The Efficacy of Traditional, Low Cost and Nonsplint Therapies for TMD a RCT

2006;137;1099-1107 J Am Dent Assoc

Lloyd Mancl and Samuel F. Dworkin Edmond Truelove, Kimberly Hanson Huggins,

disorder: A randomized controlled trialnonsplint therapies for temporomandibular The efficacy of traditional, low-cost and

jada.ada.org ( this information is current as of September 23, 2010 ):The following resources related to this article are available online at

http://jada.ada.org/cgi/content/full/137/8/1099found in the online version of this article at:

including high-resolution figures, can beUpdated information and services

http://jada.ada.org/cgi/collection/endodonticsEndodontics : subject collectionsThis article appears in the following

http://www.ada.org/prof/resources/pubs/jada/permissions.aspreproduce this article in whole or in part can be found at:

of this article or about permission toreprintsInformation about obtaining

© 2010 American Dental Association. The sponsor and its products are not endorsed by the ADA.

on September 23, 2010

jada.ada.orgD

ownloaded from

http://jada.ada.org/cgi/content/full/137/8/1099

http://www.ada.org/prof/resources/pubs/jada/permissions.asp

http://jada.ada.org

Background. Treatment recommendations for patientswith painful temporomandibular disorders (TMDs) rangefrom conservative treatments such as physiotherapy toaggressive and irreversible treatments such as restorativereconstruction and joint surgery.Methods. The authors randomized 200 subjects diagnosed withTMD into three groups: usual conservative, dentist-prescribed self-care treatmentwithout any intraoral splint appliance (UT); UT plus a conventional flat-plane hardacrylic splint (HS); and UT plus a soft vinyl (a low-cost athletic mouth guard) splint(SS). Subjects completed questionnaires and clinical examinations at three, six and12 months.Results. The authors observed no significant differences among the groups inTMD-related pain levels or other common signs and symptoms of TMD at baseline(BL) or at any follow-up. The changes from BL were comparable for all threegroups. The authors did not note any significant differences at any follow-up forcompliance with study protocols or for occurrences of adverse effects from eithersplint type. For HS versus SS, there were significant differences in rates of splintuse, but these differences were not accompanied by differences in either self-reported symptoms or in clinical findings.Conclusions. All patients improved over time, and traditional splint therapyoffered no benefit over the SS splint therapy. Neither splint therapy provided agreater benefit than did self-care treatment without splint therapy.Clinical Implications. These findings suggest that clinicians who treat patientswith TMD should consider prescribing low-cost nonsplint self-care therapy for mostpatients.Key Words. Temporomandibular disorders; myofascial pain; occlusal splint.JADA 2006;137(8):1099-107.

Temporomandibulardisorders (TMD) areassociated withcomplaints of bothacute and chronic

orofacial pain.1,2 Current con-cepts regarding the causes ofTMD-related pain includeexcessive parafunctional jawactivities, behavioral stressors,and direct or indirect traumato the jaws.3,4 Treatment rec-ommendations vary and rangefrom rest and physiotherapy tomore aggressive and irre-versible treatments, includingorthodontics, restorative careand joint surgery.5-7 Situated inbetween are approaches char-acterized as reversible andconservative, involving the useof splints, physical therapy,self-care strategies and behav-ioral therapies.8-10

Self-care approaches includethe use of heat and cold packs,jaw exercises, guidance inreduction of parafunctionaljaw activities and progressivemuscle relaxation. Outcomesfrom these interventions havedemonstrated reductions inself-reported pain, pain on pal-pation, and improvements inrange of motion and jaw function.11-13

Splint therapy, either aloneor in combination with other

A B S T R A C T

ARTICLE

2

Dr. Truelove is a professor and the chair, Department of Oral Medicine, School of Dentistry, Box 356370, Uni-versity of Washington, Seattle, Wash. 98195, e-mail “[email protected]”. Address reprint requeststo Dr. Truelove.Ms. Huggins is a research manager, Department of Oral Medicine, School of Dentistry, University of Wash-ington, Seattle.Dr. Mancl is a research associate professor, Department of Dental Public Health Sciences, School of Dentistry,University of Washington, Seattle.Dr. Dworkin is a professor emeritus, Department of Oral Medicine, School of Dentistry, and Department of Psy-chiatry and Behavioral Sciences, School of Medicine, University of Washington, Seattle.

JA D

A

CO

NT

I

NU

I N G E D UC

AT

IO

N

✷✷

The efficacy of traditional, low-cost andnonsplint therapies for temporomandibulardisorderA randomized controlled trialEdmond Truelove, DDS, MSD; Kimberly Hanson Huggins, RDH, BS; Lloyd Mancl, PhD; Samuel F. Dworkin, DDS, PhD

JADA, Vol. 137 http://jada.ada.org August 2006 1099

C O V E R S T O R Y

Copyright ©2006 American Dental Association. All rights reserved.


jada.ada.orgD

ownloaded from

http://jada.ada.org

treatments, is the most common form of TMDtreatment used, even though data comparingsplint and nonsplint therapy are not widely avail-able.14,15 Most studies of splint therapy have beenlimited by small sample size, lack of intent-to-treat analysis, use of short duration outcome(three months or less), inadequate control groupsand failure to compare splint therapy with otherforms of treatment (for example, customary andusual TMD care).16-19

Generally, studies focusing on splint therapyhave shown a reduction in orofacial pain andother symptoms associated with TMD.20-22 Thetypes of splints studied, including choice ofmaterials (for example, acrylic versus vinyl), have varied. It is difficult for dentists to makeevidence-based decisions regarding splint therapybecause few randomized controlled trials (RCT)comparing different splint designs have been con-ducted.17,23,24 To complicate the matter further,several studies have suggested that the benefit ofsplint therapy may be through a placebo effect.11,25

The cost of splint therapy varies widely,depending on specific splint designs and fabrica-tion methods (for example, jaw repositioningversus flat-plane designs; soft versus hard splint[HS] materials). The cost of splint therapy com-monly ranges from several hundred to severalthousand dollars.14,26 The pattern emerging fromthe few well-conducted studies suggests that themechanism of action of splints is unclear and pos-sibly not dependent on specificity of design.

We conducted an RCT designed to compareself-care with higher-cost, heat-processed HSs;self-care with low-cost soft vinyl splints (SS); andself-care strategies alone. The primary objectiveof the study was to determine whether addingtwo types of splint therapy to self-care offered anyadvantage for reducing TMD-related pain andself-reported TMD symptoms over self-carestrategies alone that included jaw relaxation,reduction of oral habits, heat packs, nonsteroidalanti-inflammatory drugs (NSAIDs) and jaw exercises.

SUBJECTS, MATERIALS AND METHODS

Subjects. We screened consecutive patients aged18 to 60 years seen at the Orofacial Pain Clinic,Department of Oral Medicine, School of Den-tistry, University of Washington, Seattle. All ofthe eligible subjects completed a survey thatincluded the Research Diagnostic Criteria forTemporomandibular Disorders (RDC/TMD)27

questionnaire and additional questions aboutsymptoms, treatment, and medical and behav-ioral status. The attending TMD clinical special-ists (including E.T.) performed standardizedRDC/TMD Axis I clinical examinations on thesubjects at the initial visit; study personnel con-ducted these examinations at the research follow-up visits.

Inclusion criteria. We included subjects witha RDC/TMD Axis I diagnosis of myofascial pain(Group Ia or Ib) with or without a concurrentdiagnosis of arthralgia (Group IIIa) or disk dis-placement with reduction (Group IIa), as well asan RDC/TMD Axis II Graded Chronic Pain scoreof Grade I (low pain) or Grade II (high pain), bothof which had no or minimal pain-related psy-chosocial interference.28

Exclusion criteria. We excluded from thestudy subjects who had any other RDC/TMD AxisI diagnosis (for example, arthritis, disk displace-ment without reduction), any systemic arthritisor other serious medical complications, full den-tures, major psychological disorders or theinability to communicate in English. We did notexclude patients who had a history of splint use,but we did exclude patients who used their cur-rent splints satisfactorily.

Study subjects provided written informed con-sent in accordance with human subjects require-ments of the University of Washington, Seattle.We randomized the subjects into one of threestudy groups (Figure 1). We generated random-ization assignments using randomly selectedblock sizes of six, nine or 12 and stratified themby provider. We concealed randomization to allstudy personnel until after we obtained the sub-jects’ consent. The research dental hygienists con-ducting follow-up data collection were blinded tosubject treatment group. The three groups wereas follows:dusual treatment (UT), n = 64: dentist-prescribed, conservative and reversible self-carestrategies that required the dentist to follow astandardized treatment checklist that identifiesall treatment recommendations (jaw relaxation,reduction of parafunction, thermal packs,NSAIDs, passive opening stretches and sugges-tions about stress reduction); we prescribed self-care strategies to all subjects, and we discouragedtreatments such as narcotic analgesics, antide-pressant medications and use of a nonstudy pre-scribed splint;dHS, n = 68: UT plus fabrication of a hard

1100 JADA, Vol. 137 http://jada.ada.org August 2006

C O V E R S T O R Y



jada.ada.orgD

ownloaded from

http://jada.ada.org

acrylic heat- (dental laboratory–) processed flat-plane maxillary splint, fitted by the treating dentist;dSS, n = 68: UT plus fabrication at chairside of asoft thermoplastic vinyl athletic mouthguardsplint (Figure 2) with the dentist supervising anddirecting the patient in splint fabrication.

Treatment protocols. At the baseline (BL)visit, we assessed all subjects, provided their firstusual visit TMD clinic care (UT), enrolled them inthe study and randomly assigned them to one ofthe three groups. For subjects in the HS group,we took impressions for splint fabrication. Forsubjects in the SS group, we took a bite registra-tion using dental wax to provide an oral pro-cedure of comparable duration. We instructed allof the subjects to return to the clinic in four weeksfor follow-up and to call the study coordinator ifany problems arose.

At the second visit, subjects in the HS groupreceived the splint, which was a standard full-arch flat-plane maxillary acrylic splint adjustedto centric occlusion. Subjects in the SS group par-ticipated with the dentist in fabricating theirsplints according to the same criteria as for theHSs. We instructed subjects in both groups towear their splints at night and two hours eachday while awake throughout the three-monthfollow-up and 12-month follow-up, if possible ornecessary. We also asked subjects to bring theirsplints to clinic visits and to discontinue usingthem if any problems developed. Subjects ran-domized to the two splint groups did not incurany splint-related costs or other out-of-pockettreatment costs beyond what subjects receivingUT alone incurred.

Follow-up data collection. We conductedfollow-up visits at three, six and 12 months fromthe BL visit. These visits included having thesubjects complete self-report questionnaires andreceive an RDC/TMD clinical examination con-ducted by calibrated research dental hygienistsblinded to study group assignment.

Study hypotheses and data analysis. Weformed two hypotheses. The first was that atthree months, the subjects in the HS and SSgroups would demonstrate equivalent levels ofshort-term improvement in self-reported pain andin clinical measures such as range of motion, pal-pation pain and comparable use patterns of theappliance. The subjects in the UT group wouldshow less short-term improvement than wouldthose in the HS and SS groups.

The second hypothesis was that at the six- and12-month follow-up visits, subjects in the HS, SSand UT groups would show equivalent improve-ment in TMD-related pain and other clinical andself-report measures.

Statistical power. We based a priori samplesize calculations for this trial on data from pre-vious studies of patients at the Orofacial PainClinic, School of Dentistry, University of Wash-ington, Seattle. Using the sample size calcula-


C O V E R S T O R Y

Initial VisitResearch Diagnostic Criteria for Temporomandibular

Disorders Axis I and II Examination and HistoryEligible for Randomization: n = 262

Randomized: n = 200 (76%)

Usual Treatment+ Hard Split

n = 68

Usual Treatment+ Soft Split

n = 68

Usual TreatmentOnly

n = 64

n = 54(79%)

n = 61(90%)

n = 65(96%)

n = 56(82%)

n = 58(85%)

n = 55(81%)

n = 54(84%)

n = 57(89%)

n = 48(75%)

Baseline

Three Months

Six Months

12 Months

Figure 1. Study design.

Figure 2. Soft splint. Image of Form Fit Regular mouthguardreproduced with permission of SafeTGard, Golden, Colo.



jada.ada.orgD

ownloaded from

http://jada.ada.org

tions, we determined that 68 subjects per groupwould provide at least 80 percent power to detecta difference of 1.4 (or about a 35 percent differ-ence) in characteristic pain intensity (CPI) levelsbetween any two groups based on a one-wayanalysis of variance (ANOVA) at a .05 signifi-cance level, assuming a common standard devia-tion (SD) of 2.5.

Measures and data analyses. We included inour intent-to-treat analyses all randomized sub-jects for whom follow-up data were available,whether or not they actually received treatment.The primary outcome measure was CPI (themean of present, average and worst TMD-relatedpain in the past two months).29 Additional out-come measures included self-reported clenching,bruxing or both; limitations in jaw use; changesin clinical examination findings and diagnosis;pain duration; and compliance with treatmentrecommendations.

We used one-way ANOVA to compare the meanscores for the treatment groups at follow-up forcontinuous data measures, while we used χ2

analysis to compare treatment proportions forcategorical measures. We also used ANOVA totest for group differences based on change scores,and we conducted analysis of covariance(ANCOVA) to test for group differences at 12months, adjusting for BL values. We noted that

the findings using change scores and ANCOVAwere similar to the ANOVA results; therefore, wepresent only the ANOVA results.

We collected data at three, six and 12 months.We reported short-term (three-month) and long-term (12-month) follow-up data because they are most relevant to study hypotheses, andbecause we typically found six-month data to be intermediate or equivalent to 12-month data.

RESULTS

Demographics. Table 1 shows the demographicand BL data. The average (± SD) age was 36 (± 12) years, and 86 percent of the subjects werefemale (data not shown). Through our analyses,we did not identify significant differences amongthe three treatment groups, except that therewere fewer subjects in the SS group than in theHS and UT groups with mean incomes higherthan $50,000.

Nonstudy participation and studydropouts. Of the 262 eligible patients, 200 (76.3percent) agreed to enroll in our RCT. Thirty-seven(14 percent) of the 262 patients were ineligiblebecause they already were using splints satisfactorily.

There were no statistically significant differ-ences between study subjects and nonparticipantson demographic variables, ratings of facial pain or


C O V E R S T O R Y

TABLE 1

Baseline demographics.

DEMOGRAPHIC HARD SPLINT SOFT SPLINT USUAL TREATMENT P

* SD: Standard deviation.† IQR: Interquartile range = 25th to 75th percentiles.

Age (Mean ± SD*)

Female (%)

Education More ThanHigh School (%)

Income $50,000 orGreater (%)

Race Nonwhite (%)

No. of Years With FacialPain (Mean ± SD)

No. of Prior Facial PainVisits (Median [IQR†])

Prior Splint Therapy forPain (%)

n = 68

36 ± 11

87

83

40

8

6 ± 9

2 (0-7)

28

n = 68

35 ± 12

90

71

25

13

5 ± 6

3 (1-10)

26

n = 64

36 ± 11

81

71

39

4

5 ± 5

3 (1-5)

29

.63

.41

.20

.04

.27

.69

.45

.94



jada.ada.orgD

ownloaded from

http://jada.ada.org

self-reported jaw-related symptoms. Nonpartici-pants reported more prior health care visits forfacial pain (median = 4.0) than did study subjects(median = 2.5; P = .039, Wilcoxon rank sum test).Follow-up rates were similar among groups atthree and six months, but follow-up rates differedamong groups at 12 months (P = .004, χ2 test)(Figure 1). However, we observed no significantdifferences in the primary outcome measures orthe demographic or clinical measures at BL,except that those who completed the studyreported more prior health care visits for facialpain (median = 3.0) than did study dropouts(median = 1.0; P = .059), and those who com-pleted the study reported more problems withjoint clicking (90 percent) than did studydropouts (78 percent; P = .062, χ2 test).

Self-report findings. CPI. At BL, the averageCPI was comparable across groups (HS = 5.2, SS = 6.0, UT = 5.4; P = .09). The average scoredecreased significantly from a mean (± SD) of 5.5(± 1.9) at BL to 3.1 (± 2.3) at 12-month follow-up(P < .0001). However, all of the groups showedcomparable decreases in pain across the study (P > .40) (Figure 3).

Pain duration. All of the groups showedimprovement over time for hours per day anddays per month that the subjects were in pain.There were no differences among groups for painduration at any follow-up (< .5 days versus ≥ .5the days, P = .12) or for numbers of hours per daythe subjects were in pain at BL or any follow-up(P > .05).

Self-reported TMD symptoms. There were nosignificant differences among groups at BL or atany follow-up in rates of any TMD symptomsreported (none/sometimes versus often/always)(Table 2). At the 12-month follow-up exami-nation, subjects in the UT group reported thelowest rates of nocturnal clenching (P = .076),and subjects in the HS group reported the lowestrates of tinnitus (P = .058).

Clinical examination findings. Range ofmotion. There were no significant differencesamong groups at BL or at the three- and 12-month follow-up examinations for measurementsof vertical jaw opening (P > .2) (Figure 4). We alsofound that measurements of lateral and protru-sive jaw excursions were not statistically dif-ferent across groups at any time (P > .2) (data notshown).

Joint sounds. We detected joint clicking onopening, closing or both in about one-half of all

subjects (Table 3, page 1105) and found that it didnot vary significantly among groups or over time(P ≥ .2). Also, we detected no differences amonggroups in joint sounds during excursive move-ments either at BL or at the three- and 12-monthfollow-ups (P > .2) (data not shown).

Muscle and TMJ palpation pain. We observedno significant differences across groups in pain onclinician palpation of 16 extraoral masticatorymuscle sites, four intraoral muscle sites or fourTMJ sites at any time (P > .2) (Table 3). Thenumber of painful palpation sites for extraoralmuscles and the TMJ sites decreased from BLcomparably for all groups, whereas the number ofpainful palpation sites for intraoral muscles didnot change.

RDC/TMD diagnoses. Our study inclusion cri-teria required a Group I diagnosis of myofascialpain. There was a small difference in the distribu-tion of subdiagnoses at BL (Table 4, page 1105),with the HS group having a higher rate of GroupIb diagnoses of myofascial pain with limitedopening. At the three- and 12-month follow-ups,fewer subjects had a Group I diagnosis, with nodifferences among groups (P > .3). A Group IIadiagnosis (disk displacement with reduction)occurred in about one-third of each subject groupat BL and decreased only modestly at 12 months,with no significant differences across groups (P = .47). At BL, 41 to 44 percent of subjects werediagnosed with arthralgia, but this percentage


C O V E R S T O R Y

10

9

8

7

6

5

4

3

2

1

0Baseline Three Months 12 Months

ASSESSMENT

CH

AR

ACT

ER

IST

ICPA

IN IN

TEN

SIT

Y

HS

SS

UT

+ 2.2*

+ 2.6*

Figure 3. Characteristic pain intensity. HS: Hard splint. SS: Soft splint. UT: Usual treatment. * All other standard deviations = 1.9 to 2.1.



jada.ada.orgD

ownloaded from

http://jada.ada.org

was lower at follow-up visits, except for those inthe UT group at 12 months. Few subjects (one ortwo per diagnosis) received a diagnosis of diskdisplacement without reduction (with or withoutlimitation), osteoarthritis or osteoarthrosis at anyfollow-up.

Compliance with splint use. After the initial

three months of use, self-reportedcompliance of HS or SS use wassimilar: 89 percent for HS and 75percent for SS (P = .059). At 12months, the subjects in the SSgroup reported lower splint use(30 percent) than the HS group(72 percent; P < .001). Three sub-jects in the HS group, six in theSS group and 10 in the UT groupdid not return to receive theirstudy splint or to complete asecond UT visit. No subjectsreported an adverse effect withany of the treatments. Regard-less, 10 subjects in the UT group(seven within 12 weeks of BL) andsix subjects in the SS group (two

within 12 weeks of BL) received a nonstudy HSduring the 12 months of data collection. Analysesexcluding these subjects or including them in theHS group improved the outcomes for the UT andSS groups but still did not result in differencesamong groups. The most common reasons forreceiving a nonstudy-assigned splint were the


C O V E R S T O R Y

TABLE 2

Self-reported TMD* symptoms present often/always.TMD SYMPTOMS ASSESSMENT TREATMENT GROUP

Hard Splint (%) Soft Splint (%) Usual Treatment (%)

P

* TMD: Temporomandibular disorder.† TMJ: Temporomandibular joint.

TMJ† Clicking/PoppingSounds

TMJ Grating Sounds

TMJ Locking/Catching

Tinnitus

Jaw Clenching-Diurnal

Jaw Clenching-Nocturnal

Limitations in Chewing

BaselineThree-month12-month







665337

231914

1925

171312

402617

435042

22411

533445

291317

1994

372929

282311

434141

1676

654846

301319

17812

222517

432817

423723

1988

.21

.11

.54

.64

.59

.78

.97

.21

.15

.019.12.058

.20

.93

.59

.98

.37.076

.66

.70

.59

▲ ▲●

●

● ●

●

●

●

■

■■

■■

■

■

▲

▲

▲▲ ▲

60

55

50

45

40

35

30Baseline Three Months 12 Months

ASSESSMENT

JAW

OP

EN

ING

GR

OU

P M

EA

NS (M

ILLIM

ET

ER

S)

Maximum Assisted Opening

Unassisted Opening Without Pain

HS

SS

UT

Figure 4. Jaw opening measurements. HS: Hard splint. SS: Soft splint. UT: Usual treatment.



jada.ada.orgD

ownloaded from

http://jada.ada.org

dentists’ beliefs that the HS was necessary todecrease pain. There were no differences amonggroups for occlusal changes, either by self-reportor by clinical evaluation using full-arch articu-lating paper.

Compliance with UT. Subjects reported compli-ance in following each recommendation of thestandardized treatment checklist in equal propor-tions across all groups. We noted a modest differ-ence among groups for decreasing clenching; 38percent of subjects in the HS group reported“always” compared with 20 percent of subjects

in the SS group and 40 percent of subjects inthe UT group (P = .049).

Assessment of bias due to studydropouts. To assess for a potential bias fromstudy dropout that may have been due to adeterioration of a subject’s TMD condition, wetook a conservative approach of carrying for-ward the last observation if the subjectdropped out before month 12. None of ouranalyses demonstrated any differences amongthe three treatment groups on the primary out-come measures at 12 months. For example, the


C O V E R S T O R Y

TABLE 3

Clinical examination findings.CLINICAL FINDING FOLLOW-UP TREATMENT GROUP

Hard Splint Soft Splint Usual Treatment

P

* TMD: Temporomandibular disorder.† SD: Standard deviation.‡ TMJ: Temporomandibular joint.

TMD* SymptomSubjects’ clicking onopening, closing or both(%)

Pain on PalpationMean (± SD†) number of extraoral muscle sites (0-16)

Mean (± SD) number ofintraoral muscles (0-4)

Mean (± SD) number ofTMJ‡ sites (0-4)





515257

7.3 (4.5)5.6 (5.4)3.6 (4.1)

3.0 (1.2)3.0 (1.4)2.6 (1.6)

1.5 (1.1)1.3 (1.5)0.9 (1.1)

425146

6.7 (3.8)4.7 (4.1)4.1 (4.4)

3.0 (1.2)2.8 (1.6)3.0 (1.4)

1.2 (1.1)1.1 (1.4)1.0 (1.0)

435064

6.1 (3.1)4.3 (4.0)4.5 (4.5)

2.7 (1.3)2.5 (1.6)2.6 (1.4)

1.5 (1.0)1.1 (1.3)0.8 (0.9)

.53

.98

.31

.22

.45

.69

.26

.37

.41

.31

.64

.74

TABLE 4

RDC/TMD* diagnoses.GROUP DIAGNOSES ASSESSMENT

Hard Splint (%) Soft Splint (%) Usual Treatment (%)

P

* RDC/TMD: Research Diagnostic Criteria for Temporomandibular Disorders.27

Ia (Myofascial Pain)

Ib (Myofascial PainWith Limited Opening)

IIa (Disk DisplacementWith Reduction)

IIIa (Arthralgia)





501932

501625

383629

412529

563239

441421

322823

432630

692148

311421

373636

442342

.08

.32

.33

.08

.94

.86

.74

.71

.47

.92

.94

.46

TREATMENT GROUP



jada.ada.orgD

ownloaded from

http://jada.ada.org

mean CPI was similar between the treatmentgroups (HS = 3.0, SS = 3.5, UT = 3.4; P = .45),as well as for mean maximum unassistedopening (measured in millimeters) withoutpain (HS = 44, SS = 43, UT = 44; P = .72) andmean maximum assisted opening (HS = 54, SS = 52, UT = 53; P = .70).

DISCUSSION

We originally hypothesized that compared withUT alone, the UT plus HS or SS groups woulddemonstrate the most benefit after three monthsand that all groups would show equivalent levelsof pain and symptom reduction by 12 months. Inwhat we believe to be the largest RCT of evalu-ating efficacy of splints and self-care, we did notdetect significant differences in any of the out-comes among the three treatment groups.

UT using self-care strategies alone was aseffective in reducing characteristic TMD pain asUT plus HS or SS. We observed the same patternof overall improvement, irrespective of treatmentgroup, for all self-report and physical examinationfindings recorded. The equivalent improvementacross the three groups is encouraging both for itsclinical implications and because of the signifi-cant differences in cost and chair time requiredfor the three treatment protocols. For example,the same amount of provider time was spent onUT, but there were additional time and costsassociated with providing splint therapies. Theover-the-counter soft vinyl splint blanks acquiredfrom athletic suppliers for less than $1 took onlyeight chairside minutes (on average) to fabricate.The HS, by contrast, cost approximately $400(including laboratory fees) and took 33 chairsideminutes (on average), including time for creatingstudy casts and for splint adjustment.

The results of the study may have beenaffected by a number of factors. While studyresearch hygienists blinded to treatment groupscollected all study measures, it was impossible toblind the clinicians treating the patients and todetermine whether they inadvertently modifiedtheir treatment approach. We recorded specific UTcomponents for each clinic visit and did not detectsignificant differences in treatments among groupsin our ongoing monitoring. Another limitation ofthe study could be the decision to enroll patientswith different types of TMD as long as they alsohad a concurrent diagnosis of myofascial pain.

Outcomes might have been different if we onlyincluded subjects with one type of TMD. It is well-

known, however, that in clinical practice a largepercentage of patients have more than one type ofTMD, most typically a combination of myofascialpain, arthralgia and disk displacement withreduction (clicking). We designed our study toreflect that clinical reality. Furthermore, itappears that randomization was effective in dis-tributing diagnostic subtypes of TMD comparablyacross groups. Additionally, we enrolled only sub-jects classified as “psychosocially functional” inour study, so we were not able to determinewhether people experiencing more psychosocialdysfunction would have had the same outcomes.

CONCLUSIONS

The data from this RCT and others12,13,25 lend sup-port to even more conservative treatment proto-cols than have been advocated traditionally, sinceself-reported pain and symptoms and muscle andjoint palpation scores decreased significantlyacross all three groups over time. Nonsplint, con-servative treatments also appear to be as effectiveas the two splint types we included in our studyin reducing self-reported parafunction. Theapparent benefit of the nonsplint UT protocol suggests that patients of limited means and thosewithout access to facilities that construct splintscan be treated successfully with therapies thatcan be applied outside of traditional dental settings.

This study adds to our understanding of TMDand its treatments because, unlike most studiesof splint therapy, we evaluated both splint andnonsplint treatments. To our knowledge, no studyof TMD treatments has addressed both splint andnonsplint treatment outcomes in as large an RCTwith long-term follow-up. ■

This study was supported by National Institute of Dental and Craniofa-cial Research grant P01 DE-08773.

The authors wish to acknowledge the strong support and cooperationoffered in the conduct of this randomized controlled trial by clinicians whotreat temporomandibular disorders in the Department of Oral Medicine,University of Washington, Seattle, including Dr. M. Drangsholt, Dr. B.Gandara, Dr. M. Martin, Dr. M. Schubert and Dr. E. Sommers. The authorsthank Melinda Lane, RDH, BS, and Dr. Linda LeResche for their com-ments on the manuscript. The authors also thank the staff members of theOrofacial Pain Research Group, especially Gayle Garson for her superbeffort on behalf of data management and statistical analysis and GregoryMueller for the outstanding administration of study logistics needed to con-duct this complex randomized controlled trial.

1. Dworkin SF, Huggins KH, LeResche L, et al. Epidemiology of signs andsymptoms in temporomandibular disorders: clinical signs in cases and con-trols. JADA 1990;120(3):273-81.

2. Report of the president’s conference on the examination, diagnosis andmanagement of temporomandibular disorders. JADA 1983;106(1):75-7.

3. Greene CS. Temporomandibular disorders: the evolution of concepts.In: Sarnat BG, Laskin DM, eds. The temporomandibular joint: A biological


C O V E R S T O R Y



jada.ada.orgD

ownloaded from

http://jada.ada.org

basis for clinical practice. Philadelphia: Saunders; 1992.4. McNeill C. Management of temporomandibular disorders: concepts

and controversies. J Prosthet Dent 1997;77:510-22.5. McNamara JA Jr. Orthodontic treatment and temporomandibular dis-

orders. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997;83(1):107-17.

6. Milam SB. Failed implants and multiple operations. Oral Surg OralMed Oral Pathol Oral Radiol Endod 1997;83(1):156-62.

7. Dolwick MF. The role of temporomandibular joint surgery in thetreatment of patients with internal derangement. Oral Surg Oral MedOral Pathol Oral Radiol Endod 1997;83(1):150-5.

8. Dionne RA. Pharmacologic treatments for temporomandibular dis-orders. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997;83(1):134-42.

9. Feine JS, Widmer CG, Lund JP. Physical therapy: a critique. OralSurg Oral Med Oral Pathol Oral Radiol Endod 1997;83(1):123-7.

10. Ekberg E, Vallon D, Nilner M. Occlusal appliance therapy inpatients with temporomandibular disorders: a double-blind controlledstudy in a short-term perspective. Acta Odontol Scand 1998;56(2):122-8.

11. Burgess JA, Sommers EE, Truelove EL, Dworkin SF. Short-termeffect of two therapeutic methods on myofascial pain and dysfunctionof the masticatory system. J Prosthet Dent 1988;60(5):606-10.

12. Dworkin SF, Turner JA, Wilson L, et al. Brief group cognitive-behavioral intervention for temporomandibular disorders. Pain1994;59(2):175-87.

13. Dworkin SF, Huggins KH, Wilson L, et al. A randomized clinicaltrial using research diagnostic criteria for temporomandibular disor-ders-axis II to target clinic cases for a tailored self-care TMD treat-ment program. J Orofac Pain 2002;16(1):48-63.

14. Von Korff MR, Howard JA, Truelove EL, Sommers E, WagnerEH, Dworkin S. Temporomandibular disorders: variation in clinicalpractice. Med Care 1988;26(3):307-14.

15. Le Resche L, Truelove EL, Dworkin SF. Temporomandibular dis-orders: a survey of dentists’ knowledge and beliefs. JADA 1993;124(5):90-106.

16. Drangsholt M, Mancl L, Le Resche L. Optimal study designs toinvestigate risk factors for chronic temporomandibular disorder (TMD)pain (abstract). Presented at: International Association for the Studyof Pain (IASP) Ninth World Congress on Pain, August 1999, Vienna, Austria.

17. Rubinoff MS, Gross A, McCall WD. Conventional and nonoc-

cluding splint therapy compared for patients with myofascial pain dys-function syndrome. Gen Dent 1987;35(6):502-6.

18. Dahlström L, Haraldson T. Bite plates and stabilization splints inmandibular dysfunction: a clinical and electromyographic comparison.Acta Odontol Scand 1985;43(2):109-14.

19. Lundh H, Westesson PL, Eriksson L, Brooks SL. Temporo-mandibular joint disk displacement without reduction: treatment withflat occlusal splint versus no treatment. Oral Surg Oral Med OralPathol 1992;73(6):655-8.

20. Forssell H, Kalso E, Koskela P, Vehmanen R, Puukka P, AlanenP. Occlusal treatments in temporomandibular disorders: a qualitativesystematic review of randomized controlled trials. Pain 1999;83(3):549-60.

21. Magnusson T, Syren M. Therapeutic jaw exercises and interoc-clusal appliance therapy: a comparison between two common treat-ments of temporomandibular disorders. Swed Dent J 1999;23(1):27-37.

22. Turk DC, Zaki HS, Rudy TE. Effects of intraoral appliance andbiofeedback/stress management alone and in combination in treatingpain and depression in patients with temporomandibular disorders. JProsthet Dent 1993;70(2):158-64.

23. Wright E, Anderson G, Schulte J. A randomized clinical trial ofintraoral soft splints and palliative treatment for masticatory musclepain. J Orofac Pain 1995;9(2):192-8.

24. Davies SJ, Gray RJM. The pattern of splint usage in the manage-ment two common temporomandibular disorders, part I: the anteriorrepositioning splint in the treatment of disc displacement with reduc-tion. Br Dent J 1997;183(6):199-203.

25. Dao TT, Lavigne GJ, Charbonneau A, Feine JS, Lund JP. Theefficacy of oral splints in the treatment of myofascial pain of the jawmuscles: a controlled clinical study. Pain 1994;56(1):85-94.

26. Truelove EL, LeResche L, Dworkin SF, et al. Dentists’ diagnosisand treatment of temporomandibular disorders (abstract 852). J DentRes 1991;70(special issue):372.

27. Dworkin SF, LeResche L. Research diagnostic criteria for tem-poromandibular disorders: review, criteria, examinations and specifica-tions, critique. J Craniomandib Disord 1992;6(4):301-55.

28. Von Korff M, Ormel J, Keefe FJ, Dworkin SF. Grading theseverity of chronic pain. Pain 1992;50(2):133-49.

29. Dworkin SF, Von Korff M, Whitney CW, et al. Measurement ofcharacteristic pain intensity in field research (abstract 552). Pain1990;(supplement 5):S290.


C O V E R S T O R Y



jada.ada.orgD

ownloaded from

http://jada.ada.org

The Efficacy of Traditional, Low Cost and Nonsplint Therapies for TMD a RCT

Documents

Transcript of The Efficacy of Traditional, Low Cost and Nonsplint Therapies for TMD a RCT