Cupping therapy and chronic back pain: systematic review and meta-analysis (2024)

ABSTRACT

Descriptors: Review, Chronic Pain, Back Pain, Cupping Therapy, Meta-Analysis, Nursing

Introduction

Chronic back pain causes physical, emotional and socioeconomic changes1^-3 and, consequently, high use of medicines and health resources4. The search for demedicalization leads to an increasing use of integrativeand complementary practices, such as Traditional Chinese Medicine (TCM) resources,to complement pain-related allopathic care5. Cupping therapy is one of the recommended TCM therapies for chronic painreduction6. It involves the application of cups of different materials7 in an acupoint or area of pain by means of heat or vacuum apparatus8.

The effect on pain reduction has not yet been fully elucidated9, but different mechanisms of action, based on several assumptions10, are attributed to cupping therapy, such as the metabolic, neuronalhypotheses9^,11 and TCM12. Evidence of the efficacy of this intervention is limited because of thelack of high quality, well-delineated randomized controlled trials (RCTs)6 that result in validated and efficient protocols for the treatment ofchronic back pain. Therefore, this study aims to evaluate the literature evidenceregarding the effects of cupping therapy on chronic back pain in adults compared tosham, active treatment, waiting list, standard medical treatment or no treatment,outcomes most commonly used to assess this condition, the protocol used to apply theintervention and subsequently investigate the effectiveness of cupping therapy onthe intensity of chronic back pain.

Method

A systematic review of the literature was performed, followed by meta-analysis, usedto determine the intensity of back pain in adult clients. The study was based on thecriteria of the Preferred Reporting Items for Systematic Reviews and Meta-Analyzes(PRISMA Statement)13.

The PICO (P - population; I - intervention; C - comparison; O - outcomes)14 guided the elaboration of the guiding question: “What are the effects ofcupping therapy on adults with chronic back pain?”

The search strategy, carried out by two independent reviewers from June 2017 to May2018 was based on the following databases: Medical Literature Analysis and RetrievalSystem Online (MEDLINE) via the US National Library of Medicine National Institutesof Health (PUBMED), Web of Science, The Cumulative Index to Nursing and AlliedHealth Literature (CINAHL), Physiotherapy Evidence Database (PEDro), Embase, Scopus,as well as databases indexed in the Virtual Health Library (VHL), such as LatinAmerican & Caribbean Health Sciences Literature (LILACS) and the NationalInformation Center of Medical Sciences of Cuba (CUMED). Reference lists ofsystematic reviews were also explored in the search for relevant studies related tothe guiding question.

The terms, controlled and free, were combined by means of the Boolean operators ORand AND as follows: (“Back Pain” OR “Low Back Pain” OR “Sciatica” OR “Chronic Pain”OR “Musculoskeletal Pain” OR Myalgia OR “Neck Pain” OR “Low Back Pains” OR“Musculoskeletal Pains” OR “Muscle Pain” OR “Neck Pains” OR “Cervical Pain” OR“Cervical Pains” OR Lumbago OR “lumbar pain”) AND (“cupping therapy” OR cupping ORcups).

The eligibility criteria for the selection of articles were: RCT with adults (18years or older); chronic pain (for three months or more)15 in at least one of the segments of the spine (cervical, thoracic and/orlumbar); use of cupping therapy (dry, wet, massage, flash)7 compared to one or more of the following groups: sham, active treatment,waiting list, standard medical treatment, or no treatment. We excluded studies thatdid not present online abstract in full for analysis, those that were not located byany means and studies with pregnant women.

In order to collect the information from the selected studies, we used an adaptedform16 in accordance with the recommendations of the Revised Standards forReporting Interventions in Clinical Trials of Acupuncture (STRICTA)17 and the classifications of cupping therapy7^,18.

The following data were extracted: article identification (title, author (s)/trainingarea, journal, year of publication, study country/language); objectives;methodological characteristics (design, sample size and loss of follow-up; inclusionand exclusion criteria); clinical data (number of patients by sex, mean age,diagnosis, duration of symptoms); description of interventions in the follow-upgroups (number of sessions, duration of treatment, type of technique applied (dry,wet, flash or massage cupping), application device, time of stay of the device,suction method (manual, fire, automatic-electric)/suction strength (light, medium,strong or pulsating)18; peculiarities of the intervention; application points; training area of theprofessional who carried out the intervention; years of experience in the area);outcomes and methods of evaluation (number of evaluations, intervals between them,measurement tools); data analysis; main results; and study findings.

The methodological quality of eligible studies was assessed using the Jadadscale19, which is centered on internal validity. The questions have a yes/no answeroption with a total score of five points: three times one point for the yesresponses and two additional points for appropriate randomization and concealment ofallocation methods. Two independent reviewers conducted the evaluation, and a thirdinvestigator was consulted to solve possible disagreements.

Data analyzes were performed using Stata SE/12.0 statistical software. The absolutedifference between means with 95% confidence intervals was selected to describe themean differences between the treated and control groups in the evaluation performedshortly after treatment. P-value <0.05 was considered as statisticallysignificant. Potential heterogeneity among the studies was examined using CochranQ20 and I²⁽21 statistics. Since there was statistical significance in the test forheterogeneity of the results (p <0.05) and the calculated value of I²suggested a moderate to high heterogeneity (67.7%)21, the random effects model was adopted for the analysis.

Results

A total of 614 studies were found in electronic and manual searches. Of these, 296were removed from the list because they were duplicates. After reviewing titles andabstracts, 265 studies were excluded and 53 remained for analysis of the full text.Of these, 11 studies were not found (online, via bibliographic switching or directcontact with authors) and 26 articles were excluded. Finally, 16 articles remainedin the review for the synthesis of the qualitative analysis and 10 articles enteredthe quantitative analysis (Figure 1).

All articles selected were published in English language and were conducted inGermany9^,22^-27, Taiwan28^-30, Iran31^-33, South Korea34^-35 and in Saudi Arabia36. Participants were a total of 1049 people, aged between 18 and 79 years, ofwhom 519 were in the groups receiving the experimental therapy and 530 in thecontrol groups (sham, waiting list, standard medical treatment/active treatment orno treatment). Of these, all had chronic pain conditions15, being the cervical spine/neck the most affected area9^,23^-27^,29^,34, followed by the lumbar region22^,28^,30^-33^,35^-36. Two other studies31^,33, although they did not make clear the temporality of the pain, were selectedbecause this information could be inferred with great accuracy.

The characterization of the studies regarding the objective, the interventionsapplied in the experimental and control groups, and the main findings are presentedin Figure 2.

Regarding the methodological quality of the RCTs, all reported the random sequencegeneration method and in only one study9 this process was not appropriate. In another study30 there is not enough information to infer this information. Only in fourRCTs22^,24^,28^-29⁾ there was a description of masking and in only two22^,28 this was considered appropriate. Loss of follow-up was not described in onlyone RCT29.

Therefore, 6.25% (n = 1) of the studies9 scored one on the Jadad score; 12.5% (n = 2)29^-30 scored two; 62.5% (n=10)23^,25^-27^,31^-36 scored three; 12.5% (n=2)22^,24 score four; and 6.25% (n=1)28 scored five points.

The studied outcomes, the measurement tools, the number of evaluations and theinterval between them are described in Figure3.

The most evaluated outcomes were pain intensity (100%; n=16)9^,22^-36, followed by Physical disability (62.5%; n=10)9^,23^-27^,33^-36, quality of life (37.5%; n=6)22^-27 and nociceptive threshold before the mechanical stimulus, by means of analgometer (37.5%; n=6)9^,23^-27.

The number of evaluations ranged from two (baseline and after treatment) to 18. Threestudies performed evaluations between sessions9^,28^-29; and 13 studies performed follow-up evaluations after the end of thetreatment, ranging from two days to three months 9^,22^-23^,25^-27^,30^-36 (Figure 3).

The characteristics of the intervention protocol were based on the recommendations ofthe Revised Standards for Reporting Interventions in Clinical Trials of Acupuncture(STRICTA)17 and in the classifications of cupping therapy7^,18, which are described in Figure 4.

Open in a separate window

Figure 4

Intervention protocol. Belo Horizonte, MG, Brazil, 2018(n=16)

*cm - Centimeter; †B - Bladder; ‡ml - Milliliter; §SB - Small bladder;||GB - Gallbladder; ¶LB - Large bladder; **cc - Cubic centimeter; ††mm -Millimeter; ‡‡GV - Governing Vessel

The intervention was predominantly applied by physicians (31.25%; n=5)22^,25^-28^,34; followed by nurses (18.75%; n=3)22^,29^,32 and pharmacists (6.25%; n=1)32. And 25% of the studies (n=4)9^,23^,35^-36 reported that the intervention was performed by a therapist, withoutspecifying the training area.

Only 18,75% of the studies (n = 3) presented the time of experience of theprofessional who performed the intervention, from three35^-36 to four years34; 37.5% of the studies (n=6)9^,22^-25^,27 informed only that the intervention had been performed by experienced ortrained professionals, but did not mention the time of training.

Of the 16 articles selected for the systematic review, 10 entered for meta-analysisthat investigated the effectiveness of cupping therapy on pain intensity. All ofthem approached the outcome in two comparison groups (experimental and control), inevaluations performed before and immediately after the treatment. Five studies9^,22^,29^,35^-36 did not enter because they did not have enough data for this analysis andone study33 performed the evaluation only three months after the end of treatment.

The results of the meta-analysis showed that cupping therapy was more effective inreducing pain compared to the control group (absolute difference between means:-1.59, [95% Confidence Interval: -2.07 to -1.10]; p = 0.001), with moderate to highheterogeneity (I² = 67.7%, p = 0.001) (Figure 5).

Discussion

Cupping therapy has shown positive results on chronic back pain in adults, not onlyin behavioral variables of pain, but also in physiological parameters in themajority of RCTs evaluated in this study, which contributes to the consolidation ofits use in the treatment of this clinical condition in the study population.

Regarding methodological quality, most studies23^,25^-27^,31^-36⁾ obtained a median score (three) according to the Jadad scale19. This score can be justified by the lack of masking of RCTs.

It is not feasible to conceal evaluation and intervention methods in cuppingtherapy22, since the marks left by the suction cups are often visible and may persistfor several days, making it difficult to perform a masking process27. Only one study28 achieved masking properly; however, it was true only for volunteers whor*ceived laser therapy, an intervention used concomitantly with cupping therapy,where sham laser acupuncture was performed by applying the same procedure in one ofthe groups, but without energy. In a second study24, there is a description that the masking was applied to the evaluator of theresults; however, the application of suction cups causes marks (ecchymoses,petechiae) and one of the evaluated outcomes was the pain threshold, using thealgometer; for this evaluation, as the area must be naked, the marks on the skinmake this kind of masking impossible. Finally, in another study22, the majority of participants in the minimal cupping group (84%) was able toidentify the allocation after four weeks, whereas in the cupping group 55%identified the allocation.

Regarding the evaluated outcomes, pain intensity predominated, which was measuredmostly by means of the Visual Analogue Scale (VAS)22^-25^,27^-30^,32 and the Numerical Scale26^,34^-36, followed by the Neck Pain and Disability Scale9, by the short version of the McGill Pain Questionnaire31, and by the Present Pain Intensity Scale33.

Although there are variations, the VAS usually consists of scores of 0-10 or 0-100,the extreme left being described as no pain and the extreme right as the worstpossible pain; the numerical scale has a numerical rating of 0-10, 0-20 or 0-100.These scales can be classified as: painless (0), mild (1-3), moderate (4-6), andsevere (7-10), and are frequently used in patients with chronic musculoskeletalpain37. In addition, some researchers38^-40 have pointed to these two scales as the gold standard for assessing painintensity, these being the instruments most used when evaluating adults, both inclinics and research.

Physical disability was the second most approached outcome, measured by means of theNeck Disability Index (NDI)23^-27^,34, of the Oswestry Disability Questionnaire (ODQ)33^,35^-36 and the Neck Pain and Disability Scale9. In fact, the severity and chronicity of back pain are associated withsevere functional limitations37 that imply limitations in activities of daily living41.

In addition, patients with chronic diseases, who require continuous treatment over along period, present important changes in quality of life42, being another important outcome to be evaluated, as occurred in sixstudies, through the Short Form 36 Health Survey Questionnaire (SF-36)22^-27.

Finally, the physiological parameter most evaluated in the studies was thenociceptive threshold before the mechanical stimulus, by means of a pressurealgometer9^,23^-27. It is known that individuals who have pain in the spine have highernociceptive sensitivity compared to healthy people43. However, this is still considered a subjective variable, since it is thepatient who determines his/her pain threshold. In fact, when the evaluation processis more related to the symptoms, such as subjective phenomena, especially pain, thanto physical or laboratory results, self-assessment is considered the most reliableindicator of the existence of pain44. Thus, the necessary information to carry out its evaluation has its originin the individual’s report45, who is the primary source of the assessment.

The systematized analysis of cupping therapy application methods showed that there isno standardization in the treatment protocol for chronic back pain. However, recentefforts have been made to standardize the cupping therapy procedure in general46 and specifically for chronic back pain, since the most appropriate type oftechnique, duration of treatment, number of sessions, devices, time of application,method and suction strength and application points have not been determined.

It can be observed, however, that the most applied technique was dry cupping,specifically for the lumbar22^,28^,30^-32 and cervical regions9^,27^,29^,34. This modality allows the stimulation of the acupoints in the same way asthe acupuncture needles47. Researchers18 suggest that laceration of the skin and capillaries, promoted by wetcupping, may act as another nociceptive stimulus that activates the descendinginhibitory pathways of pain control18, thus helping to treat chronic musculoskeletal conditions35. However, risk for infection, vasovagal attacks and scars are thedisadvantages of this method18. Still, compared to cupping massage, authors47 emphasize that dry cupping has a greater analgesic effect, since the use oflubricants can reduce the friction between the edge of the cup and the skin, a factcorroborated by some authors24 who used arnica oil for the realization of cupping massage.

Despite the variability in the application of the intervention, it was possible toidentify that, on average, the cupping therapy was applied in 5 sessions, withpermanence of the cups in the skin for around 8 minutes, and interval of three tofour days between the applications. According to some researchers27, at least five sessions are required for any significant effects of cuppingtreatment to appear, in addition to ensuring the feasibility of the RCT. Moreover,authors47 recommend that the cups should be left on the skin for 5 to 10 minutes ormore, which culminates in the appearance of residual marks after treatment as aresult of the rupture of small blood vessels that are painless and disappear between1 and 10 days12. Therefore, an interval between sessions is necessary in order to allow thereestablishment of the cutaneous and subcutaneous tissues.

Regarding the application cups, the disposable ones are preferable a high-levelsterilization or disinfection process is required prior to reuse, since the pressureexerted may cause extravasation of blood and fluids from the skin46. Nowadays, cupping therapy has increasingly been performed with plasticcups47. The size of the cups varies according to the place of application, but itis often applied in places with abundant muscles, such as the back48.

Regarding the suction method to create negative pressure, the use of firepredominated9^,25^,27^,29^,32, followed by manual pumping23^,34^-36 and automatic pumping22^,26^,33. Suction with fire is the traditional method used in China, however, thereis a risk of burns18. Manual vacuum is created when using a suction pump. This method allowsmicrocirculation to increase more effectively if compared to fire18. Finally, automatic pumping is created using an electric suction pump, whichallows to adjust and measure the negative pressure inside the cup, being the mostsuitable method for scientific research18.

Only three studies22^,26^,28 reported the suction strength used, which should be standardized in theapplication protocols. The suction can be light (100 and 300 millibar/one or twomanual pumpings), medium (300 and 500 milibar/three or four manual pumpings), strong(above 500 milibar/five or more manual pumpings) or pulsatile (pressure inside thecups is variable, between 100 and 200 milibar every 2 seconds)47^,49. The medium suction is often indicated for painful conditions of themusculoskeletal system18.

There was also no standardization in relation to the application points of cuppingtherapy. Despite this, the application in specific acupoints in the cervical region,mainly on the bladder, gallbladder and small intestine meridians, prevailed29^,34, and in the lumbar region on the bladder meridian30^-32^,35^-36, followed by sensitive points9^,25^-27^,30 named Ashi by TCM or trigger points by Westernmedicine.

Meridians are passages for the flow of “qi” (vital energy) and“xue” (blood), the two basic body fluids of TCM, which spreadthroughout the body surface, uniting the interior with the exterior of the body andconnecting the internal organs, the joints and the extremities, transforming thewhole body into a single organ50. Part of the meridians of the bladder, small intestine and gallbladder passthrough the dorsal region. The acupuncture points are located in the meridians;besides local action, they also play a systemic action and reestablish the energybalance of the body by adjusting the function of the organs, maintaining homeostasisand treating the disease51, so the advantage in using them.

The trigger points or Ashi are specific points of high irritability;they are sensitive to digital pressure and can trigger local and referred pain52. They may be deriving from dynamic overload, such as trauma or overuse, orstatic overload, such as postural overloads occurring during daily activities andoccupational activities53, besides emotional tension. Addressing these points can also be a way torelieve local pain54.

After the application of cupping therapy, both the acupoints of the meridians of theaffected regions and the trigger points or Ashi may presentbruising, erythema and/or ecchymoses. According to TCM, these signs representstagnation of “qi” and/or “xue” and may help thetherapist in identifying body disorders.

Finally, the meta-analysis revealed a significant reduction of the pain intensityscore in adults with chronic back pain by using cupping therapy (p = 0.001).Compared with a control group (usual care/other intervention/waiting list), thismodality has advantages in relieving pain, as can be seen in Figure 5.

Only two studies24^,30 did not present a statistically significant difference between the groups onthe benefit or harm of this intervention (Figure5). In fact, the first study24 pointed out that cupping therapy has the same effect as other intervention(progressive muscle relaxation) in reducing chronic nonspecific neck pain; despitethis, cupping therapy was better than relaxation in improving well-being anddecreasing sensitivity to pressure pain. The authors24 justify this result, among other limitations, due to the fact that cuppingtherapy was performed by patients’ relatives or friends at home. The secondstudy30, despite having found a positive result on the intensity of pain, did notobtain a result in the meta-analysis. It is believed that this may have been due tothe fact that both groups received the intervention of soft cupping and bothobtained positive results.

In the other studies23^,25^-28^,31^-32^,34, the intervention reduced the probability of the outcome, being the studywith the largest sample31 the one the most contributed (15.68% weight in the meta-analysis) for this(Figure 5). In fact, all these studiesreported promising results of intervention on pain intensity.

However, the results of the effectiveness of cupping therapy still need to beconfirmed by subgroup analyzes, based on different types of application techniquesand control groups. In addition, it is important to perform meta-regression to findthe source of heterogeneity of RCTs.

In a general way, the results showed a substantial variation in the application ofcupping therapy, especially in relation to the type of technique, as well asdifferences in the control group, which made subgroup or meta-regression unfeasible,respectively, due to the small number of studies with each of thesespecifications.

Conclusion

Cupping therapy is a promising method for the treatment and control of chronic backpain in adults, since it significantly decreases pain intensity scores when comparedto control groups. However, the high heterogeneity and the median methodologicalquality of RCTs has limited the findings.

Despite this, a protocol can be established for this clinical condition: applicationof dry cupping technique in 5 sessions, with permanence of the disposable or plasticcups on the skin for about 8 minutes, preferably automatic or manual pumping, withmedium suction strength, and three to seven days interval between applications. Itis better to opt for acupoints of the dorsal region, especially those from thebladder meridian in the lumbar region, and for the meridians of the bladder,gallbladder and small intestine in the cervical and thoracic regions, as well asAshi or trigger points. This protocol needs to be validated infuture studies. And the main outcomes evaluated for this clinical condition werepain intensity, physical disability, quality of life and nociceptive thresholdbefore the mechanical stimulus (pressure).

Referências