Journal of Cardiovascular Nursing

Vol. 31, No. 1, pp 73Y83 x Copyright B 2016 Wolters Kluwer Health, Inc. All rights reserved.

Effects of Massage on Blood Pressure in

Patients With Hypertension and
Prehypertension
A Meta-analysis of Randomized Controlled Trials
Downloaded from https://journals.lww.com/jcnjournal by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3w7SK+r5dgwMwHo5hr5RCGDEtttFxeKtzQ1hWr6vBEl3SLJcDM5qfXg== on 05/19/2019

I-Chen Liao, MSN, RN; Shiah-Lian Chen, PhD; Mei-Yeh Wang, PhD; Pei-Shan Tsai, PhD

Background: Massage may help reduce blood pressure; previous studies on the effect of massage on blood
pressure have presented conflicting findings. In addition, no systematic review is available. Objective: The aim
of this study was to evaluate the evidence concerning the effect of massage on blood pressure in patients
with hypertension or prehypertension. Methods: A search was performed on electronic database records up
to October 31, 2013, based on the following medical subject headings or keywords: hypertension, massage,
chiropractic, manipulation, and blood pressure. The methodological quality of randomized controlled trials
was assessed based on the Cochrane collaboration tool. A meta-analysis was performed to evaluate the effect
of massage on hypertension. The study selection, data extraction, and validation were performed independently
by 2 reviewers. Results: Nine randomized controlled trials met our inclusion criteria. The results of this study show
that massage contributes to significantly enhanced reduction in both systolic blood pressure (SBP) (mean difference,
j7.39 mm Hg) and diastolic blood pressure (DBP) (mean difference, j5.04 mm Hg) as compared with control
treatments in patients with hypertension and prehypertension. The effect size (Hedges g) for SBP and DBP was
j0.728 (95% confidence interval, j1.182 to j0.274; P = .002) and j0.334 (95% confidence interval, j0.560 to
j0.107; P = .004), respectively. Conclusion: This systematic review found a medium effect of massage on SBP and
a small effect on DBP in patients with hypertension or prehypertension. High-quality randomized controlled trials are
urgently required to confirm these results, although the findings of this study can be used to guide future research.
KEY WORDS: blood pressure, hypertension, massage, meta-analysis

Background drug treatments.2 Successful treatment of high blood pres-

sure is based on a combination of nonpharmacological
Hypertension is a global problem that typically presents and pharmacological therapies.3 Massage therapy is 1 of
as an asymptomatic, chronic cardiovascular disease.1 the most widely accepted complementary and alterna-
However, blood pressure goals are achieved by only tive medicine therapies, and it has become a multibillion-
25% to 40% of patients who take antihypertensive dollar industry in the United States, with 8.3% of adults
I-Chen Liao, MSN, RN receiving at least 1 massage treatment in 2007.4Y6
Doctoral Student, Graduate Institute of Nursing, College of Nursing, Although massage therapy can help reduce blood pres-
Taipei Medical University, Taipei; and Department of Nursing, College of
Medicine and Nursing, Hung Kuang University, Taichung, Taiwan. sure, previous studies have presented conflicting results.
Shiah-Lian Chen, PhD A meta-analysis showed that massage therapy can effi-
Associate Professor, Department of Nursing, National Tai-Chung ciently reduce blood pressure,7 but the effectiveness of
University of Science and Technology, Taichung, Taiwan. massage in managing hypertension remains unclear, as
Mei-Yeh Wang, PhD
Associate Professor, Department of Nursing, Cardinal Tien Junior
are the optimal type and regimen of massage therapy
College of Healthcare and Management, New Taipei City, Taiwan. for lowering blood pressure. Therefore, a meta-analysis
Pei-Shan Tsai, PhD on the effects of massage therapy on blood pressure in
Professor and Associate Dean, Graduate Institute of Nursing, College patients with hypertension and prehypertension was per-
of Nursing, Taipei Medical University; and Sleep Science Center,
Taipei Medical University Hospital, Taipei, Taiwan. formed in this study.
The authors have no conflicts of interest to disclose.
Methods
Correspondence
Pei-Shan Tsai, PhD, Graduate Institute of Nursing, College of Data Sources
Nursing, Taipei Medical University, 250 Wu-Hsing St, Taipei City,
Taiwan 110, ROC (ptsai@tmu.edu.tw). The records of 6 electronic databases were searched from
DOI: 10.1097/JCN.0000000000000217 their inception until October 31, 2013. The databases

73

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

74 Journal of Cardiovascular Nursing x January/February 2016

examined were the Cochrane Library (Issue 6 of 12, between experimental and control groups were estimated.
2013), PubMed, OVID-MEDLINE, CINAHL, Scopus, Hedges g and 95% confidence interval (CI) were cal-
and EMBASE. culated to determine the effect size. In studies involving
multiple control groups, the researchers restricted their
Inclusion and Exclusion Criteria analysis to the massage therapy group and a single con-
trol group. The I2 statistics and Cochran Q test were used
Original randomized controlled trials were included in this to assess heterogeneity, where an I2 value of 50% or more
review based on the following criteria: (1) participantsV represented substantial heterogeneity.9 To improve the
studies involving patients with hypertension or prehyper- accuracy of the results, we used a random effect model
tension were included, with no limitations on participant using variation factors among studies as a correction
age, sex, or nationality; (2) interventionVresearch where weight. We performed Egger tests to detect the existence
massage therapy was used as an intervention, using of publication bias among the included studies, with the
either the hands or mechanical devices; (3) comparisonV significance level set at .05.10 Potential publication bias
studies involving specific control interventions were was also examined using a funnel plot. An asymmetric
included (ie, use of a placebo, no treatment, standard inverted funnel-shape scatter plot of treatment effects
care, or any active treatment unrelated to massage therapy); against a measure of study size would indicate potential
and (4) outcomesVstudies were included where the main publication bias or the existence of systematic heteroge-
outcomes of interest involved systolic blood pressure (SBP) neity.11 Subgroup analyses according to type of massage,
and diastolic blood pressure (DBP). Trials were excluded sex, type of control group (active or inactive), whether
if they were not reported in English. blinding was involved, treatment sessions, and the mas-
saged area of the body were conducted. Control groups
Study Selection
were classified into inactive and active control groups.
Two reviewers independently screened the titles and Inactive controls referred to any condition that was thought
abstracts of potentially eligible articles by using a pre- to exert no active treatment effect, such as placebo or usual
determined search strategy. Subsequently, the full-text care controls. Active controls were defined as alternative
manuscripts of candidate studies were retrieved and interventions that were believed to exert active treatment
reviewed. Finally, 9 studies meeting the inclusion criteria effects on the outcome of interest, such as other forms of
were included in the meta-analysis. relaxation treatments. Comprehensive Meta-Analysis
Version 2.0 was used to perform the meta-analysis
Data Extraction (Biostat Inc, Englewood, New Jersey).
Two reviewers extracted data independently according
to the following criteria: first author, study year, sample
size, blood pressure group (hypertension or prehyperten- Results
sion), sex, mean participant age, experimental and control
Included Studies
group intervention, massage type, massage area, treatment
dose, follow-up time, and main results. Any discrepancy We identified 144 abstracts from 6 databases. Among
was discussed until the reviewers reached consensus. these, 18 articles were retrieved and reviewed in full,
and a further 9 articles were excluded. Articles were
Methodological Quality Assessment excluded only if they did not meet the inclusion criteria
The methodological quality of randomized controlled (eg, case report, participants were healthy, not relevant).
trials was assessed independently by the 2 raters based Thus, 9 randomized controlled trials were considered
on the Cochrane collaboration tool for assessing risk of eligible, all of which were reported in English.12Y20
bias, using the reporting criteria: low, high, or unclear risk Figure 1 depicts a summary of the study selection process.
of bias. This tool considers the presence of bias resulting
from random sequence generation, allocation conceal- Study Characteristics
ment, blinding of participants and personnel, incomplete The key data of the 9 included studies are summarized
outcome data (because of high discontinuation rates, in Table 1. In total, 278 participants were included. The
analysis type, or missing data), selective reporting, and per-group sample size ranged from 8 to 28 participants,
other biases.8 The 2 raters independently assessed the and 5 trials12Y16 involved only female participants. The
risk of bias; when necessary, consensus was determined number of massage therapy sessions ranged from 1 to
with the help of a third rater as the arbitrator. 20, with session durations of 10 to 60 minutes. The total
exposure over the study period was 20 to 300 minutes.
Data Synthesis and Analysis
Five studies involved using Swedish massage ther-
The mean change in outcomes (ie, SBP and DBP) from apy,12,13,15,17,18 1 study involved therapeutic chair mas-
baseline to posttest was calculated and the differences sage therapy,16 1 study used light-touch and stroking-type

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

 Effects of Massage on Blood Pressure 75

FIGURE 1. Study selection process.

massage,19 and 2 studies were unclear about the mas- and DBP (mean difference, j5.04 mm Hg) when com-
sage type.14,20 The areas massaged were categorized pared with the control treatment.
as the whole body, the head, the neck, and back area As shown in Figure 2A and B, massage therapy had
(including the face, neck, shoulders, upper chest, and a significant effect on SBP and DBP. The effect size of
back). In terms of control conditions, 4 studies used ac- SBP was j0.728 (95% CI, j1.182 to j0.274; P =
tive control groups16Y18,20 and 5 studies used inactive .006), and that of DBP was j0.334 (95% CI, j0.560
control groups.12Y15,19 Table 2 shows a summary of to j0.107; P = .004). Heterogeneity was observed among
the methodological quality of results. We minimized the studies for SBP (# 2 = 27.599, df = 8, P = 0.001; I2 =
the potential for publication bias by conducting a thorough 71%). For DBP, the test for heterogeneity showed homo-
literature search. The generated funnel plots and Egger geneity among studies (# 2 = 5.202, df = 8, P = 0.736;
test results for the primary outcomes exhibited sym- I2 = 0%).
metry and indicated no bias.
Subgroup Analysis: Systolic Blood Pressure
Massage therapy significantly improved SBP in patients
Effects of Massage Therapy on Blood Pressure
in the following: (1) 5 studies using Swedish massage
The mean change in outcomes (ie, SBP and DBP) from therapy; (2) 5 studies that included all-female samples;
baseline to posttest was calculated, and the differences (3) 4 studies that included both female and male sam-
between experimental and control groups were esti- ples; (4) 4 studies that included active control groups;
mated. One study did report insufficient data for cal- (5) 5 studies that included inactive control groups; (6)
culating the mean difference.17 The mean difference was 4 studies that blinded outcome assessors; (7) 7 studies in-
thus calculated using the Vernier Caliper by measuring volving multiple massage sessions; and (8) 6 studies of mas-
the figure that showed the baseline and posttest data sage therapy involving the head, neck, or back. Moderator
for the massage group and the comparison group. analysis indicated that the studies that blinded outcome
Overall, the pooled results from the 9 studies indicated assessors yielded a larger effect size than did those that
that massage contributed to a significantly greater re- did not involve blinding (P G .001); studies that used an
duction in both SBP (mean difference, j7.39 mm Hg) inactive control group had a significantly larger effect

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

 TABLE 1 Summary of Controlled Clinical Studies of Massage for Hypertension
Time
Point for
First Author (Year) Sample Intervention Massage Area Dose Posttest
1. Supa’at (2013) 16 subjects Massage (Swedish massage), n = 8 Whole body 60 min once a week 48 h
Prehypertension and stage I Placebo (rest at the same environment 1 session
hypertension with no massage), n = 8 Dose: 60 min
Women
Mean age, 51 y
2. Givi (2013) 50 subjects Massage (Swedish massage), n = 25 Face, neck, shoulders, and upper chest 10 min 72 h
Prehypertensive Placebo (relaxed at the same environment 3 times a week
Women with no massage), n = 25 10 sessions
18Y60 y Dose: 100 min
3. Ju (2013) 55 subjects Massage group (unclear type), n = 28 Whole body, back, posterior legs, 60 min Immediate
Hypertensive patients No-treatment control group, n = 27 anterior legs, abdomen, arms, 5 sessions in 4 wk
Women and shoulders Dose: 300 min
Mean age, 53 y
4. Moeini (2011) 50 subjects Massage (Swedish massage), n = 25 Face, neck, shoulders, and upper chest 10 min 3 times a week Immediate
Prehypertensive Placebo (relaxed at the same environment 10 sessions
Female with no massage), n = 25 Dose: 100 min
76 Journal of Cardiovascular Nursing x January/February 2016

18Y60 y
5. Jefferson (2010) 34 subjects Group 1 (therapeutic chair massage Back 20 min Immediate
Hypertension intervention), n = 17 1 session
Female Group 2 (received patient teaching on Dose: 20 min
Mean age, 53 y diaphragmatic breathing), n = 17
6. Olney (2008) 29 subjects Treatment group (Swedish massage), n = 15 Back 10 min Immediate
Prehypertension or controlled Control group (relaxation sessions using 10 times over 3.5 weeks
hypertension learned relaxation techniques), n = 14 10 sessions
Men and women Dose: 100 min
Mean age, 48 y
7. Olney (2005) 14 subjects Treatment group (Swedish massage), n = 8 Back 10 min 3 times a week 48 h
Hypertension for at least 1 y Control group (self-determined 10 sessions
Men and women relaxation method), n = 6 Dose: 100 min

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Mean age, 54 y
8. Plaugher (2002) 14 subjects Treatment group (light-touch, stroking-type Localized regions of the spine 10 min Immediate
Hypertension massage), n = 8 3 times/week/first
Men and women Control group (rested alone), n = 6 month, 2 times/week/
Mean age, 37 y second month
20 sessions
Dose: 200 min
9. Hernandez-Reif (2000) 30 subjects Massage (unclear type), n = 15 Whole body, head/arms/torso/legs/back 30 min Immediate
Hypertension Progressive (muscle relaxation), n = 15 2 times a week
Male and female 10 sessions
Mean age, 51 y Dose: 300 min
 Effects of Massage on Blood Pressure 77

TABLE 2 Summary of Cochrane’s Risk of Bias

Random Blinding of Blinding Incomplete Other
Sequence Allocation Participants of Outcome Outcome Selective Sources
First Author (year) Generation Concealment and Personnel Assessment Data Reporting of Bias
1. Supa’at (2013) V V X ? V V V
2. Givi (2013) V V X V V V V
3. Ju (2013) ? ? V ? V V V
4. Moeini (2011) V V X V V V V
5. Jefferson (2010) V V X ? V V ?
6. Olney (2008) V V X ? V V V
7. Olney (2005) V V V V V V V
8. Plaugher (2002) V V X V X X V
9. Hernandez-Reif (2000) V V V ? X X V

X = high risk; V = low risk; ? = unclear risk.

size than did those that included an active control group Subgroup Analysis: Diastolic Blood Pressure
(P = .003). The results are illustrated in Figure 3AYH and Massage therapy significantly improved DBP in patients
summarized in Table 3. Meta-regression revealed that the in the following: (1) 4 studies that used non-Swedish
effect of massage on SBP was not significantly associated massage therapy; (2) 4 studies that enrolled both female
with total massage time (P = .058). and male participants in the sample; (3) 4 studies involving

FIGURE 2. Forest plots of the overall effect of massage on hypertension. (A) Systolic. Test for overall effect: Z = 3.142 (P =
.002). (B) Diastolic. Test for overall effect: Z = 2.889 (P = .004). SBP indicates systolic blood pressure; DBP, diastolic
blood pressure; CI, confidence interval.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

78 Journal of Cardiovascular Nursing x January/February 2016

FIGURE 3. Continues.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

 Effects of Massage on Blood Pressure 79

FIGURE 3. Forest plots of the subgroup effect of massage on systolic blood pressure in hypertension. (A) Swedish massage. Test
for overall effect: Z = j2.901 (P = .004). (B) Female. Test for overall effect: Z = j2.291 (P = .022). (C) Female and male. Test for
overall effect: Z = j2.499 (P = .049). (D) Active. Test for overall effect: Z = j2.126 (P = .034). (E) Inactive. Test for overall
effect: Z = j2.161 (P = .031). (F) Blinding. Test for overall effect: Z = j5.004 (P = .000). (G) Multiple sessions. Test for overall
effect: Z = j2.999 (P = .003). (H) Head, neck, or back. Test for overall effect: Z = j3.331 (P = .001). A forest plot is a
graphical representation of component studies within a meta-analysis. A horizontal line runs through the square to show its
95% confidence interval. The squares represent point estimates for each study, and the overall estimate from the meta-analysis
and its confidence interval are noted at the bottom (represented as a diamond). Each study is marked by a square (placed where
the data measure the effect). The size of the each square was proportional to the weight that the individual study contributed to
the meta-analysis. SBP indicates systolic blood pressure; DBP, diastolic blood pressure; CI, confidence interval.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

80 Journal of Cardiovascular Nursing x January/February 2016

TABLE 3 Results of Systolic Blood Pressure Subgroup Effect Analyses: Number of Trials per
Subgroup, Effect Sizes, Heterogeneity, and Subgroup Differences
Subgroup Category Number of Trials Effect Size (Confidence Interval) Heterogeneity Pa
Type of massage
Swedish 5 j1.051 (j1.761 to j0.341)b I2 = 75% .079
Non-Swedish 4 j0.320 (j0.655 to 0.015) I2 = 0%
Sex
Female 5 j0.776 (j1.440 to j0.047)b I2 = 80% .809
Female and male 4 j0.536 (j0.957 to j0.116)b I2 = 50%
Control group
Active 4 j0.605 (j1.163 to j1.103)b I2 = 50% .003a
Inactive 5 j0.779 (j1.486 to j0.073)b I2 = 80%
Blinding of outcome assessors
Yes 4 j1.382 (j1.923 to j0.841)b I2 = 43% G.001a
No 5 j0.269 (j0.569 to 0.032) I2 = 0%
Session
Single session 2 j0.331 (j0.872 to 0.209) I2 = 0% .334
Multiple sessions 7 j0.853 (j1.411 to j0.296)b I2 = 75%
Area
Whole body 3 j0.230 (j0.613 to 0.152) I2 = 0% 0.080
Head, neck, or back 6 j0.988 (j1.570 to j0.407)b I2 = 71%
a
Moderator analyses for subgroup differences.
b
Significant subgroup effects, P G .05.

active control groups; (4) 5 studies with blinded out- Meta-regression results revealed a nonsignificant associ-
come assessors; (5) 7 studies reporting multiple treatment ation between doses and effect size (P = .735).
sessions; (6) 3 studies that involved whole body massage;
and (7) 6 studies that reported head, neck, or back massage Discussion
therapy. As shown in Table 4, moderator analysis was
Overview of Findings
nonsignificant for all moderators except for the moderating
effect of control conditions. Studies that used an active When the results of all 9 studies in this review were
control group reported significantly larger effect sizes than pooled, we observed that massage therapy was asso-
did those involving an inactive control group (P = .011). ciated with a significant reduction in SBP and DBP in

TABLE 4 Results of Diastolic Blood Pressure Subgroup Effect Analyses: Number of Trials per
Subgroup, Effect Sizes, Heterogeneity, and Subgroup Differences
Subgroup Category Number of Trials Effect Size (Confidence Interval) Heterogeneity Pa
Type of massage
Swedish 5 j0.301 (j0.608 to 0.006) I2 = 8% .759
Non-Swedish 4 j0.373 (j0.708 to j0.037)b I2 = 0%
Sex
Female 5 j0.240 (j0.510 to 0.030) I2 = 0% .212
Female and male 4 j0.556 (j0.972 to j0.140)b I2 = 0%
Control group
Active 4 j0.518 (j0.906 to j0.130)b I2 = 0% .011a
Inactive 5 j0.238 (j0.528 to 0.052) I2 = 0%
Blind of outcome assessors
Yes 4 j0.177 (j0.538 to 0.183) I2 = 0% .275
No 5 j0.435 (j0.726 to j0.145)b I2 = 0%
Session
Single session 2 j0.508 (j1.054 to 0.037) I2 = 0% .491
Multiple sessions 7 j0.297 (j0.546 to j0.049)b I2 = 0%
Area
Whole body 3 j0.392 (j0.777 to j0.006)b I2 = 0% .715
Head, neck, or back 6 j0.303 (j0.583 to j0.023)b I2 = 0%
a
Moderator analyses for subgroup differences
b
Significant subgroup effect, P G .05.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

 Effects of Massage on Blood Pressure 81

patients with hypertension and prehypertension. Mas- minutes) as a continuous variable, the treatment effect
sage therapy contributed to a 7Ymm Hg reduction in SBP was not significantly associated with dose. However, a
and 5Ymm Hg reduction in DBP as compared with control marginally significant association between the dose and
conditions. A recent meta-analysis reported that admin- effect size was observed for SBP (P = .058). Further stud-
istering blood pressure-lowering treatment did not con- ies are needed to investigate the optimal dose of massage
tribute to a significant reduction in cardiovascular risk therapy.
among patients with mild hypertension.21 However, linear Previous meta-analyses or systematic reviews have
associations were observed between blood pressure assessed the effect of massage therapy28Y30; however,
and angina, myocardial in farction, or subarachnoid those reviews were limited in scope, and the partic-
hemorrhage.22 Treatments that reduce blood pressure, ipants and outcome measures differed from those dis-
therefore, can potentially reduce the risk of various cussed in this study. Previous reviews have involved
cardiovascular diseases. Moreover, a lifetime burden of healthy men, hospice patients, postYmyocardial infarc-
hypertension has been reported. Compared with partici- tion patients, breast cancer patients, or shoulder pain
pants with normal blood pressure, those with blood patients. The outcome measures in those studies included
pressure equal to or higher than 140/90 mm Hg ex- pain, anxiety, depression, stress, fatigue, and physio-
hibited increased overall cardiovascular disease at logical components of relaxation. Only 1 meta-analysis
30 years of age, and they developed cardiovascular disease concluded that applied massage therapy reduced blood
5 years earlier.22 A 7Ymm Hg reduction in SBP and pressure; however, the participants reviewed in that study
5Ymm Hg reduction in DBP are thus potentially of included healthy adults, hypertensive adults, healthy
clinical significance. women, patients undergoing cardiac catheterization,
Subgroup analyses of the type of massage revealed po- and soldiers.7
tentially paradoxical findings for SBP and DBP. Swedish For SBP, heterogeneity was observed among the re-
massage therapy effectively reduced SBP, whereas non- viewed studies. To explore the factors contributing to
Swedish massage therapy effectively reduced DBP. Raised heterogeneity, we examined whether the type of mas-
SBP is modulated partially by a change in balance of sage, the sex of participants, control group (active or
sympathetic activity and parasympathetic activity.23 inactive), quality of studies (blinding of outcome as-
Swedish massage is known for its therapeutic relaxation sessors), treatment sessions, and the massaged area of
effects, and hypertension is associated with stress.12 the body affected the effectiveness of the treatment. As
Swedish massage is classified as a superficial type of expected, moderator analyses revealed that the bene-
massage, consisting of 5 primary stroking actions that ficial effect of massage therapy for SBP was greater for
stimulate the circulation of blood through the soft tissues studies involving inactive control groups compared with
of the body.24,25 The significant reduction in heart rate those that had active control groups, and larger for
induced by Swedish massage supports the link between studies with better quality (blinding assessors). Numerous
enhanced parasympathetic activity and reduced SBP.25 potentially interesting moderator variables remain un-
On the other hand, DBP was influenced by peripheral examined in primary massage therapy research, and
vascular resistance26,27; the mechanical techniques of only 1 study has determined change in blood pressure
the Swedish massage may not have direct effect on changes and associated factors.31 Systolic blood pressure seemed
in peripheral vascular resistance. Of note, 2 studies14,20 to decrease after Swedish massage therapy, but the as-
that we classified as using non-Swedish massage therapy sociation between changes in blood pressure and the
reported inadequately detailed information to determine massaged area of the body, duration of the massage, or
the exact treatment components. Thus, misclassifica- pressure applied was not statistically significant.31 Previous
tion of massage type might have occurred. studies lack a clear definition of the types of massage
Significant reductions were observed in both SBP therapy, and they have a limited number of participants.
and DBP when massage therapy was compared with Unfortunately, the use of different massage therapy tech-
active control treatments. The superior effect of mas- niques by different practitioners may affect the associated
sage therapy in comparison with active control treat- therapeutic outcomes.
ments further supports the notion that massage therapy
exerts specific treatment effects in reducing SBP and
Limitations of the Review
DBP. Furthermore, our results provide evidence of mod-
erating factors that could influence the effectiveness of Some limitations were encountered while conducting
treatment effects. For example, massage therapy involv- this systematic review. This review may have been
ing the head, neck, or back was effective in reducing both affected by the heterogeneity in massage therapy char-
SBP and DBP. Significant reductions in SBP and DBP acteristics, including frequency, duration, number of
were observed for both sexes. Multiple sessions of mas- sessions, and massage technique. The number of in-
sage therapy were more effective than 1 session alone. cluded studies was small, with each study also having
When considering the total exposure of massage (in a relatively small sample. Bias was also possible, and

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

82 Journal of Cardiovascular Nursing x January/February 2016

Self-monitoring and other non-pharmacological interventions

What’s New and Important to improve the management of hypertension in primary care:
a systematic review. Br J Gen Pract. 2010;60(581):e476Ye488.
h Massage therapy significantly reduces SBP and DBP 4. Suzuki S, Ohta T. Non-pharmacological treatment of hyper-
compared with control. tension in the elderly. Nihon Rinsho. 2005;63(6):1010Y1015.
h Swedish massage therapy causes the largest reduction
5. Moraska A, Pollini RA, Boulanger K, Brooks MZ,
in SBP in patients with hypertension and prehypertension. Teitlebaum L. Physiological adjustments to stress measures
h Future studies of higher quality are needed, with a
following massage therapy: a review of the literature. Evid
particular emphasis on Swedish massage therapy. Based Complement Alternat Med. 2010;7(4):409Y418.
6. Prasad K, Sharma V, Lackore K, Jenkins SM, Prasad A,
Sood A. Use of complementary therapies in cardiovascular
blinding subjects and therapists for treatments is difficult. disease. Am J Cardiol. 2013;111(3):339Y345.
However, assessor blinding and allocation concealment 7. Moyer CA, Rounds J, Hannum JW. A meta-analysis of mas-
are critical for reducing bias. Only 4 studies have used sage therapy research. Psychol Bull. 2004;130(1):3Y18.
8. Higgins JPT, Green S (Eds). Cochrane Handbook for Sys-
blinding in their outcome assessments.
tematic Reviews of Interventions Version 5.1.0 [updated
Moderator analyses were performed to explain grad- March 2011]. The Cochrane Collaboration. 2011. http://
ients in treatment effect among subgroups. How- www.cochrane-handbook.org. Accessed October 31, 2013.
ever, because moderator variables cannot be assumed 9. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring
to be statistically independent,32 inferences drawn from inconsistency in meta-analyses. BMJ. 2003;327:557Y560.
these analysis results should be interpreted with caution. 10. Egger M, Davey Smith G, Schneider M, Minder C. Bias in
meta-analysis detected by a simple, graphical test. BMJ. 1997;
315:629Y634.
Methodological Recommendations for 11. Sterne JA, Egger M. Funnel plots for detecting bias in meta-
Future Studies analysis: guidelines on choice of axis. J Clin Epidemiol. 2001;
54(10):1046Y1045.
Future studies considering the mechanism of massage 12. Supa’at I, Zakaria Z, Maskon O, Aminuddin A, Nordin NA.
therapy are necessary. To strengthen our evidence, fu- Effects of Swedish massage therapy on blood pressure, heart
rate, and inflammatory markers in hypertensive women. Evid
ture studies should consider extending this review after
Based Complement Alternat Med. 2013;2013:171852.
the publication of new eligible randomized controlled 13. Givi M. Durability of effect of massage therapy on blood
trials concerning the effect of massage therapy on hy- pressure. Int J Prev Med. 2013;4(5):511Y516.
pertension. However, future research must include larger 14. Ju MS, Lee S, Bae I, Hur MH, Seong K, Lee MS. Effects of
samples to increase their explanatory power. In addi- aroma massage on home blood pressure, ambulatory blood
pressure, and sleep quality in middle-aged women with hy-
tion, blinding the researcher to the data to prevent pos-
pertension. Evid Based Complement Alternat Med. 2013;
sible bias will be critical. Moreover, most studies measured 2013:403251.
BP immediately before and after massage. Future studies 15. Moeini M, Givi M, Ghasempour Z, Sadeghi M. The effect
evaluating the intermediate or long-term effects of mas- of massage therapy on blood pressure of women with pre-
sage on blood pressure are needed. Finally, future designs hypertension. Iran J Nurs Midwifery Res. 2011;16(1):61Y70.
16. Jefferson LL. Exploring effects of therapeutic massage and
should consider the massage type, regimen, massaged
patient teaching in the practice of diaphragmatic breath-
area, patient type (prehypertension versus established ing on blood pressure, stress, and anxiety in hypertensive
hypertension), and application of massage therapy for African-American women: an intervention study. J Natl Black
nonpharmacological control of hypertension. Nurses Assoc. 2010;21(1):17Y24.
17. Olney CM. The effect of therapeutic back massage in hy-
pertensive persons: a preliminary study. Biol Res Nurs. 2005;
Conclusion 7(2):98Y105.
18. Olney CM. Determining long term effects and dosage of
Significant differences between the control and mas- back massage for persons with elevated blood pressure: a
sage therapy groups in SBP and DBP were identified, clinical trial. South Online J of Nurs Res. 2008;8(2):66Y76.
with massage therapy shown to reduce both parame- 19. Plaugher G, Long CR, Alcantara J, et al. Practice-based
ters. Massage therapy is a simple nonpharmacological randomized controlled-comparison clinical trial of chiro-
practic adjustments and brief massage treatment at sites of
stress-reduction method that appears to positively impact subluxation in subjects with essential hypertension: pilot
blood pressure. This review of massage therapy for hyper- study. J Manipulative Physiol Ther. 2002;25(4):221Y239.
tension provides crucial guidelines for future research. 20. Hernandez-Reif M, Field T, Krasnegor J, Theakston H,
Hossain Z, Burman I. High blood pressure and associated
symptoms were reduced by massage therapy. J Psychosom
REFERENCES Obstet Gynaecol. 2000;4:31Y38.
1. Chockalingam A. World Hypertension Day and global aware- 21. Diao D, Wright JM, Cundiff DK, Gueyffier F. Pharmaco-
ness. Can J Cardiol. 2008;24(6):441Y444. therapy for mild hypertension. Cochrane Database Syst Rev.
2. Burnier M. World Health Organization/International Soci- 2012;8:CD006742.
ety of Hypertension, Guidelines: blood pressure control and 22. Rapsomaniki E, Timmis A, George J, et al. Blood pressure
the implementation of guidelines in clinical practice: can we and incidence of twelve cardiovascular diseases: lifetime
fill the gap? J Hypertens. 2002;20(7):1251Y1253. risks, healthy life-years lost, and age-specific associations
3. Glynn LG, Murphy AW, Smith SM, Schroeder K, Fahey T. in 1.25 million people. Lancet. 2014;383(9932):1899Y1911.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

 Effects of Massage on Blood Pressure 83

23. Biaggioni I. Interventional approaches to reduce sympa- 28. Kong LJ, Zhan HS, Cheng YW, Yuan WA, Chen B, Fang M.
thetic activity in resistant hypertension: to ablate or stimulate? Massage therapy for neck and shoulder pain: a systematic
Hypertension. 2012;59(2):194Y195. review and meta-analysis. Evid Based Complement Alternat
24. Aourell M, Skoog M, Carleson J. Effects of Swedish mas- Med: eCAM. 2013;2013:613279.
sage on blood pressure. Complement Ther Clin Pract. 2005; 29. Smith JM, Sullivan SJ, Baxter GD. Massage therapy: more
11(4):242Y246. than a modality. N Z J Physiother. 2010;38(2):44Y51.
25. Cowen VS, Burkett L, Bredimus J. A comparative study of Thai 30. Labyak SE, Metzger BL. The effects of effleurage backrub
massage and Swedish massage relative to physiological and psy- on the physiological components of relaxation: a meta-analysis.
chological measures. J Bodyw Mov Ther. 2006;10:266Y275. Nurs Res. 1997;46(1):59Y62.
26. Pinto E. Blood pressure and ageing. Postgrad Med J. 2007; 31. Cambron JA, Dexheimer J, Coe P. Changes in blood pres-
83:109Y114. sure after various forms of therapeutic massage: a prelim-
27. Guo X, Pantoni L, Simoni M, et al. Blood pressure components inary study. J Altern Complement Med. 2006;12:65Y70.
and changes in relation to white matter lesions: a 32-year pro- 32. Lipsey MW. Those confounded moderators in meta-analysis:
spective population study. Hypertension. 2009;54(1):57Y62. good, bad, and ugly. Ann Am Acad Pol Soc Sci. 2003;587:69Y81.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.