Content uploaded by Jerrilyn Cambron
Author content
All content in this area was uploaded by Jerrilyn Cambron
Content may be subject to copyright.
THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 12, Number 1, 2006, pp. 65–70
© Mary Ann Liebert, Inc.
Changes in Blood Pressure After Various Forms of
Therapeutic Massage: A Preliminary Study
JERRILYN A. CAMBRON, D.C., M.P.H., Ph.D.,
1
JENNIFER DEXHEIMER, L.M.T.,
1
and PATRICIA COE, D.C., C.M.T.
2
ABSTRACT
Objectives: The objective of this study was to determine the change in blood pressure (BP) in normoten-
sive and prehypertensive adults resulting from a therapeutic massage, and the factors associated with such
changes, including demographic and massage characteristics.
Design: settings/location: National University of Health Sciences Massage Therapy Clinic, Lombard, IL.
Subjects: The subjects were 150 current adult massage therapy clients with BP lower than 150/95.
Interventions: BP was measured before and after a therapeutic massage
Outcome measures: Change in BP and potential associated factors such as type of massage, duration of
massage, specific body area massaged, amount of massage pressure, and demographic characteristics were stud-
ied.
Results: Overall, systolic BP decreased by 1.8 mm Hg and diastolic BP increased by 0.1 mm Hg. Demo-
graphic factors associated with BP decrease included younger age (p 0.01) and taller stature (p 0.09). Type
of massage was associated with change in BP: Swedish massage had the greatest effect at BP reduction. Trig-
ger point therapy and sports massage both increased the systolic BP, and if both forms of massage were in-
cluded in a session, both the systolic and diastolic BP readings significantly increased. No other massage fac-
tors were associated with a significant change in BP.
Conclusions: Type of massage was the main factor affecting change in BP. Increases in BP were noted for
potentially painful massage techniques, including trigger point therapy.
65
INTRODUCTION
M
assage is a commonly used form of treatment, ac-
counting for almost half of all visits to complemen-
tary and health care providers.
1
This form of manual ther-
apy is mainly used for musculoskeletal complaints and
relaxation. As a result of the relaxation response and through
the increase in blood flow throughout the body, an overall
decrease in blood pressure (BP) is thought to
occur. However, various studies have found conflicting ev-
idence as to whether massage can produce a relaxation
response and thus the decrease in BP. Kaufmann
2
and Long-
worth
3
show no significant difference in BP before and af-
ter massage, whereas several other studies demonstrate sig-
nificant decreases.
4–10
Such variability in outcomes may re-
sult from the limited number of subjects within most previ-
ous studies along with a lack of clearly defined form of
massage used. The purpose of this preliminary study is to
measure the change in BP before and after receipt of a
therapeutic massage in normotensive and prehypertensive
adults, and to determine if BP changes are associated with
certain patient characteristics or the type, duration, or force
of massage. This is the first study to assess what specific
characteristics of massage may affect BP, and these data will
assist in determining a massage treatment for hypertensive
clients in future clinical studies.
1
Department of Research, National University of Health Sciences, Lombard, IL.
2
Massage Therapy Program, College of Allied Health Sciences, National University of Health Sciences, Lombard, IL.
Hypertension is defined as systolic blood pressure 140
mmHg or diastolic blood pressure 90 mmHg. According
to the 1999–2000 U.S. National Health and Nutrition Ex-
amination Survey (NHANES), nearly 29% of the U.S. adult
population has hypertension, amounting to approximately
58.4 million Americans.
11
The majority of Americans take
medications to control their hypertension; however, not all
hypertensive patients can or are willing to take prescribed
medications. An alternative, nonpharmacologic treatment
for hypertension may be massage therapy.
Few previous studies have assessed the change in BP with
massage. One clinical pilot study assessed chiropractic ma-
nipulation versus a massage control group for subjects with
essential hypertension.
12
In this study, eight subjects re-
ceived 5 to 10 second effleurage massages in two areas along
the spinal column, for up to three visits per week during
month 1 and two visits per week during month 2. Both sys-
tolic and diastolic BPs decreased at some point during the
course of the massage treatments, however no statistical
analyses were completed.
A second clinical trial compared massage therapy (n
15) with a muscle relaxation control group (n 15) to de-
crease hypertension.
8
Subjects in the massage group re-
ceived ten 30-minute massages specifically formatted for
this study including several techniques to the head/neck,
arms, torso, legs, and back. The results of this study demon-
strated that the massage group had significantly lower sys-
tolic and diastolic BP readings over time compared to the
control group.
In both of these clinical trials, the number of subjects was
small and the reader was left with the question of which
form of massage was the most beneficial to the hyperten-
sive client. The purpose of this study was to assess several
demographic and massage factors potentially associated
with change in BP in normotensive and prehypertensive
adults. These data will assist in development of a massage
protocol for future clinical trials on reduction of BP.
MATERIALS AND METHODS
Participants
Returning clients who were scheduled for a massage ther-
apy appointment at the National University of Health Sci-
ences Massage Therapy Clinic were approached for study
participation. A description of the study and an Institutional
Review Board (IRB) approved written consent form were
provided to eligible clients. The subjects who consented to
participate were asked to provide demographic information
(i.e., height, weight, age), have an initial BP measurement
taken, receive a therapeutic massage for
1
/
2
to 1
1
/
2
hours,
and then have a second BP measurement taken.
Subjects were excluded if they were not 18 years of age
or older, not fluent and literate in the English language, had
no time or interest to participate, or were first-time clients.
New massage therapy clients were excluded from participa-
tion because it was expected that their BP would be elevated
prior to the massage because of apprehension, rather than a
true physiologic elevation, and could lead to false results.
Based on clinic regulations, subjects with a BP of 150/95
or greater are not allowed to be treated within the Massage
Therapy Clinic without written consent from their primary
care physician and therefore were not eligible for this study
without consent.
Interventions
Massages were provided by 25 massage therapy (MT)
students, 12 were in their second trimester of coursework
and 13 were in their third (final) trimester. MT students were
trained to provide massage based on client need, rather than
structuring the massage for this study. Each massage was
between 30 and 90 minutes, and integrated six possible types
of massage, including Swedish massage, deep tissue mas-
sage, myofascial release, sports massage, trigger point ther-
apy, and/or craniosacral therapy.
We defined the type of massage used in the study as fol-
lows: Swedish massage is considered the most “traditional”
form of massage and uses compression, holding, gliding (ef-
fleurage), kneading (petrissage), shaking, tapping, and fric-
tion. Deep tissue massage is designed to reach the deep lay-
ers of muscle and uses a combination of compression,
cross-fiber, and friction strokes. Myofascial release is a tech-
nique in which the fascia connecting and surrounding mus-
cle is manipulated by the therapist. Sports massage is typi-
cally a more vigorous type of massage used to prepare
athletes for peak performance and uses a combination of
techniques including joint mobilization, stretching and/or
postisometric relaxation, cross-fiber friction, and pressure
point massage. Trigger point massage is a neuromuscular
technique used to break the pain-spasm-pain cycle and uses
focused ischemic pressure at sites of myofascial trigger
points. Finally, craniosacral massage uses gentle touch on
the head and sacral areas to release restrictions.
Details about the massage were collected from each ther-
apist after each client’s massage.
Objectives
The objective of this study was to determine the change
in BP resulting from a therapeutic massage and the factors
associated with such changes.
Outcomes
For this study, BP was measured using an automated cuff
(similar to the cuff at a local drug store) because of the ease
of use for the massage therapy students collecting the BP.
Although automated cuffs may be less accurate than man-
ual/standard methods (e.g., reading slightly high or low), the
CAMBRON ET AL.
66
authors were interested in the change in BP and were not
concerned with an exact BP reading. Future studies will in-
corporate the use of manual BP measures.
A survey was developed to collect the details about the
massage including: type(s) of massage provided, overall
amount of pressure used during massage as perceived by the
massage therapist, body areas massaged, and the length of
time each massage session lasted. These data were collected
from the MT students immediately after each massage.
Statistical methods
Analysis of variance (ANOVA) models were developed
to determine continuous differences in systolic and diastolic
BP changes (dependent variable) resulting from various cat-
egoric or ordinal independent variables including: (1) de-
mographic patient characteristics such as gender, race, and
medication use; (2) duration of massage; (3) pressure used
during massage; (4) body area massaged; or (5) type of mas-
sage. Linear regression models were developed to assess as-
sociations between BP and continuous demographic vari-
ables such as age, height, and weight. For this preliminary
study, significance values were set at p 0.10.
RESULTS
Participant flow
From February 1, 2005 to April 29, 2005, 213 massage
therapy clients were approached to participate in this re-
search study. Of the 213 clients, 60 were not interested or
asked the massage therapist to inquire later but subsequently
did not participate. Of the 153 who agreed to participate,
three were excluded leaving 150 active participants in the
study.
Baseline data
The average BP at baseline was 124 (range: 90–187) for
systolic blood pressure (SBP) and 73 (range: 50–105) for
diastolic blood pressure (DBP). Because of the possible va-
lidity issues in the BP monitor readings (consistent false-
high or -low), change in BP readings rather than actual read-
ings are discussed.
As demonstrated in Table 1, the subjects were primarily
female (61.3%), white (88.7%), and not taking BP medica-
tions (88%). The average age was 42.5 years (range: 19–79),
MASSAGE AND BLOOD PRESSURE
67
T
ABLE
1. B
ASELINE
C
HARACTERISTICS
A
SSOCIATED WITH
C
HANGE IN
B
LOOD
P
RESSURE
Systolic change Diastolic change
(before-after) (before-after)
Sample size p-value p-value
Gender 0.45 0.93
Male 58 2.60 0.05
Female 92 1.29 0.17
Race 0.96 0.61
Caucasian 133 1.80 0.38
Black 2 1.00 4.00
Hispanic 4 5.25 1.50
Other 11 0.73 3.09
Medication for high 0.93 0.10
blood pressure
No 132 1.83 0.52
Yes 18 1.61 2.78
Age 0.01
a
0.67
19–31 40 3.95 0.00
32–41 36 1.25 0.53
42–53 40 1.95 0.18
54–79 33 0.64 0.30
Height 0.09
a
0.88
60–64 41 0.27 1.02
65–66 35 1.09 0.86
67–69 38 3.82 0.05
70–77 36 2.72 0.25
Weight 0.33 0.10
105–150 43 0.81 1.37
151–175 32 1.22 0.94
176–195 38 4.21 0.03
196–387 36 2.50 2.06
a
Indicates significant group difference at p 0.10.
average height was 67.0 inches (range: 60–77), and average
weight was 176.2 pounds (range: 105–387).
Characteristics associated with blood
pressure changes
Overall, the average decrease in systolic BP was 1.8
mmHg (range: 24–34) and the average change in diastolic
BP was negligible increase of 0.1 (range: 53–18).
As noted in Table 1, some specific baseline characteris-
tics demonstrated significantly greater changes in SBP, in-
cluding younger age (p 0.01) and taller stature (p 0.09).
Items that did not significantly affect the systolic or dias-
tolic pressure in this study included gender, race, usage of
BP medications, or weight; although there did appear to be
a trend of decreased pressure readings in males, Hispanics,
and heavier individuals.
The authors then assessed specific treatment factors,
demonstrating no association with change in BP and dura-
tion of massage, amount of pressure used during the mas-
sage, and the trimester of the massage therapy intern (Table
2). Nor was there any association between change in BP and
the specific body areas massaged (Table 3).
Finally, we assessed the type of massage, which appeared
to be the significant factor affecting the change in BP. For
the majority of subjects, some Swedish massage was pro-
vided, which appeared to be associated with a decrease in
systolic BP, although no statistically significant findings
were noted. Trigger point therapy and sports massage both
increased the systolic BP, and if both forms of massage were
included in a session, both the systolic and diastolic BP read-
ings significantly increased. No significant effect in either
direction of BP change was noted with deep tissue massage,
myofascial release, or CranioSacral therapy.
DISCUSSION
The purpose of this preliminary study was to measure the
changes in BP before and after receipt of a therapeutic mas-
CAMBRON ET AL.
68
T
ABLE
2. T
REATMENT
F
ACTORS
A
SSOCIATED WITH
C
HANGE IN
B
LOOD
P
RESSURE
Systolic change Diastolic change
(before-after) (before-after)
Sample size p-value p-value
Duration of massage 0.45 0.51
Unknown 2 4.50 0.50
30 minutes 12 1.00 2.50
60 minutes 134 2.10 0.11
90 minutes 2 4.00 2.50
Pressure 0.41 0.53
Light 1 2.00 6.00
Light/medium 25 3.60 2.12
Medium 48 1.33 0.38
Medium/heavy 64 2.44 0.84
Heavy 12 3.17 0.67
Trimester of massage 0.37 0.72
therapy intern
Trimester 2 46 2.94 0.48
Trimester 3 104 1.30 0.03
T
ABLE
3. M
ASSAGED
B
ODY
A
REAS AND
A
SSOCIATION WITH
C
HANGE IN
B
LOOD
P
RESSURE
Sample size
a
Systolic change (before-after) Diastolic change (before-after)
Received treatment? No Yes No Yes p-value No Yes p-value
Back 1 149 1.00 1.81 0.94 10.00 0.06 0.21
Head and/or neck 8 142 0.88 1.85 0.80 3.63 0.07 0.20
Upper extremity 21 129 2.38 1.71 0.78 0.57 0.24 0.67
Lower extremity 29 121 0.59 2.09 0.48 0.72 0.02 0.65
Posterior hip 38 112 0.68 2.18 0.44 1.87 0.46 0.12
Face 61 89 1.90 1.73 0.92 0.16 0.10 0.96
a
Most subjects received massage on more than one body part during the massage session.
sage in 150 massage clients, and determine the factors as-
sociated with such changes. It was determined that some
baseline demographic characteristics were associated with a
significant decrease in systolic BP such as decreased age
and increased height. It was also determined that certain
types of massage were significantly associated with an in-
crease in systolic BP, such as trigger point therapy and
sports massage; and that there appeared to be a decrease in
systolic BP with Swedish massage although this result was
not statistically significant. No significant association was
noted between change in BP and body area massaged, mas-
sage duration, or pressure used during massage.
Previous research studies on the topic of BP change
through massage therapy demonstrate either no change or a
significant decrease in both systolic and diastolic BP; how-
ever, results of this study demonstrate that the change in BP
may be based on massage type, with certain forms of mas-
sage actually increasing the systolic BP. The most signifi-
cant increase in BP within this study was through trigger
point therapy, which may cause a pain response and thus an
increase in sympathetic nerve activity and an increase in BP.
Interestingly, a recent study by Delaney et al.
9
demonstrated
a decrease in BP with trigger point therapy; however, the
subjects in this study received a 20-minute massage that also
included “linear stroking to the sternocleidomastoid mus-
cles,” which may have stimulated the carotid sinus, thereby
causing a BP reduction. Future studies on BP changes in
massage may include a survey question on pain felt during
the massage in order to determine if this is the factor asso-
ciated with BP increase.
Overall, these data are important for therapists in the field
who treat hypertensive clients, and who are focused on re-
duction of BP. These data demonstrate that certain forms of
massage (particularly those forms that may cause pain) may
increase the client’s BP rather than decrease it. As the U.S.
population ages, hypertension will become a heavier health
care burden within U.S. society and most therapists (know-
ingly or unknowingly) will be treating clients with this dis-
order.
As in all studies, this study did have some limitations.
First, all independent data were self-reported and therefore
may contain errors. For example, therapists were asked to
indicate what forms of massage they used (Table 4), what
body parts were massaged, and the pressure applied during
the massage. Future studies may consider using more ob-
jective criteria to more specifically measure these factors,
as well as train therapists to only apply one form of mas-
sage using a specific amount of pressure or only work on
one specific body area. Likewise, as stated, the dependent
outcome of BP was measured using automated BP cuffs that
were not calibrated for validity; however, the authors be-
lieve that the equipment was reliable and therefore trusted
the measure of BP change. Future studies should use man-
ual BP readings to avoid this issue.
Second, this study was a case series and therefore was
not powered for multiple statistical tests. It is acknowledged
that the statistics in this study were overused and that some
findings may result from chance alone because of multiple
statistical testing. However, this was the first step in deter-
mining a change in BP based on different massage charac-
teristics, and future studies should be powered based on a
specific research question.
Beyond these limitations, this study was completed on a
large sample size (n 150) and did provide some useful in-
formation. Future studies are encouraged.
CONCLUSIONS
A case series of 150 massage therapy clients was com-
pleted to determine change in BP and factors associated with
this change. It was determined that certain types of massage
MASSAGE AND BLOOD PRESSURE
69
T
ABLE
4. T
YPE OF
T
REATMENT AND
A
SSOCIATION WITH
C
HANGE IN
B
LOOD
P
RESSURE
Sample size
a
Systolic change (before-after) Diastolic change (before-after)
Received treatment? No Yes No Yes p-value No Yes p-value
Swedish massage 15 135 0.07 2.01 0.46 1.07 0.26 0.54
Deep tissue 88 62 2.33 1.05 0.46 0.26 0.06 0.81
Trigger point (TP) 118 32 2.79 1.84 0.02
b
0.41 2.09 0.12
Myofascial release (MR) 136 14 2.16 1.71 0.18 0.07 0.64 0.80
Sports massage (Sp) 142 8 2.16 4.63 0.07
b
0.09 0.75 0.82
CranioSacral 149 1 1.91 14.00 0.12 0.09 5.00 0.54
TP MR 142 8 2.09 3.25 0.16 0.16 5.13 0.07
b
TP Sp 145 5 2.07 6.00 0.09
b
0.11 7.00 0.05
b
MR Sp 145 5 1.91 1.40 0.48 0.15 0.60 0.84
TP MR Sp 147 3 1.84 0.00 0.76 0.01 6.67 0.15
a
Most subjects received more than one type of massage during the massage session.
b
Indicates significant group difference at p 0.10
were significantly associated with an increase in systolic
BP, such as trigger point therapy and sports massage; and
that there appeared to be a decrease in systolic BP with
Swedish massage, although this result was not statistically
significant. Likewise, a combination of trigger point ther-
apy and sports massage was associated with an increase in
systolic as well as diastolic BP. No significant association
was noted between change in BP and body area massaged,
duration, or pressure used during massage.
ACKNOWLEDGMENTS
The authors thank the National University of Health Sci-
ences for their financial support and many student massage
therapists for their help with this study.
REFERENCES
1. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alterna-
tive medicine use in the United States, 1990–1997: Results of
a follow-up national survey. JAMA 1998;280:1569–1575.
2. Kaufmann MA. Autonomic responses as related to nursing
comfort measures. Nursing Res 1964;13:45–55.
3. Longworth JCD. Psychophysiological effects of slow stroke
back massage in normotensive females. Advan Nurs Sci
1982;50:44–61.
4. Barr JS, Taslitz N. The influence of back massage on auto-
nomic functions. Phys Ther 1970;50:1679–1691.
5. Bauer WC, Dracup KA. Physiologic effects of back massage
in patients with acute myocardial infarction. Focus Crit Care
1987;14:42–46.
6. Cady SH, Jones GE. Massage therapy as a workplace inter-
vention for reduction of stress. Percept Mot Skills 1997;84:
157–158.
7. Fakouri C, Jones P. Relaxation RX: Slow stroke back rub. J
Geront Nurs 1996;13:32–35.
8. Hernandez-Reif M, Field T, Krasnegor J, et al. High blood
pressure and associated symptoms were reduced by massage
therapy. J Bodywork Mov Ther 2000;4:31–38.
9. Delaney JPA, Leong KS, Watkins A, Brodie D. The short-
term effects of myofascial trigger point massage therapy on
cardiac autonomic tone in healthy subjects. J Advan Nurs
2002;37:364–371.
10. McNamara ME, Burnham DC, Smith C, Carroll DL. The ef-
fects of back massage before diagnostic cardiac catheteriza-
tion. Altern Ther 2003;9:50–57.
11. Hajjar I, Kotchen TA. Trends in prevalence, awareness, treat-
ment, and control of hypertension in the United States,
1988–2000. JAMA 2003;290:199–206.
12. Plaugher G, Long CR, Alcantara J, et al. Practice-based ran-
domized controlled-comparison clinical trial of chiropractic
adjustments and brief massage treatment at sites of subluxa-
tion in subjects with essential hypertension: Pilot study. J Man
Physiol Ther 2002;25:221–239.
Address reprint requests to:
Jerrilyn Cambron, D.C., M.P.H., Ph.D.
Department of Research
National University of Health Sciences
200 East Roosevelt Road
Lombard, IL 60148
E-mail: jcambron@nuhs.edu
CAMBRON ET AL.
70
