Acupuncture for Temporomandibular Disorders: A Systematic Review
Temporomandibular disorders (TMD) include a group of conditions that affect the temporomandibular joint (TMJ),masticatory muscles, and associated head and neck musculo -
skeletal structures, and may present as a cluster of joint and muscledisorders.1 TMD are also known as craniomandibular disorders(CMD) and are a frequent cause of facial pain problems.2 TMDusually manifest as one or more of the following signs or symp-toms: pain, joint sounds, limitation in jaw movement, muscle ten-derness, and joint tenderness.3 It is also commonly associated withother symptoms affecting the head and neck region such asheadache, ear-related symptoms, and cervical spine disorders.4,5
Patients with persistent TMD frequently report symptoms ofdepression, poor sleep quality, and low energy.6 Prevalence studieshave reported approximately 75% of the population have at leastone sign of joint dysfunction that includes abnormal jaw move-ment, joint noises, and tenderness on palpation, and approximately33% of people have at least one symptom such as facial pain andjoint pain.7,8
Seung-Hun Cho, KMD, PhDAssistant ProfessorHospital of Korean MedicineKyung Hee University Medical CenterKyung Hee UniversitySeoul, Korea
Wei-Wan Whang, KMD, PhDProfessorHospital of Korean MedicineKyung Hee University Medical CenterKyung Hee UniversitySeoul, Korea
Correspondence to: Dr Seung-Hun ChoHospital of Korean Medicine Kyung Hee University Medical Center Kyung Hee University#1 Hoegi-Dong, Dongdaemun-GuSeoul 130-701, Republic of KoreaFax: (+82)-2-958-9187Email: [email protected]
Aims: To assess the effectiveness of acupuncture for the symp-tomatic treatment of temporomandibular disorders (TMD) from areview of studies using randomized controlled trials (RCTs).Methods: Electronic databases were systematically searched forarticles reporting RCTs investigating acupuncture for TMD. Themethodological qualities of eligible studies were assessed using thecriteria described in the Cochrane Handbook. Results: Nineteenreports were systematically reviewed. There was moderate evi-dence that classical acupuncture had a positive influence beyondthose of placebo (three trials, 65 participants); had positive effectssimilar to those of occlusal splint therapy (three trials, 160 partici-pants); and was more effective for TMD symptoms than physicaltherapy (four trials, 397 participants), indomethacin plus vitaminB1 (two trials, 85 participants), and a wait-list control (three trials,138 participants). Only two RCTs addressed adverse events andreported no serious adverse events. Conclusion: This systematicreview noted moderate evidence that acupuncture is an effectiveintervention to reduce symptoms associated with TMD. There is aneed for acupuncture trials with adequate sample sizes thataddress the long-term efficacy or effectiveness of acupuncture.J OROFAC PAIN 2010;24:152–162
The pathogenesis of pain in TMD is unclear, withphysical, biochemical, and psychological factors allpotentially playing a role.9 Currently managementof TMD include reassurance (patient education, selfcare, and behavior therapy), acu puncture, physio-therapy (such as ultrasound, mega pulse, short-wavetherapy, diathermy, laser, heat, cold, transcutaneouselectrical nerve stimulation, mobilizations, mas-sages, stretching, instructions, exercises, andbiofeedback), splint therapy, occlusal adjustment,surgery (arthrocentesis, arthroscopy), pharmacologi-cal intervention, and combined approaches.1,10–13
Published surveys on complementary and alter-native medicine (CAM) use suggest that CAM issought most frequently for musculoskeletal andpain disorders.14–16 A recent survey has docu-mented the relatively frequent concurrent use ofCAM and conventional therapies by patients withTMD.17 Nearly two-thirds of the respondents(62.5%) reported using CAM therapies for TMDor related conditions.17 Among those respondentswho reported using acupuncture, the techniquewas reported as the satisfactory CAM therapy forTMD, with 72.8% of respondents “extremely sat-isfied” or “very helpful.”17
Since the most recent systematic review18 on fourrandomized control trials (RCTs) on acupuncturefor TMD, 10 RCTs19–28 (two23,24 incorporating anonpenetrating needle placebo control) have beenpublished. Therefore, the aim of this systematicreview was to assess the effectiveness of acupunc-ture for the symptomatic treatment of TMD from areview of studies using RCTs.
Materials and Methods
Search Strategy
The following sources were searched up to July2008: The Cochrane Library including theCochrane Central Register of Controlled Trials(CENTRAL, 2008); MEDLINE, EMBASE, Alliedand Com plementary Medicine Database (AMED),CINAHL, PsycInfo, Korean medical databases(including the National Assembly Library,KoreaMed, Korean Studies Information ServiceSystem, DBpia, and Korea Institute of ScienceTechnology Information and Research InformationService System), a Japanese database (Japan Scienceand Technology Information AggregatorElectronic), Chinese databases (which included theChina Academic Journal, Century Journal Project,China Doctor/ Master Dissertation Full Text DB,and China Proceedings Conference Full Text DB),
BIREME (Latin American and Caribbean Center onHealth Sciences Information), and MEDPILOT(German medical databases). The authors alsosearched databases of clinical trials such as CurrentControlled Trials (http://www.controlled-trial.com),National Center for Complementary andAlternative Medicine (NCCAM) at the US NationalInstitutes of Health (NIH) (http://nccam.nih.gov/),and the Comple mentary and Alternative MedicineSpecialist Library at the National Health ServiceNational Library for Health (http://www.library.nhs.uk/cam/) for on-going studies, funded research,and protocols. The reference lists of articles werechecked for further relevant publications andexperts in complementary medicine, acupunctureresearch groups, or oral health groups were askedfor information concerning any additional trials. Afurther manual search was conducted of relevantjournals, symposia, and conference proceedings,and relevant trials were retrieved; all identified pub-lications were cross-referenced (Journal of OralRehabilitation, Journal of Oral and Maxillo facialSurgery, Journal of Craniomandibular Practice,Acta Odontologica Scandinavica, Journal of theAmerican Dental Association, Journal of Cranio -mandibular Disorders, Journal of Orofacial Pain,Journal of Korean Academy of Cranio mandibularDisorders). If necessary, personal contact was madewith the authors of the published studies to requestadditional data.
Key words used to search RCTs were (“acupunc-ture” OR “electroacupuncture” OR “meridian”OR “acupoint” OR “acupoint injection” OR“auricular acupuncture”) AND (“temporomandibu-lar joint disorders” OR “temporomandibular jointdysfunction syndrome” OR “craniomandibular dis-orders” OR “myofascial pain syndromes” OR“temporomandibular joint” OR “TMJ” OR“CMD” OR “TMD” OR “TMDs” OR “temporo-mandibular”). All of the various databases utilizedfor this study possessed their own subject headingsand each database was searched independently.
Study Selection
Types of Studies. The review was restricted to RCTsthat compared acupuncture with a control group toassess the efficacy of acupuncture for the treatmentof TMD. No restriction was imposed on studieswith respect to language, publication types, blind-ing, and the type of design such as parallel orcrossover. Crossover trials were included as long asoutcome data were available for each treatment seg-ment prior to crossover. The review excluded quasi-randomized trials.
Types of Participants. The study included allpatients with TMD diagnosed by clinical and/orimaging criteria regardless of their age, race, gen-der, profession, or residential location. Trials withpatients having congenital abnormalities, concomi-tant inflammatory or neoplastic conditions, or witha recent history of acute trauma were excluded.
Types of Intervention. Clinical trials evaluatingclassical acupuncture, electroacupuncture, electricalauricular acupuncture, auricular acupuncture,warm-needle acupuncture, and acupoint injectionwere included. Both traditional acupuncture (classi-cal meridian points) and contemporary acupuncture(nonmeridian, trigger points, or electroacupuncture-points according to Voll) were included. Regardingauricular acupuncture, trials with a traditionalChinese type or a European type (auriculo-therapy)were included. Trials with acupuncture-relatedstimulation (for example, seed, laser, acupressure,magnetic devise, or moxibustion) were excluded.Studies that assessed the combined effect ofacupuncture with other therapies (for example,acupuncture plus short-wave diathermy, oracupuncture plus moxibustion therapy) wereexcluded because the purpose of the review was toassess the effects of acupuncture alone. Trials thatcompared different forms of acupuncture to eachother were also excluded. Types of control inter-ventions considered in this review included notreatment (wait-listed or treatment as usual),placebo treatment (such as nonpenetrating needle,or placing either short-wave or ultrasonic physio-therapy appliances in proximity to the TMJ but notactivating the TMJ) superficial acupuncture (so-called sham-acupuncture or minimal acupuncture),actual physiotherapy, relaxing appliances, pharma-cological interventions, any occlusal appliance,orthodontic treatment, or surgery.
Types of Outcome Measures. The primary out-comes were pain in the TMJ and masticatory mus-cles (pain intensity or pain relief recorded using avisual analog scale [VAS] or a validated categoricalscale, data on frequency, severity, or duration ofpain), tenderness on palpation of TMJ and masti-catory muscles, global measures (such as theHelkimo anamnestic index and/or dysfunctionindex), mandibular movement (range of motion,maximum interincisal opening, quantitative mea-surements of lateral movement, and protrusion),joint sounds, proportion of patients who improvedafter treatment as self-assessed by the patient orvia a clinical assessment carried out by the clini-cian (categorical data were converted to binaryoutcomes; eg, improved/not improved), and sub-jective assessments by the patients (such as pain on
face and jaw, clicking of the joints, and dysfunc-tion). Secondary outcomes assessed were otherclinically important outcomes (eg, headache) andadverse effects from treatment (incidence and typeof side effects).
Data Abstraction and Quality Assessment
Each study identified by the search strategy wasassessed against the inclusion criteria by one of thereviewers. Where there was uncertainty regardingeligibility, a second reviewer also assessed thestudy and a decision was reached through discus-sion and consensus. Both reviewers independentlyassessed whether the studies met the inclusion cri-teria with disagreements resolved by discussion.Further information was sought from the authorswhere papers contained insufficient information tomake a decision about eligibility. Data extractionwas undertaken for relevant articles through theuse of a properly designed data extraction form.The data extraction form was based on severalpapers and was modified as needed before use.The quality assessment of all studies was under-taken by two reviewers following the detaileddescriptions of these categories provided in theCochrane Handbook for Systematic Reviews ofInterventions.29 The following questions wereassessed: (1) Was the allocation sequence ade-quately generated? (2) Was allocation adequatelyconcealed? (3) Was knowledge of the allocatedinterventions adequately prevented during thestudy? (4) Were incomplete outcome data ade-quately addressed? (5) Were reports of the studyfree of suggestions of selective outcome reporting?(6) Was the study apparently free of other prob-lems that could put it at a risk of bias? A “Yes”answer indicated a low risk of bias (A), “Unclear”indicated a uncertain risk of bias (B), and a “No”answer indicated a high risk of bias (C).
Data Analysis
The RCTs were clinically heterogeneous withrespect to the type of the interventions (methods ofacupuncture, type of control) and outcomes.Furthermore, the outcomes were poorly presentedin some studies. Therefore, it was decided not topool the data statistically, but to perform a quali-tative review. The method of best evidence synthe-sis30 was used to formulate conclusions on theeffectiveness of acupuncture for each type of con-trol. This method consists of five levels of evidenceand takes into account the methodological qualityand the outcome of the studies:31
• Level 4: conflicting evidence—inconsistent find-ings among multiple trials (RCTs)
• Level 5: no evidence—no RCTs
An RCT was considered to be of high quality ifthe methodological quality satisfied all sixCochrane categories. “Multiple” was defined asmore than one.
Results
Study Description
Fourteen RCTs covering 808 patients wereincluded in this systematic review. An initial searchidentified 115 potentially relevant articles. Ninety-five articles were initially excluded because theydid not meet the inclusion criteria. Among them,20 papers involved a combined intervention ofacupuncture with other therapies in the experi-mental group, for example, acupuncture–moxibus-tion and acupuncture–spinal tuina therapy. Twostudies were laser acupuncture trials.32,33 The
remaining 20 studies were further evaluatedregarding randomization; one trial was found tobe quasi-randomized trials.34 The remaining 19studies met the inclusion criteria and were system-atically reviewed. Among them, one trial wasreported three times35–37 and another trial wasreported four times,38–41 and each report con-cerned a different end point or measurement point.These reports were included in this review for suf-ficient data, but the results were considered as onestudy. Figure 1 summarizes the search resultsbased on the quality of reporting of meta-analyses(QUOROM) flow diagram.42
The key data are summarized in Table 1. Of the14 trials, 6 were conducted in mainlandChina,19,20,25–28 2 were conducted in Sweden,38,43
2 in the United States,23,44 1 in Austria,22 1 inFinland,35 1 in the United Kingdom,24 and 1 inSouth Korea.21 The trials were published in 6reports between 2006 and 2008. There was con-siderable diversity in the clinical presentation anddiagnosis of participants with TMD among theincluded studies. Two of the studies used theresearch diagnostic criteria established by Dworkinand LeResche45 to classify the patients as havingmyogenous TMD. Most of the studies used theirown diagnostic criteria, based on signs and symp-toms of the patients. The number of participantsranged from 15 to 170 patients. The number ofpatients per study group was less than 20 in sixstudies.21–24,43,44 Except for two trials,20,35 12 ofthe 14 studies gave detailed criteria for inclusion
Fig 1 Flow diagram showing the num-ber of studies included and excluded fromthe systematic review.
Potentially relevant articlesreviewed and retrieved for moredetailed evaluation (n = 115)
Excluded (n = 95)• Case studies, not a clinical trial (n = 12)
• Not TMD (n = 24)• Review, survey articles (n = 14)• Comparing different forms ofacupuncture to each other (n = 23)
• Intervention included other treatments with acupuncture (n = 20)
Table 1 Characteristics of RCTs of Acupuncture for TMD
Intervention type, No. of subjects treatment frequency
[range or Inclusion (treatment period); Type of Quality Study mean age] Location criteria treated acupoints control group assessment*
Goddard et al, 200244 18 [22 – 52 y] USA RDC/TMD by AT, 1 se; LI4, ST6 Superficial AT A-B-A-A-A-ADworkin and (Sham AT at LeResche45 nonacupoints)
Johansson et al, 199143 45 [NR] Sweden Signs and symptoms AT, 6 se (3 months); (a) Occlusal splints B-B-A-B-A-Aof CMD, headache 3 to 7 adjunctive (b) Wait-listor facial pain acupoints and LI4
Li and Rong, 200319 40 [12 – 50 y] China TMD with muscular EAT plus AI, 20 se Indomethacin B-B-C-B-B-Cdysfunction or (22 days); ST6, 25 mg and articular dysfunction SI18, LI4 Vitamin B1 0.2
tablet 3 times/dayList et al, 199238 110 [19 – 76 y] Sweden Signs and symptoms AT plus EAT, 6 to 8 se (a) Occlusal splints B-B-B-C-A-AList et al, 199341 of CMD of primarily (6 to 8 weeks); ST6,7, (b) Wait-list List and Helkimo, 1992a39 muscular origin, SI18,19, BL2,10, GV20, (3 months)List and Helkimo,1992b40 pain > 6 months GB20,21, EX Qianzheng,
EX2 Taiyang, LI4, ST36,and adjunctive points
Luo et al, 200120 41 [19 – 47 y] China Signs and symptoms AT, 20 se (25 days); Ultrasound therapy, B-B-C-B-B-Cof TMD GB2,20, ST7, SI19, 20 se
TE17, LI4 Park et al, 199921 38 [13 – 54 y] Korea Signs and symptoms EAT, 6 se (2 weeks); Wait-list B-B-C-B-A-A
of TMD LI4, SI19, TE17,BL10, GB2,21, ST5,6,7
Raustia et al,198535 50 (27.8 y) Finland TMJ dysfunction AT, 3 se (1 month); Standard B-B-B-B-A-ARaustia et al, 198636 Adjunctive points stomatognathicRaustia and Pohjola, 198637 treatmentsSchmid-Schwap et al, 200622 23 [17 – 59 y] Austria Female patients with AT, 1 se; LI4, SI2,3, Sham laser B-A-A-A-A-A
TMJ pain and ear, sternum treatment (SI2,3,tenderness of (acupuncture needles) ear, and maxilla craniomandibular and intraoral points and mandible musculature (insulin syringes with retromolar without
0.5 mL) contact and being activated)
Shen and Goddard, 200723 15 (43.1 y) USA Chronic myofascial AT, 1 se; LI4 Placebo AT (non- B-B-A-A-A-Apain syndrome of penetrating masticatory muscles; needling)pain ≥ 12 weeks
Smith et al, 200724 27 (40.5 y) UK RDC/TMD by AT, 6 se (over Placebo AT A-A-A-A-A-ADworkin and 3 weeks); ST7 (non-penetrating LeResche45 needling)
Wang, 199625 66 [15 – 50 y] China Signs and symptoms EAT, NR; ST6,7, SI19 Massage, NR B-B-C-B-B-Cof TMJ dysfunction
Zhang, 200826 45 [18 – 55 y] China TMD with muscular Warm-AT, 20 se Indomethacin B-B-C-B-B-Cdysfunction or (20 days); ST7, 25 mg 3 times articular dysfunction TE17, SI19 and Vitamin B1
10 mg 3 times/dayZhong et al, 200727 120 [20 – 50 y] China Signs and symptoms Warm-AT, 10 se Ultrasound therapy A-B-C-B-B-C
Zhu, 200728 170 [15 – 48 y] China Signs and symptoms EAT, 30 se (30 days); Manual therapy, B-B-C-B-B-Cof TMJ dysfunction SI19, GB2, ST7, TE17, 30 se
LI4, TE5
Abbreviations: se = sessions; NR = not reported; AT = classical acupuncture; AI = acupoint injection; EAT = electroacupuncture; RDC/TMD = Research Diagnostic Criteria for TMD.*(1) Was the allocation sequence adequately generated? (2) Was allocation adequately concealed? (3) Was knowledge of the allocated interventions adequately prevented during the study? (4) Were incomplete outcome data adequately addressed? (5) Were reports of the study free of suggestion ofselective outcome reporting? (6) Was the study apparently free of other problems that could put it at a risk of bias? Key: (A) indicates Yes; (B), Unclear; (C), No.
and/or exclusion of patients in the study. The typesof acupuncture techniques used in the trialsincluded classical acupuncture, warm-needleacupuncture, electroacupuncture, acupoint injec-tion, and intraoral acupuncture; warm-needleacupuncture is one of the needling acupuncturetechniques in which a needle is inserted into anacupoint and moxa (Artemisia vulgaris) is attachedand burned on top of the needle to provide heatvia the needle. Varied styles of acupuncture wereused in the included RCTs: individualized (7%),standardized (72%), and semistandardized (21%)acupuncture. Semistandardized acupuncture hasbeen defined as a set formula of points supple-mented by some additional points individuallychosen for each patient. Various acupoints foracupuncture treatments were used in the includedRCTs; the ST6,7, LI4, or SI19 acupoints werecommonly selected in over 5 trials. The frequencyof overall treatment sessions ranged from 1 to 30.Comparison groups included placebo needle, shamtreatment, superficial acupuncture (sham acupunc-ture), occlusal splints, physical therapy, pharmaco-logical treatments, and wait-list. There wasvariation in the type of measurement used for themain outcomes. Pain was measured using a VAS,numerical analog scale, and presence or absence ofheadache. Other measured outcomes included clin-ical dysfunction scores, tenderness, range in move-ment, sounds, locking and deviation in opening ofmouth, and overall improvement.
Methodological Quality
Sequence Generation. Three of the included trialsdescribed adequate methods of randomization. Twostudies27,44 referred to a random number table andone study24 used a computer-generated randomiza-tion for sequence generation. The other trials didnot describe the sequence generation process.
Allocation Concealment. Two studies22,24
ensured that allocations were concealed by usingenvelopes. The other trials received allocationscores of “B” as they did not have clear descrip-tions of their method of allocation concealment.
Blinding. Four studies reported blinding of theassessor and participants by using nonpenetratingplacebo needle,23,24 sham acupuncture at nonacu-points,44 and sham inactive laser intervention.22 Inone trial,43 participants were not blinded; how-ever, investigators were blind to treatment groupassignment and outcome assessments. Two trialsreported an independent assessor but did not men-tion the blinding of the assessor.35,38 The otherstudies did not blind participants or acupuncturists
or outcome assessment; the outcome measure-ments were likely to be influenced by this lack ofblinding.
Incomplete Outcome Data. Three studies hadno participant losses, or the missing data balancedin numbers across intervention groups.22,23,44 Inone study,24 “intention-to-treat” analysis was donewith one dropout. The risk of bias in the otherincluded trials is unclear because the numbers ran-domized into each intervention group were notclearly reported.
Selective outcome reporting. Six studies ana-lyzed continuous outcomes as a dichotomous vari-able, with the further possibility of selecting frommultiple cut-off points.19,20,25–28
Other sources of bias. The six trials used insen-sitive and subjective criteria measured as thewomen with pain relief, reduced pain, or noimprovement.19,20,25–28 An insensitive instrumentcan lead to under- or overestimation of the effects.
Data Analysis
Acupuncture Versus Inactive Treatment. Onehigh-quality RCT demonstrated that classicalacupuncture had greater influences on VAS forpain intensity and the number of areas of pain,headache, VAS scores for functional impairment,tenderness, maximum opening and maximum painfree opening, and joint sounds than those ofplacebo acupuncture (Table 2).24 The other trialsusing placebo needles showed significant improve-ments of the classical acupuncture group in paintolerance of the masseter muscle (P = .027), butcould not find significant differences in facial pain,headache, and neck pain between both groups.23 Acomparison with sham laser treatment in onestudy reported significant reduction in VAS forpain (P = .033) and tenderness, and pain on pres-sure in neck and masticatory muscles (P < .05).22
The results from the three trials involving 65 par-ticipants provided moderate evidence of a positiveinfluence of acupuncture beyond those of placebo.
Acupuncture Versus Superficial Acupuncture(Sham Acupuncture). One study compared classicalacupuncture with superficial acupuncture (“shamacupuncture”) at nonacupoints.44 Both groupsshowed significant reduction in VAS (P = .001), butthere was not a significant difference betweengroups (Table 2). The limited evidence from thissingle trial involving 18 participants was consistentwith the ability of classical acupuncture and super-ficial acupuncture to reduce pain evoked bymechanical stimulation of the masseter muscle inTMD patients.
Acupuncture Versus Occlusal Splint Therapy.With respect to pain, clinical dysfunction score, ten-derness, and subjective symptoms, two trials com-pared acupuncture to occlusal splints. The trialsshowed significant reductions of both groups frombaseline.38,43 There were not significant differencesbetween groups (Table 3). One of the trials usingclassical acupuncture plus eletroacupuncturereported similar results in pain, clinical dysfunctionscore, and subjective symptoms at 6 and 12 monthsfollow-up.39 One study also reported similar resultsin modified clinical dysfunction score compared toocclusal splint therapy combined with counselingand muscular exercise for the lower jaw.35 The datafrom these three trials involving 160 participantsprovide moderate evidence of the positive effects ofacupuncture similar to those of conventionalocclusal splint therapy.
Acupuncture Versus Wait-list Control. Three tri-als reported significant differences in pain betweenacupuncture and wait-list (Table 3). With respectto clinical dysfunction score and subjective symp-toms, two trials using classical acupuncture,38 orclassical acupuncture plus electroacupuncture,43
also showed significant differences between both
groups. One study reported significant improve-ments in pressure pain threshold.41 One studyfailed to reach a significant level in pressure painthreshold (P = .055), but showed significant resultsin noise frequency (P = .016) and limitation ofmotion (P = .004) between electroacupuncture andwait-list groups.21 The data from the three trialsinvolving 138 participants provide moderate evi-dence that acupuncture is more effective than await-list control at relief for patients with TMD.
Acupuncture Versus Physical Therapy. Threestudies showed significant differences in responderrate with improvement of TMD symptoms compar-ing warm-needle acupuncture versus ultrasoundtherapy,27 electroacupuncture versus manual ther-apy,28 or electroacupuncture versus massage.25 Onestudy,20 provided only as an abstract, reported adifference in responder rate with improvement ofTMD symptoms without mentioning statistical val-ues comparing classical acupuncture versus ultra-sound therapy (Table 4). The findings providemoderate evidence (four trials, 397 participants)that acupuncture is more effective for TMD’ssymptoms than physical therapy such as ultrasoundtherapy, manual therapy, or massage.
Table 2 Outcomes of RCTs of Acupuncture Versus Inactive Treatment or Sham Acupuncture
Placebo ATSmith et al, 200724 Reduced VAS for pain Reduced VAS for Improved tenderness Improved maximum opening
intensity (P = .001) from functional impairment of left masseter, right (P = .02), maximum pain baseline, number areas from baseline in AT temporalis, and both free opening (P < .0001) of pain (P = .003) (P = .001) lateral pterygoid muscles from baseline in AT; improved presence of headache in AT joint sounds values in only (P = .014) in AT one subject for AT,
no changes in placeboShen and Goddard, No between-group Significant difference in 200723 differences in numeric pain tolerance of the
scales for facial pain, masseter muscle with AT headache, or neck pain versus placebo (P = .027)
Sham laser treatmentSchmid-Schwap Significant higher reduction Significant differences in Significant difference in et al, 200622 in VAS with AT (P = .033) tenderness and pain on mouth opening of patients
pressure in neck and with restricted openingmasticatory muscles with between groups (P = .037)AT versus sham (P < .05)
Superficial AT (Sham AT)Goddard et al, Reduced VAS from baseline 200244 within both groups (P = .001);
Occlusal splintsJohansson et al, Reduced VAS for facial pain Reduced clinical dysfunction Reduced number of Improved subjective 199143 and headache within both score within both groups from masticatory muscles tender symptoms within both
groups from baseline baseline (P < .01); no within both groups from groups; no between-group (P < .05); no between-group between-group differences baseline (P < .05); no differencesdifferences between-group differences
List et al, 199238 Reduced frequency in AT, Improved clinical dysfunction Improved pressure pain Improved subjective List et al, 199341 intensity of pain (VAS) in score in both groups from threshold in both groups evaluation (98% versus 65%; List and Helkimo, both groups from baseline; baseline; no between-group from baseline: no between- P < .001), Activity of Daily 199239 no between-group differences. differences. Similar results at group differences; no Living (P < .01) with AT List and Helkimo, Similar results at 6-, and 6-, and 12-months follow-up significant differences in versus occlusal splints. 199240 12-months follow-up both groups compared with Significant differences from
short-term results at baseline at 6-, and 12-months6-month follow-up follow-up
Standard somatognathic treatmentsRaustia et al, 198535 Similar improved modified No between-group Raustia et al, 198636 clinical dysfunction score in differences in responder rate Raustia and Pohjola, both groups; no between- with subjective patient 198637 group differences estimates at 3-month follow-upWait-listJohansson et al, Significant difference in VAS Significant difference in clinical Significant difference in 199143 for facial pain and headache dysfunction score with AT subjective symptoms with AT
with AT versus wait-list versus wait-list (P < .01) versus wait-list (P < .01)(P < .01)
List and Helkimo, Reduced frequency and Significant difference in clinical Improved pressure pain Improved activity of daily 199240 intensity of pain (VAS) with dysfunction score with AT threshold with AT versus living with AT versus wait-listList et al,199341 AT versus wait-list (P < .01) versus wait-list wait-list (P < .05) (P < .01)List et al,199238
List and Helkimo,199239
Park et al, 199921 Reduced pain numerical Improved pressure pain Reduced noise frequencyanalogue scale with EAT threshold in EAT from (P = .016), limitation of versus wait-list (P = .001) baseline; no between-group motion (P = .004) with EAT
differences on ST7 point versus wait-list; reduced(P = .055) maximum comfortable
opening and active range of motion (P < .001) from baseline within EAT
Table 4 Outcomes of RCTs of Acupuncture Versus Physical Therapy or Pharmacologic Treatments
Control/interventions Study Overall improvement
Physical therapyUltrasound therapy Luo et al, 200120 Difference in responder rate with improvement of TMD symptoms between
AT versus control (95.1% versus 75% )Zhong et al, 200727 Significant difference in responder rate with improvement of TMD symptoms
between warm-AT versus control (91.7% versus 66.7%, P < .05)Manual therapy Zhu, 200728 Significant difference in responder rate with improvement of TMD symptoms
between EAT versus control (95.5% versus 82.5%, P < .05)Massage Wang, 199625 Significant difference in responder rate with improvement of TMD symptoms
between EAT versus control (97.8% versus 91.2%, P < .001)Pharmacological treatmentsIndomethacin and Vitamin B1 Li and Rong, 200319 Significant difference in responder rate with improvement of TMD symptoms
between EAT plus AI versus control (95.0% versus 80.0%, P < .05)Zhang, 200826 Significant difference in responder rate with improvement of TMD symptoms
between warm-AT versus control (95.7% versus 68.2%, P < .05)
Acupuncture Versus Pharmacologic Treatment.Two studies found significant differences in respon-der rate with improvement of TMD symptoms infavor of warm-needle acupuncture, or elec-troacupuncture plus acupoint injections comparedwith indomethacin plus vitamin B1 (Table 4).19,26
This is moderate evidence (two trials, 85 partici-pants) that acupuncture is more effective for TMDsymptoms than pharmacologic treatment such asindomethacin plus vitamin B1.
Adverse Events
Adverse events were addressed only in two studies.One trial40 reported minimal adverse events, whilethe other trial22 reported no serious adverse eventsor complications in the classical acupuncturegroup. In one of the trials,40 classical acupunctureplus electroacupuncture was associated with anenhanced relaxed feeling, improved sleep, tem-porarily increased pain, and eleven cases of minorhematoma.
Discussion
This review has revealed moderate evidence for theefficacy of acupuncture in management of TMD.The results support the view that the specificeffects of acupuncture are short-term but haveimportant clinical treatment benefits. Given theplethora of treatments for TMD, it is important tocontextualize the results of the current review withrespect to current guidelines. The effects ofacupuncture are equivalent to the results for treat-ments that are currently advocated (eg, occlusalsplints, physical therapy such as manual therapy).
There was moderate evidence that acupuncturehas a positive influence beyond those of placebo,which updates the previous evidence based on onesham acupuncture RCT that showed no differencebetween classical acupuncture and sham acupunc-ture.44 Important aspects of the acupuncture andsham acupuncture control need to be considered tointerpret properly this finding. This lack of differ-ence between sham and real acupuncture raises con-sideration of how appropriate controls can bechosen. First, although a superficial insertionmethod was used in the control group, this is not aninert placebo method because it elicits peripheralsensory stimulation. It does not seem to be possibleto insert needles without any sensorial stimula-tion.46,47 In fact, even a very gentle form of placeboreferred to as minimal or microacupuncture, wherethe needle is superficially inserted and left for a very
short time with no further stimulation, seems toexert an effect.48 Second, although nonacupointswere used in the control group, these may also notbe inert. A current ongoing discussion in theacupuncture field is whether the acupuncture pointsshould be referred to as acupuncture areas or zones,as needling in the areas around the traditionalacupuncture points may also be efficient.48 Indeed,some evidence suggests that sham acupuncture maynot be inert.49,50 Finally, it appears that the effectinduced from the superficial insertion is strongenough to induce significant reduction in symptomsof TMD in the responders, negating any differencebetween the two methods.44 Whether the resultswere due to the effectiveness of the noninert shamneedling or a placebo effect could not be deter-mined. Until recently, attempts at providing a con-trol group for acupuncture, reflecting a similartherapeutic setting as well as being neurophysiologi-cally inert, have been unsuccessful. Recent studieshave provided evidence of a method which enabledadequate blinding of the participant while maintain-ing an identical therapeutic setting to that of realacupuncture.51–53 Indeed, in this review, the twostudies favoring real classical acupuncture overplacebo acupuncture used a nonpenetratingneedling as the control.23, 24 In the present study’squalitative synthesis, these two latter comparisonswere separated out to show moderate evidence thatacupuncture alone is a positive influence fromplacebo acupuncture (based on the addition of twonew trials), whereas the findings for acupuncture/sham acupuncture provide conflicting evidence.
This review also has provided moderate evidencethat there is a significant difference betweenacupuncture and indomethacin plus vitamin B1,and physical therapy. For other comparisons, theaddition of the one RCT either strengthened orconfirmed the previous conclusions by providingmoderate evidence favoring acupuncture over wait-list. However, the review is unable to make manyfirm statements about the strength of the evidence,since the RCTs had poor quality and used varioustypes of acupuncture. The included articles sufferedfrom methodological flaws including incompletereporting of randomization procedure, the lack ofblinding, and follow-up results. Variations in out-come variables made comparison across trialsproblematic.
This systematic review has several limitations.The trials satisfying the inclusion criteria were clin-ically as well as methodologically heterogeneouswith respect to the TMD diagnosis, type of TMD,different acupoints, type of acupuncture variants,and variants of outcome measures for evaluating
treatments of TMD. There was a limitation of gen-eralization in the control interventions.Pharmacotherapy for TMD includes nonsteroidalanti-inflammatory agents, tricyclic antidepressants,and selective serotonin-reuptake inhibitors.However, two trials included in this review usedonly indomethacin as the control intervention. Useof the qualitative method for synthesizing the evi-dence has its limitations. Consistency of evidencewas determined by group consensus, but thismethod is sensitive to how studies are categorized,since meeting the criterion of a certain level of evi-dence depends on the number of studies present ina category. Many of the reviewed studies were oflow quality and had methodological shortcomingssuch as an inadequate level of blinding and powercalculations. Although blinding of the therapistwho applies acupuncture would be difficult, blind-ing of patients and other care providers as well asoutcome assessors should be attempted to mini-mize the performance and assessment bias of trials.It should also be emphasized that trials withacupuncture should be randomized, blinded(including assessor blinding), well-controlled forplacebo effects, have adequately concealed alloca-tions, and utilize an appropriate level of powerthrough sample size determination.
In conclusion, the results of this systematicreview provide moderate evidence that acupunc-ture is an effective intervention to reduce symp-toms associated with TMD. There is a need foracupuncture trials with adequate sample sizes thataddress the long-term efficacy or effectiveness ofacupuncture.
References
1. Scrivani SJ, Keith DA, Kaban LB. Temporomandibulardisorders [review]. N Engl J Med 2008;359:2693–2705.
2. Dworkin S. Personal and Societal Impact of OrofacialPain. New York: Raven Press, 1995.
3. Benoit P. History and physical examination for TMD. In:Kraus SL (ed). Temporomandibular Disorders, ed 2. NewYork: Churchill Livingstone, 1994.
4. De Wijer A, De Leeuw JRJ, Steenks MH, Bosman F.Temporomandibular and cervical spine disorders: Self-reported signs and symptoms. Spine 1996;21:1638–1646.
5. Gremillion HA. The prevalence and etiology of temporo-mandibular disorders and orofacial pain. Tex Dent J 2000;117:30–39.
6. Morris S, Benjamin S, Gray R, Bennett D. Physical, psy-chiatric and social characteristics of the temporomandibu-lar disorder pain dysfunction syndrome: The relationshipof mental disorders to presentation. Br Dent J 1997;182:255–260.
7. Rugh JD, Solberg WK. Oral health status in the UnitedStates: Temporomandibular disorders. J Dent Educ 1985;49:398–406.
8. Schiffman EL. Epidemiology of TMJ and craniofacialpain. In: Fricton JR, Kroening RJ, Hathaway KM, (eds).TMJ and Craniofacial Pain: Diagnosis and Management.St Louis: IEA Publications, 1998.
9. Koh H, Robinson P. Occlusal adjustment for treating andpreventing temporomandibular joint disorders. J OralRehabil 2004;31:287–292.
10. Goudot P, Jaquinet AR, Hugonnet S, Haefliger W, RichterM. Improvement of pain and function after arthroscopyand arthrocentesis of the temporomandibular joint: A com-parative study. J Craniomaxillofac Surg 2000;28:39–43.
11. Yuasa H, Kurita K. Treatment group on temporomandibu-lar disorders. Randomized clinical trial of primary treat-ment for temporomandibular joint disk displacementwithout reduction and without osseous changes: A combi-nation of NSAIDs and mouth-opening exercise versus notreatment. Oral Surg Oral Med Oral Pathol Oral RadiolEndod 2001;91:671–675.
12. Crider AB, Glaros AG. A meta-analysis of EMG biofeed-back treatment of temporomandibular disorders. J OrofacPain 1999;13:29–37.
13. Komiyama O, Kawara M, Arai M, Asano T, KobayashiK. Posture correction as part of behavioural therapy intreatment of myofascial pain with limited opening. J OralRehabil 1999;26:428–435.
14. Bausell RB, Lee WL, Berman BM. Demographic andhealth-related correlates of visits to complementary andalternative medical providers. Med Care 2001;39:190–196.
15. Eisenberg DM, Davis RB, Ettner SL, et al. Trends in alter-native medicine use in the United States, 1990-1997:Results of a follow-up national survey. J Am Med Assoc1998;280:1569–1575.
16. Palinkas LA, Kabongo ML. The use of complementaryand alternative medicine by primary care patients. J FamPract 2000;49:1121–1130.
17. DeBar LL, Vuckovic N, Schneider J, Ritenbaugh C. Use ofcomplementary and alternative medicine for temporo-mandibular disorders. J Orofac Pain 2003;17:224–236.
18. Fink M, Rosted P, Bernateck M, Stiesch-Scholz M, KarstM. Acupuncture in the treatment of painful dysfunction ofthe temporomandibular joint: A review of the literature.Forsch Komplement Med 2006;13:109–115.
19. Li W, Rong J. Electroacupuncture and acupoint injectionon temporomandibular disorder. Xinjiang J Trad ChineseMed 2003;21:40–41.
20. Luo L, Liu K, Ou Y. Clinical observation on treatment oftemporomandibular disorder by acupuncture. Med J NatDefend Forces Southwest China 2001;11:274.
21. Park T, Park J, Ko M. A clinical effect of electro-acupunc-ture stimulation therapy on TMD patients. J KoreanAcademy Craniomandib Disord 1999;11:103–111.
22. Schmid-Schwap M, Simma-Kletschka I, Stockner A, et al.Oral acupuncture in the therapy of craniomandibular dys-function syndrome: A randomized controlled trial. WienKlin Wochenschr 2006;118:36–42.
23. Shen YF, Goddard G. The short-term effects of acupunc-ture on myofascial pain patients after clenching. PainPractice 2007;7:256–264.
24. Smith P, Mosscrop D, Davies S, Sloan P, Al-Ani Z. Theefficacy of acupuncture in the treatment of temporo-mandibular joint myofascial pain: A randomised con-trolled trial. J Dent 2007;35:259–267.
25. Wang H. Acupuncture for 46 cases with temporo-mandibular disorder. Chinese Naturopathy 1996;6:21.
26. Zhang H. Clinical observations on treatment of temporo-mandibular joint disorder by warm needling. Shanghai JAcupunct Moxib 2008;27:26–27.
27. Zhong D, Huang Y, Zhou J. Warm acupuncture on 60cases with temporomandibular disorder. J Gansu CollegeTrad Chinese Med 2007;24:26–27.
28. Zhu X. Electro acupuncture on 90 cases with temporo-mandibular joint disorder. Fujian J Trad Chinese Med2007;38:13.
29. Higgins JPT, Green S. Cochrane Handbook for SystematicReviews of Interventions, Version 5.0.1. Available fromwww.cochrane-handbook.org: The Cochrane Collabo -ration; 2008.
30. Slavin RE. Best evidence synthesis: An intelligent alterna-tive to meta-analysis. J Clin Epidemiol 1995;48:9–18.
31. Van Tulder MW, Gherkin DC, Berman B, Lao L, KoesBW. The effectiveness of acupuncture in the managementof acute and chronic low back pain: A systematic reviewwithin the framework of the Cochrane collaboration backreview group. Spine 1999;24:1113–1123.
32. Mazzetto MO, Carrasco TG, Bidinelo EF, De AndradePizzo RC, Mazzetto RG. Low intensity laser application intemporomandibular disorders: A phase I double-blindstudy. Cranio J Craniomandibular Prac 2007;25:186–192.
33. Zeng L, Xiao Y. Laser acupuncture on 80 cases with tem-poromandibular joint dysfunction syndrome. Shanghai JAupunct Moxib 1996;15:194–195.
34. Yang T, Zhou Q. Clinical observation on treatment oftemporomandibular disorders by electroacupuncture andtuina therapy. J Trad Chinese Med Chinese MateriaMedica Jilin 2004;24:43–44.
35. Raustia AM, Pohjola RT, Virtanen KK. Acupuncturecompared with stomatognathic treatment for TMJ dys-function. Part I: A randomized study. J Prosthet Dent1985;54:581–585.
36. Raustia AM, Pohjola RT, Virtanen KK. Acupuncturecompared with stomatognathic treatment for TMJ dys-function. Part II: Components of the dysfunction index. J Prosthet Dent 1986;55:372–376.
37. Raustia AM, Pohjola RT. Acupuncture compared withstomatognathic treatment for TMJ dysfunction. Part III:Effect of treatment on mobility. J Prosthet Dent 1986;56:616–623.
38. List T, Helkimo M, Andersson S, Carlsson GE.Acupuncture and occlusal splint therapy in the treatmentof craniomandibular disorders. Part I. A comparativestudy. Swed Dent J 1992;16:125–141.
39. List T, Helkimo M. Acupuncture and occlusal splint ther-apy in the treatment of craniomandibular disorders. II. A1-year follow-up study. Acta Odontol Scand 1992;50:375–385.
40. List T, Helkimo M. Adverse events of acupuncture andocclusal splint therapy in the treatment of craniomandibu-lar disorders. Cranio 1992;10:318–324.
41. List T, Helkimo M, Karlsson R. Pressure pain thresholds inpatients with craniomandibular disorders before and aftertreatment with acupuncture and occlusal splint therapy: Acontrolled clinical study. J Orofac Pain 1993;7:275–282.
42. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D,Stroup DF. Improving the quality of reports of meta-anal-yses of randomised controlled trials: The QUOROMstatement. The Lancet 1999;354:1896–1900.
43. Johansson A, Wenneberg B, Wagersten C, Haraldson T.Acupuncture in treatment of facial muscular pain. ActaOdontol Scand 1991;49:153–158.
45. Dworkin SF, LeResche L. Research diagnostic criteria fortemporomandibular disorders: Review, criteria, examina-tions and specifications, critique. J Craniomandib Disord1992;6:301–355.
46. Lewith GT, Vincent C. On the evaluation of the clinicaleffects of acupuncture: A problem reassessed and a frame-work for future research. J Altern Complement Med1996;2:79–90.
47. Vincent CA, Richardson PH. The evaluation of therapeuticacupuncture: Concepts and methods. Pain 1986;24:1–13.
48. Mann F. A new system of acupuncture. In: Filshie J, WhiteA (eds). Medical Acupuncture: A Western ScientificApproach. London: Churchill Livingstone, 1998.
49. Birch S, Hesselink JK, Jonkman FAM, Hekker TAM, BosA. Clinical research on acupuncture. Part 1: What havereviews of the efficacy and safety of acupuncture told usso far? J Altern Complement Med 2004;10:468–480.
50. Lund I, Lundeberg T. Are minimal, superficial or shamacupuncture procedures acceptable as inert placebo con-trols? Acupunct Med 2006;24:13–15.
51. Streitberger K, Kleinhenz J. Introducing a placebo needleinto acupuncture research. Lancet 1998;352:364–365.
52. Park J, White A, Lee H, Ernst E. Development of a newsham needle. Acupunct Med 1999;17:110–112.
53. Fink M, Gutenbrunner C, Rollnik J, Karst M. Credibilityof a newly designed placebo needle for clinical trials inacupuncture research. Forsch Komplementarmed KlassNaturheilkd 2001;8:368–372.
![Revisão Pirazol[1]](https://static.documents.pub/doc/80x56/563db82a550346aa9a91217b/revisao-pirazol1.jpg)
