhood. Acta Med Scand
Suppl 1950;246:187–203.
Sheard C. Dermatomyositis. Arch Intern Med 1951;88:640–658.
Simonton CO, Mathews-Simonton S, Creighton J. Getting Well
Again. New York: Bantam, 1980.
Smith GR, McKenzie JM, Marmar DJ, Steele RW. Psychologic
modulation of the human immune response to varicella zoster.
Arch Intern Med 1985;145:2110–2112.
Solberg E, Halvorsen R, Sundgot-Borgen J, Ingjer F, Holen A.
Meditation: A modulator of the immune response to physical
stress? A brief report. Br J Sports Med 1995;29:255–257.
Taylor DN. Effects of a behavioral stress-management program on
anxiety, mood, self-esteem, and T-cell count in HIV-positive
men. Psychol Rep 1995;76:451–457.
Unverricht H. Dermatomyositis acuta. Dtsch Med Wochenschr
1891;17:41–44.
Unverricht H. Polymyositis acuta progressiva. Z Klin Med
1887;12:533–549.
Vivekananda N. Raja-Yoga. New York: Ramakrishna-Vivekanada
Center, 1982.
Wagner E. Ein Fall von acuter Polymyositis. Dtsch Arch Klin Med
1887;40:241–266.
Walton JN, Adams RD. Polymyositis. London: E & S Livingstone,
1958.
Wedgwood RJ, Cook CD, Cohen J. Dermatomyositis. Report of
26 cases in children with a discussion of endocrine therapy in
13. Pediatrics 1953;12:447–466.
Winkelmann RK, Mulder DW, Lambert EH, Howard FM, Diess-
ner GR. Course of dermatomyositis-polymyositis: comparison
of untreated and cortisone-treated patients. Mayo Clin Proc
1968;43:545–556.
Zachariae R. Hypnosis and immunity. In: Ader R, Felten DL, Co-
hen N, eds. Psychoneuroimmunology, 3rd ed. San Diego: Aca-
demic Press, 2001:133–160.
Zachariae R, Hansen JB, Andersen M, Jinquan T, Petersen KS, Si-
monsen C, Zachariae C, Thestrup-Pedersen K. Changes in cel-
lular immune function after immune specific guided imagery and
relaxation in high and low hypnotizable healthy subjects. Psy-
chother Psychosom 1994;61:74–92.
Address reprint requests to:
Michael P. Collins, M.D.
Department of Neurosciences
Marshfield Clinic
1000 North Oak Avenue
Marshfield, WI 54449
E-mail: collins.michael@marshfieldclinic.org