Prevalence of Temporomandibular Disorder in children and adolescents with Juvenile Idiopathic Arthritis – a Norwegian cross sectional multicenter study

Background. Children and adolescents with JIA may suffer pain from temporomandibular disorder (TMD) and routines for the assessment of temporomandibular joint (TMJ) pain for the health and dental care are lacking. The aims were to examine the prevalence of TMD in children and adolescents with juvenile idiopathic arthritis (JIA) as compared to their healthy peers and to examine potential associations between JIA and TMD. Methods. This comparative cross-sectional study is part of a longitudinal multicenter study performed during 2015 - 2020, including 228 children and adolescents aged 4-16 years, with a diagnosis of JIA according to the ILAR criteria. This sub-study draws on a subset of data from the rst study visit, including assessments of TMD as part of a broader oral health examination. Participants with JIA were matched with healthy controls according to gender, age, and center site. Calibrated examiners performed the clinical oral examinations according to a standardized protocol, including shortened versions of diagnostic criteria for TMD (DC/TMD) and the former EuroTMJoint Recommendations for Clinical TMJ Assessment in Patients Diagnosed with JIA. Symptoms were recorded and followed by a clinical examination assessing the masticatory muscles and TMJ`s. Results. In our cohort of 221 participants with JIA and corresponding controls, 88 participants with JIA (39.8%) and 25 controls revealed TMD based on symptoms and clinical signs. Painful TMD during the last 30 days was reported in 59 (26.7%) participants with JIA vs. 10 (5.0%) of the healthy controls (p<0.001). Vertical unassisted jaw movement was lower in JIA than in controls; mean 46.2 mm vs. 49.0 mm, respectively (p <0.001). Among participants with JIA, a higher proportion of those using synthetic disease-modifying antirheumatic-drugs (sDMARDs) and biologic (bDMARDs) presented with painful masticatory and TMJs at palpation.


Background
Juvenile idiopathic arthritis (JIA) is currently the most common chronic rheumatic disease in children and adolescents [1,2]. The International League of Associations of Rheumatology (ILAR) de nes JIA as arthritis of unknown etiology, starting before the age of 16 with a duration of at least six weeks [3]. It encompasses seven categories, including systemic arthritis, oligoarthritis (persistent or extended), rheumatoid factor negative polyarthritis, rheumatoid factor positive polyarthritis, psoriatic arthritis, and enthesitis-related arthritis, with different, though overlapping characteristics. Cases that t none or more than one of these categories are de ned as undifferentiated arthritis. The burden of JIA is characterized by short and long-term functionalisability and pain. Common features at presentation are morning stiffness, swelling of one or more joints, functional disturbances, and sometimes pain. The reported prevalence is around 1-2 per 1000 children, with girls more frequently affected than boys [1,2]. Temporomandibular disorder (TMD), known as anmbrella or collective term for muscle pain and jaw dysfunction, masks a heterogeneous group of conditions [4]. TMD is associated with various clinical signs and symptoms, involving masticatory muscles, teeth, tongue, temporomandibular joint (TMJ) and/or its supportive tissues [5][6][7]. Changes in motor behavior, caused by musculoskeletal pain and painrelated movement disorders, re ect sustained pain perception. In two recent studies from Western Norway [9,10], the prevalence of painful TMD amongst otherwise healthy adolescents was reported to be around seven percent, based on self-reported pain screening questionnaires, adopted by Nilsson and colleagues [11]. In the study by Graue and colleagues [10] the prevalence of TMD was 11.9%, when using the Diagnostic Criteria for Temporomandibular Disorders (DC/TMD). In all three studies, girls were more frequently affected than males.
In children and adolescents with JIA the reported gures are substantially higher. A Swiss study amongst school children [12] and a Danish study of children [13] with JIA reported TMD prevalences of 38.6% and 53%, respectively.
Previous studies revealed that children and adolescents irrespective of their JIA category are prone to develop TMJ arthritis [14,15]. Also younger children with JIA may suffer pain from TMJs, either caused by in ammation and/or destructive change, by muscular tensions from the surrounding muscles as a component of TMD, or by a combination of the two [13]. Symptoms indicating TMJ arthritis include decreased mouth opening and/or ear-ache and pain during eating, chewing or yawning [16][17][18] (16,(25)(26)(27). In the younger children, clinical assessment of painful TMD symptoms may be biased by indirect input from their parents.
The aims of the present study were to examine the prevalence of TMD in children and adolescents with JIA as compared to their healthy peers and to examine potential associations between JIA and TMD.

Study design and participants
This cross-sectional study is part of a longitudinal multicenter study, the NorJIA study, performed during 2015 -2020, including 228 children and adolescents. Inclusion criteria were a diagnosis of JIA according to the International League of Associations for Rheumatology (ILAR) [3] and age 4-16 years Exclusion criteria were lack of written informed consent, congenital facial anomalies, or major medical comorbidities.
This particular substudy (2015-2018), using a matched comparative cross-sectional design, draws on a subset of data from the rst study visit, including assessments of TMD as part of a broader oral health examination. Children and adolescents were matched (1:1) with healthy controls according to gender, age and center site. The healthy controls were recruited from seven different Public Dental Service clinics (PDS) representing both rural and urban areas in the western, middle and northern part of Norway. The sample size estimate was based on a Swedish study, reporting a TMD prevalence of 26% in children with JIA [25]. Each group, for both, participants with JIA and for healthy controls required a precision of 5% for the 95% con dence interval a sample size of 296.

Data collection
At the study visits, children and adolescents with JIA were examined by experienced pediatric rheumatologists at Haukeland University Hospital (HUS), Bergen, University Hospital of North Norway (UNN), Tromsø and St. Olavs University Hospital, Trondheim. Registered data included background characteristics in terms of age at disease onset, disease category, previous and on-going medication, disease status on the day of examination, a thorough joint examination, blood tests, and validated measures for patient-reported disability, general body pain and health assessments. Both children and adolescents with JIA and controls underwent a thorough clinical oral examination performed by experienced odontologists, including a TMD assessment.

TMD screening and assessment
The assessment procedures were standardized and based on two shortened versions of diagnostic tools; "Axis I Clinical Examination for DC/TMD" [26] and the self-assessment questionnaire Recommendations for Clinical TMJ Assessment in Patients Diagnosed With JIA" [27]. The latter was used to enhance the operational speci cation of DC/TMD due to the fact that the DC/TMD tool alone is reported to show a weak validity for TMJ assessment, e.g. disc displacement diagnosis (low sensitivity) and degenerative joint disease diagnosis (low sensitivity and speci city) [26].
Prior to, and during the study period, calibration sessions for the ve participating oral examiners were performed, including four calibration exercises according to procedures previously described by our research team [28]. Further details on the calibration results are presented in Supplementary material (Supplementary Tables S1 and S2).

Variables and outcomes
The demographic variables were age, gender, JIA categories and medication status. The symptom outcomes were TMD pain the last 30 days (n, %) and how many of the individuals that experienced pain during jaw movement in the clinical examination (n, %). The clinical signs outcomes were vertical and lateral unassisted jaw movements (mm); palpation of the masticatory muscles and the TMJ was noted if painful (n), and if the TMJ disc was clicking in a painful manner (n).

Statistical methods
A two-way mixed intraclass correlation coe cient (ICC) and percent agreement were used for calibration measurements. Differences between groups were tested using Chi-square statistics or a two sample t-test as appropriate. All statistical tests were performed using SPSS version 25 (IBM, Chicago, IL). The level of statistical signi cance was set at 5% (P ≤ 0.05).

Ethical considerations
The study was approved by the regional ethics committee (2012/542/REK vest). Written informed consents were obtained from all parents and/or participants as appropriate. The study was registered in ClinicalTrials.gov (No: NCT03904459).

Results
A total of 360 children and adolescents with JIA were eligible for the main study, of whom 228 accepted the invitation to participate, yielding a response rate of 63.3%. The acceptance rate for healthy controls was 224 / 294 (76.2%). Mean age for both participants with JIA and healthy controls was 12.0 years (SD 3.17 and 3.21, respectively) (p=0.98). Mean age of the 228 participants with JIA was higher than for the 132 eligible patients that did not participate, 12.0 years vs. 10.6 years, respectively (p <0.001). The proportion of girls with JIA was also higher than among the 132 not participating (59.2% vs 58.3%, p=0.027). Among 228 participating children with JIA, 224 underwent an oral examination, and 221 underwent the TMD assessment and were thus included in the present substudy ( Figure 1).
Clinical oral examination Self-reported pain in the jaws during the last 30 days was reported in 59 (26.7%) (44 girls) participants with JIA vs. 10 (5%) (8 girls) of the healthy controls (p<0.001). Pain during jaw movements at the clinical examination was reported in 112 (51%) (67 girls) participants with JIA vs. in 59 (26.8%) (34 girls) of the healthy controls (p<0.001) ( Figure 2); ranging from 28.6% to 50% in the different JIA categories (Table 1). No statistically signi cant differences in the presence of TMD according to JIA categories was found (p=0.848) (results not shown).
The clinical examination revealed that the mean vertical unassisted jaw movement was lower for participants with JIA than for controls; 46.2 mm vs. 49.0 mm, respectively (p< 0.001) ( Table 2). There were 88 (61 girls) participants with JIA (39.8%) and 25 (17 girls) (11.3%) healthy controls that had both symptoms and clinical signs of TMD (Figure 2). When assessing the jaw muscles and TMJ, 111 (50.2%) (75 girls) participants with JIA had both painful masticatory muscles and TMJs on palpation, vs. 62 (28.2%) (39 girls) of the healthy controls (p< 0.001) ( Table 3). A higher proportion of participants on current sDMARDs and/or bDMARDs treatment presented painful masticatory muscle and TMJ at palpation compared to participants with no biologic treatment (Table 4).
Among participants with JIA, there were no signi cant differences in vertical unassisted jaw movement according to medication, mean 46.4 mm (SD 7.1) in the JIA group and 45.8 mm (SD 7.1) among those not using DMARDs (p=0.986). (Results not shown). However, in both groups more than half of the participants had a vertical unassisted jaw movement of more than 40 mm. The proportion without this medication treatment is slightly higher (82.7%) as compared to those on current sDMARDs and/or bDMARDs (77.4%).

Discussion
We have shown, using a comparative cross-sectional multicenter design, that around one third of participants with JIA in this cohort had TMD. Half of children and adolescents with JIA reported pain during jaw movements and pain on palpation of the masticatory muscles and TMJs as compared to one fourth of their healthy peers; palpatory pain was associated with sDMARDs and bDMARDs treatment, and children and adolescents with JIA had a signi cantly lower mean vertical jaw movement. Moreover, TMJrelated clinical signs and vertical jaw movement ≤40mm had the highest association to the JIA group.
The reported prevalence of TMD in children with JIA varies between 38% to 83% according to the de nitions and methods of ascertainment used, to the cohort examined and to differences in populations (15,(32)(33)(34)(35). Ferraz and colleagues, in their study of fteen children with JIA, age ranged between six and twenty-eight years and mean age 16.3 years, reported a high prevalence of 83%, however, without describing the method of ascertainment, eg. whether the gures were based on self-reporting or on clinical examination [29]. A previous study from Rongo and colleagues based on fty participants with JIA aged between nine and sixteen years, found the prevalence of TMJ damage from 100 joints to be 74% as assessed on magnetic resonance imaging (MRI) [34]. Others have reported prevalences of 55% based on a questionnaire [32] and of 72% based on clinical signs [33]. Both studies are not based on DC/TMD, and the children were older than those in our study. In contrast, a longitudinal study by Zwir et al, including seventy-ve children, revealed a prevalence of 38% based on symptoms and 47% based on clinical examination [30]. Their results are in line with ours.
In our study, the prevalence of TMD, either based on symptoms or clinical signs, in the healthy peers were quite high, 28 and 29%. It was higher than in earlier studies among adolescents, reported by Graue and collegues (7 and 12%) and Østensjø and collegues (7%) [9,10]. A combination of both symptoms and clinical signs for TMD, was more in line with the mentioned studies. Studies from Finland and Brazil con rm our results with a high prevalence of TMD in the normal population. Vierola et al [35] presented a TMD prevalence of 35% (mean age 7.9 years) and de Paiva Bertoli 34% (mean age 11.0 years) [36]. Reasons for the difference in TMD prevalence in the normal population of children and adolescents are probably due to the use of different diagnostic used tools, number of participants, age in the studied population, different countries involved and different study design. In studies from Norway, Graue and collegues [10] used two screening questions for pain related to TMD [11] and DC/TMD [26] form for symptoms and clinical signs in a population of 210 children and adolescents, age 12-19 years. Østensjø et al [9] used the same two screening questions of TMD symptoms [11] for screening in a population of 560 adolescents, age 13-19 years, and then a modi ed research diagnostic criteria for TMD (RDC/TMD) examination [37] in those that answered yes to 1) having pain in the temples, face, temporo-mandibular joints, or jaws once a week or more, and 2) having pain when open the mouth wide or chew once a week or more. The Finish group [35] used RDC/TMD [37] form for clinical signs on 483 children of 6-8 years and the Brasilian group [36] used the American Academy of Orofacial Pain [38] form for screening and RDC/TMD [37] for clinical examination on a population of 934 children between 10-14 years. It is clear that it can be di cult to get an exact gure on the prevalence of TMD in the normal population. A previous meta-analysis conducted by da Silva and colleagues revealed the overall prevalence of intraarticular joint disorder of 16% [39].
In our study, approximately half of the JIA subjects had clinical ndings consistent with TMD, with no differences according to JIA category. Since the numbers for three of the categories, including systemic arthritis, rheumatoid factor positive polyarthritis, psoriatic arthritis were relatively low, these results should be interpreted with caution.
he sensitivity and speci city of the clinical orofacial examination in relation to TMJ has been debated, as displacement of the disc, although eliciting a clicking sound, may be asymptomatic [40][41][42]. Based on the DC/TMD criteria, asymptomatic TMJ clicking is still de ned as TMD. However, several studies have shown that pain-free clicking represents a normal variant, typically seen in girls during puberty [16].
Recently, a clinical examination protocol for JIA was developed by the Temporomandibular joint juvenile arthritis working (TMJaw) group. This examination protocol focuses on three general items: TMJ symptoms, TMJ dysfunction, and dentofacial deformity in JIA and shows acceptable reliability and construct validity [7].
We found, in accordance with other studies, that the TMJ area and the masseter muscle region were common locations for pain in JIA [32]. However, a recent study from Koos and colleagues revealed a lower frequency of masticatory pain on palpation [16], and Kristensen and colleagues stated that masticatory pain complaints could develop over time [43]. In the present study, more than half of the participants with JIA revealed clinical signs in the TMJ region and in the masseter region, and more than one-fourth of the participants with JIA had TMD. A longitudinal multicenter approach could elucidate the development of masticatory muscle pain, as Kristensen and collegues have pointed out [43].
The vertical unassisted jaw movement has been widely used as a valid marker for TMJ arthritis [44]. We showed that participants with JIA have lower vertical movements as compared to their healthy peers, however, the differences were relatively small, questioning its clinical signi cance. Differently viewed, for children and adolescents aged <11 years, the cutoff value 40mm ranges as normal vertical jaw movement [45]. Further, our ndings suggest that lateral movement did not differ signi cantly between the two groups, which is in line with Twilt and colleagues [46] and by Küseler and colleagues (22). In the latter study, fteen children with JIA with a mean age of 12 years, recorded decreased lateral movements ≤ 5 mm with no signi cant relevance [20].
We found no statistically differences in the presence of TMD to JIA categories. However, we found a signi cantly higher occurrence of clinical signs in participants with JIA currently on DMARDs medication (whether synthetic or biologic) than in those off such medication. A high risk of developing clinical signs of TMD was associated with a severe disease course, indicated by the use of DMARDs.
The strengths of the study is the relatively large number of participants, that the study groups were well matched and the meticulous standardization of the clinical TMJ assessment performed prior to and during the study period. However, the large number of participants should not hide the fact that we are dealing with an underpowered sample size with a lack of 75 participants. There are some more limitations to the study. The overall response rate of 63%, although considered acceptable, might have in uenced the results because the group which not participated, on average, was slightly younger and had a slightly lower proportion of girls. Secondly, the shortened version of DC/TMD used in this study is not directly comparable with studies having used the full DC/TMD score. In the present study children and adolescents with JIA with TMD involvement were de ned based on self-reported pain and clinical examination. Further studies may focus on the role of imaging on the diagnosis of TMJ arthritis in children with JIA. This is particularly important in JIA with non-symptomatic TMJ involvement since hard tissue loss in the condyle may hinder the growth of the mandible and subsequently affect chewing function and cause aesthetic problem [16].

Conclusion
Symptoms or clinical signs of TMD were seen in approximately half of the participants with JIA compared to about one fourth of their healthy peers. Painful palpation to masticatory muscles and decreased vertical unassisted jaw movement are more frequent in children with JIA than in healthy controls and should be part of both medical and dental routine examinations in the follow-up of JIA.

Declarations
The datasets generated and analysed during the current study are not publicly available due to restrictions their containing information that could compromise the individual privacy of research participants but are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
The study was approved by the regional ethics committee (2012/542/REK vest). Written informed consents were obtained from the parents /legal representatives and the adolescents.The study was registered in ClinicalTrials.gov (No: NCT03904459).

Consent for publication
Not applicable.   Prevalence of TMD in children and adolescents with JIA vs. healthy peers, >10 years and <10 years of age, 1) symptoms: pain the last 30 days and pain at jaw movements; 2) clinical signs: pain at palpation of masticatory muscles and TMJ and 3) a combination of symptoms (1) and clinical signs (2).