High prevalence of temporomandibular joint arthritis at disease onset in children with juvenile idiopathic arthritis as detected by magnetic resonance imaging but not by ultrasound.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 58, No. 4, April 2008, pp 1189–1196 DOI 10.1002/art.23401 © 2008, American College of Rheumatology High Prevalence of Temporomandibular Joint Arthritis at Disease Onset in Children With Juvenile Idiopathic Arthritis, as Detected by Magnetic Resonance Imaging but Not by Ultrasound Pamela F. Weiss,1 Bita Arabshahi,2 Ann Johnson,2 Larissa T. Bilaniuk,2 Deborah Zarnow,2 Anne Marie Cahill,2 Chris Feudtner,1 and Randy Q. Cron3 arthritis was diagnosed in 69% by MRI and in 28% by US. Findings of both acute and chronic TMJ disease were detected by MRI in 53% of the patients. Of those with acute TMJ arthritis, 71% were asymptomatic, and 63% had normal findings on jaw examination. Fifty-six percent of patients with acute disease had an improved maximal incisal opening after corticosteroid injection. Among these responders, 56% had been asymptomatic and had normal jaw examination findings. Conclusion. TMJ arthritis was present in the majority of patients with new-onset JIA. Findings on MRI along with responses to treatment among asymptomatic patients with normal jaw examination findings suggest that a history review and physical examination are not sufficient to screen for TMJ disease. Our results also suggest that MRI and US findings are not well correlated, and that MRI is preferable for the detection of TMJ disease in new-onset JIA. Objective. To determine the prevalence of temporomandibular joint (TMJ) disease in a cohort of children with new-onset juvenile idiopathic arthritis (JIA), and to compare magnetic resonance imaging (MRI) with ultrasound (US) for the detection of acute and chronic changes of TMJ arthritis. Methods. Between January 2005 and April 2007, children with newly diagnosed JIA were prospectively evaluated for TMJ arthritis. Prior to imaging, jaw pain and disability were assessed with questionnaires and physical examination. The TMJs of all patients were imaged with MRI and US within 8 weeks of diagnosis. Results. Of the 32 patients enrolled, 78% were female, and the median age was 8.6 years (range 1.5– 17.2 years). Acute TMJ arthritis was diagnosed in 75% of the children by MRI and in none by US; chronic Supported by grants from the Ethel Brown Foerderer Fund for Excellence and the Nickolett Family Awards Program for JRA Research. 1 Pamela F. Weiss, MD, Chris Feudtner, MD, PhD, MPH: Children’s Hospital of Philadelphia and University of Pennsylvania Center for Clinical Epidemiology and Biostatistics, Philadelphia, Pennsylvania; 2Bita Arabshahi, MD (current address: Fairfax Hospital for Children, Fairfax, Virginia), Ann Johnson, MD, Larissa T. Bilaniuk, MD, Deborah Zarnow, MD, Anne Marie Cahill, MD: Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; 3Randy Q. Cron, MD, PhD: Immunology Graduate Group, University of Pennsylvania, Philadelphia, Pennsylvania. Dr. Cron has received speaking fees (less than $10,000) from the 52nd Annual Lowe Conference on Rheumatic Diseases, Nauvoo, Alabama. Address correspondence and reprint requests to Pamela F. Weiss, MD, Division of Rheumatology, Children’s Hospital of Philadelphia, Children’s Seashore House Room 236, 34th and Civic Center Boulevard, Philadelphia, PA 19104-4399. E-mail: weisspa@email. chop.edu. Submitted for publication August 3, 2007; accepted in revised form December 14, 2007. Juvenile idiopathic arthritis (JIA) affects ⬃1 of 1,000 children worldwide (1). Active arthritis of the temporomandibular joint (TMJ) in patients with JIA was recognized as early as 1897 (2), but until recently, it has been relatively ignored (3). The reported prevalence of TMJ involvement in patients with an established diagnosis of JIA varies widely (range 17–87%) (4), and the prevalence of TMJ involvement has not been reported previously in patients with a new diagnosis of JIA as determined by magnetic resonance imaging (MRI). Arthritis of the TMJ is a concern, particularly in patients who are actively growing, because the mandibular growth plate is located below the fibrocartilage and is therefore susceptible to damage from inflammation. Damage to the mandibular growth plate leads to asym1189 1190 WEISS ET AL metric growth, micrognathia, retrognathia, malocclusion, and a decreased maximal incisal opening (MIO) (5,6). In addition, the presence of TMJ arthritis in childhood may predispose patients to jaw pain and dysfunction in adulthood. In 2 recent longitudinal prospective studies, the prevalences of TMJ symptoms and jaw dysfunction were significantly higher in patients with JIA than in matched control subjects after 15 years and 26 years of followup, respectively (7,8). The detection of TMJ arthritis in patients with JIA is challenging, because the symptoms and clinical examination findings may be misleading. Pain with jaw excursion was found to be a significant predictor of TMJ disease in one study (9); however, pain at rest, crepitation, clicking, and swelling were not statistically significant predictors of TMJ disease (9). In another case series, it was reported that only 57% of patients with active disease, as detected by MRI, had jaw discomfort (10). Consequently, the diagnosis of TMJ arthritis has increasingly been based on evidence obtained by imaging. Ultrasound (US), MRI, and orthopantomogram have all been used in the evaluation and diagnosis of both active TMJ arthritis and chronic TMJ arthritis (11–14). In a prospective study of 66 subjects with TMJ internal derangement, the sensitivity, specificity, and accuracy of detection of chronic changes, effusions, and disk placement were compared between high-resolution US and MRI; with MRI as the gold standard, US had a sensitivity of 81% and a specificity of 100% for detection of effusions (15). A prospective study of 48 patients with JIA showed that 55% had destructive changes that were detected by US; effusions and pannus formation were not evaluated (14). Another group of investigators used US alone to assess TMJ involvement at the time of diagnosis of JIA and found that 65% of patients with JIA had effusions (16). We therefore sought to prospectively determine the prevalence of TMJ disease and to compare the ability of US with that of MRI for the detection of acute and chronic TMJ changes in a cohort of children with new-onset JIA. PATIENTS AND METHODS Protection of human subjects. The protocol for the conduct of this study was reviewed and approved by the Children’s Hospital of Philadelphia Committee for the Protection of Human Subjects. Patients. Any child (ⱕ17 years of age) with a diagnosis of JIA was eligible for participation in this study. Children with cardiac pacemakers, metal implants or clips, or dental braces were excluded. Forty-six patients with newly diagnosed JIA, ages 1–17 years, who attended the pediatric rheumatology clinic at Children’s Hospital of Philadelphia were prospectively enrolled between January 2005 and April 2007. For 3 of these patients, who were between the ages of 16 years and 17 years, the diagnosis did not meet the formal definition of JIA (required age at diagnosis ⬍16 years), but they were nevertheless included in the analysis. A total of 131 new diagnoses of JIA were made at the Children’s Hospital of Philadelphia during the recruitment period; however, not all providers actively recruited patients for enrollment. New patients are randomly assigned to attending physicians, and those patients who were enrolled were recruited irrespective of jaw symptoms or dysfunction. Of the 46 recruited patients, 2 were ineligible for the study because they subsequently received alternate diagnoses (1 inflammatory bowel disease, 1 pigmented villonodular synovitis), 2 were excluded from analysis because of sedation problems and incomplete imaging evaluation, and the families of 10 patients decided against participation after consent had been provided. Data collection. Prior to imaging, jaw pain and dysfunction in each patient were assessed with a parental questionnaire and a physical examination by the physician. Overall JIA activity and disability were assessed with parent’s and physician’s visual analog scales (range 0–100 mm), the Childhood Health Assessment Questionnaire (C-HAQ) (scale 0–3) (17), and a physical examination. The examination was performed by 1 of 2 physicians (PFW or BA) and consisted of evaluation of anterior translation, clicking, crepitus, pain with palpation and opening, asymmetric opening, micrognathia, and the MIO. Objective values for the MIO were obtained from direct measurements, using the Therabite Measuring Scale (Atos Medical, Milwaukee, WI); normal values for the MIO were obtained from a previously published study in which the MIO was determined in normal healthy children up to age 18 years (18). Accurate measurement of the MIO is difficult in children between the ages of 5 years and 7 years, due to the loss of deciduous teeth; only 3 patients in this study were in that age range, and none had missing deciduous teeth at the time of evaluation. The mean MIO was not significantly different between the patients who underwent a physical examination by either of the 2 physicians. Although corticosteroid injections were not part of the study protocol, a large proportion of the patients received intraarticular injections of corticosteroid for active TMJ arthritis; the majority of these patients underwent a repeat MIO measurement after injection. Imaging examination. All patients were screened with MRI and US of the TMJ within 8 weeks of diagnosis. MRI was performed on a 1.5T scanner utilizing a high-resolution surface coil. At the onset of the study, axial T1-weighted images were obtained at the level of the TMJs to determine the orientation of the mandibular condyles. The MRI evaluation included coronal T1- and T2-weighted images, sagittal T1-weighted and dual-echo T2-weighted images, and, following intravenous administration of gadolinium contrast material, sagittal and coronal fat-saturated T1-weighted images. All MR images of each TMJ were obtained with the patient in a closed-mouth position. A focused US evaluation of the TMJs was performed using a 12.5-MHz linear array transducer, and all evaluations DETECTION OF TMJ ARTHRITIS BY MRI VERSUS US Figure 1. Chronic and acute changes in the temporomandibular joints (TMJs) of patients with juvenile idiopathic arthritis, as detected by magnetic resonance imaging (MRI). Chronic changes in the TMJs are shown on T1-weighted sagittal MR images (A and B), involving condylar flattening (A), condylar erosions (arrows in A and B), and anterior disk displacement (B). Acute changes in the TMJs are shown on fat-saturated T2-weighted MR images (C and D), involving joint effusions in the anterior superior, inferior, and posterior superior compartments (arrows in C and D) and condylar erosions. C ⫽ condyle. were interpreted by a single radiologist (AJ) who is experienced in pediatric US. The US images were obtained in the coronal and axial planes, with the patient in either an openmouth or closed-mouth position. For detection by MRI, chronic TMJ arthritis was defined as condylar changes or erosions demonstrated on T1-weighted images (Figures 1A and B). Acute TMJ arthritis was defined as effusions or synovial thickening demonstrated on fat-saturated T2-weighted MR images (Figures 1 C and D). Synovial enhancement on MRI was not included as a criterion for acute arthritis, because experience from our institution has shown that normal healthy children (those without any identifiable TMJ disorder) have various degrees of TMJ enhancement on fat-saturated T1-weighted images following intravenous injection of contrast material (Weiss PF, et al: unpublished data). In addition, in one study of adult subjects, MR images from asymptomatic volunteers and from subjects with TMJ dysfunction were evaluated for evidence of TMJ effusions and condyle marrow alterations, with the results showing that marked effusions and condylar abnormalities were not detected in any of the asymptomatic volunteers (19). For detection by US, chronic TMJ arthritis was defined as condylar changes or erosions demonstrated on multiplanar imaging, in either an open-mouth or closed-mouth view. Acute TMJ arthritis was defined as fluid collection in either joint, 1191 detected on multiplanar US imaging in either an open-mouth or closed-mouth view. Radiologists at our institution have not routinely performed TMJ ultrasonography in normal healthy children; however, several investigators have previously reported US findings in the TMJs of normal healthy subjects (20–22). All results of MRI studies were read by an experienced neuroradiologist (LB), and 8 of the MRI studies were also read by a second neuroradiologist (DZ), to evaluate for interobserver agreement. Results of all US studies were read by an experienced ultrasonographer (AJ). Children younger than age 8 years were sedated intravenously for the MRI, using a combination of intravenous midazolam hydrochloride (0.1–0.3 mg/kg), fentanyl citrate (1–3 mg/kg), and pentobarbital sodium (2–4 mg/kg), as previously described (23). Treatment of active TMJ arthritis was not part of the study protocol. Patients with active TMJ arthritis were either referred for injection or followed up for symptoms, depending on the routine practice of the patient’s attending physician. Twenty-one of 24 patients with active TMJ arthritis underwent TMJ injection(s). The corticosteroid injections were performed by 1 of 4 experienced pediatric interventional radiologists, using either a 22-gauge or a 25-gauge needle, under computed tomography guidance as previously described (23). Once the position of the needle within the joint was confirmed, 0.5 cc (10 mg) of triamcinolone hexacetonide (Aristospan-20; SAB-Pharma, Lake Forest, IL) was injected. All patients were sedated with the same agents as used for MRI. Statistical analysis. Stata (version 9.2; StataCorp, College Station, TX) was used for all statistical analyses. Cohen’s kappa coefficients were used to compare interrater agreement for MRI interpretation and for agreement between MRI and US findings of acute and chronic TMJ arthritis (24). Cohen’s kappa coefficient is the preferred measure of agreement, as compared with simple percent agreement, since the kappa statistic accounts for agreement that may occur by chance. A kappa value of 1 denotes perfect observed agreement, and a kappa value of 0 indicates a lack of agreement other than by chance. RESULTS Demographic and clinical characteristics. Thirty-two patients were enrolled in the study. The frequency of JIA subtype, distributions of age and sex, and serologic findings, as shown in Table 1, were consistent with those observed in studies of larger JIA populations (25). Baseline and subjective symptoms. C-HAQ scores, used for the assessment of disease activity and disability, were low, with a median C-HAQ score of 0.3 (range 0–2) among the 32 patients with JIA. For overall disease severity, the median scores on the parent’s and physician’s visual analog scales were 41 mm (range 0–100) and 25 mm (range 0–67), respectively. Disease severity according to sex and JIA subtype is shown in 1192 WEISS ET AL Table 1. Characteristics of the new-onset juvenile idiopathic anthritis (JIA) study population at diagnosis* Characteristic Sex Female Male JIA subtype Oligoarticular Polyarticular, RF⫹ Polyarticular, RF⫺ Systemic Psoriatic ERA Other Positive serologic findings† ANA (n ⫽ 30) RF (n ⫽ 29) CCP (n ⫽ 22) HLA–B27 (n ⫽ 17) Overall disease severity No. (%) of patients Age, median (range) years Duration of symptoms, median (range) months 25 (78) 7 (22) 8.4 (1.9–17.2) 9.9 (1.5–13.8) 3 (0.8–18) 3 (0.5–12) 3 (1–46) 2 (1–32) 0.1 (0–1.5) 0.4 (0–2) 10 (0–61) 28 (0–67) 11 (34) 3 (9) 7 (22) 1 (3) 4 (13) 5 (16) 1 (3) 4.3 (2–14.6) 12.6 (8.4–16.5) 7.3 (1.5–13.3) 3.4 12.9 (9.9–14.6) 13.9 (2.9–17.2) 16.9 3 (0.5–8) 3 (3–8) 3 (1–12) 3 11.5 (1.5–18) 6 (2.5–7) 6 1 (1–4) 15 (4–25) 7 (1–46) 3 4 (1–32) 2 (1–2) 9 0 (0–1.6) 1.3 (0–1.6) 0.6 (0.3–2) 1.37 1 (0–1.5) 0.1 (0–0.9) 1.9 12 (0–31) 52 (0–66) 48 (11–61) 56 61 (0–67) 25 (0–49) 66 Number of joints with active disease, median (range) C-HAQ score, median (range) Physician VAS, median (range) mm 15 (50) 3 (10) 4 (18) 3 (18) * C-HAQ ⫽ Childhood Health Assessment Questionnaire; VAS ⫽ visual analog scale; RF ⫽ rheumatoid factor; ERA ⫽ enthesitis-related arthritis; ANA ⫽ antinuclear antibody; CCP ⫽ anti–cyclic citrullinated peptide. † Serologic testing was not performed on all patients; results are reflective of those in whom serologic findings were known. Table 1. TMJ symptoms were absent in 25 patients (81%) (Table 2). In the 6 patients with TMJ symptoms, Table 2. Jaw symptoms and physical examination findings of acute or chronic temporomandibular joint (TMJ) disease at diagnosis of juvenile idiopathic arthritis* Finding Subjective symptoms† None Any Clicking Locking Pain at rest Pain with chewing Physical examination‡ Normal Abnormal (any) MIO below normal Clicking Tenderness to palpation Tenderness with opening Absent translation Asymmetric opening Micrognathia Acute TMJ disease, no. (%) Chronic TMJ disease, no. (%) 25 (81) 6 (19) 4 (13) 1 (3) 0 (0) 5 (16) 17 (68) 6 (100) 16 (64) 6 (100) 19 (59) 13 (41) 7 (22) 2 (6) 4 (13) 4 (13) 2 (6) 6 (19) 0 15 (79) 9 (69) No. (%) of patients 11 (58) 11 (85) * Acute and chronic changes were defined by magnetic resonance imaging results. MIO ⫽ maximal incisal opening. † n ⫽ 31. ‡ n ⫽ 32. clicking and pain with chewing were the most common (Table 2). Objective signs of TMJ arthritis. Nineteen patients (59%) had normal findings on jaw examination, including an MIO in the normal range (Table 2). In contrast, 7 (22%) of the 32 patients had an MIO that was below the normal range for age-matched normal healthy subjects (17), and 6 (19%) had asymmetric jaw opening. Baseline medications. At the time of JIA diagnosis, the majority of patients were receiving nonsteroidal antiinflammatory agents for symptomatic relief; none of the patients were receiving disease-modifying agents. At the time of imaging evaluation, treatment consisted of Table 3. Agreement between magnetic resonance imaging (MRI) and ultrasound (US) for the detection of acute or chronic temporomandibular joint arthritis in patients with new-onset juvenile idiopathic arthritis Acute changes Unilateral Bilateral Chronic changes Unilateral Bilateral MRI, no. (%) of patients US, no. (%) of patients 24 (75) 4 (17) 20 (83) 22 (69) 7 (32) 15 (68) 0 0 0 9 (28) 5 (56) 4 (44) Agreement, % Kappa 23 0.00 50 0.12 DETECTION OF TMJ ARTHRITIS BY MRI VERSUS US nonsteroidal antiinflammatory agents (15 of 32 patients), methotrexate (12 of 32 patients, with a median treatment duration of 22.5 days [range 9–54 days]), and etanercept (3 of 32 patients, with a median treatment duration of 12 days [range 9–20 days]). Imaging findings. Acute TMJ arthritis was found in 24 patients (75%) by MRI, with the majority of these patients (20 [83%]) having bilateral disease. Acute TMJ arthritis was not found in any of the patients by US. Chronic TMJ arthritis was found in 22 patients (69%) by MRI and in 9 (28%) by US (Table 3). Findings of both acute and chronic disease were obtained in 17 patients (53%) by MRI. In comparing MRI and US findings, the 2 techniques had only 23% agreement and a kappa coefficient of 0 for the detection of acute TMJ arthritis, indicating that the agreement between techniques was no more than what would be expected by chance (Table 3). For the detection of chronic TMJ changes, findings from the 2 imaging techniques displayed 50% agreement and a kappa value of 0.12. All MR images were read by the same experienced pediatric neuroradiologist, and 8 (25%) of the randomly selected images were read by a second pediatric radiologist to evaluate for interobserver agreement. For any findings of acute or chronic disease obtained by MRI, there was 100% agreement and a kappa coefficient of 1.0 between the 2 radiologists. For detection of specific signs, the degree of agreement was lower; specifically, for effusion there was 75% agreement ( ⫽ 0.38), and for synovial thickening there was 62.5% agreement ( ⫽ 0.33). Utility of reported symptoms and examinationdetected abnormalities for imaging-defined acute TMJ arthritis. Table 4 shows the sensitivity, specificity, positive predictive value, and negative predictive value for the reported TMJ symptoms and for having abnormal physical examination findings in the detection of TMJ Table 4. Sensitivity, specificity, positive predictive, and negative predictive values of symptoms and abnormal physical examination findings for the detection of acute temporomandibular joint arthritis in new-onset juvenile idiopathic arthritis* Sensitivity Specificity Positive predictive value Negative predictive value Symptoms Abnormal physical examination findings 26 (10–48) 100 (63–100) 100 (54–100) 32 (15–54) 38 (19–59) 50 (16–84) 69 (39–91) 21 (6–46) * Values are the percent (95% confidence interval). 1193 Figure 2. Change in the maximal incisal opening (MIO) from pre– to post–corticosteroid injection in patients with juvenile idiopathic arthritis (JIA), by age category. Bars show the mean ⫾ SEM MIO in patients with JIA as compared with that in normal healthy subjects (see ref. 18). disease. TMJ symptoms were highly specific, but not sensitive, for acute TMJ arthritis. Physical examination findings were not predictive of either acute or chronic TMJ arthritis. Response to TMJ corticosteroid injection. Although not part of the study protocol, computed tomography–guided injections of corticosteroid were administered as part of routine care to 21 of the 24 patients with acute TMJ arthritis. In 16 of these 21 patients, the MIO was measured again at followup, after corticosteroid injection (median followup 2.4 months postinjection, range 1.4–7.3 months). Nine (56%) of 16 patients showed improvement in the MIO, with 5 (56%) of the 9 showing an increase in MIO of at least 0.5 cm. Among the 9 responders, 5 (56%) had been asymptomatic and had normal findings on jaw examination. Changes in the MIO from pre– to post–corticosteroid injection are shown by age category in Figure 2. In all categories except the youngest age range, the MIO in all patients was in the normal range after injection. As part of the routine followup evaluation, 5 of 21 patients who underwent TMJ injection(s) with corticosteroid were evaluated with a repeat MRI of the TMJs within 6 months postinjection. At this followup time point, 5 of 6 patients had MRI evidence of decreased joint effusions and/or decreased synovial thickening; the sixth patient had stable findings. Of the 5 patients with improvement on MRI, 2 had an improved MIO, 1 had no change in MIO, and 1 had a decreased MIO after injection. The sixth patient, whose MRI findings were 1194 WEISS ET AL stable, also had no change in the MIO after injection. No interval improvement in condylar changes or in erosions were noted with any of the evaluations. DISCUSSION The present study is the first to investigate the prevalence of MRI findings of acute and chronic TMJ arthritis in patients with JIA, detected at the time of diagnosis, and to directly compare MRI with US for detection of TMJ disease. TMJ arthritis as diagnosed by MRI was present in the majority of patients with newonset JIA. Seventy-five percent of our subjects had acute TMJ arthritis at the time of diagnosis of JIA, which is at the higher end of previously reported ranges of 17–87% (4,6,14). This prevalence at the time of JIA diagnosis indicates that the TMJ is one of the most commonly involved joints in children with JIA. Among those patients with TMJ arthritis, 83% had bilateral acute disease, a prevalence that is higher than the 53% reported in a recent study (6). Furthermore, 69% of patients with newly diagnosed JIA were demonstrated by MRI to have chronic TMJ arthritis, which is higher than the 55% reported in a study that used US to assess chronic destructive changes (14). More than half of the patients in our cohort had both acute and chronic findings diagnosed by MRI. Of the patients with acute TMJ arthritis, the majority were asymptomatic and had normal jaw examination findings. In addition, although treatment and followup of response were not part of the study, we have obtained data to suggest that corticosteroid injection is beneficial for acute disease, even in those patients who are asymptomatic. Among patients with acute disease who received corticosteroid injections and who underwent subsequent MIO measurements, the MIO improved in most, with half of the patients showing improvement of at least 0.5 cm. Even among patients with the greatest response, 4 of the 8 were asymptomatic and had normal jaw examination findings. Further studies are warranted to evaluate the response to corticosteroid injection. Given the limited ability to follow up TMJ arthritis by clinical examination and MIO measurements, future studies should assess the response using repeat imaging evaluations shortly after TMJ injection. Our results suggest that MRI and US findings are not well correlated, and that MRI has greater sensitivity for the detection of TMJ disease. The majority of recent studies utilizing high-resolution US for detection of TMJ disease in JIA have been from European centers that have considerable experience with US techniques (14,15,26). In the United States, ultrasonography is frequently used to identify synovial thickening and effusions in joints other than the TMJs, but experience with this technique in assessing the TMJs is limited. Given the bony anatomy and small size of the TMJs, it may be more difficult to master techniques and to adequately visualize effusions and synovial thickening. As experience with TMJ ultrasonography grows, the utility of the technique may increase, particularly given the lower costs and less invasive nature of such studies. In accordance with the results from prior studies, we found that subjective symptoms and abnormalities on physical examination were unreliable predictors of acute and chronic TMJ arthritis as detected by MRI. Among the patients with acute TMJ arthritis, 71% were asymptomatic, and 63% had normal jaw examination findings. The positive predictive value of symptoms was high (100%), indicating that if a patient reports symptoms of jaw pain or dysfunction, he or she is very likely to have active TMJ disease. The sensitivity of symptoms, however, was very low (26%), making it a poor screening tool for detection of acute TMJ disease. Similarly, abnormalities on jaw examination had a high positive predictive value (69%) and low sensitivity (38%). In addition, the negative predictive values for both symptoms and physical examination findings were low (32% and 21%, respectively), indicating that a lack of symptoms and a lack of abnormalities on examination did not preclude the presence of TMJ disease. Furthermore, a beneficial response to treatment among asymptomatic patients with normal jaw examination findings suggests that the history review and physical examination are not sufficient to screen for TMJ disease. Given the results of 2 recent longitudinal studies showing that the prevalences of TMJ dysfunction and pain were increased in adult patients with a history of JIA, the implications of early treatment of acute TMJ arthritis may be substantial (7,8). One group of investigators reported a high prevalence of TMJ involvement (67%) in women by an average of 26 years after diagnosis of oligoarticular or polyarticular JIA (8). In addition, detection of active TMJ disease by MRI in asymptomatic patients may be a predictor of arthritis extension; a recent study demonstrated that MRI findings of a clinically unaffected knee distinguished between patients with persistent monarthritis and those in whom extended arthritis developed (27). However, this theory would apply only to the small subset of patients DETECTION OF TMJ ARTHRITIS BY MRI VERSUS US with acute, and not chronic, TMJ arthritis detected by MRI. Given the high prevalence of TMJ disease at the time of JIA diagnosis, and given that TMJ disease is often undetectable by jaw examination and history review, universal screening with MRI for TMJ arthritis at the time that JIA is diagnosed should be considered. Several questions regarding effective treatment and ongoing disease surveillance, however, remain unanswered, especially since children with JIA-associated TMJ arthritis may be undergoing treatment for their JIA with both methotrexate and a tumor necrosis factor inhibitor (10). Specifically, the following issues should be considered. 1) Should children of all ages who have acute TMJ arthritis be treated with intraarticular corticosteroid injections? 2) Should patients with only chronic TMJ disease be similarly treated? 3) Should the treatment of TMJ arthritis vary depending on the concurrent systemic treatment of JIA? 4) Should a child with arthritis in 2 TMJs and 3 other joints with active disease be classified as having polyarticular arthritis? 5) How often and with what techniques should physicians screen for activity and resolution of TMJ disease? As the awareness of the frequency and severity of TMJ disease in children with JIA increases, we will need to improve our ability to screen, diagnosis, treat, and monitor TMJ arthritis. 1195 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. ACKNOWLEDGMENTS We thank Dr. David D. Sherry and Dr. Esi Morgan DeWitt for critical review of the manuscript. 15. AUTHOR CONTRIBUTIONS 16. Dr. Weiss had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Weiss, Arabshahi, Feudtner, Cron. Acquisition of data. Weiss, Arabshahi. Analysis and interpretation of data. Weiss, Johnson, Bilaniuk, Zarnow, Cahill, Feudtner, Cron. Manuscript preparation. Weiss, Johnson, Feudtner, Cron. Statistical analysis. Weiss, Feudtner. 17. 18. 19. REFERENCES 1. Schneider R, Passo MH. Juvenile rheumatoid arthritis [review]. Rheum Dis Clin North Am 2002;28:503–30. 2. Ronchezel MV, Hilario MO, Goldenberg J, Lederman HM, Faltin K Jr, de Azevedo MF, et al. Temporomandibular joint and mandibular growth alterations in patients with juvenile rheumatoid arthritis. J Rheumatol 1995;22:1956–61. 3. Arabshahi B, Cron RQ. Temporomandibular joint arthritis in 20. 21. juvenile idiopathic arthritis: the forgotten joint. Curr Opin Rheumatol 2006;18:490–5. Pedersen TK, Jensen JJ, Melsen B, Herlin T. Resorption of the temporomandibular condylar bone according to subtypes of juvenile chronic arthritis. J Rheumatol 2001;28:2109–15. Grosfeld O, Czarnecka B, Drecka-Kuzan K, Szymanska-Jagiello W, Zyszko A. Clinical investigations of the temporomandibular joint in children and adolescents with rheumatoid arthritis. Scand J Rheumatol 1973;2:145–9. Twilt M, Schulten AJ, Nicolaas P, Dulger A, van Suijlekom-Smit LW. Facioskeletal changes in children with juvenile idiopathic arthritis. Ann Rheum Dis 2006;65:823–5. Engstrom AL, Wanman A, Johansson A, Keshishian P, Forsberg M. Juvenile arthritis and development of symptoms of temporomandibular disorders: a 15-year prospective cohort study. J Orofac Pain 2007;21:120–6. Bakke M, Zak M, Jensen BL, Pedersen FK, Kreiborg S. Orofacial pain, jaw function, and temporomandibular disorders in women with a history of juvenile chronic arthritis or persistent juvenile chronic arthritis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001;92:406–14. Twilt M, Mobers SM, Arends LR, ten Cate R, van Suijlekom-Smit L. Temporomandibular involvement in juvenile idiopathic arthritis. J Rheumatol 2004;31:1418–22. Arabshahi B, Dewitt EM, Cahill AM, Kaye RD, Baskin KM, Towbin RB, et al. Utility of corticosteroid injection for temporomandibular arthritis in children with juvenile idiopathic arthritis. Arthritis Rheum 2005;52:3563–9. Rohlin M, Petersson A. Rheumatoid arthritis of the temporomandibular joint: radiologic evaluation based on standard reference films. Oral Surg Oral Med Oral Pathol 1989;67:594–9. Ingervall B. Range of movement of mandible in children. Scand J Dent Res 1970;78:311–22. Larheim TA. Role of magnetic resonance imaging in the clinical diagnosis of the temporomandibular joint. Cells Tissues Organs 2005;180:6–21. Jank S, Haase S, Strobl H, Michels H, Hafner R, Missmann M, et al. Sonographic investigation of the temporomandibular joint in patients with juvenile idiopathic arthritis: a pilot study. Arthritis Rheum 2007;57:213–8. Jank S, Rudisch A, Bodner G, Brandlmaier I, Gerhard S, Emshoff R. High-resolution ultrasonography of the TMJ: helpful diagnostic approach for patients with TMJ disorders? J Craniomaxillofac Surg 2001;29:366–71. Melchiorre D, Simonini G, Vierucci S, Giani T, Falcini F. Ultrasonography of the temporomandibular joint is helpful in the early diagnosis of the oligoarticular and monoarticular-onset juvenile idiopathic arthritis [abstract]. Arthritis Rheum 2005;52 Suppl 9:S87. Singh G, Athreya BH, Fries JF, Goldsmith DP. Measurement of health status in children with juvenile rheumatoid arthritis. Arthritis Rheum 1994;37:1761–9. Sheppard IM, Sheppard SM. Maximal incisal opening: a diagnostic index? J Dent Med 1965;20:13–5. Larheim TA, Katzberg RW, Westesson PL, Tallents RH, Moss ME. MR evidence of temporomandibular joint fluid and condyle marrow alterations: occurrence in asymptomatic volunteers and symptomatic patients. Int J Oral Maxillofac Surg 2001;30:113–7. Elias FM, Birman EG, Matsuda CK, Oliveira IR, Jorge WA. Ultrasonographic findings in normal temporomandibular joints. Braz Oral Res 2006;20:25–32. Vilanova JC, Barcelo J, Puig J, Remollo S, Nicolau C, Bru C. Diagnostic imaging: magnetic resonance imaging, computed tomography, and ultrasound. Semin Ultrasound CT MR 2007;28: 184–91. 1196 22. Stefanoff V, Hausamen JE, van den Berghe P. Ultrasound imaging of the TMJ disc in asymptomatic volunteers: preliminary report. J Craniomaxillofac Surg 1992;20:337–40. 23. Cahill AM, Baskin KM, Kaye RD, Arabshahi B, Cron RQ, Dewitt EM, et al. CT-guided percutaneous steroid injection for management of inflammatory arthropathy of the temporomandibular joint in children. AJR Am J Roentgenol 2007;188:182–6. 24. Cohen J. A coefficient of agreement for nominal scales. Educ Psychol Meas 1960;20:37–46. WEISS ET AL 25. Weiss JE, Ilowite NT. Juvenile idiopathic arthritis [review]. Pediatr Clin North Am 2005;52:413–42, vi. 26. Jank S, Emshoff R, Norer B, Missmann M, Nicasi A, Strobl H, et al. Diagnostic quality of dynamic high-resolution ultrasonography of the TMJ: a pilot study. Int J Oral Maxillofac Surg 2005;34:132–7. 27. Gardner-Medwin JM, Killeen OG, Ryder CA, Bradshaw K, Johnson K. Magnetic resonance imaging identifies features in clinically unaffected knees predicting extension of arthritis in children with monoarthritis. J Rheumatol 2006;33:2337–43.