Stabilization Splint Therapy for the Treatment of Temporomandibular Myofascial Pain: A Systematic Review

Transcription

1 Evidence-Based Dentistry Stabilization Splint Therapy for the Treatment of Temporomandibular Myofascial Pain: A Systematic Review Ziad Al-Ani, B.D.S., M.Sc., Ph.D.; Robin J. Gray, B.D.S., M.D.S., Ph.D., M.F.G.D.P.; Stephen J. Davies, B.D.S., M.D.Sc., D.G.D.P.; Philip Sloan, B.D.S., Ph.D., F.R.C. Path. F.D.S.; Anne-Marie Glenny, B.Sc., M.Med.Sci. Abstract: The aim of this review is to establish the effectiveness of stabilization splint (SS) therapy in reducing symptoms in patients with myofascial pain. Searching of electronic databases, handsearching of relevant key journals, and screening of reference lists of included studies were undertaken. There was no language restriction, and unpublished research was sought. The selection criteria were randomized controlled trials comparing splint therapy to either no treatment or another active treatment. Data extraction and validity assessment were carried out independently and in duplicate. Studies were grouped according to treatment type. Twenty potentially relevant Randomized Controlled Trials (RCTs) were identified. Only twelve met the inclusion criteria. There is insufficient evidence either for or against the use of stabilization splint therapy over other active interventions for the treatment of temporomandibular myofascial pain. However, it appears that stabilization splint therapy may be beneficial for reducing pain severity at rest and on palpation and depression when compared to no treatment. The authors suggested the need for well conducted RCTs that pay attention to method of allocation, blind outcome assessment, sample size, and duration of follow-up. Various measures were adopted to assess the outcomes of treatment. Standardization of the methods used to measure outcomes of the treatment of myofascial pain should be established in future RCTs. Dr. Al-Ani is Clinical Lecturer in Restorative Dentistry, School of Dentistry; Dr. Gray is an Honorary Fellow in Dental Medicine and Surgery, School of Dentistry; Prof. Davies is a Lecturer in Dental Practice, School of Dentistry; Prof. Sloan is a Professor of Experimental Oral Pathology, School of Dentistry; and Ms. Glenny is a Lecturer in Evidence-Based Oral Health Care, Cochrane Oral Health Group, MANDEC, School of Dentistry all at the University of Manchester, UK. Direct correspondence and requests for reprints to Dr. Ziad Al-Ani, TMD Unit, School of Dentistry, University of Manchester, Higher Cambridge Street, Manchester, UK M15 6FH; or 6788 phone; fax; ziad.al-ani@manchester.ac.uk. Key words: temporomandibular joint disorders, myofascial pain, systematic review, myofacial pain, stabilization splint, pain dysfunction syndrome Submitted for publication 8/3/05; accepted 8/22/05 Myofascial pain is the most common temporomandibular disorder (TMD). There are many synonyms for this condition including facial arthromylagia, TMJ dysfunction syndrome, myofacial pain dysfunction syndrome, craniomandibular dysfunction, pain dysfunction syndrome (PDS), and myofacial pain dysfunction. 1 The etiology of myofascial pain is multifactorial. Consequently, many different therapies, some conservative and reversible, others irreversible, have been advocated for patients with myofascial pain. A number of successful treatment outcomes have been reported, including occlusal splints, physiotherapy, muscle-relaxing appliances, and pharmacological interventions. There are various types of occlusal splints (bite plates or intra-oral appliances of variable designs used in the management of TMD) described in the literature; they have different indications and functions. The stabilization splint (SS), one such type of occlusal splint, is also known as the Tanner appliance, the Fox appliance, the Michigan splint, or the centric relation appliance. The stabilization splint is a hard acrylic splint that provides a temporary and removable ideal occlusion (ideal contact between the teeth for the muscles and the temporomandibular joints). 1,2 Providing an ideal occlusion by the use of splint therapy reduces abnormal muscle activity and produces neuromuscular balance. 2 Normally, it is suggested that 1242 Journal of Dental Education Volume 69, Number 11

2 patients wear the splint only at night. The splint needs to be adjusted (rebalancing of the splint to the new position of the jaw by grinding some of its surface points, since the lower jaw will adopt a new position as a result of wearing the splint) over several visits as the masticatory muscles relax until a consistent jaw relationship is reached. The patients then should be reviewed at regular intervals. After a period of successful splint therapy (normally between two to three months), patients can be weaned off the splint. 2 The splint is constructed after taking impressions of the upper and lower dental arches, face bow registration, and recording of centric relation. A face bow is a calliper-like device used to record the spatial relationship of the upper teeth to some anatomic reference point or points and then enable transfer of this relationship to an articulator. It orients the dental cast in the same relationship to the opening axis of the articulator as the teeth relate to the jaw joints in the patient. A number of clinical studies have specifically evaluated the treatment of myofascial pain by SS therapy, and articles demonstrating clinical success have been published. 3-5 When properly adjusted, the SS delivers a good method of providing centric relation occlusion (the position of the jaw relative to the skull when the muscles are at their most relaxed and least strained position), eliminating posterior interferences (any predominant contacts between the back teeth that interfere with or hinder harmonious jaw movement), providing anterior guidance on anterior teeth (the contact between the anterior teeth without any posterior contact during jaw movements), reducing neuromuscular activity, and obtaining stable occlusal relationships with uniform tooth contacts throughout the dental arch. 1,2 The objective of this review was to determine the effectiveness of stabilization splint therapy in reducing symptoms in patients with myofascial pain, compared with any control group. The review was undertaken with the Cochrane Oral Health Group; a more detailed version of the review is available on the Cochrane Library ( Method To be included in the review, the identified studies had to meet the following four criteria: 1. Study Design. Randomized or quasi-randomized controlled trials (RCTs), in which splint therapy is compared concurrently to no treatment, physiotherapy, relaxing appliances, pharmacological interventions, other occlusal appliances, or any other active intervention. Data from conference abstracts were not included in the review. 2. Types of Participants. All patients with myofascial pain of all degrees of severity were included. A diagnosis of myofascial pain can be made if the patient exhibits more than one of the following signs and/or symptoms in any combination: 5 Pain on palpation of the temporomandibular joint. Pain on palpation of associated mandibular muscles. Limitation and/or deviation of mandibular movement. This was assessed by measuring the range of jaw movement, which is the only measurable parameter that can be objectively recorded in relation to temporomandibular disorder. Joint sounds and headache. Headache alone or joint sounds alone are not diagnostic of myofascial pain. Joint sounds can be intermittent. Studies examining temporomandibular disorders other than those of primarily muscular origin were excluded. 3. Types of Interventions. Stabilization splint (Tanner appliance, the Fox appliance, the Michigan splint, or the centric relation appliance). Any control group (no treatment, physiotherapy, pharmacological intervention, other occlusal appliances, or any other active intervention). 4. Types of Outcome Measures. The outcome measures considered were improvement in pain of the temporomandibular joint, improvement in pain of associated mandibular muscles, improvement of the range of movement to a normal range of movement, and, when relevant, disappearance of joint sounds. Depression, dysfunction scores, treatment credibility, and quality of life were also examined. Response to treatment must be assessed subjectively and clinically. Subjective assessment requires patients to have recorded their present overall state at review as being cured, improved, static, or deteriorating. Since discrepancies between subjective and objective assessments have been reported, 6 improvement in pain should also be confirmed clinically by palpating the muscles and the joints. Search Strategy for Identification of Studies For the identification of studies included or considered for this review, detailed search strategies November 2005 Journal of Dental Education 1243

3 were developed for each database searched (Cochrane Oral Health Group Trials Register; The Cochrane Central Register of Controlled Trials [CENTRAL], 2003, Issue 2; MEDLINE, 1966 to June 2001; EMBASE, 1966 to June 2001). This search was combined with the Cochrane Sensitive Search Strategy for RCTs. 7 The search strategy developed for MEDLINE was revised appropriately for each database to take into account differences in controlled vocabulary and syntax rules. The search attempted to identify all relevant studies irrespective of language. The reference lists of all relevant trials obtained were checked, along with the reference lists of relevant review articles. In addition, reference lists from prosthetic dentistry textbooks on temporomandibular disorders and splint therapy were also checked. The following journals have been identified as being important to be handsearched for this review for the period 1960 to present: Journal of Prosthetic Dentistry, Acta Odontologica Scandinavica, Journal of the American Dental Association, and Journal of Oral Rehabilitation. The results of the searches were screened independently and in duplicate by two reviewers (MZA and RG). The full articles of all studies meeting, or potentially meeting, the defined inclusion criteria were obtained for further assessment. Data Assessment and Synthesis All potentially relevant articles and reports were assessed independently by two reviewers (MZA and RG) using a previously prepared data extraction form. The two reviewers knew the names of the authors, institutions, journal of publication, and results when applying both the inclusion criteria and during data extraction. After assessment of the studies, the results were compared and discussed until consensus was achieved. Disagreements were handled by discussion and by consulting a third reviewer (PS). Additional information was sought from the authors when necessary. To reduce pre-formed opinions of experts in the area that can bias the assessment of the relevance of articles, one reviewer was not an expert in the area. Each study included was quality assessed independently by two reviewers (MZA and RG). In the case of discrepancies, the authors were contacted for details of randomization where necessary. Three main quality criteria were assessed: 1) randomization and allocation concealment; 2) blind outcome assessment participants and investigators cannot be blinded to splint therapy; however, it is feasible to blind the outcome assessor and all included studies were assessed for blinding of the outcome assessment; and 3) completeness of the follow-up. Uncertainty about methodology employed during the quality assessment phase was resolved by contacting the authors when necessary. The significance of discrepancies in the estimates of treatment effects from the different trials was assessed by inspection of a graphical display and by means of Cochrane s test for heterogeneity. The Cochrane Oral Health Group s statistical guidelines were followed and relative risk (RR) values calculated along with 95 percent confidence intervals (CI) for binary data. Weighted mean difference (WMD) was used for continuous data. Meta-analysis was to be undertaken using a random effects model in the absence of clinical and statistical heterogeneity. Data were analyzed using Review Manager program and the main results of the studies presented separately for pain (TMJ, muscles and jaw movements), movement of the jaw (deviation in mouthopening, range of mouth opening, lateral movements), TMJ clicking, depression level, Helkimo dysfunction score (an index for anamnestic and clinical dysfunction and occlusal state), treatment credibility, and quality of life. If data allowed, a sensitivity analysis had to be conducted to see how the quality of the studies affects the findings. Results Twenty RCTs were identified as being potentially relevant. All were published in English. Eight trials were excluded due to use of splint therapy other than stabilization splint or due to the participants not meeting the criteria for myofascial pain Twelve RCTs fulfilled the inclusion criteria. Details are presented in Table 1. Apart from one study with vague descriptions of the splint used, 16 all other studies provided some information about splint construction and adjustment to centric relation. Comparison groups included acupuncture, 16,17 bite-plates, 18 biofeedback, 19,20 visual feedback, 21 non-occluding splints, relaxation/hypnorelaxation, 25,26 jaw exercises, 27 and minimal/no treatment. 17,20,22,25 The study participants consisted of patients who had been referred for treatment for myofascial pain to a special clinic. In two studies, 22,24 patients 1244 Journal of Dental Education Volume 69, Number 11

4 were recruited through announcements published in a local journal. The number of participants ranged from twenty 18,21 to eighty patients. 20 The number of patients per study group was twenty or less in four studies. Five studies gave detailed criteria for inclusion and/or exclusion of patients in the study. 17,20,22-24 Information about previous TMD treatment was reported in only one study. 17 The number of follow-up visits was standardized in both the study and control treatments in all studies included. However, the period of treatment/ follow-up varied from four to twelve weeks across included studies. The type of outcomes measures varied between the studies (Table 1). Pain was measured using a Visual Analogue Scale (VAS) 17,22,25 and the Pain Severity Scale (PSS), Muscle Palpation Index (PPI), 20 and a pain diary. 24 Two studies reported number of patients reporting pain on movement of mandible or pain on retrusion of the mandible. 16,22 Other outcomes measured included electromyographic analyses, 18 quality of life, 22 clinical dysfunction scores, 17,19,27 depression, 20 range of jaw movement, sound, locking, and deviation in opening of mouth. 16,21,25 The methodological quality of included studies is as follows: randomization and concealment procedures one of the included studies reported on the method used to generate the randomization sequence or allocation concealment; blind outcome assessment blind outcome assessment was clearly stated in only two trials; 22,25 and drop-outs drop-outs were reported in six studies, 7 none of which undertook a full intention to treat analysis. 2,18,20,22-24 Minimal/No Treatment Pain. In the study by Turk et al., 20 SS was compared to a waiting list control group. A statistically significant lower pain score, using the PPI, was shown in the SS patients with a WMD (95% CI: -4.81, -1.59). Similarly, using PSS, a statistically significant difference in pain scores was found in favor of the SS group with a WMD (95% CI: -2.19, -0.61). A statistically significant difference was also seen when SS was compared to a minimal treatment group 25 in terms of change in present pain (WMD ; 95% CI: , ) and pain on palpation (WMD -0.77; 95% CI: -0.90, -0.64) measured using a VAS. However, no statistically significant difference was shown when SS was compared to a passive control group in terms of the number of patients showing an improvement in intensity of pain at rest. 22 Johansson et al. compared SS with acupuncture and a no intervention control group. 17 Both of the treatment groups showed a statistically significant improvement in pain post-treatment compared to the non-intervention group (p<0.01). Depression. A statistically significant difference was shown between the same groups in terms of depression level evaluated by the Centre for Epidemiologic Studies-Depression (CED-S) (WMD -3.90; 95% CI: -6.29, -1.52) but not the Profile of Mood States (POMS) index (WMD -3.90; 95% CI: -7.74, -0.06). 20 Non- Occluding Splints Pain. No statistically significant difference in palpation score or pain diary score was seen, however, when SS was compared with a non-occluding splint. 24 The number of painful muscles on palpation in the SS group in one study did not differ significantly from the number in the palatal splint group after six weeks of treatment. 23 One trial investigated improvement in unpleasantness and intensity of pain at rest and showed no statistically significant difference between the two groups with an RR 0.75 (95% CI: 0.41, 1.37) and RR 1.44 (95% CI: 0.81, 2.58) for the improvement in the unpleasantness and intensity of pain respectively. 22 Movement. There was no statistically significant difference in the increase in maximal opening (mm) between groups receiving SS or a non-occluding splint (WMD 1.45; 95% CI: -1.47, 4.37). 24 Overall Improvement. There was no statistically significant difference in the number of participants reporting an overall improvement of symptoms between groups receiving SS or a non-occluding splint (RR 1.13; 95% CI: 0.79, 1.61). 24 Similarly, in another study that compared SS to a non-occluding splint, no statistically significant improvement was found for any of the functional outcomes measured (swallowing, chewing, yawning, drinking, etc.). 23 Quality of Life. In the study by Dao et al., quality of life was examined in terms of improvement in sleep, efficiency at work, social activities, feeling depressed, feeling anxious, and poor appetite. 22 A statistically significant difference in favor of SS therapy was shown for efficiency at work (RR 0.12; 95% CI: 0.02, 0.90). However, a statistically significant November 2005 Journal of Dental Education 1245

5 Table 1. Characteristics of included studies Turk et al. 20 Winocur et al. 25 Dao et al. 22 Rubinoff et al. 24 Raphael and Marbach 23 Design Single Center RCT Single Center RCT Single Center RCT Single Center RCT Single Center RCT Duration 6 weeks 49 days 10 weeks 6 weeks 6 weeks Allocation Concealment Unclear Unclear Unclear Not used Unclear Blinding Unclear Unclear Double-Blind Unclear Unclear Participants 80 myofascial pain patients 40 myofascial pain patients 63 myofascial pain patients 30 myofascial pain patients 68 myofascial pain patients Diagnostic Criteria Unclear Unclear Stated Stated Stated Mean Age (y) Range Group 1 SS SS SS SS SS Group 2 Biofeedback/ Hypnorelaxation U-shaped non-occluding Non-occluding palatal Palatal splint Stress Management splint appliance Group 3 No treatment Minimal treatment SS worn for only 30 mins at each appointment Group 4 Results A statistically significant Statistically significant lower pain score, using the PPI in the SS patients. Using PSS, a statistically significant difference in pain scores in favor of SS group. SS patients had lower pain scores for PSS. However, the difference between the two groups was not shown to be statistically significant. No statistically significant difference was found between SS and BF/SM for the reduction in muscle severity as evaluated by (PPI) Index. A statistically significant benefit in BF/SM group was found in depression level evaluated by (CES-D) index. Evaluation of depression using the Profile of Mood States (POMS) index also showed a statistically significant difference between the SS. No statistically significant difference regarding the increase in treatment credibility between SS and BF/SM. difference was seen when SS was compared to a minimal treatment group in terms of change in present pain (and pain on palpation measured using a VAS). No statistically significant difference was shown when SS was compared to a passive control group in terms of the number of patients showing an improvement in intensity of pain at rest. No statistically significant difference between the two groups for the improvement in the unpleasantness and intensity of pain respectively. A statistically significant difference in favor of bite plates was shown for improvement when examining participants social activities and feeling of depression. No statistically significant difference in palpation score or pain diary score was seen. No statistically significant difference in the increase in maximal opening (mm) and in overall improvement of symptoms between the two groups. Number of dropouts 2 None Group 1: 2 2 Group 1: 2 Group 2: 1 Group 2: 2 Group 3: 1 Reason for dropouts Not stated Group1: Not stated Not stated Not stated Group 2: Couldn t wear the splint Group 3: Diagnosed as migraine The number of painful muscles on palpation in the SS group did not differ significantly from the number in the palatal splint group after six weeks of treatment. No statistically significant improvement was seen for any of the functional outcomes measured Journal of Dental Education Volume 69, Number 11

6 Raustia and Pohjola 16 Johansson et al. 17 Dahlstrom and Haraldson 18 Design Single Center RCT Single Center RCT Single Center RCT Duration 3 months 3 months 6 weeks Allocation Concealment Unclear Unclear Unclear Blinding Unclear Unclear Unclear Participants 50 myofascial pain patients 45 myofascial pain patients 20 myofascial pain patients Diagnostic Criteria Unclear Unclear Unclear Mean Age (y) 26.4 Not stated 26.3 Group 1 SS SS SS Group 2 Acupuncture Acupuncture Bite plate with a frontal plateau Group 3 Control Group 4 Results No statistically significant difference between the two groups for the pain on palpation or clicking. No statistically significant difference was shown for pain on movement (retrusion) or movement (opening). A statistically significant difference with regard to deviation to the right side in mouth-opening movement after treatment was shown in SS. No significant difference between these two groups in terms of deviation to the left or to the right side in the mouth-opening movement after treatment. No statistically significant difference was found in the SDS although both groups showed a statistically significant decrease in SDS and VAS scores after treatment. Both groups showed a statistically significant decrease in dysfunction score although no betweengroup differences were found. No statistically significant difference was found in the number of patients showing a high Helkimo dysfunction score in both groups at final examination (after six weeks of appliance treatment). Number of dropouts Not stated None 1 Reason for dropouts Refused to undergo second EMG recording analysis. Dahlstrom et al. 19 Monteiro and Clark 21 Magnusson and Syren 27 Okeson et al. 26 Design Single Center RCT Single Center RCT Single Center RCT Single Center RCT Duration 6 weeks 4 weeks 6 months 6 weeks Allocation Concealment Unclear Unclear Unclear Unclear Blinding Unclear Unclear Unclear Unclear Participants 30 myofascial pain patients 20 myofascial pain patients 26 myofascial pain patients 24 myofascial pain patients Diagnostic Criteria Unclear Stated Unclear Stated Mean Age (y) Group 1 SS SS SS SS Group 2 Biofeedback Visual feedback Jaw exercises Relaxation therapy Group 3 Group 4 Results No statistically significant difference was shown with regard to the number of patients achieving a decrease in the dysfunction score. Subjective rating of symptoms using a 5-point scale showed a statistically significant reduction in both groups. The difference between groups was not statistically significant. Statistically significant decrease was found in lateral mandibular movement scores. An improvement in all outcomes measured for both groups at both three month and six month assessment was shown. No statistically significant differences were observed between the groups at any time. A statistically significant reduction in pain on palpation was shown in favor of SS. A statistically significant difference in favor of SS was found regarding changes in maximal mouth opening. Number of dropouts None None Group 1: 2 None Group 2: 1 Reason for dropouts Not Stated November 2005 Journal of Dental Education 1247

7 difference in favor of bite plates was shown when examining participants social activities and feelings of depression (RR 1.88; 95% CI: 1.07, 3.27 and RR 2.00; 95% CI: 1.07, 3.75 respectively). No other statistically different differences were shown. Acupuncture Pain. One study reported the number of patients with pain on palpation three months after treatment. 16 There was no statistically significant difference between the two groups with a relative risk (RR) of 0.63 (95% CI: 0.24, 1.65) for the pain on palpation of the right side and RR 2.00 (95% CI: 0.19, 20.67) for pain on palpation of the left side. Similarly, no statistically significant difference was shown for pain on retrusion of the mandible (RR 0.60; 95% CI: 0.16, 2.25) or jaw opening (RR 0.86; 95% CI: 0.34, 2.19). One study assessed change in severity of pain, described by a subjective symptom score (SDS). 17 No statistically significant difference was found in the SDS when comparing the SS and acupuncture groups with RR 1.50 (95% CI: 0.29, 7.73), although both groups showed a statistically significant decrease in SDS and VAS scores after treatment. Movement. Raustia and Pohjola 16 showed a statistically significant difference with regard to deviation to the right side in mouth-opening movement after treatment in the SS group when compared with the acupuncture group with an RR 0.55 (95% CI: 0.35, 0.84). Interestingly, the analysis showed no significant difference between these two groups in terms of deviation to the left side in the mouth-opening movement after treatment with RR 5.00 (95% CI: 0.63, 39.79). There was no statistically significant difference between groups with regard to range of lateral movements (to the right WMD 0.80; 95% CI: -0.43, 2.03; to the left WMD 0.40; 95% CI: -0.86, 1.66) or range of mouth opening (WMD 2.60; 95% CI: -0.86, 6.06). Clicking. The same study compared the TMJ clicking in the SS and acupuncture groups. 16 No statistically significant benefit was found. The RR for TMJ clicking on the right side was 1.00 (95% CI: 0.56, 1.78) and that for the TMJ clicking on the left side was 0.58 (95% CI: 0.28, 1.23). Dysfunction Score. Johansson et al. 17 assessed the clinical signs by means of the Helkimo clinical dysfuction score. Both the SS and acupuncture groups showed a statistically significant decrease in dysfunction score (p<0.01) although no betweengroup differences were found. Biteplates In the trial by Dahlstrom et al., 18 no statistically significant difference was found in the number of patients showing a high Helkimo dysfunction score in SS and bite plates groups with an RR 0.68 (95% CI: 0.20, 2.23) at final examination (after six weeks of appliance treatment). Feedback Pain. One trial 20 investigated reduction in pain severity level using PSS. A comparison between biofeedback/stress management (BF/SM) and SS was undertaken. The results of this trial showed that SS patients had lower pain scores for PSS. However, the difference between the two groups was not statistically significant (WMD 0.30; 95% CI: -0.48, 1.08). Similarly, no statistically significant difference was found between SS and BF/SM for the reduction in muscle severity as evaluated by the Palpation Pain Index (PPI) (WMD 0.90; 95% CI: -0.26, 2.06). Depression. In the same study by Turk et al., 20 a statistically significant benefit in BF/SM group was found, when compared with the SS group, in depression level evaluated by CES-D index with a WMD 5.50 (95% CI: 1.49, 9.51). Evaluation of depression using the POMS index also showed a statistically significant difference between the SS and BF/SM groups with a WMD 6.50 (95% CI: 2.12, 10.88). Treatment Credibility. No statistically significant difference was reported in the study by Turk et al. 20 regarding the increase in treatment credibility between SS and BF/SM with a WMD (95% CI: -0.69, 0.49). Dysfunction Score. One trial, 19 comparing SS and biofeedback, presented data for clinical signs of dysfunction according to the Helkimo index D1. No statistically significant difference was shown with regard to the number of patients achieving a decrease in the dysfunction score, RR 1.11 (95% CI: 0.64, 1.92). Subjective rating of symptoms using a five-point scale showed a statistically significant reduction in both groups, with a median change in score from three to one in both groups. However, the difference between groups was not statistically significant Journal of Dental Education Volume 69, Number 11

8 Movement. In the study by Monteiro and Clark, 21 both the SS and visual feedback groups showed statistically significant decrease in their lateral mandibular movement scores (p<0.01). However, data were unavailable for analysis. The authors gave no details of the data obtained. Jaw Exercises Magnusson and Syren 27 found improvements in all outcomes measured (mandibular mobility, TMJ function, TMJ pain, muscle pain, pain on movement, clinical dysfunction, and anamnestic dysfunction) for both groups at both three-month and six-month assessment. A follow-up questionnaire one to four years after the first clinical assessment showed a lasting treatment effect in most patients, although many patients continued to perform their jaw exercises or wear their occlusal splint. No statistically significant differences were observed between the groups at any time point. Relaxation Pain. Two trials compared SS with relaxation. 25,26 Both studies assessed reduction in pain and changes in the maximal mouth opening. Okeson et al. 26 showed a statistically significant reduction in pain on palpation in favor of SS; however, this was not supported by the Winocur trial. 25 Significant statistical heterogeneity was found between the two trials for reduction of pain on palpation (p<0.01; I 2 =98.3%); it was therefore felt inappropriate to pool the data. Movement. Similarly, for changes in maximal mouth opening, significant heterogeneity was found between the two studies both for active/comfortable opening (p<0.01; I 2 =92.9%) and assisted/painful opening (p<0.01; I 2 =94%). Again, it was felt inappropriate to pool the data from the trials. Okeson et al. 26 found a statistically significant difference in favor of SS, but these results were not supported by Winocur et al. 25 Discussion Myofascial pain is the most common temporomandibular disorder (TMD) and the aetiology is multifactorial. Consequently, many different therapies, some conservative and reversible, others irreversible, have been advocated for patient with myofascial pain. A number of successful treatment outcomes have been reported. 3-5 Various types of occlusal splints described in the literature have different indications and functions. The stabilization splint (SS) is one such type of occlusal splint. The stabilization splint is a hard acrylic splint that provides a temporary and removable ideal occlusion. Providing an occlusion by the use of splint therapy reduces abnormal muscle activity and produces neuromuscular balance. Our systematic review of the literature showed that there is only weak evidence to suggest that SS therapy may be beneficial in comparison to minimal or no treatment in terms of pain as measured using the PPI, PSS, VAS, and CED-S. However, there was no evidence of a statistically significant difference in the effectiveness of stabilization splint therapy in reducing symptoms in patients with myofascial pain when compared with other conventional treatments such as acupuncture, biteplates, and biofeedback. In general, the comparisons were based on a small number of patients with no standardization of the outcomes measured. Little attention was given to the randomization process and concealment of allocation in the included studies, increasing the risk of selection bias. Given the subjective nature of the outcome assessment in this area, blind outcome assessment is imperative to reduce bias within the RCTs. Only two of the included studies stated that blind outcome assessment was used. 22,25 On the basis of our results, we conclude that there is insufficient evidence either for or against the use of stabilization splint therapy over other active interventions for the treatment of temporomandibular myofascial pain. However, it appears that stabilization splint therapy may be beneficial for reducing pain severity at rest and on palpation and depression when compared to no treatment. Well-conducted RCTs are needed to clarify the effectiveness of stabilization splint therapy for patients with myofascial pain. Trials should pay attention to the generation of the randomization sequence, method of allocation, blind outcome assessment, sample size, and duration of follow-up. Clear inclusion/exclusion criteria may help identify whether or not the patients entering a study had myofascial pain or a different diagnosis of temporomandibular disorder. A standardization of the outcomes used to assess the treatment of myofascial pain should be established in the RCTs. The Cochrane version of this November 2005 Journal of Dental Education 1249

9 systematic review will be updated periodically to include any newly emerging, relevant RCTs. Acknowledgments The reviewers would like to thank members of the Cochrane Oral Health Group for their support, particularly Professor Helen Worthington (coordinating editor and statistician), Emma Tavender (review group coordinator), and Sylvia Bickley (trials search coordinator). REFERENCES 1. Gray RJM, Davies SJ, Quayle AA. A clinical guide to temporomandibular disorders. London: BDJ Books, 1997: Gray RJM, Davies SJ. Occlusal splints and temporomandibular disorders: why, when, how? Dent Update 2001;28: Tsuga K, Akagawa Y, Sakaguchi R, Tsuru H. A short-term evaluation of the effectiviness of stabilisation therapy for specific symptoms of temporomandibular joint dysfunction syndrome. J Prosthet Dent 1989;61: Gray RJM, Davies SJ, Quale AA. A comparison of two splints in the treatment of TMJ myofascial pain: can occlusal analysis be used to predict success of splint therapy? Br Dent J 1991;170: Davies SJ, Gray RJM. The pattern of splint usage in the management of two common temporomandibular disorders. Part III: Long-term follow-up in an assessment of splint therapy in the management of disc displacement with reduction and pain dysfunction syndrome. Br Dent J 1997;183: Gray RJM. How reliable is your patient? A comparison of subjective complaints and clinical findings in a group of temporomandibular patients. J Dent 1986;14: Clarke A, Oxman AD. Cochrane reviewer s handbook. In: The Cochrane Library. Issue 4. Oxford: Update Software, Pettengil CA, Growney MR, Schoff R, Kenworthy CR. A pilot study comparing the efficacy of hard and soft stabilizing appliances in treating patients with temporomandibular disorders. J Prosthet Dent 1998;79: Linde C, Isacsson G, Jonsson BG. Outcome of 6-week treatment with transcutaneous electric nerve stimulation compared with splint on symptomatic temporomandibular joint disk displacement without reduction. Acta Odontol Scand 1995;53: Lundh H, Westesson P, Kopp S. Anterior repositioning splint in the treatment of temporomandibular joints with reciprocal clicking: comparison with a flat occlusal splint and an untreated control group. Oral Surg 1985;60: Lundh H, Westesson P-L, Jisander S, Eriksson L. Diskrepositioning onlays in the treatment of temporomandibular joint disk displacement: comparison with a flat occlusal splint and with no treatment. Oral Surg Oral Med Oral Pathol 1988;66: Lundh H, Westesson PL, Eriksson L, Brooks SL. Temporomandibular joint disk displacement without reduction: treatment with flat occlusal splint versus no treatment. Oral Surg Oral Med Oral Pathol 1992:73: Ekberg E, Vallon D, Nilner M. Occlusal appliance therapy in patients with temporomandibular disorders: a doubleblind controlled study in short-term perspective. Acta Odontol Scand 1998;56: Wenneberg B, Nystrom T, Carlsoon GE. Occlusal equilibration and other somatognathic treatment in patients with mandibular dysfunction and headache. J Prosthet Dent 1988;59: Wright E, Anderson G, Schulte J. A randomized clinical trial of intraoral soft splints and palliative treatment for masticatory muscle pain. J Orofac Pain 1995;9: Raustia AM, Pohjola RT. Acupuncture compared with stomatognathic treatment for TMJ dysfunction. Part III: effect of treatment on mobility. J Prosthet Dent 1986; 56: Johansson A, Wenneberg B, Wagersten C, Haraldson T. Acupuncture in treatment of facial muscular pain. Acta Odontol Scand 1991;49: Dahlstrom L, Haraldson T. Bite plates and stabilisation splints in mandibular dysfunction: a clinical and electromyographic comparison. Acta Odontol Scand 1985;43: Dahlstrom L, Carlsson GE, Carlsson SG. Comparison of effects of electromyographic biofeedback and occlusal splint therapy on mandibular dysfunction. Scand J Dent Res 1982;90: Turk DC, Zaki HS, Rudy TE. Effects of intraoral appliance and biofeedback/stress management alone and in combination in treating pain and depression in patients with temporomandibular disorders. J Prosthet Dent 1993;70: Monteiro AA, Clark GT. Mandibular movement feedback vs. occlusal appliances in the treatment of masticatory muscle dysfunction. J Craniomandib Disord 1988;2: Dao TT, Lavigne GJ, Charbonneau A, Feine JS, Lund JP. The efficacy of oral splints in the treatment of myofacial pain of the jaw muscles: a controlled clinical trial. Pain 1994;56: Raphael KG, Marbach JJ. Widespread pain and the effectiveness of oral splints in myofascial face pain. J Am Dent Assoc 2001;132: Rubinoff MS, Gross A, McCall WD. Conventional and non-occluding splint therapy compared for patients with myofascial pain dysfunction syndrome. Gen Dent 1987;35: Winocur E, Gavish A, Emodi-Perlman A, Halachmi M, Eli I. Hypnorelaxation as treatment for myofascial pain disorder: a comparative study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002;93: Okeson JP, Kemper JT, Moody PM, Haley JV. Evaluation of occlusal splint therapy and relaxation produces in patients with temporomandibular disorders. J Am Dent Assoc 1983;107: Magnusson T, Syren M. Therapeutic jaw exercises and interocclusal appliance therapy: a comparison between two common treatments of temporomandibular disorders. Swed Dent J 1999;23: Journal of Dental Education Volume 69, Number 11