Journal of Bodywork & Movement Therapies 24 (2020) 503e518

Contents lists available at ScienceDirect

Journal of Bodywork & Movement Therapies

journal homepage: www.elsevier.com/jbmt

Myofascial Pain and Treatment

Dry cupping for musculoskeletal pain and range of motion: A

systematic review and meta-analysis
Sarah Wood a, *, Gary Fryer b, Liana Lei Fon Tan c, Caroline Cleary d
a
The Institute for Health and Sport, Victoria University, Melbourne, Australia
b
College of Health & Biomedicine, Victoria University, Melbourne, Australia
c
Endeavour College of Natural Health, Melbourne, Australia
d
Torrens University, Melbourne, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: This review evaluated the efficacy and safety of western dry cupping methods for the
Received 28 April 2019 treatment of musculoskeletal pain and reduced range of motion.
Received in revised form Methods: A systematic literature search was performed until April 2018 for randomised controlled trials
16 March 2020
(RCTs) pertaining to musculoskeletal pain or reduced range of motion, treated with dry cupping. Out-
Accepted 7 June 2020
comes were pain, functional status, range of motion and adverse events. Risk of bias and quality of
evidence was assessed using the modified Downs & Black (D&B) checklist and GRADE.
Results: A total of 21 RCTs with 1049 participants were included. Overall, the quality of evidence was fair,
with a mean D&B score of 18/28. Low-quality evidence revealed dry cupping had a significant effect on
pain reduction for chronic neck pain (MD,
21.67; 95% CI,
36.55, to
6.80) and low back pain
(MD,
19.38; 95%CI,
28.09, to
10.66). Moderate-quality evidence suggested that dry cupping
improved functional status for chronic neck pain (MD,
4.65; 95%CI,
6.44, to
2.85). For range of
motion, low quality evidence revealed a significant difference when compared to no treatment
(SMD,
0.75; 95%CI,
0.75, to
0.32).
Conclusion: Dry cupping was found to be effective for reducing pain in patients with chronic neck pain
and non-specific low back pain. However, definitive conclusions regarding the effectiveness and safety of
dry cupping for musculoskeletal pain and range of motion were unable to be made due to the low-
moderate quality of evidence. Further high-quality trials with larger sample sizes, long-term follow
up, and reporting of adverse events are warranted.
© 2020 Elsevier Ltd. All rights reserved.

1. Introduction practice (Rozenfeld and Kalichman, 2016). Dry cupping involves the
use of glass, plastic or bamboo cups that are placed over localised
Musculoskeletal pain is highly prevalent and a significant areas of skin. A vacuum suction is achieved with heat from a flame,
contributor to global disability and disease, with most countries a manual handheld pump or electrical pumping devices to create a
reporting neck and low back pain as a leading cause of disability negative pressure, drawing localised skin and soft tissue structures
(Vos et al., 2017). Improving function and controlling pain are key into the cup (Rozenfeld and Kalichman, 2016; Tham et al., 2006).
aims for the treatment of musculoskeletal pain, which typically Depending on geographical region and culture, there are variations
consists of a combination of physical therapy, self-management and in nomenclature to describe the numerous cupping approaches (Al
short-term analgesic medication (Babatunde et al., 2017). The Bedah et al., 2016). Wet cupping is the most commonly docu-
application of dry cupping therapy for reducing musculoskeletal mented application of cupping described in traditional medical
pain and improving mobility has become an increasingly frequent literature; whereby superficial skin incisions are made using a
practice among manual and physical therapists in western clinical surgical instrument to promote blood-letting (Al Bedah et al., 2016;
Farhadi et al., 2009). In contrast, dry cupping does not involve in-
cisions or penetrate the skin barrier.
Recent systematic reviews have investigated the use of cupping
* Corresponding author. in musculoskeletal pain conditions (Azizhani et al., 2018 Kim et al.,
E-mail address: [email protected] (S. Wood).

https://doi.org/10.1016/j.jbmt.2020.06.024
1360-8592/© 2020 Elsevier Ltd. All rights reserved.
504 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

2018). However, there remains an absence of reviews specifically 2.4. Trial selection
examining the use of dry cupping for the treatment of musculo-
skeletal pain and range of motion. Therefore, this study aims to Only RCTs relating to the effects of dry cupping therapy and
critically evaluate the evidence from randomised controlled trials published in the English language were included in the systematic
(RCTs) to determine the efficacy and safety of western dry cupping review. Trials published in the form of dissertations or grey litera-
methods for the treatment of musculoskeletal pain and range of ture were included. A flowchart depicting the trial selection process
motion. using the preferred reporting items for systematic review and
meta-analysis (PRISMA) is shown in Fig. 1.
2. Methods
2.5. Data extraction
The protocol for this systematic review was registered on
PROSPERO (Registration Number: CRD42018088855). The protocol Author SW independently extracted data from the included
for this review followed the Preferred Reporting Items for Sys- trials using a Cochrane Collaboration standardised data extraction
tematic Reviews and Meta-analyses (PRISMA). form LT verified the extracted data.

2.6. Assessment of heterogeneity

2.1. Criteria for considering trials for the current review
Assessment of heterogeneity was based on the calculation of I2.
2.1.1. Types of trials
The Cochrane Collaboration provides the following interpretation
Randomised controlled trial (RCTs) pertaining to the effects of
of I2: 0%e30%, might not be important; 30%e60%, may represent
dry cupping therapy for musculoskeletal pain and range of motion
moderate heterogeneity; 50%e75%, may represent substantial
were included.
heterogeneity; and 75%e100%, considerable heterogeneity
(Higgins and Green, 2011).
2.1.2. Types of participants
Participants were limited to adults (18 years and older) that 2.7. Assessment of risk of bias and methodological quality
received dry cupping treatment for musculoskeletal pain or re-
striction in range of motion. There were no restrictions to duration The Downs & Black (D&B) quality assessment scale was applied
of pain period. to evaluate the methodological quality of the included trials
(Downs and Black, 1998). SW and CC completed an independent
2.1.3. Types of intervention evaluation of the included trials; disagreement and discrepancies
Treatment was limited to dry cupping therapy for a musculo- between author evaluations were resolved through discussion or
skeletal condition as the sole intervention or combined with by consulting a third review author, GF. The D&B assessment scale
another modality. Any other form of cupping (e.g., wet cupping) is a validated risk of bias tool and has been found to have good
was excluded. inter-rater reliability (Downs and Black, 1998). The modified D&B
assessment scale provides individual scores for each section and an
2.1.4. Types of comparison overall numeric rating, out of a possible 28-points. The following
Western medicine, sham cupping treatment or cross over score ranges were given to corresponding quality levels: excellent
intervention used as a control were included. Trials were excluded (26e28), good (20e25), fair (15e19), and poor (14).
if the comparison was not relevant to Western Medicine.
2.8. Data analysis
2.2. Included outcome measures
A quantitative meta-analysis of the included trials was conducted
using RevMan 5.3 software (RevMan, Version 5.3., Cochrane Centre).
2.2.1. Primary outcomes
For binary outcomes, data were summarised using risk ratios (RR)
The primary outcomes were pain and functional status. Pain was
with 95% CI, and for continuous outcomes, data were summarised
measured by visual analogue scale (VAS), numerical rating scale
using mean difference (MD) or standard mean difference (SMD) with
(NRS), Short Form McGill Pain Questionnaire (SMPQ) and pain
95% CI. When heterogeneity I2 statistic was less than 25% a fixed-effect
pressure thresholds (PPT). Functional status was measured by
model was applied, greater than 25%, a random-effects model was
validated self-reported functional ability questionnaires, SF-36,
applied (Higgins and Green, 2011). Scores for outcome measures, such
neck disability index, Roland-Morris Disability Questionnaire,
as visual analogue scale were converted to a 100-point scale.
Oswestry Disability Index and WOMAC.
2.9. Assessment of clinical relevance
2.2.2. Secondary outcomes
Secondary outcomes were range of motion and adverse events. Assessment of clinical relevance was made using the recom-
Range of motion was measured by goniometer, inclinometer or mendations of the Cochrane Back Review Group, defined as follow:
digital software. Any adverse events reported for each trial were small effect as MD less than 10% of the scale (e.g., 10 mm on a
recorded. 100 mm VAS) and SMD or 'd' scores less than 0.5. Medium effect as
MD 10%e20% of the scale and SMD or 'd' scores from 0.5 to 0.8.
2.3. Literature searches Large effect as MD greater than 20% of the scale and SMD or 'd'
scores greater than 0.8 (Furlan et al., 2009).
The following databases were searched from their inception
until April 2018: Cochrane Central Register of Controlled Trials, 2.10. Quality of evidence
EBSCO Host, PEDro, ProQuest, PubMed, Science Direct and Scopus.
In addition, a manual search for citation references in relevant The overall quality of evidence for each outcome was assessed
literature was applied. Search strategies are outlined in Appendix A. using the Grading of Recommendations, Assessment, Development
 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518 505

Fig. 1. Flowchart of study selection depicted by preferred reporting items for systematic review and meta-analysis (PRISMA). Abbreviations: RCT, randomised controlled trial; MSK,
musculoskeletal; TCM, Traditional Chinese Medicine.

and Evaluation (GRADE) approach. Quality of evidence is specified inclusion criteria. Significant characteristics of the included trials
as high, moderate, low and very low. Key factors are (1) limitations are summarised in Table 1. Sample sizes varied from 14 to 141, with
in study design, (2) inconsistency of results, (3) indirectness or a median of 40 participants. Of the 21 included trials, 7 trials
generalisation of findings, (4) imprecision, and (5) other (such as originated from Germany, 6 from the United States, 2 from both
publication bias). Depending on the quality of evidence evaluated, India and Korea, and 1 each from Turkey, Taiwan, Iran and Egypt.
levels of evidence can be downgraded to moderate, low, or even The treated musculoskeletal conditions were neck pain (Arslan
very low (Ryan and Hill, 2016). et al., 2015; Chi et al., 2016; Cramer et al., 2011; Lauche et al 2011,
2013; Saha et al., 2017; Yim et al., 2017); low back pain (Akbarzadeh
et al., 2014; El Rahim et al., 2017; Singh et al., 2016); knee osteo-
3. Results arthritis (Khan et al., 2013; Teut et al., 2012); plantar fasciitis (Ge
et al., 2017); and fibromyalgia (Lauche et al., 2016). Numerous tri-
3.1. Included trials als investigated the use of dry cupping for increasing range of
motion for the hamstring muscle (Barger, 2016; Kim et al., 2017;
The literature search identified 262 trials (Fig. 1). Duplicates LaCross, 2014), iliotibial band (Biehl, 2017; Doozan, 2015) and
were removed, and 232 trials remained for title and abstract shoulder (Smith, 2015). The number of treatment sessions ranged
screening. A total of 21 RCTs including 1049 participants met the
Table 1

506
Overview of included clinical trials.

Study ID, year Aim Sample Participants a. Intervention b. Control Treatment Total Follow Outcome Measurement Authors conclusions D&B
and country size/ (a. duration No. of up measures times score
drop Intervention; sessions
outs b. Control

Akbarzadeh To investigate the effect of dry 100/0 Condition: Dry cupping (fire Routine care 4 days 4 2 VAS (cm), Baseline: “The study results showed Good
et al. (2014) cupping therapy at BL23 point on low back pain cupping): and referral to weeks SMPQ before cupping therapy to be effective quality
Iran the intensity of low back pain in post labour Performed on the lumbar specialist in intervention in sedation of pain. Thus, it can 21/28
primiparous women Gender erector spinae muscle case of severe Post: after be used as an effective
(female): (BL23) for 15e20 min pain intervention, treatment for reducing the low
a: 50; b: 50 every day at 2-week back pain”
Mean age follow up
(SD):
a: 25.0 (4.2)
b: 27.0 (3.8)
“Cupping therapy is a non-

S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

Arslan et al. To investigate the effectiveness of 40/0 Condition: Dry moving cupping Unclear 5 weeks 10 Nil VAS Baseline: Poor
(2015) active dry cupping of the upper upper (vacuum cupping): before invasive and harmless quality
Turkey shoulder and neck to alleviate shoulder and Performed twice weekly intervention therapeutic application, and it 11/28
pain neck pain 30 min cupping session to Post: after can be confidently used to
Gender upper shoulders and neck intervention reduce upper shoulder and
(female) neck pain …”
a: 20; b: 20
Mean age
(SD):
Not reported
Barger (2016) To examine the effects of 20/0 Condition: Dry cupping (vacuum Graston Single 1 Nil ROM, Baseline: “The results of this study Poor
United States compressive or decompressive hamstring cupping): Technique for Session Muscle before suggest that both GT and MFD quality
(thesis soft tissue techniques on flexibility 6 stationary cups over the 4 min Strength, intervention are effective therapies for 14/28
publication) hamstring flexibility, strength, Gender hamstrings for 3 min, PFAQ, GROC Post: after improving hamstring flexibility
and perceived function (male): 1 min of moving cupping intervention and strength and decreasing
a: 10; b: 10 pain immediately following the
Mean age therapy”
(SD)
21.35 (1.76)
Biehl (2017) To determine if dry cupping is an 40/0 Condition: Dry cupping (vacuum Sham cupping Single 1 24 h AROM, Baseline: “… This study supports the Fair
United States effective treatment intervention ITB flexibility cupping): session PROM before assumption that dry cupping quality
(thesis in releasing ITB tightness and Gender 4 stationary cups placed intervention may be a safe and effective 15/28
publication) increasing hip and knee range of (male/ along the ITB for 7 min Post: after treatment option to combat ITB
motion in a physically active female): 17/ intervention, tightness seen by clinicians.”
population. 23 at 24 h follow
Mean age up
(SD):
21 (1.8)
Chi et al. To investigate the effectiveness of 60/0 Condition: Dry cupping (fire Rest for 20 min Single 1 Nil VAS, SST, BP Baseline: “One treatment of cupping Good
(2016) cupping therapy in changes of chronic neck cupping): session before therapy is shown to increase quality
Taiwan skin surface temperature for and shoulder Cups placed on 3 intervention SST and reduce systemic blood 23/28
relieving chronic neck and pain acupoints for 10 min Post: after pressure. Cupping therapy
shoulder pain Gender before repeating on intervention mimics an analgesic effect
(male/ opposite side which has no known negative
female) side effects and may be
a: 3/27; b: 2/ considered safe”
28
Mean age
(SD):
a: 43.6 (8.0)
b: 42.5 (7.4)
To investigate the effect of 50/2 Condition: Dry cupping (pneumatic Usual care 14 days 5 18 days Baseline: “Pneumatic pulsation therapy
pneumatic pulsation therapy on chronic neck pulsatile cupping): before appears to be a safe and
Cramer et al. chronic neck pain compared to pain Glass cupping massage VAS, NDI, intervention effective method to relieve pain Good
(2011) standard medical care Gender over neck and shoulders PPT, MDT, Post: after and to improve function and quality
Germany (male/ for 10e15min with a VDT, SF-36 intervention quality of life in patients with 24/28
female) mechanical device, chronic neck pain”
a: 4/20; b: 6/ followed with 4 stationary
18 cups applied to the
Mean age trapezius muscle for 5
(SD): e10min
a: 44.46
(10.79)
b: 47.88
(13.50)
Doozan (2015) To evaluate the effectiveness of 32/3 Condition: Dry cupping (vacuum Non-treatment 6 weeks 6 2 ROM Baseline: “The study results showed Poor
United States Chinese cupping in increasing ITB flexibility cupping): leg used as weeks before Chinese cupping is beneficial as quality
(thesis iliotibial band range of motion Gender a1: 8 cups over ITB left for control intervention a short term and a possible 11/28
publication) (male) 5e10min þ stationary During: 2 long-term therapeutic
a1: 10; a2: 9; bike for 10 min. weeks, 4 technique that can be used to

S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

a3: 10 a2: Cupping for weeks increase ROM in athletes”
Mean age 10 min þ stretch protocol. Post: after
(SD): a3: 10 min stationary intervention,
a1: 20.6 (1.3) bike þ stretch protocol at 2 weeks
a2: 19.9 (1.2) follow up
a3: 20.3 (1.2)
El Rahim et al. To investigate the effect of 60/0 Condition: Dry cupping (fire 30 min of 4 weeks 12 Nil ROM, SMPQ, Baseline: “Cupping therapy and Good
(2017) cupping therapy with inferential mechanical cupping): traditional RMDQ before interferential therapy in quality
Egypt therapy on mechanical low back LBP a1: Cupping physical intervention addition to traditional physical 20/28
pain Gender therapy þ physical therapy Post: after therapy can be used as an
(male/ therapy intervention effective treatment in patients
female) a2: Cupping with mechanical low back
a1: 10/10 therapy þ interferential pain”
a2: 10/10 therapy for
b: 10/10 30min þ physical therapy
Mean age
(SD):
a1: 27.35
(4.23)
a2: 28.8
(4.57)
b: 27.3 (4.32)
Ge et al. (2017) To determine the effects of dry 29/0 Condition: Dry cupping (vacuum 10 min of 4 weeks 8 Nil VAS, FAAM, Baseline: “The results support that both Fair
United States cupping on pain and function plantar cupping): interferential LEFS, PPT before dry cupping therapy and quality
with plantar fasciitis fasciitis Cupping applied to the therapy intervention electrical stimulation therapy 16/28
Gender most painful site for During: at could reduce pain and increase
(male/ 10 min each session function in the population
female) Post: after tested”
a: 4/10; b: 10/ intervention
5
Mean age
(SD):
a: 40.1 (14.6)
b: 39.3 (13.5)
Khan et al. To evaluate the effect of cupping 62/22 Condition: Dry cupping (fire Medication 15 days 11 Nil Grading Baseline: “The study proved cupping to Fair
(2013) therapy in a clinical setting for knee OA cupping): Acetaminophen scales (pain, before be a good analgesic and anti- quality
India knee osteoarthritis Gender Stationary cups applied stiffness, intervention inflammatory with efficacy 17/28
(male/ around the knee for crepitus, Post: after better than acetaminophen.
female) 15 min oedema, intervention Thus, cupping can be
a: 6/14; b: 8/ movement, recommended for other painful
12 tenderness) conditions besides being a line
Mean age of treatment for osteoarthritis”

507
(continued on next page)
Table 1 (continued )

508
Study ID, year Aim Sample Participants a. Intervention b. Control Treatment Total Follow Outcome Measurement Authors conclusions D&B
and country size/ (a. duration No. of up measures times score
drop Intervention; sessions
outs b. Control

(SD)
Unable to
determine
Kim et al. To measure the effects of cupping 30/0 Condition: Dry cupping (fire Passive stretch Single 1 Nil ROM, PPT, Baseline: “It was evident from findings of Fair
(2017) on flexibility, muscle activity, and hamstring cupping): to the session EMG before this study that cupping therapy quality
Korea pain threshold of the hamstring flexibility Cups applied to the hamstring intervention has as much positive effect on 19/28
muscle compared to passive Gender hamstring muscle for muscle and Post: after flexibility, pain threshold, and
stretching in healthy subjects (male/ 5 min held for 10 s x 9 intervention muscle contraction as passive
female) reps stretching”
a:12/3
Mean age

S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

(SD)
30.10 (5.52)
LaCross (2014) To examine the effectiveness of 17/0 Condition: Dry cupping (vacuum Heat pack for Single 1 Nil ROM, PFAQ, Baseline: “Results of this study suggest Fair
United States MFD and moist heat pack with hamstring cupping): 10min þ foam session GROC before that either treatment may be quality
(thesis foam roller on hamstring pathology Cups applied to roll intervention beneficial for ROM increases in 17/28
publication) pathology Gender hamstrings þ active Post: after patients with hamstring
(male/ movement protocol with intervention injuries”
female) cups in place
a: 8/1; b:5/3
Mean age
(SD)
Unable to
determine
Lauche et al. To determine whether a series of 50/4 Condition: Dry cupping (fire Wait list 2 weeks 5 18 days VAS, NDI, Baseline: “A series of five dry cupping Good
(2011) cupping treatments effectively non-specific cupping): PD, SF-36, before treatments appeared to be quality
Germany relieves chronic non-specific neck neck pain Cups retained on affected PPT, MDT, intervention effective in relieving chronic 24/28
pain Gender areas for 10e20 min every VDT During: at non-specific neck pain. Not
(male/ 3e4 days each session only subjective measures
female) Post: after improved, but also mechanical
a: 7/15; b: 4/ intervention pain sensitivity differed
20 significantly between the two
Mean age groups, suggesting that
(SD) cupping has an influence on
a: 48.6 (1.2) functional pain processing”
b: 53.0 (11.4)
Lauche et al. This study aimed to test the 61/7 Condition: Dry cupping: Progressive 12 weeks 24 Nil VAS, NDI, Baseline: “… cupping massage is no more Good
(2013) efficacy of 12 weeks of a partner- non-specific Cupping massage twice muscle PD, SF-36, before effective than progressive quality
Germany delivered home-based cupping neck pain weekly for 10e15 min relaxation PPT, HADS, intervention muscle [relaxation] in reducing 23/28
massage, compared to the same Gender FEW-16, Post: after chronic non-specific neck pain.
period of progressive muscle (male/ PSQ-20, intervention Both therapies can be easily
relaxation in patients with female) patient used at home and can reduce
chronic non-specific neck pain. a: 16/45; b: 6/ expectations pain to a minimal clinically
24 relevant extent.”
Mean age
(SD)
a: 54.16
(12.7)
b: 54.5 (12.3)
Lauche et al. This study aimed to investigate 141/48 Condition: Dry cupping (pneumatic b1: Sham 18 days 5 6 VAS, FIQ, SF- Baseline: “Five cupping treatments were Excellent
(2016) the efficacy of cupping therapy fibromyalgia pulsatile cupping): cupping months 36, PPT, before more effective than usual care quality
Germany compared to usual care and a Gender 4-8 cups retained on b2: usual care MFI-20, intervention to improve pain intensity and 26/28
sham procedure to improve (male/ trapezius, levator scapula, PSQI, PPT, Post: after quality of life in patients
symptoms and quality of life in female) latissimus dorsi, or Blinding intervention, diagnosed with the
 patients diagnosed with a: 1/46 gluteus maximus for success, at 6-month fibromyalgia syndrome. Given
fibromyalgia syndrome. b: 1/47 30 min, twice weekly satisfaction, follow up that effects were small, and
b2: 0/46 safety cupping was not superior to
Mean age sham cupping treatments
(SD) currently no recommendation
a:54.35 (10.6) for cupping in the treatment of
b1: 56.3 (8.7) fibromyalgia can be made.”
b2: 56.8 (7.7)
Saha et al. To test the efficacy of cupping 50/5 Condition: Dry cupping (vacuum Usual care 2 weeks 5 3 VAS, NDI, Baseline: “Cupping massage appears to Good
(2017) massage in patients with chronic chronic neck cupping): weeks SF-36, PPT, before be effective in reducing pain quality
Germany non-specific neck pain pain Cupping massage twice MVD, VDT intervention and increasing function and 25/28
Gender weekly for 10 min Post: at 3- quality of life in patients with
(male/ week follow chronic non-specific neck pain”
female) up
a: 4/21; b 0/
25
Mean age

S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

(SD)
a: 54.3 (8.6)
b: 53.3 (11.1)
Singh et al. Clinical evaluation of Hijamat Bila 40/12 Condition: Dry cupping (vacuum Diclofenac 15 days 6 Nil VAS Baseline: “The result was clinically Poor
(2016) Shurt (Dry cupping) in the low back pain cupping): sodium 50 mg before significant in both therapies, quality
India management of Waja ul zahar Gender Cupping over lumbosacral orally twice a intervention but hijamat bila shurt was more 8/28
(Low Back Pain) (male/ region for 20min day During: at effective in comparison to
female) each session diclofenac sodium.”
4:1 ratio Post: after
Mean age intervention
(SD)
20 (60)
Smith (2015) To assess the effectiveness of MFD 30/0 Condition: Dry cupping (vacuum Rest for 7 min Single 1 Nil ROM, Baseline: “Due to lack of statistical Poor
United States on shoulder ROM and strength in shoulder cupping): session strength before significance in all variables quality
(thesis healthy overhead athletes ROM IASTM þ cupping to intervention except [external rotation] ROM, 13/28
publication) Gender rotator cuff þ active Post: after this study demonstrates little
(male/ movement protocol with intervention to no clinical relevance to the
female) cups in place for a total of use of MFD for the purpose of
15/15 10e15 min increase immediate shoulder
Mean age ROM and strength”
(SD)
22.5 (2.21)
Teut et al. To investigate the effectiveness of 40/0 Condition: Dry cupping (pneumatic No intervention 4 weeks 8 12 WOMAC, Baseline: “Dry cupping with a pulsatile Good
(2012) cupping in relieving the knee OA pulsatile cupping): weeks VAS, SF-36 before cupping device relieved quality
Germany symptoms of knee osteoarthritis Gender Silicone dry cupping via intervention symptoms of knee OA 24/28
(male/ mechanical cupping Post: after compared to no intervention”
female) device over low back and intervention,
a: 5/16; b 8/ knee joint for 10 min at 12-week
11 follow up
Mean age
(SD)
a: 68.1 (7.2)
b: 69.3 (6.8)
Teut et al. “The aim of our study was to 110/0 Condition: Dry cupping (pneumatic Medication 4 weeks 8 12 VAS, SF-36 Baseline: “Both forms of cupping were Excellent
(2018) investigate the effectiveness of chronic LBP pulsatile cupping): paracetamol weeks before effective in cLBP without quality
Germany dry pulsatile cupping in reducing Gender Silicone dry cupping via 4  500mg/day intervention showing significant differences 26/28
pain and improving back function (male/ mechanical cupping Post: after in direct comparison after four
and quality of life in patients with female) device over low back for intervention, weeks, only pulsatile cupping
nonspecific cLBP” a1: 43.2/16 8 min, twice weekly at 12-week showed effects compared to
a2: 36.1/13 a: 150e350 mbar follow up control after 12 weeks.”
b: 32.4/12 a2: 70 mbar (minimal
Mean age cupping)

509
(continued on next page)
 510 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

from 1 to 24, with a duration of 4e30 min per session. There was

detection threshold; MFD Myofascial Decompression; MFI-20 The Multidimensional Fatigue Inventory; NDI Neck Disability Index; OA Osteoarthritis; PFAQ Perceived functional ability scale; PSQ-20 Perceived Stress Ques-
rating of change scale; GT Graston Technique; HADS Hospital Anxiety and Depression Scale, IASTM Instrument-assisted soft tissue mobilisation; LBP Low back pain; LEFS Lower Extremity Functional Scale; MDT Mechanical
Abbreviations: AKE Active Knee Extension Test; cLBP Chronic Low Back Pain; FAAM Foot and Ankle Ability Measure; FEW-16 (German) Assessment of Physical Wellbeing; FIQ Fibromyalgia Impact Questionnaire; GROC Global

tionnaire; Pain pressure threshold; PSQI Pittsburgh Sleep Quality Inventory; RMDQ Rowland Morris Disability Questionnaire; ROM Range of motion; SD Standard Deviation; SF-36 36-Item Short Form Survey; SST Skin surface
variation in dry cupping methods implemented, including vacuum

effective in improving the ROM quality

of the cervical spine and pain 14/28
cupping, fire cupping, stationary or static cupping, moving cupping,
score

“… Cupping treatment is more Poor

D&B

and myofascial decompression with active movements.

McKenzie stretching method

thresholds compared to the
3.2. Excluded trials
Measurement Authors conclusions

A total of 232 titles and abstracts were screened for eligibility.

Seventy-eight trials did not include RCT methodology; 21 used wet
cupping application; 30 were not related to a musculoskeletal
condition; 24 trials were not relevant; 33 trials pertained to
Traditional Chinese Medicine (TCM) theory; 19 were not published
…”

in the English language, and 1 trial was unable to be accessed.

Twenty-six articles were accessed in full, with a further 5 trials
intervention

intervention

excluded; 3 did not include RCT methodology, 1 was unable to be

Post: after
Baseline:

accessed in full, and 1 used wet cupping application.

before
times

3.3. Risk of bias and methodological quality

ROM, PPT
measures
Follow Outcome

Methodological quality of the trials ranged from excellent to fair.

Overall, the quality of evidence was found to be fair, with a mean
D&B score of 18/28. Two trials from Germany were rated excellent
quality with a D&B score of 26/28 and were found to have low
temperature; VAS Visual analogue scale; VDT Vibration detection threshold; WOMAC Western Ontario McMasters University Osteoarthritis Index.
Nil
up

internal bias (Lauche et al., 2016; Teut et al., 2018). Seven trials were
sessions

rated as being good, with scores ranging from 20 to 25 (Akbarzadeh

No. of
Treatment Total

et al., 2014; Chi et al., 2016; Cramer et al., 2011; Lauche et al 2011,
1

2013; Saha et al., 2017; Teut et al., 2012). Six trials were rated fair,
with a score ranging from 15 to 19 (Biehl, 2017; El Rahim et al.,
duration

stretch protocol session

2017; Ge et al., 2017; Khan et al., 2013; Kim et al., 2017; LaCross,
Single

2014). Six trials were rated poor quality, with high internal bias
and scores ranging from 8 to 14 (Arslan et al., 2015; Barger, 2016;
Doozan, 2015; Singh et al., 2016; Smith, 2015; Yim et al., 2017).
McKenzie's
b. Control

Cupping applied to upper for 8 min

Blinding of both the participant and outcome assessors was only

carried out by 2 trials (Biehl, 2017; Lauche et al., 2016); 2 trials
attempted to blind the participants only (Teut et al., 2018; Yim et al.,
2017) and 2 trials attempted to blind the outcome assessors (Kim
scapulae muscle to 8 min

et al., 2017; Lauche et al., 2013). All 21 trials were RCTs; however,
Dry cupping (vacuum

trapezius & levator

only 11 of the trials described adequate randomisation techniques

and only 8 of the 21 trials sufficiently concealed randomised
Participants a. Intervention

intervention assignment. Inadequate reporting of the random

cupping):

variability in the data were observed in 10 of the trials, with failure

to report confidence intervals for normally distributed data or
interquartile ranges for non-normally distributed data.
b: 50.7 (10.7)
a1: 49.0 (3.7)
Intervention;

Meta-analysis results are depicted in the forest plots (Figs. 2e10)

22.66 (2.98)
Neck ROM
Condition:
b. Control

Mean age

and GRADE summaries are presented in Tables 2e7. All results are
a2: 47.5

female)
Gender
(male/

based on the short-term effects of dry cupping (<3 months).

(13.8)

12/6
(SD)

(SD)
(a.

3.4. Dry cupping therapy for chronic non-specific neck pain

Sample

drop
size/

“… to investigate the differences 18/0

outs

3.4.1. Effects of dry cupping vs. no treatment on pain for non-

the cervical vertebrae by applying
in the angle of the cervical spine

specific neck pain

and the pain thresholds around

the McKenzie exercise and the

Five trials including 239 participants were analysed for the ef-
fect of dry cupping for pain relief in chronic non-specific neck pain
(Arslan et al., 2015; Chi et al., 2016; Cramer et al., 2011; Lauche et al.,
2011; Saha et al., 2017). The 5 trials compared cupping therapy to
cupping therapy.”

no intervention (Arslan et al., 2015; Chi et al., 2016), standard

medical care (Cramer et al., 2011), wait-list (Lauche et al., 2013), and
usual care (Saha et al., 2017). All 5 trials reported a statistically
significant effect on reducing pain, in favour of dry cupping. Meta-
Aim
Table 1 (continued )

analysis of the 5 trials (Fig. 2) revealed a statistically significant

effect on pain relief in favour of dry cupping, with a MD of
21.67
Study ID, year
and country

(95% CI,
36.55,
6.80; I2 ¼ 94%) and a large effect was observed
Yim et al.
(2017)

with a SMD of
1.04 (95% CI,
1.79,
0.28). Heterogeneity was

Korea

considerable between the 5 trials; omission of the outlying trial by

Chi et al. (2016) reduced heterogeneity and resulted in a MD
 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518 511

of
12.40 (95% CI,
15.99,
8.81; I2 ¼ 0%). For chronic neck pain, and imprecision) that dry cupping had a significant effect on pain
there was low-quality evidence (downgraded due to inconsistency relief (Table 3).
and imprecision) that dry cupping had a significant effect on pain
relief (Table 2).
3.5.2. Effects of dry cupping vs. comparative or control group on
low back pain
3.4.2. Effects of dry cupping vs. no treatment on functional status
Two trials including 160 participants were analysed for the ef-
for non-specific neck pain
fect of dry cupping for pain relief in non-specific low back pain
Four trials including 191 participants were analysed for the ef-
(Akbarzadeh et al., 2014; El Rahim et al., 2017). The trials compared
fect of dry cupping on functional status in non-specific neck pain,
cupping therapy to routine care (Akbarzadeh et al., 2014), inter-
measured by the neck disability index. All 4 trials reported a sig-
ferential therapy and traditional physical therapy (El Rahim et al.,
nificant effect on disability, in favour of dry cupping Cramer et al.,
2017). Both trials reported a significant effect on pain, in favour of
(2011); Lauche et al., (2011); Lauche et al., (2013); Saha et al.,
dry cupping. Meta-analysis of the trials (Fig. 6) displayed a statis-
(2017). Meta-analysis of the 4 trials (Fig. 3) displayed a statisti-
tically significant effect on pain relief in favour of dry cupping, with
cally significant effect on functional status in favour of dry cupping,
a MD of
11.20 (95%CI,
13.76,
8.64) and a large effect was found
with a MD of
4.65 (95%CI,
6.44,
2.85; I2 ¼ 14%); though, only a
with a SMD of
2.60 (95%CI,
3.48,
1.72); however, considerable
medium effect was observed (SMD,
0.77; 95%CI,
1.07,
0.48). For
heterogeneity was observed between the two trials, I2 ¼ 76%. For
functional status in chronic neck pain, there was moderate-quality
low back pain, there was low-quality evidence (downgraded due to
evidence (downgraded due to imprecision) that dry cupping had a
inconsistency and imprecision) that dry cupping had a significant
significant effect (Table 2).
effect on pain relief (Table 3).

3.4.3. Effects of dry cupping vs. no treatment on pressure pain

sensitivity for non-specific neck pain 3.5.3. Effects of dry cupping therapy on pressure pain sensitivity
Four trials including 191 participants were analysed for the ef-
3.5.3.1. Effects of dry cupping on pressure pain sensitivity in partic-
fect of dry cupping on pressure pain sensitivity in non-specific neck
ipants with symptomatic musculoskeletal conditions. Six trials
pain, measured by pressure algometry. All 4 trials reported a sig-
including 306 participants were analysed for the effect of dry
nificant effect on pressure pain thresholds, in favour of dry cupping
cupping on pressure pain sensitivity in symptomatic participants,
(Cramer et al., 2011; Lauche et al 2011, 2013; Saha et al., 2017).
measured by pressure algometry. The 6 trials used dry cupping
Meta-analysis of the 4 trials (Fig. 4) displayed a statistically sig-
therapy for the treatment of chronic neck pain (Cramer et al., 2011;
nificant effect on pressure pain thresholds in favour of dry cupping,
Lauche et al 2011, 2013; Saha et al., 2017), plantar fasciitis (Ge et al.,
with a SMD of
0.40 (95%CI,
0.69,
0.11; I2 ¼ 0%); however, only a
2017) and fibromyalgia (Lauche et al., 2016). Meta-analysis of the 6
small effect was observed. For pressure pain sensitivity in chronic
trials (Fig. 7) displayed a statistically significant effect on pressure
neck pain, there was moderate-quality evidence (downgraded due
pain thresholds in favour of dry cupping, with a SMD of
0.40 (95%
to imprecision) that dry cupping had a significant effect (Table 2).
CI,
0.63,
0.17; I2 ¼ 0%); however, only a small effect was
observed. For pressure pain sensitivity in symptomatic participants,
3.5. Dry cupping therapy for low back pain there was moderate-quality evidence (downgraded due to impre-
cision) that dry cupping had a significant effect (Table 4).
3.5.1. Effects of dry cupping vs. comparative or control group on low
back pain
Two trials including 196 participants were analysed for the ef- 3.5.3.2. Effects of dry cupping on pressure pain sensitivity in partic-
fect of dry cupping for pain relief in non-specific low back pain ipants without symptomatic musculoskeletal conditions (healthy
(Akbarzadeh et al., 2014; Teut et al., 2018). The trials compared controls). Two trials including 66 participants were analysed for
cupping therapy to routine care (Akbarzadeh et al., 2014), minimal the effect of dry cupping on pressure pain sensitivity in asymp-
cupping (low vacuum suction) and medication (Teut et al., 2018). tomatic participants, measured by pressure algometry. The 2 trials
Both trials reported a significant effect on pain, in favour of dry compared dry cupping therapy to passive stretching (Kim et al.,
cupping. Meta-analysis of the trials (Fig. 5) displayed a statistically 2017) and active stretching (Yim et al., 2017). Meta-analysis of the
significant effect on pain relief in favour of dry cupping, with a MD 2 trials (Fig. 8) displayed a statistically significant effect on pressure
of
19.38 (95%CI,
28.09,
10.66) and a large effect was found with pain thresholds in favour of dry cupping, with a SMD of
0.63 (95%
a SMD of
1.08 (95%CI,
2.04,
0.12). Moderate heterogeneity was CI,
1.13,
0.14; I2 ¼ 0%); however, only a medium effect was
observed between the two trials, I2 ¼ 59%. The mean difference observed. For pressure pain sensitivity in asymptomatic partici-
of
19.38 mm on the VAS was found to exceed the clinically pants, there was low-quality evidence (downgraded due to limi-
important range of 15 mm (Ostelo et al., 2008). For low back pain, tations and imprecision) that dry cupping had a significant effect
there was low-quality evidence (downgraded due to inconsistency (Table 5).

Fig. 2. Effects of dry cupping vs. no treatment on pain for non-specific neck pain (Outcome Measure: Visual Analogue Scale 100 mm).
512 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

Fig. 3. Effects of dry cupping vs. no treatment on functional status for non-specific neck pain (Outcome Measure: Neck Disability Index).

Fig. 4. Effects of dry cupping vs. no treatment on pressure pain sensitivity for non-specific neck pain (Outcome measure: Pressure Pain Threshold).

Fig. 5. Effects of dry cupping vs. comparative or control group on low back pain (Outcome Measure: Visual Analogue Scale 100 mm).

Fig. 6. Effects of dry cupping vs. comparative or control group on low back pain (Outcome Measure: Short Form McGill Pain Questionnaire).

Fig. 7. Effects of dry cupping on pressure pain sensitivity in symptomatic participants (Outcome Measure: Pressure Pain Threshold).

3.6. Effects of dry cupping on range of motion stretching (Kim et al., 2017) and active stretching (Yim et al., 2017).
Meta-analysis of the 3 trials (Fig. 9) displayed no significant effect
3.6.1. Dry cupping vs. active control on range of motion with a SMD of
1.13 (95%CI,
2.57, þ0.31), with
Three trials including 126 participants were analysed for the considerable heterogeneity observed between the two trials,
effect of dry cupping on range of motion, measured by active range I2 ¼ 92%. For range of motion, there was very low-quality evidence
of motion using a goniometer. The 3 trials compared dry cupping (downgraded due to limitations, inconsistency and imprecision)
therapy to an active control group, including interferential therapy that dry cupping had no significant effect (Table 6).
and traditional physical therapy (El Rahim et al., 2017), passive
 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518 513

Fig. 8. Effects of dry cupping on pressure pain sensitivity in asymptomatic participants (Outcome Measure: Pressure Pain Threshold).

Fig. 9. Dry cupping vs. active control (Outcome Measure: Range of Motion).

Fig. 10. Dry cupping vs. no treatment (Outcome Measure: Range of Motion).

Table 2
Dry cupping therapy vs. no intervention, standard care, wait-list and usual care for chronic non-specific neck pain.

No. of Quality assessment No. of patients Treatment Effect Quality of the evidence
studies (95%CI) (GRADE)
Limitations Inconsistency Indirectness Imprecision Other Intervention Control
considerations

Pain measured with: VAS from 0 to 100 (worse pain)

5 no serious seriousa no serious seriousb none 118 121 MD,
21.67 44OO
limitations indirectness [-36.55,
6.80] LOW
Functional status measured with: neck disability questionnaire
4 no serious no serious no serious seriousb none 93 98 MD -4.65 444O
limitations limitations limitations [-6.44,
2.85] MODERATE
Pressure pain sensitivity measured with: pressure algometry
4 No serious no serious no serious seriousb none 93 98 SMD -0.40 444O
limitations limitations limitations [-0.69,
0.11] MODERATE

Abbreviations: CI Confidence Interval, MD Mean Difference, SMD Standard Mean Difference, VAS Visual Analogue Scale.
a 2
I ¼ 94%.
b
Sample size<400.

3.6.2. Dry cupping vs. no treatment mention adverse events. Of the 11 trials that reported on adverse
Two trials including 88 participants were analysed for the effect events, 2 trials reported that no adverse events occurred during the
of dry cupping on range of motion, measured by active range of trial duration and 9 trials reported a total of 47 adverse events
motion using a goniometer. The 2 trials compared dry cupping occurred in the dry cupping group, with an overall relative risk ratio
therapy to no treatment. Meta-analysis of the 2 trials (Fig. 10) dis- of 1.88 (95%CI, 1.11, 3.20).All adverse events are summarised in
played a statistically significant effect on range of motion with a Table 8. Most symptoms were of mild to moderate severity,
SMD of
0.75 (95%CI,
1.19,
0.32; I2 ¼ 0%; however, only a me- resolving within 48-h, and included mild muscular soreness
dium effect was observed. For range of motion, there was low- (18.9%), increase in pain (13.79%), and an onset of a headache
quality evidence (downgraded due to limitations and impreci- (3.45%). Mild hematomas were also reported, as was blister for-
sion) that dry cupping had a significant effect versus no treatment mation e often associated with fire cupping.
(Table 7). There were 2 serious adverse events reported in the dry cupping
group; however, both authors concluded was not a consequence of
3.6.3. Adverse events the intervention (Lauche et al., 2013; Saha et al., 2017). Firstly,
Of the 21 RCTs included in this review, an adverse event state- Lauche et al. (2013) reported a participant was diagnosed with a
ment was reported in 11 trials; the remaining 10 trials failed to prolapsed intervertebral disc in the home-based cupping massage
514 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

Table 3
Dry cupping therapy vs. routine care, minimal cuppingd and medication for non-specific low back pain.

No. of Quality assessment No. of patients Treatment Effect Quality of the evidence
studies (95%CI) (GRADE)
Limitations Inconsistency Indirectness Imprecision Other Intervention Control
considerations

Pain measured with: VAS from 0 to 100 (worse pain)

2 no serious seriousa no serious seriousb None 115 81 MD -19.38 44OO

limitations limitations [-28.09,
10.66] LOW
Pain measured with: Short Form McGill Pain Questionnaire
4 no serious seriousc no serious seriousb none 90 70 MD -11.20 44OO
limitations limitations [-13.76,
8.64] LOW

Abbreviations: CI Confidence Interval, MD Mean Difference, SMD Standard Mean Difference, VAS Visual Analogue Scale.
a 2
I ¼ 59%.
b
Sample size<400.
c b
I ¼ 76%.
d
minimal cupping (low vacuum suction).

Table 4
Effects of dry cupping therapy on pressure pain sensitivity in symptomatic participants.

No. of Quality assessment No. of patients Treatment Effect Quality of the evidence
studies (95%CI) (GRADE)
Limitations Inconsistency Indirectness Imprecision Other Intervention Control
considerations

Pressure pain sensitivity measured with: pressure algometry

6 no serious no serious no serious seriousa none 155 151 SMD -0.40 444O
limitations limitations limitations [-0.63,
0.17] MODERATE

Abbreviations: CI Confidence Interval, MD Mean Difference, SMD Standard Mean Difference, VAS Visual Analogue Scale.
a
Sample size<400.

Table 5
Effects of dry cupping therapy on pressure pain sensitivity in asymptomatic participants.

No. of Quality assessment No. of patients Treatment Effect Quality of the evidence
studies (95%CI) (GRADE)
Limitations Inconsistency Indirectness Imprecision Other Intervention Control
considerations

Pressure pain sensitivity measured with: pressure algometry

2 seriousa no serious no serious seriousb none 33 33 SMD -0.63 44OO

limitations limitations [-1.13,
0.14] LOW

Abbreviations: CI Confidence Interval, MD Mean Difference, SMD Standard Mean Difference, VAS Visual Analogue Scale.
a
Serious risk of bias.
b
Sample size<400.

Table 6
Effects of dry cupping therapy vs. active control on range of motion.

No. of Quality assessment No. of patients Treatment Effect (95% Quality of the evidence
studies CI) (GRADE)
Limitations Inconsistency Indirectness Imprecision Other Intervention Control
considerations

Range of motion measured with: goniometer

3 seriousa seriousb no serious seriousc none 53 7b SMD -1.13 4OOO

limitations [-2.57, þ0.31] VERY LOW

Abbreviations: CI Confidence Interval, MD Mean Difference, SMD Standard Mean Difference, VAS Visual Analogue Scale.
a
Serious risk of bias.
b 2
I ¼ 92%.
c
Sample size<400.

group. Secondly, a participant from the Saha et al. (2017) trial was 4. Discussion
diagnosed with a lipoma after the first cupping session that
required surgical removal. The authors concluded that it was un- The purpose of this study was to evaluate the efficacy and safety
likely the cupping massage caused the lipoma, although, it may of dry cupping therapy for the treatment of musculoskeletal pain
have elicited the visibility. A case-study report is available; how- and range of motion. To our knowledge, this review is the first
ever, it is only available in German (Schumann et al., 2012). Overall, systematic review and meta-analysis specifically examining the
the adverse events were mild to moderate, with 2 serious events e effects of dry cupping therapy for musculoskeletal pain and range
not directly resulting from the dry cupping treatment. of motion.
 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518 515

Table 7
Effects of dry cupping therapy vs. no treatment on range of motion.

Quality assessment No. of patients Treatment Effect Quality of the evidence

(95%CI) (GRADE)
No. of Limitations Inconsistency Indirectness Imprecision Other Intervention Control
studies considerations

Range of motion measured with: goniometer

2 seriousa no serioua no serious seriousb none 44 4b SMD -0.75 44OO

limitations limitations [-1.19,
0.32] LOW

Abbreviations: CI Confidence Interval, MD Mean Difference, SMD Standard Mean Difference, VAS Visual Analogue Scale.
a
Serious risk of bias.
b
Sample size<400.

Meta-analyses were conducted for 15 trials. When the included providing important information when judging an intervention as
trials were pooled and analysed, a significant large effect was clinically meaningful. A total of 11 trials included an adverse event
observed for dry cupping therapy on pain intensity in chronic neck statement, with a total of 47 adverse events reported for the dry
pain and non-specific low back pain. A significant medium effect cupping group. Most of the events were mild to moderate and
was found for dry cupping therapy on neck function. Despite the abated within 48-h. Two serious events were reported but were not
significant effect, the quality of evidence to support the use of dry associated with the treatment (Lauche et al., 2013; Saha et al.,
cupping for chronic neck pain and low back pain was low-quality, 2017). The most frequent events reported in the dry cupping
due to high heterogeneity and small sample sizes (<400 accord- groups were mild muscular soreness, increased pain, headache, and
ing to the GRADE recommendations (Ryan and Hill, 2016)). The blister formation. Dry cupping can be performed with either a
analysis for chronic neck pain revealed high heterogeneity; when manual handheld pump (or mechanical device) or heat from an
an outlying trial (Chi et al., 2016) was omitted from the analysis, ignited cotton ball and glass cups (fire cupping). Blister formations
heterogeneity was low. This trial reported high baseline VAS scores and burns have been associated with fire cupping, and numerous
with little to no improvement in the control group, resulting in a severe adverse events have been documented through case reports
large effect and variability in the meta-analysis data (Chi et al., (Seifman et al., 2017). This current review found no reports of burns
2016). For changes in pressure pain sensitivity and functional sta- in the included trials that used fire cupping (Akbarzadeh et al.,
tus, the quality of evidence was moderate, due to serious limita- 2014; Chi et al., 2016; El Rahim et al., 2017; Khan et al., 2013; Kim
tions associated with small sample sizes. For pressure pain et al., 2017; Lauche et al., 2011); however, blister formation was
sensitivity, there were different effects between symptomatic and reported by Khan et al. (2013). Vacuum cupping using a manual
asymptomatic participants, with moderate-level evidence of a handheld pump has far less risks than the use of fire cupping;
small effect in symptomatic patients and low-level evidence of a however, it is unknown whether the benefits of heat outweigh the
medium effect in asymptomatic participants. Most of the symp- risks associated with fire cupping; therefore, caution is warranted
tomatic participants were diagnosed with chronic neck pain; when in the use of fire cupping.
a separate analysis was conducted for chronic neck pain only, a Previous systematic reviews and meta-analysis have examined
similar small effect was observed. the efficacy of all cupping methods, rather than dry cupping alone.
The reported minimal important change for low back pain on a Cao et al. (2014) conducted a meta-analysis on 2 dry cupping trials
visual analogue scale (VAS) is 15 mm on a 100 mm scale (Ostelo et al., that produced a significant effect for reducing pain and improving
2008). A change of over 20 mm, as seen in the results for low back quality of life. The analysis combined data from two separate
pain in this review, exceeds the clinically important range. This musculoskeletal conditions, knee osteoarthritis and chronic neck
suggests that the changes in pain from the dry cupping treatment pain, and therefore may not be generalisable to a specific condition.
were clinically meaningful; however, these results do not take into Recent systematic reviews have investigated cupping therapy on
consideration patient perspectives, risks and costs; therefore, chronic neck pain (Azizhani et al., 2018; Kim et al., 2018) and
cannot be deemed clinically important until further research with chronic back pain (Moura et al., 2018); however, the recent reviews
high-quality trials has been undertaken (Ferreira, 2012). A previous included all types of cupping therapy. Nonetheless, Kim et al. (2018)
review by Lauche et al. (2013) reanalysed 4 cupping trials to assess reported similar results in their meta-analysis to this current re-
the minimal clinical important differences (MCID) for chronic neck view for the effects of cupping on chronic neck pain and function.
pain patients. Lauche et al. (2013) observed comparable results in Kim et al. (2018) observed a significant reduction in pain and
pain reduction to other conventional therapies. Results revealed a improved function in patients treated with cupping compared to no
MCID of -8mm (
21%) on the VAS and substantial clinical benefit treatment. Additionally, when compared to active treatment, there
(SCB) of
26.5 mm (
66.8%). For a clinical benefit, a percentage was also a significant reduction in pain and improved quality of life.
change of over 50% for SCB is recommended (Dworkin et al., 2008). The results reported by Kim et al. (2018) were not limited to dry
The number of dry cupping session and duration of treatment cupping therapy; however, they do concur with the results pro-
varied among trials. Most of the trials investigating the efficacy of duced from this current review.
dry cupping on non-specific neck pain conducted a total of 5 ses- From this review, 8 trials investigated the use dry cupping as a
sions over a 2-week period, with the session duration lasting form of myofascial release to increase range of motion (Barger,
10e15 min. For low back pain, there were no standardised treat- 2016; Biehl, 2017; Doozan, 2015; El Rahim et al., 2017; Kim et al.,
ment duration or number of sessions; however, El Rahim et al., 2017; LaCross, 2014; Smith, 2015; Yim et al., 2017). Meta-analysis
(2017) and Teut et al., (2018) both conducted their trials over 4 of dry cupping compared to an active control group did not
weeks, with a total of 8e10 sessions. Further studies are required to display significant differences. Dry cupping was found to have a
standardise the optimal number of sessions required in the treat- medium effect when compared to no treatment; however, the
ment of musculoskeletal conditions. quality of evidence was low. The individual trials found dry cupping
This current systematic review assessed the risks and safety of to be just as effective as passive stretching (Kim et al., 2018) and
dry cupping therapy for the treatment of musculoskeletal pain, thus self-myofascial release with the use of a foam roller (LaCross, 2014),
516 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

Table 8
Reported adverse events for dry cupping in musculoskeletal conditions.

Study ID Sample No. of events Details of the reported adverse events

Size T ¼ Treatment
group
C ¼ Control
group

Chi et al. (2016) (n ¼ 60) T: (n ¼ 2) T: mild low back pain related to the seated position (n ¼ 2)
Cramer et al. (n ¼ 50) T: (n ¼ 5) T: muscle soreness for 1e2 days (n ¼ 2); minor hematoma at the treated site for 2 days (n ¼ 1); increased neck pain for 1e5 h
(2011) (n ¼ 2)
Khan et al. (n ¼ 62) T: (n ¼ 11) T: blister formation (n ¼ 5); ecchymosis (n ¼ 6)
(2013) C: (n ¼ 8) C: GI symptoms (n ¼ 8)
Lauche et al. (n ¼ 50) T: (n ¼ 1) T: symptoms temporarily worsened (n ¼ 1)
(2011)
Lauche et al. (n ¼ 61) T: (n ¼ 3) T: increased muscular tension and pain (n ¼ 1); pain in shoulder (n ¼ 1); prolapsed intervertebral disc (n ¼ 1)
(2013)
Lauche et al. (n ¼ 141) T: (n ¼ 4) T: severely increased pain (n ¼ 1); bruised ribs (n ¼ 1); flu (n ¼ 1); acute torticollis (n ¼ 1)
(2016) C: (n ¼ 3) C: torn meniscus (n ¼ 1); persistent pain after spinal operation (n ¼ 1); flu (n ¼ 1)
Saha et al. (n ¼ 50) T: (n ¼ 5) T: headache <1hr (n ¼ 2); upper back pain <48hrs (n ¼ 1); vertigo <48hrs (n ¼ 1); lipoma (n ¼ 1)
(2017)
Teut et al. (n ¼ 40) T: (n ¼ 6) T: mild hematoma (n ¼ 3); self-limiting light tingling sensations for a few minutes in the legs after cupping the knee (n ¼ 2);
(2012) increased LBP <24hrs (n ¼ 1)
Teut et al. (n ¼ 110) T: (n ¼ 10) T: aggravation of LBP <24hrs (n ¼ 2); light muscular backache (n ¼ 8)
(2018)

Abbreviations: GI Gastrointestinal; LBP Low back Pain.

with both trials reporting no significant differences between the the literature (Aboushanab and Ravalia, 2017; Emerich et al., 2014;
interventions. Furthermore, dry cupping was found to be more Larsson et al., 1990; Pomeranz and Stux, 2001; Tham et al., 2006).
effective than the Mc Kenzie's cervical stretch protocol (Yim et al., Future trials should continue to investigate the mechanisms of dry
2017) for increasing cervical range of motion and traditional cupping therapy and how the application of cupping may benefit
physical therapy (El Rahim et al., 2017) for increasing lumbar range musculoskeletal pain conditions, additional to the clinical impor-
of motion. tance of dry cupping therapy from the perspectives of patients to
D&B scores varied greatly between 21 RCTs, ranging from assess the MCID. Cost analysis should be conducted to determine
excellent to poor quality, with almost half the trials found to be the benefits of dry cupping treatment compared to other in-
good quality. For internal validity, many trials suffered due to lack of terventions currently used for the treatment of musculoskeletal
blinding of participants and outcome assessors. Most trials did not conditions. Adverse event statements should continue to be re-
perform any blinding of the intervention to participants; however, ported to monitor the safety and risks of dry cupping therapy.
a sham cupping device was used in 2 trials (Biehl, 2017; Lauche Furthermore, future trials should examine the long-term effects of
et al., 2016). The reliability of the sham device was tested in a pi- dry cupping and ensure the sample size is appropriate, and the trial
lot study (Lee et al., 2011) and it was reported that the device was is considerably powered.
valid; however, in contrast, Lauche et al. (2016) observed a lack of
success in their trial, with 73.2% of patients correctly identifying the
sham treatment, resulting in questionable validity of the sham
5. Conclusion
device for blinding participants.
External validity was weak for more than half of the trials
To our knowledge, this current systematic review is the first to
examined, with most of the trials using convenient samples or
analyse western dry cupping methods in the treatment of muscu-
healthy college athletes that are not generalisable to the population
loskeletal pain and range of motion. The results suggest that dry
at large. Treatment representation was weak for many studies, as
cupping may be effective in reducing pain and improving func-
dry cupping is not representative of a usual intervention to treat
tional status in patients with chronic neck pain when compared
musculoskeletal conditions in the facilities where participants were
with no intervention. A significant reduction in pain for non-
treated. Other domains of limitations included statistical power,
specific low back pain was observed, although the quality of evi-
with less than half of the included trials including a power analysis
dence was found to be moderate to low. The adverse events asso-
in the methodology. Small sample sizes due to underpowered trials
ciated with dry cupping were typically mild to moderate and
can lead to an overestimation of the treatment effects or fail to
resolved within 48 h. Considering the low quality of evidence,
detect a clinically important effect (Akobeng, 2005).
further higher-quality RCTs with larger sample sizes and long-term
There were several limitations associated with this current re-
outcomes are warranted to provide definitive conclusions
view. A language restriction of English resulted in the exclusion of
regarding the effectiveness and safety of dry cupping for the
19 trials and may have resulted in a possible selection bias. Many of
treatment of musculoskeletal pain and range of motion.
the included trials had small sample sizes which may lead to sta-
tistical heterogeneity and overestimation of the effect size. Addi-
tional sources of heterogeneity may have been from multiple
interventions performed, variation in comparison groups between CRediT authorship contribution statement
trials, and patient characteristics, including pain duration. This
study only investigated the short-term (less than 3 months) effects Sarah Wood: conceived the study, Formal analysis, Validation,
of cupping and the long-term effects remain unknown. Writing - original draft, Methodology, results and discussion, All
The specific physiological mechanisms underpinning dry authors critically reviewed, edited and approved the final manu-
cupping remain unclear; however, proposed theories prevail within script. Gary Fryer: conceived the study. Liana Lei Fon Tan: Vali-
dation. Caroline Cleary: Formal analysis.
 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518 517

Declaration of competing interest Ferreira, et al., 2012. A critical review of methods used to determine the smallest
worthwhile effect of interventions for low back pain. Journal of Clinical
Epidemiology 65 (3), 253e261. https://doi.org/10.1016/j.jclinepi.2011.06.018.
GF is a member of the journal's International Advisory Board but Furlan, A.D., Pennick, V., Bombardier, C., van Tulder, M., 2009. 2009 updated method
had no role in reviewing this article. The authors declare that they guidelines for systematic reviews in the Cochrane Back Review Group. Spine
have no other competing interests. (Phila Pa 1976) 34, 1929e1941. https://doi.org/10.1097/BRS.0b013e3181b1c99f.
Ge, W., Leson, C., Vukovic, C., 2017. Dry cupping for plantar fasciitis: a randomized
controlled trial. J. Phys. Ther. Sci. 29 (5), 859e862. https://doi.org/10.1589/
Appendix A. Supplementary data jpts.29.859.
Higgins, J.P.T., Green, S., 2011. Cochrane Handbook for Systematic Reviews of In-
terventions Version 5.1.0. The Cochrane Collaboration. Retrieved from. http://
Supplementary data to this article can be found online at handbook.cochrane.org.
https://doi.org/10.1016/j.jbmt.2020.06.024. Khan, A.A., Jahangir, U., Urooj, S., 2013. Management of knee osteoarthritis with
cupping therapy. "J. Adv. Pharm. Technol. Research"" (JAPTR)" 4 (4), 217e223.
https://doi.org/10.4103/2231-4040.121417.
References Kim, J.E., Cho, J.E., Do, K.S., Lim, S.Y., Kim, H.J., Yim, J.E., 2017. Effect of cupping
therapy on range of motion, pain threshold, and muscle activity of the
Aboushanab, T., Ravalia, M., 2017. Genetic theory e a suggested cupping therapy hamstring muscle compared to passive stretching. J.Korean Soc.Phys. Med 12
mechanism of action. Research Gate. https://doi.org/10.7490/ (3), 23e32. https://doi.org/10.13066/kspm.2017.12.3.23.
f1000research.1114877.1. Kim, S., Lee, S.H., Kim, M.R., Kim, E.J., Hwang, D.S., Lee, J., Shin, J.S., Ha, I.H., Lee, Y.J.,
Akbarzadeh, M., Ghaemmaghami, M., Yazdanpanahi, Z., Zare, N., Azizi, A., 2018. Is cupping therapy effective in patients with neck pain? A systematic
Mohagheghzadeh, A., 2014. The effect dry cupping therapy at acupoint BL23 on review and meta-analysis. BMJ Open 8 (11). https://doi.org/10.1136/bmjopen-
the intensity of postpartum low back pain in primiparous women based on two 2017-021070.
types of questionnaires, 2012; A randomized clinical trial. Int. J. Community.- LaCross, Z., 2014. Treatment Outcomes of Myofascial Decompression on Hamstring
Base Nurs.Midwifery 2 (2), 112e120. Retrieved from: https://www.ncbi.nlm.nih. Pathology. Oklahoma State University, Ann Arbor. https://hdl.handle.net/11244/
gov/pmc/articles/PMC4201191/. 14951.
Azizhani, M., Ghorat, F., Soroushzadeh, S.M.A., Karimi, M., Yekaninejad, S., 2018. The Larsson, S.E., Bodegard, L., Henriksson, K.G., Oberg, P.A., 1990. Chronic trapezius
effect of cupping on non-specific neck pain: a systematic review and meta- myalgia. Morphology and blood flow studied in 17 patients. Acta Orthop. Scand.
analysis. Iran. Red Crescent Med. J. 20 (7), e55039 https://doi.org/10.5812/ 61, 394e398. https://doi.org/10.3109/17453679008993548.
ircmj.55039. Lauche, R., Cramer, H., Choi, K.E., Rampp, T., Saha, F.J., Dobos, G.J., Musial, F., 2011.
Akobeng, A.K., 2005. Understanding randomised controlled trials. Arch. Dis. Child. The influence of a series of five dry cupping treatments on pain and mechanical
90 (8), 840e844. https://doi.org/10.1136/adc.2004.058222. thresholds in patients with chronic non-specific neck pain-a randomised
Al Bedah, A.M., Khalil, M.K., Posadzki, P., Sohaibani, I., Aboushanab, T.S., AlQaed, M., controlled pilot study. BMC Compl. Alternative Med. 11 (63), 1e11. https://
Ali, G.I., 2016. Evaluation of wet cupping therapy: systematic review of ran- doi.org/10.1186/1472-6882-11-63.
domized clinical trials. J. Alternative Compl. Med. 22, 768e777. https://doi.org/ Lauche, R., Materdey, S., Cramer, H., Haller, H., Stange, R., Dobos, G., Rampp, T., 2013.
10.1089/acm.2016.0193. Effectiveness of home-based cupping massage compared to progressive muscle
Arslan, M., Yaman, G., Ilhan, E., Alemdag, M., Bahar, A., Dane, S., 2015. Moving dry relaxation in patients with chronic neck pain e a randomized controlled trial.
cupping therapy reduces upper shoulder and neck pain in office workers. Clin. PloS One 8 (6), e65378. https://doi.org/10.1371/journal.pone.0065378.
Investigative Med. 38 (4), E217eE220. Lauche, R., Spitzer, J., Schwahn, B., Ostermann, T., Bernardy, K., Cramer, H., Dobos, G.,
Babatunde, O.O., Jordan, J.L., Van der Windt, D.A., Hill, J.C., Foster, N.E., Protheroe, J., Langhorst, J., 2016. Efficacy of cupping therapy in patients with the fibromyalgia
2017. Effective treatment options for musculoskeletal pain in primary care: a syndrome e a randomised placebo controlled trial. Sci. Rep. 6, 37316. https://
systematic overview of current evidence. PloS One 12 (6), e0178621. https:// doi.org/10.1038/srep37316.
doi.org/10.1371/journal.pone.0178621. Lee, M.S., Kim, J.I., Kong, J.C., Lee, D.H., Shin, B.C., 2011. Developing and validating a
Barger, K.M., 2016. Compressive versus Decompressive Soft Tissue Therapy on Acute sham cupping device. Acupunct. Med. 28 (4), 200e204. https://doi.org/10.1136/
Hamstring Flexibility and Pain in Male Athletes with Perceived Hamstring aim.2010.002329.
Tightness. Oklahoma State University, Ann Arbor. https://hdl.handle.net/11244/

Moura, C.C., Chaves, E.C.L., Cardoso, A.C.L.R., Nogueira, D.A., Corre ^a, H.P., 2018.
49038. Chianca TCM 2018 Cupping therapy and chronic back pain: systematic review
Biehl, M.M., 2017. The Therapeutic Effects of Dry Cupping on Iliotibial Band Tight- and meta-analysis. Rev. Latino-Am. Enferm. 26, e3094 https://doi.org/10.1590/
ness. Illinois State University, Ann Arbor. https://doi.org/10.30707/ 1518-8345.2888.3094.
ETD2017.Biehl.M. Ostelo, R.W., Deyo, R.A., Stratford, P., Waddell, G., Croft, P., Von Korff, M.,
Cao, H., Li, X., Yan, X., Wang, N., Bensoussan, A., Liu, J., 2014. Cupping therapy for Bouter, L.M., de Vet, H.C., 2008. Interpreting change scores for pain and func-
acute and chronic pain management: a systematic review of randomized tional status in low back pain: towards international consensus regarding
clinical trials. J. Tradit. Chin. Med. Sci. 1, 49e61. https://doi: 10.1016/j.jtcms.2014. minimal important change. Spine (Phila Pa 1976) 33 (1), 90e94. https://doi.org/
11.003. 10.1097/BRS.0b013e31815e3a10.
Chi, L.M., Lin, L.M., Chen, C.L., Wang, S.F., Lai, H.L., Peng, T.C., 2016. The effectiveness Pomeranz, B., Stux, G., 2001. Scientific Bases of Acupuncture. Springer-Verlag, Berlin
of cupping therapy on relieving chronic neck and shoulder pain: a randomized Heidelberg.
controlled trial. Evidence. base Compl.Alternative Med. 2016, 7358918. https:// Rozenfeld, E., Kalichman, L., 2016. New is the well-forgotten old: the use of dry
doi.org/10.1155/2016/7358918. cupping in musculoskeletal medicine. J. Bodyw.Movment .Ther 20, 173e178.
Cramer, H., Lauche, R., Hohmann, C., Choi, K.E., Rampp, T., Musial, F., Langhorst, J., https://doi.org/10.1016/j.jbmt.2015.11.009.
Dobos, G., 2011. Randomized controlled trial of pulsating cupping (pneumatic Ryan, R., Hill, S., 2016. How to GRADE the Quality of Evidence. Cochrane Consumers
pulsation therapy) for chronic neck pain. Forsch Komplementmed 18 (6), and Communication Group. Retrieved from. https://cc.cochrane.org/sites/cc.
327e334. https://doi.org/10.1159/000335294. cochrane.org/files/public/uploads/how_to_grade.pdf.
Doozan, A., 2015. The Use of Cupping as a Myofascial Release Tool to Increase Saha, F.J., Schumann, S., Cramer, H., Hohmann, C., Choi, K.E., Rolke, R., Langhorts, J.,
Iliotibial Band Flexibility in Collegiate Football Athletes. Lamar University - Rampp, T., Lauche, R., 2017. The effects of cupping massage in patients with
Beaumont, Ann Arbor. chronic neck pain - a randomised controlled trial. Complementary Med.Res. 24,
Downs, S.H., Black, N., 1998. The feasibility of creating a checklist for the assessment 26e32. https://doi.org/10.1159/000454872.
of the methodological quality both of randomised controlled and non- Schumann, S., Lauche, R., Hohmann, C., Zirbes, T., Dobos, G., Saha, F.J., 2012.
randomised studies of health care interventions. J. Epidemiol. Community 52, Development of lipoma following a single cupping massage - a case report.
377e384. https://doi.org/10.1136/jech.52.6.377. Forsch Komplementmed 19, 202e205. https://doi.org/10.1159/000341869.
Dworkin, R.H., Turk, D.C., Wyrwich, K.W., Beaton, D., Cleeland, C.S., Farrar, J.T., Seifman, M.A., Alexander, K.S., Lo, C.H., Cleland, H., 2017. Cupping: the risk of burns.
Haythornthwaite, J.A., et al., 2008. Interpreting the clinical importance of Med. J. Aust. 206 (11), 500. https://doi.org/10.5694/mja17.00230.
treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. Singh, B., Rajesh, Ahmad, N., 2016. Clinical evaluation of Hijamat Bila Shurt (Dry
J. Pain 9 (2), 105e121. https://doi.org/10.1016/j.jpain.2007.09.005. cupping) in the management of Waja ul zahar (Low Back Pain)-A randomized,
El Rahim, A.A.A., ELkabalawy, M.A., Al-Din Hassan, A.S., Atta, H.K., 2017. Cupping single blind, standard controlled study. Journal of Medicinal Plants 4 (5),
therapy versus interferential cupping therapy on mechanical low back pain. Int. 189e192. Retrieved from. http://www.plantsjournal.com/archives/2016/
J.ChemTech. Res. 10 (2), 253e264. Retrieved from. http://www.sphinxsai.com/ vol4issue5/PartC/4-5-16-123.pdf.
2017/ch_vol10_no2/1/(253-264)V10N2CT.pdf. Smith, K., 2015. Effect of Myofascial Decompression on Shoulder Range of Motion
Emerich, M., Braeunig, M., Clement, H.W., Ludtke, R., Huber, R., 2014. Mode of action and Strength of Healthy Overhead Athletes. Oklahoma State University, Ann
of cupping–local metabolism and pain thresholds in neck pain patients and Arbor. https://hdl.handle.net/11244/45207.
healthy subjects. Compl. Ther. Med. 22, 148e158. https://doi.org/10.1016/ Tham, L.M., Lee, H.P., Lu, C., 2006. Cupping: from a biomechanical perspective.
j.ctim.2013.12.013. J. Biomech. 39, 2183e2193. https://doi.org/10.1016/j.jbiomech.2005.06.027.
Farhadi, K., Schwebel, D.C., Saeb, M., Choubsaz, M., Mohammadi, R., Ahmadi, A., Teut, M., Kaiser, S., Ortiz, M., Roll, S., Binting, S., Willich, S.N., Brinkhaus, B., 2012.
2009. The effectiveness of wet-cupping for nonspecific low back pain in Iran: a Pulsatile dry cupping in patients with osteoarthritis of the knee - a randomized
randomized controlled trial. Compl. Ther. Med. 17, 9e15. https://doi.org/ controlled exploratory trial. BMC Compl. Alternative Med. 12 (184) https://
10.1016/j.ctim.2008.05.003. doi.org/10.1186/1472-6882-12-184.
518 S. Wood et al. / Journal of Bodywork & Movement Therapies 24 (2020) 503e518

Teut, M., Ullmann, A., Ortiz, M., Rotter, G., Binting, S., Cree, M., Lotz, F., Roll, S., systematic analysis for the global burden of disease study 2016. Lancet 390,
Brinkhaus, B., 2018. Pulsatile dry cupping in chronic low back pain - a ran- 1211e1259. https://doi.org/10.1016/S0140-6736(17)32154-2.
domized three-armed controlled clinical trial. BMC Compl. Alternative Med. 18 Yim, J., Park, J., Kim, H., Woo, J., Joo, S., Lee, S., Song, J., 2017. Comparison of the
(115) https://doi.org/10.1186/s12906-018-2187-8. effects of muscle stretching exercises and cupping therapy on pain thresholds,
Vos, T., Abajobir, A.A., Abbafati, C., Abbas, K.M., Abate, K.H., Abd-Allah, F., et al., 2017. cervical range of motion and angle: a cross-over study. Phys. Ther.Rehabil. Sci. 6
Global, regional, and national incidence, prevalence, and years lived with (2), 83e89. https://doi.org/10.14474/ptrs.2017.6.2.83.
disability for 328 diseases and injuries for 195 countries, 1990-2016: a