Treatment outcome of supervised exercise, home exercise and bite splint therapy, respectively, in patients with symptomatic disc displacement with reduction: A randomised clinical trial

Abstract The best treatment strategy for disturbing temporomandibular clicking sounds is not known. The aim was to evaluate the effect of exercise and bite splint therapy, respectively, in patients with symptomatic disc displacement with reduction. The study was a randomised clinical trial of subjects with temporomandibular joint (TMJ) clicking sounds with a reported severity/intensity of ≥4 on a numerical rating scale (0‐10) and signs fulfilling the Research Diagnostic Criteria (RDC/TMD) for disc displacement with reduction. Thirty subjects each were randomised to bite splint, home exercise, or supervised exercise programme at the clinic. Two examiners (authors), blinded to the treatment modality, examined the same subject at baseline and at a 3‐month follow‐up. Non‐parametric statistical methods were applied for analyses. A P‐value <.05 was considered statistically significant. The dropout rate was highest in the home exercise group. About 50% of the participants reported improvement of their TMJ sounds with no significant difference between treatments. In the supervised exercise and the bite splint groups, approximately 2/3 of the patients reported 30% or more improvement of their TMJ sounds and half reported 50% improvement or more. The supervised exercise group also showed reductions in TMD pain, neck disability, mood disturbances and somatisation. Jaw exercise programmes and bite splint treatments had positive effects on TMJ clicking. The supervised exercise programme had an additional effect on the subject's well‐being and thus may help to encourage patient's empowerment and coping strategies.


| BACKG ROU N D
Temporomandibular disorders (TMD) involve symptoms such as pain around the temporomandibular joint (TMJ) and jaw muscles, pain on jaw movements, impaired jaw mobility, TMJ sounds, and temporary or persistent locking of the jaw. 1 The symptoms affect individual's wellbeing and quality of life. 2,3 Most intervention studies in this field have been directed towards pain conditions and only a few 4,5 have focused on the most commonly reported TMD condition in the population, that is symptomatic disc displacement with reduction.
This condition commonly involves clicking sounds caused by interference of the TMJ disc, and treatment may in some specific cases end up in discectomy. TMJ clicking sounds with or without associated pain can thus significantly disturb the individual's normal jaw function and generate a demand for treatment. Reversible treatment modalities are recommended as primary intervention for patients with TMD. These modalities include advice, bite splints, jaw exercises and over-the-counter medication. The National Guidelines for dental health care in Sweden 6  The overall objective was to abate the current uncertainties about the effects of treatment modalities for symptomatic disc displacement with reduction. The primary aim was to evaluate the effect of supervised and home exercise programmes, respectively, among patients with symptomatic disc displacement with reduction. The second aim was to evaluate whether there is a significant difference in outcomes between a supervised exercise programme, home exercise programme and bite splint therapy.
Bite splint therapy was used as "standard treatment." Our first hypothesis was that a supervised exercise programme has a moderate to high effect on symptoms frequency and intensity, as well as on jaw opening capacity. Further, this effect was expected to be significantly better than a home exercise programme. The second hypothesis was that treatment with a bite splint has a moderate effect on TMJ symptoms and does not differ significantly from a supervised exercise programme; also, a low effect on jaw opening capacity was expected. The third hypothesis was that a home exercise programme has low-to-moderate effect on symptom relief that is significantly lower than bite splint and a supervised exercise programme.

| Study design
The study design was a randomised controlled clinical trial (RCT) with three parallel treatment groups of 30 subjects in each group.
The participants were examined at baseline and re-examined after 3 months. The study was approved by the Regional Ethical Review Board in Umeå, Sweden, (Dnr 2011-219-31M) and carried out in accordance with the Declaration of Helsinki.

| Study population
Participants were recruited from patients referred to the Clinical Oral Physiology department in Umeå, Sweden as well as among those who responded to targeted advertising. After a screening procedure, all subjects fulfilling the inclusion criteria were given oral and written information about the study. Those who were willing to participate signed an informed consent. In total, 593 individuals were screened and 90 subjects were included in the trial.
The general inclusion criteria were age between 18 and 70 years, accommodation in Umeå Municipality's proximity, and able to understand Swedish, orally and written. They should have no major psychiatric diagnosis, no ongoing dental, medical or physiotherapeutic treatments related to the patient's symptom that may interfere with the study, no active rheumatologic disease and no malignant disease.
The specific inclusion criteria for being allocated to symptomatic disc displacement with reduction were that TMJ clicking sounds were presented as their major symptom. They should fulfil the Research Diagnostic Criteria for TemporoMandibular Disorders (RDC/TMD) for symptomatic disc displacement (ie, reproducible TMJ clicking sounds during jaw opening and closing with the opening click registered at >5 mm interincisal distance and with TMJ clicking sounds that cannot be recognised when jaw opening is performed in a protruded position). The participants should indicate the severity of the TMJ clicking sounds at ≥4 on a Numerical Rating Scale (NRS 0-10) and the frequency of TMJ clicking sounds once a week or more.

| Study settings
The study was performed at the Department of Odontology/Clinical Oral Physiology at Umeå University Sweden in collaboration with the Public Dental Health service. Two specialists in clinical oral physiology/TMD performed the examinations. Each participant had the same examiner at baseline and at 3 months follow-up. The examiner was always blinded to the participant's intervention. Two assistants were engaged in the treatments.

| Questionnaire
After enrolment, the participants received the Swedish transla- The questionnaire was filled out at the subject's home and was taken to the clinic where subjects were randomised to intervention.

| Randomising process and blinding
Those who fulfilled the inclusion criteria were randomised into three different treatment groups. There were 30 individuals in each group. The randomisation was done with the aid of SPSS 20 (randomised numbers) before the study started. When a subject fulfilled the inclusion criteria and accepted to participate, an assistant contacted the subject and assigned the participant to intervention. Participants received written and oral information about the study and signed an informed consent. Each participant was carefully instructed to not disclose his/her treatment allocation to the examiner at follow-up. All treatments were performed by trained assistants that were not involved in the examination and evaluation of the respective participant. The two examiners (authors) were blinded to the subject's intervention trial. They examined the same subject clinically at follow-up the same as at baseline.

| Interventions
All patients received brief oral information of the TMJ structure and function that was performed by the same assistant. Subjects were instructed to avoid food with a tough texture and to try to not provoke TMJ clicking sounds during the day.

| Outcomes
The outcomes were based on the domains recommended by

| Sample size
A power calculation based on the results by Burgess et al 8 revealed that 21 subjects in each subgroup would be sufficient to detect a difference when α = 0.05 and β = 0.8. With an expected dropout of 25%, the number for each trial was decided at 30 subjects. Thus, a total of 90 subjects were registered for the trials.

| Statistical methods
Analyses were based on an intention-to-treat approach. The individual baseline data were imputed for dropouts regardless of reason. Changes were calculated as the difference between baseline data and data at the 3-month follow-up. Variables with normal distribution were analysed with independent sample t test, and the remaining with non-parametrical Wilcoxon for paired observations and Chi-square. A P-value <.05 was considered statistically significant.  . Wilcoxon's test for paired observations. b Independent sample t test.

| RE SULTS
Flow chart of number of participants, loss to follow-up and reasons for dropout are presented in Figure 1. The number of dropouts was highest among the home exercise group (n = 10), compared with the bite splint group (n = 2) and supervised exercise group (n = 3).
There were more women (n = 63) than men (n = 27) included in the study, with no significant difference between treatment groups.   (Table 3). In the bite splint group and in the supervised exercise group, approximately 2/3 reported 30% or more improvement of their TMJ sounds and half reported 50% improvement or more TA B L E 2 Pre-and post-treatment values for registered temporomandibular disorders clicking sounds, and maximal jaw mobility, associated with the following treatment modalities in a blinded randomised control trial: bite splint, home exercise and supervised exercise

P-value a (n) (%) (n) (%) (n) (%)
TMJ clicking sounds Impaired ( Table 3). The relationship between PGIC for TMJ clicking sound and the percentage reduction of TMJ clicking severity is presented in Figure 2. The relationship was significant (P < .001; ANOVA).
Temporomandibular disorders pain was a co-morbid symptom for 60% of these cases with TMJ clicking sounds as predominate disorder and 46% reported at the start of the study awareness of tooth clenching or tooth grinding at least once a week or more. No neck disability (NDI <10%) was reported by 63%, and no limitation of jaw function (JFLS <5) was reported by 1/3 of the total sample.
After the 3-month intervention period showed the supervised exercise group the highest individual improvements for the concomitant symptoms; TMD pain, neck disability, mood disturbances and somatisation (Table 1).

| D ISCUSS I ON
In this study, we specifically addressed the condition symptomatic disc displacement with reduction, with specific focus on the ques- The treatment modalities tested have two basically opposing approaches. The bite splint has a passivating intention with the aim to decrease parafunctional activity and to alter the biomechanical loading on the TMJ structures. Exercise has an activating approach with the aim to mobilise and challenge dysfunction and pain. Both approaches share the common circumstances that they depend heavily on the patient's compliance, decisions and motivation to complete the treatments in accordance with the instructions and intentions. The supervised exercise model is a method to strengthen the patient's empowerment and self-efficacy, which hopefully is favourable in a long-term perspective.
Those who were assigned to supervised exercise showed significant improvements also in other indicators of health and well-being, such as neck disability, mood disturbance and somatisation. These results are in accordance with a previous study on TMD pain patients. 10 In the study by Yoda et al, 5 those who reported improvement in their TMJ clicking sounds also reported less discomfort and interference with daily life. The positive effect may thus be assigned both the supportive arrangement and to the exercise in itself. The direct underlying mechanisms are not disclosed by this study but may be related to both F I G U R E 2 95% confidence interval (CI) of the percentage reduction of temporomandibular joint (TMJ) severity in relation to patient's global impression of change (PGIC). Worse = −1, no change = 0, minimally improved = 1, much improved = 2, very much improved = 3 unspecified effects of an intervention and changed biomechanical loadings of the structures involved TMJ movements.
The strength of the current study was that the participants were symptomatically reasonably homogenised, randomised to treatment and evaluated intra-individually by the same examiner who was also blinded to treatment. A weakness was the unbalanced dropout. On the other hand, the higher dropout in the home exercise group may disclose that this choice of intervention calls for a high motivational effort. Our results are thus in line with a previous study on patients with jaw muscle pain that concluded that extensive communication between patient and doctor may be more effective than an occlusal appliance. 11 Our results are also in accordance with an observational study with counselling and physical therapy that resulted in significant improvement in pain and jaw function in patients with myofascial pain. 12

| CON CLUS ION
Jaw exercise programmes and bite splint therapy had similar positive effects on perceived severity of TMJ clicking sounds. The supervised exercise programme had additional effects on subject's well-being and may thus help to encourage subjects' empowerment and coping strategies.

ACK N OWLED G M ENTS
The study was supported by funding The Swedish Research Council (2011-300), Västerbotten County Council (7001272) and Swedish Dental Society. A special thanks to assistants Inga-Liza Lundström and Elinor Fowler.

CO N FLI C T O F I NTE R E S T
No conflict of interest.