Vol. 44 No.

1 July 2012 Journal of Pain and Symptom Management 95

Review Article

Guided Imagery for Non-Musculoskeletal

Pain: A Systematic Review of Randomized
Clinical Trials
Paul Posadzki, PhD, Wendy Lewandowski, PhD, APRN, BC, Rohini Terry, BSc, PhD,
Edzard Ernst, MD, PhD, FMedSci, FSB, FRCP, FRCPEd, and Anthony Stearns, PhD
Department of Complementary Medicine (P.P., R.T., E.E.), University of Exeter, Peninsula Medical
School, Exeter, Devon, United Kingdom; and Kent State University College of Nursing (W.L., A.S.),
Kent, Ohio, USA

Abstract
Context. Our previous review of the literature concluded that there is encouraging
evidence that guided imagery alleviates musculoskeletal pain, but the value of guided
imagery in the management of non-musculoskeletal pain remains uncertain.
Objectives. The objective of this systematic review was to assess the effectiveness
of guided imagery as a treatment option for non-musculoskeletal pain.
Methods. Six databases were searched from their inception to February 2011.
Randomized clinical trials were considered if they investigated guided imagery in
human patients with any type of non-musculoskeletal pain in any anatomical
location and assessed pain as a primary outcome measure. Trials of motor imagery
and hypnosis were excluded. The selection of studies, data extraction, and
validation were performed independently by two reviewers.
Results. Fifteen randomized clinical trials met the inclusion criteria. Their
methodological quality was generally poor. Eleven trials found that guided
imagery led to a significant reduction of non-musculoskeletal pain. Four studies
found no change in non-musculoskeletal pain with guided imagery in comparison
with progressive relaxation, standard care, or no treatment.
Conclusion. The evidence that guided imagery alleviates non-musculoskeletal pain
is encouraging but remains inconclusive. J Pain Symptom Manage 2012;44:95e104.
2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.

Key Words
Complementary and alternative medicine, pain, systematic review, non-musculoskeletal pain

Introduction was 20.4%.1 To the best of our knowledge, there

is no standard definition of NMSP. A proposed
The prevalence of non-musculoskeletal pain
operational definition is as follows: NMSP
(NMSP) in the U.S. population in 1992e1994

Address correspondence to: Paul Posadzki, PhD, Depart- Salmon Pool Lane, Exeter, Devon EX2 4SG, United
ment of Complementary Medicine, University of Kingom. E-mail: Paul.Posadzki@pcmd.ac.uk
Exeter, Peninsula Medical School, Veysey Buliding, Accepted for publication: July 20, 2011.

2012 U.S. Cancer Pain Relief Committee 0885-3924/$ - see front matter
Published by Elsevier Inc. All rights reserved. doi:10.1016/j.jpainsymman.2011.07.014
96 Posadzki et al.
generates nociceptive stimuli from structures
other than the musculoskeletal system such as
the viscera, skin, cardiovascular, genitourinary,
or nervous systems; it may be acute, continu-
ous, recurrent, chronic, or postsurgical and
may be located in a wide variety of anatomical
regions.2 Harding and Yelland2 noted that
NMSP is more likely to occur when a patient
has a history of non-musculoskeletal causes,
current systemic symptoms, skin problems,
or deep tenderness in the abdomen. However,
it is often difficult to differentiate non-
musculoskeletal causes from musculoskeletal
causes. Pain of musculoskeletal origin is more
likely to be triggered by movement of the af-
fected part.
Guided imagery (GI) is a widely used com-
plementary therapy, and its use for pain man-
agement has increased over the past two
decades. However, definitions of GI used in
various health science disciplines are inconsis-
tent. In this review, we define GI as follows: GI
involves the generation or recall of different
mental images, such as perception of objects
or events, and can engage mechanisms used
in cognition, memory, and emotional and mo-
tor control.3 The images are typically visualized
within a state of relaxation, possibly with a spe-
cific outcome in mind (e.g., pain relief).4 Wey-
dert et al.5 suggested that during GI, all the
senses should be used because the more detail
with which the image is sensed, the more po-
tential for pain relief it has. GI has been sug-
gested as an effective treatment for cancer
pain.6 A recent systematic review concluded
that the evidence of the effectiveness of GI in
alleviating musculoskeletal pain (MSP) is en-
couraging but inconclusive.7 The therapeutic
value of GI in the treatment of pain other
than musculoskeletal and cancer pain, how-
ever, remains unclear. The rationale for this re-
search was to continue investigations into the
possible analgesic effects of GI. Therefore,
the aim of this systematic review was to criti-
cally evaluate the evidence for the effectiveness
of GI as a treatment for NMSP.
Methods
Literature searches were performed to iden-
tify all controlled clinical trials of GI for human
patients with any type of NMSP. The Cochrane
Vol. 44 No. 1 July 2012
Central Register of Controlled Trials, MEDLINE,
EMBASE, CINAHL, AMED, and PsycINFO were
searched from their inception to February 2011.
A manual search was carried out using the refer-
ence lists of articles located through a scoping
search in major electronic databases and
through scanning our own files. Reference lists
of all retrieved articles were hand-searched for
relevant studies. No language restrictions were
imposed.
All retrieved studies, including uncontrolled
trials, case studies, and preclinical and observa-
tional studies, were reviewed for safety infor-
mation. Only randomized controlled trials
(RCTs) testing GI in human patients with
NMSP of any duration and intensity were con-
sidered for evaluation of the effectiveness of
GI. Trials of motor imagery and hypnosis
were excluded.8,9 Trials also were excluded if
pain was not a primary outcome measure10e12
or if the trials were related to MSP13e15 or if GI
was bundled into a complex therapeutic pack-
age.16 However, studies of GI combined with
relaxation were included in the review because
relaxation is commonly used during GI to
achieve a state of relaxed focus. All included
articles were read in full.
Two of the authors extracted data indepen-
dently using a custom-made data extraction
form. For each study, trial design, randomiza-
tion, blinding, dropout rate, inclusion and ex-
clusion criteria, details of treatment method
and control groups, main outcome measures,
and main results were extracted. Methodolog-
ical quality of studies was assessed using the
5-point Jadad score.17 Clinical trials scoring
3 or more points were considered high qual-
ity. Differences in scoring between reviewers
were resolved through discussion. The mean
change in pain ratings from baseline was
considered the primary outcome measure.
Changes in pain ratings were used to assess
the differences between the intervention
and control condition.
Effect sizes were calculated for the effect of GI
on pain outcome measures. Difference scores be-
tween experimental and control groups were cal-
culated. If other statistics, such as F, t, mean, and/
or standard deviations, were reported, they were
converted to correlations using commonly avail-
able formulas.18e20 The difference scores were
then converted to eta-squares and d statistics
using Cohens21 formulas.
Vol. 44 No. 1 July 2012 Guided Imagery for Non-Musculoskeletal Pain
Results
Our search strategy generated a total of
2478 references, of which 155 were potentially
relevant (Fig. 1). A total of 43 clinical trials
were retrieved for further evaluation, and 15
of these, involving 1172 human patients with
NMSP, were eligible for review.5,22e34 In all
of these trials, GI was used as the only
treatment5,22,24e31,33,35 or was combined with
relaxation34 and standard care.23 Some studies
used pleasant imagery,32 whereas others used
pain control imagery5,22,24e31,33,35 (Table 1).
Populations of patients with NMSP were het-
erogeneous, with pain types ranging from
postoperative pain23e32,35 to abdominal
pain5,22,34 (Table 2). Eight (53.3%) of the
RCTs involved adult subjects; the remainder
(46.7%) involved pediatric subjects. Control
groups received standard or usual
care,22e24,26,28,31 relaxation,29 breathing exer-
cises,5 or no intervention.25,27,30,32e35 Primary
outcome measures included the visual ana-
logue scale,23e26,29e33 numeric rating scale,27
the Facial Affective Scale,5,28,34 Likert scales,26
and the Abdominal Pain Index.22 The quality
of the RCTs ranged from 1 to 4 points on
the Jadad scale.17
Eleven RCTs showed significantly greater
reduction of NMSP with GI than with no in-
tervention, standard care, or breathing exer-
cises.5,22e24,27,28,30,32e34 Four RCTs showed no
significant effect of GI over progressive relaxa-
tion,29 standard care,31 or no treatment.25,26
Potentially relevant
studies (n=155)
Publications excluded after
initial screening of title and
abstract (n =112)
Full text articles
retrieved for further
evaluation (n=43)
Not related to pain (n=11)
Related to musculoskeletal pain
(n=9)
Not clinical trials ( n=8)
Included studies (n=15)
Fig. 1. Flow chart diagram of eligibility criteria.
97
In seven of the 15 reviewed trials, statistics
needed for effect size calculation were not re-
ported and/or could not be derived or recre-
ated. Effect sizes (Cohens d ) of GI to reduce
NMSP calculated in the remainder of the trials
ranged from 0.05 (small) to 1.93 (large)
(Table 1).
Effects of GI on Adults
Carrico et al.,33 who explored the effects of
GI on pelvic pain and urinary symptoms in
women with interstitial cystitis, reported signif-
icant improvements in mean pain scores at the
end of the study in the GI group. This RCT,
however, lacked appropriate blinding and allo-
cation concealment and did not use intention-
to-treat analysis and power calculations. We
scored this study a 3.
Daake and Gueldner32 examined the effec-
tiveness of pleasant imagery in the manage-
ment of acute postsurgical pain. The GI
group reported significantly less postsurgical
pain than a control group and consumed sig-
nificantly less pain medication. This study,
however, lacked appropriately described ran-
domization, blinding, power calculations,
dropouts, and intention-to-treat analysis. This
study was scored a 1.
Danhauer et al.31 measured differences in
state anxiety and perceived pain level between
a GI group and a standard care group after col-
poscopy and found no differences in either
anxiety or pain. This study lacked appropri-
ately described power calculations, randomiza-
tion, blinding, and intention-to-treat analysis
and was scored a 2.
Gonzales et al.30 evaluated the effects of GI
on postoperative outcomes in patients under-
going same-day surgical procedures. They re-
ported significantly less pain in the GI group
at two hours after surgery. However, the study
did not report an appropriate sampling
method or numbers of dropouts. We scored
it a 3.
Haase et al.29 looked at whether brief psy-
chological interventions to reduce periopera-
tive stress improved the postoperative course
of patients undergoing abdominal surgery.
They reported no benefits of GI, as compared
with a control group. This study lacked appro-
priately described randomization and sam-
pling methods. We scored it a 4.
 98
Table 1
Controlled Clinical Studies of GI for the Treatment of Non-Musculoskeletal Pain
Pain
First Author Condition/ Measurement Adverse Authors
(year) Study Design Sample Size Intervention Control Methods Main Result Effects Conclusion Effect Size (Cohens d)

Ball 34
(2003) RCT with two 11 pediatric GI relaxation No intervention FACES 67% decrease in NIP GI is an effective and Insufficient data
groups patients with pain in safe treatment for
recurrent GI relaxation childhood RAP
abdominal group
pain
Carrico33 RCT with two 30 patients with GI No intervention VAS Significant NIP GI may be a useful tool Insufficient data
(2008) groups interstitial (rest) improvements in to offer women with
cystitis mean pain interstitial cystitis for
scores pain and symptom
(P 0.039) management
Daake32 (1989) RCT with two 32 patients with Pleasant No intervention VAS 100 mm Significantly less NIP These findings 1.76
groups postsurgical imagery postsurgical pain suggest that nurses
pain in the GI group can enhance the
(P < 0.05) management of
postoperative pain
by teaching patients
to use pleasant
imagery.
Danhauer31

Posadzki et al.
RCT with three 170 patients after 1. Music SC VAS 100 mm No between-group NIP Mind-body 0.20
(2007) groups colposcopy 2. GI differences interventions had no
statistically
significant impact
on reported anxiety,
perceived pain, or
satisfaction with care
(.)
Gonzales30 Single-blind 44 patients with GI No intervention vVAS Significantly less NIP The use of guided 0.49 at one hour; 0.53
(2010) RCT with postoperative pain in GI group imagery in the at two hours
two groups pain (P 0.041) at ambulatory surgery
two hours setting can (.)
result in less
postoperative pain
(.)
Haase29 (2005) Single-blind 74 patients with GI 1. Relaxation VAS 100 mm No between-group NIP (.)guided imagery Insufficient data
RCT with postsurgical 2. No differences and relaxation
three groups pain intervention yielded a very
positive patient
response but did not

Vol. 44 No. 1 July 2012

show a clinically
relevant influence
(.)
Huth28 (2004) RCT with two 73 pediatric GI SC 1. Oucher Significantly lower Two children GI should be used to 0.44 at one to four
groups patients with Scale pain in GI group allocated to reduce postoperative hours after surgery;
postoperative 2. FACES the imagery pain 0.22 at 22e27 hours
pain intervention after discharge
were
withdrawn
from the
 Vol. 44 No. 1 July 2012
study after
they became
distressed
when they
listened to
the tape
Lambert27 RCT with two 52 pediatric GI No intervention NRS Significantly lower NIP Positive effects of GI on Insufficient data
(1996) groups patients with pain ratings pediatric
surgical pain postsurgery patients
Laurion26 Single-blind 84 patients with GI 1. Music Verbal report No between-group NIP These findings Insufficient data
(2003) RCT with postoperative 2. SC on a VAS-like differences suggest that both
three groups pain scale guided imagery and
(10 cm) music are effective
strategies in
improving pain
(.)
Omlor35 RCT with two 208 patients with GI No intervention VAS Significantly less NIP (.) a decrease of Insufficient data
(2000) groups postsurgical pain in the analgesic

Guided Imagery for Non-Musculoskeletal Pain

pain visualization requirements after
group (P < 0.05) surgical treatment
was observed.
Pederson25 RCT with two 24 pediatric GI No treatment VAS 100 mm No changes in pain NIP GI may assist in coping Insufficient data
(1995) groups patients with scores with anxiety
cardiac
catheterization
olkki24 (2008)
P RCT with two 60 pediatric GI SC VAS Significantly less NIP GI can reduce 0.43 immediately after
groups patients with pain in GI group postoperative pain surgery; 0.05 at one
postoperative hour
pain postintervention
Tusek23 (1997) RCT with two 130 patients with GI SC SC VAS 100 mm Significantly less NIP Guided imagery 1.93
groups postoperative pain in the GI significantly reduces
pain group postoperative
(P < 0.001) anxiety, pain, and
narcotic
requirements of
colorectal surgery
and increases
patient satisfaction.
van Tilburg22 RCT with two 34 pediatric GI SC API (two Significant NIP GI SC was superior to Insufficient data
(2009) groups patients with questions) improvement in SC only
functional abdominal pain
abdominal
pain
Weydert5 RCT with two 22 pediatric GI Breathing FACES Significantly None occurred GI has the potential to 0.37
(2006) groups patients with exercises greater decrease benefit children with
RAP in the number of RAP
days with pain
RCT randomized controlled trial; GI guided imagery; FACES Facial Affective Scale; NIP no information provided; RAP recurrent abdominal pain; VAS visual analogue scale; SC standard care;
vVAS vertical visual analogue scale; NRS numeric rating scale; API Abdominal Pain Index.

99
100 Posadzki et al. Vol. 44 No. 1 July 2012

Table 2
Details of GI Intervention
First Author (Year) GI Intervention (quote)
34
Ball (2003) .learning to feel the difference between tense and relaxed muscles and using your imagination to tell your body
what to do.
33
Carrico (2008) One group (treatment) listened to a 25-minute guided imagery compact disc (CD), that was created specifically for
women with pelvic pain and IC, twice a day for 8 weeks. [.]you may begin to feel healing warmth melt into your
bladder . coating the inside and outer surfaces . letting each and every cell relax and release its tension .
soothing any painful areas with its gentle golden glow.
Daake32 (1989) Professionally prepared tape recordings of three preselected scenes (a beach scene, a mountain cabin scene, and an
autumn scene) were made available to each patient in the experimental group for practice throughout the day and
evening prior to surgery. The scenes were designed to engender the use of all five senses. Patients in the imagery
group were also encouraged to use their own pleasant past experiences to form an image. Each individual in the
treatment group was instructed to use the imagery technique three times per day after surgery (upon first waking,
midafternoon, and evening), for a period of 15 to 20 minutes each time.
Danhauer31 (2007) Participants in the imagery group listened to a 20-minute audio recording entitled A Meditation to Help You Be Relaxed
and Awake During Medical Procedures by Naparstek (www.healthjourneys.com).
Gonzales30 (2010) This CD led the patient through a progressive relaxation and guided imagery exercise. (.) This CD consisted of
soothing biorhythmic music combined with positive, encouraging statements. [28 minutes]
Haase29 (2005) The guided imagery tape taught patients to become calm and activate their inner resources. Patients were sent on an
imaginative journey to a special place where they could feel safe, comforted, and supported. They were encouraged
to confront and work through feelings of anxiety and stress. [12 minutes]
Huth28 (2004) The videotape presentation (deep breathing and imagery techniques), audiotape (deep breathing, muscle
relaxation, music, and suggestions for picturing a park), and booklets were given to the child and parent at home
2e22 days prior to the scheduled surgery to teach the child imagery skills for postoperative pain.
Lambert27 (1996) Each child was asked to select an enjoyable image that felt good. Each childs image was incorporated into an
individually tailored relaxation exercise. During the imagery practice, the investigator guided the child through a
rehearsal of the impending surgical experience and included suggestions for healing, uncomplicated recovery, and
minimal pain.
Laurion26 (2003) Tape I, side A, contains imagery geared for preparing patients for surgery. Tape I, side B, consists of 10 minutes of
positive affirmations that can be used to prepare for surgery but are also used for healing after surgery. Tape II,
which consists of the musical score from tape I, was played continuously on an auto-reverse Walkman during the
patients surgery (.)
Omlor35 (2000) Patients received preoperatively under supervision of a psychologist a visualization therapy in connection with 45 min
relaxation exercises.
Pederson25 (1995) Based on the childs previously identified favorite activities and places, the intervener guided the childs imagery of
going to the childs favorite places and doing his or her favorite activities (Kuttner, 1986). While guiding the childs
imagery the intervener asked the child to notice the sights and sounds in the childs favorite place, or how it felt to
perform favorite activities (.). The intervener also provided suggestions of feeling comfortable, competent, safe,
and happy.
olkki24 (2008)
P . allowed scope for the childs own pleasant imagery, and reassurance and repeatable music with sounds of nature
provided possibilities to achieve a deep state of relaxation. .the child was encouraged to mentally choose a
favorite place (.), and .proceeded toward this place along a comfortable path (downward). At the end of the
trip, the child returned along this path (upward).
Tusek23 (1997) The guided imagery tape taught techniques that allowed them to become calm and focused. They were then brought
to a special place in their mind that was safe, secure, protected, supported, and relaxed. The imagery story
encouraged patients to confront and work through any feelings of fear, anxiety, and negativity. Patients were
instructed to listen to the tape without interruption twice per day, once in the morning and once in the evening.
Tapes were 20 minutes long and had a soft, soothing, musical background. [3 days before, during, and after
surgery]
van Tilburg22 (2009) Each session includes induction imagery to produce relaxation, followed by imagery and suggestions for decreased
discomfort and healing.
Weydert5 (2006) Once achieving relaxation, subjects were asked to invite an image to come to mind that represented their pain. They
were encouraged to describe the image in detail using all the senses as the more detailed the image is sensed, the
more potential the pain reliever it could be. Once this image was established, they were then asked to invite a
second image to come that would get rid of the pain (first image).

Laurion and Fetzer26 examined the effects of
GI or music therapy on postoperative pain, nau-
sea and vomiting, and length of stay for gyneco-
logic laparoscopic patients. They reported
significantly less pain at discharge in both GI
and music therapy groups than in controls.
This study, however, lacked appropriately de-
scribed randomization, an appropriate sampling
method, power calculations, and intention-to-
treat analysis. This study was scored a 1.
Omlor et al.35 evaluated the influence of pre-
operative GI on postoperative pain in adults.
They reported significantly less pain and anal-
gesic use in the GI group than in the control
group. However, several secondary outcome
measures, including postoperative nausea,
infections, and fever, showed no significant dif-
ferences. This RCT lacked appropriately de-
scribed eligibility criteria, randomization,
blinding, allocation concealment, dropout
rate, and intention-to-treat analysis. We scored
this study a 1.
Tusek et al.23 looked at whether GI used in the
perioperative period improved the outcomes of
Vol. 44 No. 1 July 2012 Guided Imagery for Non-Musculoskeletal Pain
colorectal surgery patients. They reported signif-
icantly less pain and opioid use in the GI group
than in the control group, but the study lacked
blinding, dropout rate, and intention-to-treat
analysis. This study was scored a 2.
Effects of GI on Pediatric Patients
Ball et al.34 investigated the use of relaxa-
tion and GI for children suffering from recur-
rent abdominal pain. They found that the
children experienced a 67% decrease in pain
during the therapy. This RCT, however, lacked
appropriately described eligibility criteria, ran-
domization, blinding, allocation concealment,
dropout rate, and intention-to-treat analysis.
We scored this study a 1.
Huth et al.28 investigated the effectiveness
of imagery, in addition to routine analgesics,
in reducing tonsillectomy and adenoidectomy
pain and anxiety after ambulatory surgery and
at home. They found that by controlling for
trait anxiety and analgesic intake one to four
hours before the pain measures, the analyses
showed significantly less pain in the treatment
group one to four hours after surgery. This
study, however, lacked appropriate sampling
methods, blinding, formal power and sample
size calculations, and intention-to-treat analy-
sis. We scored this study a 3.
Lambert27 examined the effects of hypnosis/
GI on the postoperative course of pediatric sur-
gical patients. Significantly lower postoperative
pain ratings and shorter hospital stays were re-
ported in the hypnosis/GI group. Yet, this study
lacked appropriate sampling methods, explicit
eligibility criteria, appropriately described ran-
domization, blinding, dropout rate, power cal-
culations, and intention-to-treat analysis. This
study was scored a 1.
Pederson25 examined the effects of imagery
on childrens pain and anxiety during cardiac
catheterization. No significant differences in
pain were reported between groups. This RCT
lacked fully described randomization, blinding,
power and sample size calculations, exclusion
criteria, loss-to-follow-up ratio, and intention-
to-treat analysis. We scored this study a 1.
Polkki et al.24examined the efficacy of imagery
and relaxation for postoperative pain relief in
hospitalized children. They reported that the
children in the experimental group had signifi-
cantly less postoperative pain than the control
children. This RCT lacked sufficiently described
101
randomization, concealed allocation, clearly de-
scribed eligibility criteria, power calculations,
loss to follow-up rate, and intention-to-treat anal-
ysis. It was scored a 1.
van Tilburg et al.22 developed and assessed
a home-based, GI treatment protocol for chil-
dren with functional abdominal pain. They re-
ported significant improvements in abdominal
pain in the GI group. However, this study
lacked blinding and had a small sample. We
scored this study a 3.
Finally, Weydert et al.5 compared the efficacy
of GI and breathing exercises alone in the
treatment of recurrent abdominal pain in chil-
dren. They reported that children who learned
GI with muscle relaxation had significantly
fewer days with pain than those who learned
breathing exercises alone after one and two
months. This study had a small sample size,
however, and lacked therapist blinding. We
scored this study a 4.
Discussion
In a previous review that focused on GI for
MSP, we included nine RCTs and concluded
that the evidence that GI alleviates MSP is
encouraging but not conclusive.7 The present
review evaluated the evidence for the effective-
ness of GI as a treatment for NMSP. Fifteen
RCTs met our eligibility criteria. Eleven of these
(73.3%) suggested that GI is effective for
NMSP, and four (26.7%) showed no significantly
greater effects on NMSP than progressive relaxa-
tion,29 standard care,31 or no treatment.25,26 The
evidence from RCTs of GI for treating NMSP is
thus encouraging but inconclusive. There are
several reasons for this. The patients who re-
ceived GI were heterogeneous in terms of clini-
cal condition treated. There also was
heterogeneity in design and methods including
types of GI, use of control vs. attention control
groups, exclusion and inclusion criteria, and pri-
mary pain outcome measures. Given such vari-
ability in GI intervention and control groups, it
is difficult to draw any definite conclusions.
In most studies included in the present review,
methodological quality was low, with only six of
15 trials scoring 3 or more out of a possible 5
points on the Jadad scale for methodological
quality (Table 3).5,22,28e30,33 Of these high-
quality trials, five of six favored GI,5,22,28,30,33
102 Posadzki et al. Vol. 44 No. 1 July 2012

Table 3
Quality Assessment of the Included Studies
Blinding of
Random Sequence Appropriate Participants Blinding of Outcome Withdrawals and Sum (Jadad
First Author (Year) Generation Randomization or Personnel Assessors Dropouts Score)

Ball34 (2003) 1 0 0 0 0 1
Carrico33 (2008) 1 1 0 0 1 3
Daake32 (1989) 1 0 0 0 0 1
Danhauer31 (2007) 1 0 0 0 1 2
Gonzales30 (2010) 1 1 0 1 0 3
Haase29 (2005) 1 0 1 1 1 4
Huth28 (2004) 1 1 0 0 1 3
Lambert27 (1996) 1 0 0 0 0 1
Laurion26 (2003) 1 0 0 0 0 1
Omlor35 (2000) 1 0 0 0 0 1
Pederson25 (1995) 1 0 0 0 0 1
olkki24 (2008)
P 1 0 0 0 0 1
Tusek23 (1997) 1 1 0 0 0 2
van Tilburg22 (2009) 1 1 0 0 1 3
Weydert5 (2006) 1 1 0 1 1 4
Note: A 1 indicates the lowest and 5 indicates the highest results on the scale.

but one showed no significant effect over relaxa-
tion.29 Five of nine low-quality trials (less than
3 on the Jadad score) reported a significant re-
duction in NMSP after the GI intervention, and
four low-quality RCTs reported no change in
pain scores.
Guidelines for designing and reporting
RCTs note that descriptions of adverse effects
are an ethical imperative in clinical research.36
Our review suggests that in GI research, this
imperative is frequently ignored. Two of the
15 RCTs reviewed reported adverse events,5,28
but 13 failed to provide that information
(Table 1).22e27,29e34
Our review had several limitations. We at-
tempted to identify all RCTs on the subject,
but it is conceivable that some negative RCTs
remain unpublished, distorting the overall pic-
ture. Other limitations of our review include
the poor quality of the primary data, poor re-
porting of results, and lack of statistical pool-
ing as a result of the high heterogeneity of
the studies. Also, difficulties with conceptuali-
zation of NMSP may have affected the findings
of the review. Finally, the total number of trials
included in our review and the total numbers
in samples prevent any definitive judgments.
To definitively establish the effects of GI on
pain in NMSP patients, adequately designed tri-
als are needed. Studies of GI need to follow the
standards of trial design and reporting as per
the Consolidated Standards of Reporting Trials.
More specifically, studies need to have adequate
samples based on power calculations, use vali-
dated outcome measures, control for nonspe-
cific effects, and minimize other sources of bias
related to internal validity. Furthermore, GI in-
terventions should be standardized. Although
we assessed outcome measures related only to
pain, other outcomes such as analgesic con-
sumption, anxiety, depression, disability, func-
tion, length of hospital stay, or quality of life
should be taken into account in future research.
Finally, the inclusion of biobehavioral mediating
variables can lead to greater understanding of
the mechanisms by which GI produces its effects.
In addition to psychological mediators, such as
improved pain self-efficacy,33 possible modes of
action of GI may include modulation of immune
and nervous system through the release of en-
kephalins, endorphins, cholecystokinin, and do-
pamine and/or the inhibition of prostaglandin
and cortisol.37,38
In conclusion, the evidence that GI allevi-
ated NMSP overall is positive, but no definite
judgments can be made. Further rigorous re-
search seems warranted.
Disclosures and Acknowledgments
Dr. Posadzki was supported by a Fellowship
from the Pilkington Family Trust. The authors
declare no conflicts of interest.
References
1. Yelin EH, Trupin LS, Sebesta DS. Transitions in
employment, morbidity, and disability among per-
sons ages 51-61 with musculoskeletal and non-
musculoskeletal conditions in the US, 1992-1994.
Arthritis Rheum 1999;42:769e779.
Vol. 44 No. 1 July 2012 Guided Imagery for Non-Musculoskeletal Pain
2. Harding G, Yelland M. Back, chest and abdomi-
nal paindis it spinal referred pain? Aust Fam Physi-
cian 2007;36:422e423, 425, 427e429.
3. Kosslyn SM, Ganis G, Thompson WL. Neural
foundations of imagery. Nat Rev Neurosci 2001;2:
635e642.
4. Astin JA, Goddard TG, Forys K. Barriers to the
integration of mind-body medicine: perceptions of
physicians, residents, and medical students. Explore
(NY) 2005;1:278e283.
5. Weydert JA, Shapiro DE, Acra SA, et al. Evalua-
tion of guided imagery as treatment for recurrent
abdominal pain in children: a randomized con-
trolled trial. BMC Pediatr 2006;6:29.
6. Ernst E, Cohen L, Gerner J, et al. Alternative
cancer cures: looking for common ground. Lancet
Oncol 2000;1:54e59.
7. Posadzki P, Ernst E. Guided imagery for muscu-
loskeletal pain: a systematic review of randomized
clinical trials. Clin J Pain 2011;27:648e653.
8. Moseley GL. Graded motor imagery is effective
for long-standing complex regional pain syndrome:
a randomised controlled trial. Pain 2004;108:
192e198.
9. Grondahl JR, Rosvold EO. Hypnosis as a treat-
ment of chronic widespread pain in general prac-
tice: a randomized controlled pilot trial. BMC
Musculoskelet Disord 2008;9:124.
10. Manyande A, Berg S, Gettins D, et al. Preopera-
tive rehearsal of active coping imagery influences
subjective and hormonal responses to abdominal-
surgery. Psychosom Med 1995;57:177e182.
11. Lahmann C, Nickel M, Schuster T, et al. Func-
tional relaxation and guided imagery as comple-
mentary therapy in asthma: a randomized
controlled clinical trial. Psychother Psychosom
2009;78:233e239.
12. Lin MF, Hsu MC, Chang HJ, et al. Pivotal mo-
ments and changes in the Bonny Method of Guided
Imagery and Music for patients with depression.
J Clin Nurs 2010;19:1139e1148.
13. Baird CL, Murawski MM, Wu J. Efficacy of
guided imagery with relaxation for osteoarthritis
symptoms and medication intake. Pain Manag
Nurs 2010;11:56e65.
14. Menzies V, Taylor AG, Bourguignon C. Effects
of guided imagery on outcomes of pain, functional
status, and self-efficacy in persons diagnosed with
fibromyalgia. J Altern Complement Med 2006;12:
23e30.
15. Antall GF, Kresevic D. The use of guided imag-
ery to manage pain in an elderly orthopaedic popu-
lation. Orthop Nurs 2004;23:335e340.
16. Gaston-Johansson F, Fall-Dickson JM, Nanda J,
et al. The effectiveness of the comprehensive coping
strategy program on clinical outcomes in breast
103
cancer autologous bone marrow transplantation.
Cancer Nurs 2000;23:277e285.
17. Jadad AR, Moore RA, Carrol D, et al. Assessing
the quality of reports of randomized clinical trials-
dis blinding necessary? Control Clin Trials 1996;
17:1e12.
18. Hunter J, Schmidt F. Methods of meta-analysis:
Correcting error and bias in research findings. Bev-
erly Hills, CA: Sage Publications, 1990.
19. Kubeck J, Delp N, Haslett T, McDaniel M. Does
job-related training performance decline with age?
Psychol Aging 1996;11:92e107.
20. Rosenthal R. Parametric measures of effect size.
In: Cooper H, Hedges LV, eds. The handbook of re-
search synthesis. New York: Russell Sage Founda-
tion, 1994:231e244.
21. Cohen J. Statistical power analysis for the behav-
ioral sciences, 2nd ed. Hillsdale, NJ: Erlbaum, 1988.
281e285.
22. van Tilburg MA, Chitkara DK, Palsson OS, et al.
Audio-recorded guided imagery treatment reduces
functional abdominal pain in children: a pilot study.
Pediatrics 2009;124:e890ee897.
23. Tusek DL, Church JM, Strong SA, Grass JA,
Fazio VW. Guided imagery: a significant advance
in the care of patients undergoing elective colo-
rectal surgery. Dis Colon Rectum 1997;40:
172e178.
24. Polkki T, Pietila A, Vehvilainen-Julkunen K,
Laukkala H, Laukkala H. Imagery-induced relaxa-
tion in childrens postoperative pain relief: a ran-
domized pilot study. J Pediatr Nurs 2008;23:
217e224.
25. Pederson C. Effect of imagery on childrens
pain and anxiety during cardiac catheterization.
J Pediatr Nurs 1995;10:365e374.
26. Laurion S, Fetzer SJ. The effect of two nursing
interventions on the postoperative outcomes of gy-
necologic laparoscopic patients. J Perianesth Nurs
2003;18:254e261.
27. Lambert SA. The effects of hypnosis/guided im-
agery on the postoperative course of children. J Dev
Behav Pediatr 1996;17:307e310.
28. Huth MM, Broome ME, Good M. Imagery re-
duces childrens post-operative pain. Pain 2004;
110:439e448.
29. Haase O, Schwenk W, Hermann C, Muller JM.
Guided imagery and relaxation in conventional co-
lorectal resections: a randomized, controlled, par-
tially blinded trial. Dis Colon Rectum 2005;48:
1955e1963.
30. Gonzales EA, Ledesma RJ, McAlister DJ, et al.
Effects of guided imagery on postoperative out-
comes in patients undergoing same-day surgical
procedures: a randomized, single-blind study.
AANA J 2010;78:181e188.
104 Posadzki et al.
31. Danhauer SC, Marler B, Rutherford CA, et al.
Music or guided imagery for women undergoing
colposcopy: a randomized controlled study of ef-
fects on anxiety, perceived pain, and patient satisfac-
tion. J Low Genit Tract Dis 2007;11:39e45.
32. Daake DR, Gueldner SH. Imagery instruction
and the control of postsurgical pain. Appl Nurs
Res 1989;2:114e120.
33. Carrico DJ, Peters KM, Diokno AC. Guided im-
agery for women with interstitial cystitis: results of
a prospective, randomized controlled pilot study.
J Altern Complement Med 2008;14:53e60.
34. Ball TM, Shapiro DE, Monheim CJ, Weydert JA.
A pilot study of the use of guided imagery for the
treatment of recurrent abdominal pain in children.
Clin Pediatr (Phila) 2003;42:527e532.
Vol. 44 No. 1 July 2012
35. Omlor G, Kiewitz S, Pietschmann S, Roesler S.
Effect of preoperative preoperative visualization
therapy on postoperative outcome after inguinal
hernia surgery and thyroid resection. Zentralbl
Chir 2000;125:380e385.
36. CONSORT Group. The CONSORT statement.
2010. Available from http://www.consort-statement.
org/. Accessed December 2010.
37. Zachariae R, Kristensen J, Hokland P, et al. Ef-
fect of psychological intervention in the form of re-
laxation and guided imagery on cellular immune
function in normal healthy-subjectsdan overview.
Psychother Psychosom 1990;54:32e39.
38. Sauro M, Greenberg R. Endogenous opiates
and the placebo effect: a meta-analytic review. J Psy-
chosom Res 2005;58:115e120.