3-D sonography for diagnosis of osteoarthrosis and disk degeneration of the temporomandibular joint, compared with MRI

Ultrasound in Med. & Biol., Vol. 32, No. 5, pp. 627– 632, 2006 Copyright © 2006 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/06/$–see front matter

doi:10.1016/j.ultrasmedbio.2006.01.014

● Original Contribution 3-D SONOGRAPHY FOR DIAGNOSIS OF OSTEOARTHROSIS AND DISK DEGENERATION OF THE TEMPOROMANDIBULAR JOINT, COMPARED WITH MRI CONSTANTIN A. LANDES,* WOJCIECH GORAL,* MARTIN G. MACK,† and ROBERT SADER* *Department of Oral Maxillofacial and Plastic-Facial Surgery; and †Department of Diagnostic and Interventional Radiology, Frankfurt University Medical Centre, Frankfurt, Germany (Received 22 September 2005, revised 10 January 2006, in final form 31 January 2006)

Abstract—This study determined the value of three-dimensional (3-D) sonography for the assessment of osteoarthrosis and disk degeneration of the temporomandibular joint (TMJ). Sixty-eight patients (136 TMJ) with clinical dysfunction were examined by 272 sonographic 3-D scans. An 8- to 12.5-MHz motor-angulated transducer positioned inferior-parallel to the zygomatic arch scanned the region-of-interest. 3-D condylar morphology was compared with subsequent magnetic resonance imaging (MRI). Fifty-three datasets were complete, i.e., 106 TMJ, 212 examinations. 3-D sonographic examination took 5 min and attained 70% sensitivity/76% specificity/75% accuracy; positive predictive value was 44%%; negative predictive value was 90%. Disk degeneration was diagnosed synonymously with 64%/73%/71%/42%/ 87%. 3-D sonography proved to be reliable for exclusion of osteoarthrosis as disk degeneration compared with MRI, whereas the presence of osteoarthrosis and disk dislocation cannot be reliably diagnosed. Prospective use will include routine screening, using more sophisticated equipment with higher frequency in real-time 3-D viewing. (E-mail: [email protected]) © 2006 World Federation for Ultrasound in Medicine & Biology. Key Words: Temporomandibular joint osteoarthrosis, Disk degeneration, Temporomandibular dysfunction, MRI, 3-D sonography, Four-dimensional sonography.

INTRODUCTION

cost hinder routine screening of TMD by MRI (depending on country, investigator and clinical setting). Two-dimensional (2-D) sonography permits fast reliable assessment of condylar translation and considerable 2-D information of hard tissue structures (Emshoff et al. 2003; Landes et al. 2000). TMJ sonography is comfortable to the patient, with low cost and high availability (Emshoff 2003; Landes 2004; Landes and Sterz 2003). This study evaluated the benefit of 3-D sonography in the diagnosis of TMJ osteoarthrosis and disk degeneration, and compared 3-D sonograms with MRI after a previous study concerning correct assessment of disk position (Landes et al. 2006).

Temporomandibular joint dysfunction (TMD) is a common disorder, and osteoarthrosis has been repeatedly associated with TMD (Stegenga et al. 1993), characterized by a deterioration of the articular surfaces and simultaneous remodeling of the underlying bone, including surface irregularity, underlying sclerosis, flattening and osteophyte formation. At the same time, the articular disk becomes flatter, scarified and eventually perforated (De Leeuw et al. 1995; Emshoff et al. 2003; Sokolow 1979). Magnetic resonance imaging (MRI) of the temporomandibular joint (TMJ) is widely used and yields fine anatomic detail in static examinations with high sensitivity, specificity and accuracy (Westesson et al. 1987). Rapid sequences permit reconstructed motion cycles (Eberhard et al. 2000); however, availability and

PATIENTS AND METHODS From July 2002 to May 2003, 68 patients with TMD (136 TMJ) had 3-D sonography. These were 44 females, 24 males, with ages ranging from 14 to 77 y, mean age 32 y, referred to the TMJ Clinic at our department (Landes et al. 2006). All reported pain and dysfunction of the temporomandibular area as their primary complaint. When patients agreed to the study procedure with

Address correspondence to: Dr. Dr. C. Landes, Oral Maxillofacial and Plastic-Facial Surgery, The Frankfurt University Medical Centre, Theodor-Stern-Kai 7, 60596 Frankfurt, Germany. E-mail: [email protected] 627

628

Ultrasound in Medicine and Biology

informed consent, sonography was performed singleblind, i.e., sonograms were analyzed, ignorant of the MRI result beforehand, by the first author. The first author’s results regarding disk position were reported to the second author for statistical evaluation, who then compared the results with those assessed by the radiologist in the identical joint. MRI examination was performed directly after the sonographic examination. To maintain comparable jaw position, the closed-mouth and maximum opening positions were used for comparison. All 3-D sonographic trials were performed with 3-D sonographic equipment (V530, General Electric-Kretz, Solingen, Germany). The transducer was an 8- to 12.5MHz linear array, moved by step motor. A 2-D positioning in axial transection of the target volume over the joint and lateral mandibular condyle was performed first, when the transducer was tilted to optimum visualization. The orientation of the transducer was standardized according to the standard planes of head and neck sonography parallel-inferior to the zygomatic arch (Siegert 1987). Therefore, the transducer intersected axially the lateral superior and inferior joint compartment. Sections parallel to the route of condylar translation were obtained by positioning the transducer in anterioposterior orientation parallel to the zygomatic arch (Landes et al. 2006). The examination was held in occlusion and maximal mouth opening. While sitting relaxed but erect, the patient moved the mandible from occlusion to maximal opening. Images were scrutinized for condylar surface irregularity, underlying sclerosis, flattening, osteophytes, effusion and thickness of the lateral articular space as a thickened or irregular joint capsule. The condyle was seen on its lateral aspect, together with the articular eminence as parallel hyperechogenic lines (Fig. 1). The MR images were judged independently by an experienced radiologist. A 1.5-T MR tomograph (Magnetom Vision; Siemens, Erlangen, Germany) with a dedicated surface TMJ coil was used to acquire simultaneously bilateral sagittal oblique and coronal oblique images. The imaging protocol included sagittal oblique and coronal oblique T1-weighted spin-echo (SE) sequences (450/15 [repetition time ms/echo time ms], imaging matrix 256 ⫻ 256, field of view 128 mm, pixel size 0.5 ⫻ 0.5 mm, slice thickness 3 mm) and sagittal oblique T2- and proton density-weighted turbo spin-echo (TSE) sequence (2840/103 to 15, matrix 512 ⫻ 512, field of view 128 mm, pixel size 0.3 ⫻ 0.3 mm, slice thickness 3 mm). All sagittal oblique sections were orientated perpendicular to the long axis of the condyle in transverse plane. The patient was positioned in supine position. Sequential bilateral images were obtained in the closed-mouth and maximum opening position. The MR images were interpreted without knowledge of the findings from the sonography. MRI diagnosis

Volume 32, Number 5, 2006

Fig. 1. (a) Shows the lateral condylar pole and articular capsule viewed as hyperechogenic bands, divided by the hypoechogenic disk in situ in a normal joint. Above left lies 2-D frontal, above right 2-D top and below left the 2-D side view. Below right shows the deliberately cut 3-D volume, not yet rotated for viewing; and (b) shows the identical picture with outlined condyle, disk and capsule.

of TMJ osteoarthrosis was defined by the presence of flattening, surface irregularities, erosion or presence of condylar deformities, osteophytes and subcondylar sclerosis. The statistical parameters sensitivity, specificity, accuracy, positive and negative predictive value of 3-D sonography were compared with MRI. The statistical significance was assessed by ␹2 test. RESULTS The datasets of 53 patients (106 TMJ) were complete. Nine (14% of 68 joints) MRI examinations and four (6%) sonographics dropped out, as the images were lost. Two (3%) patients developed claustrophobia during MRI and interrupted the procedure. One patient had communication problems and did not properly open her mouth on demand. Altogether, 212 of 272 images were analyzed (105 in the closed-mouth position, 107 in the open-mouth position). The 3-D sonography revealed 16 of 23 osteoarthroses, verified on MRI. 3-D sonography therefore evinced 70% sensitivity to detect osteoarthrosis if present on MRI. Specificity for osteoarthrosis was 76% and accu-

3-D TMJ-sonography for osteoarthrosis ● C. A. LANDES et al.

629

Table 1. Cross-table of 3-D sonographic vs. MRI findings, regarding osteoarthrosis Osteoarthrosis ⫹ 3-D sonography

–

⫹

16 True positives (TP)

20 False-positives (FP)

All with positive test 36

⫺

7 False-negatives (FN)

63 True negatives (TN)

All with negative test 70

All with osteoarthrosis 23

All without osteoarthrosis 83 Specificity ⫽ TN/(FP ⫹ TN) ⫽ 0.76

Everyone TP ⫹ FP ⫹ FN ⫹ TN Pre-test probability ⫽ (TP ⫹ FN)/(TP ⫹ FP ⫹ FN ⫹ TN) ⫽ 0.22

Sensitivity ⫽ TP/(TP ⫹ FN) ⫽ 0.70

racy 75% (see Tables). Thirty-six TMJ diagnosed with osteoarthrosis by 3-D sonography (Figs. 2 and 3) were confirmed by MRI in 16 instances (positive predictive value 44%). Of the 70 TMJ with 3-D sonographic diagnosis of normal morphology, seven had osteoarthrosis on MRI (negative predictive value 90%). To detect disk degeneration if present on MRI, 3-D sonography correctly diagnosed 16 of 25 positives, therefore evincing 64% sensitivity. Specificity for disk degeneration was 73% and accuracy 71%. Thirty-eight TMJ diagnosed with disk degeneration by 3-D sonography were confirmed by MRI in 16 instances (positive predictive value 42%). Of the 68 TMJ with 3-D sonographic diagnosis of normal morphology, nine had osteoarthrosis noted on MRI (negative predictive value 87%). For the diagnosis, it proved valuable in 3-D sonography to cut the data block in several transections. Initially an axial cut just transecting the upper condylar pole and disk was preferred, which was later substituted by a coronary and a sagittal cut. The latter sagittal aspect finally became the most important for diagnosis of anterioposterior condylar degeneration. The 3-D sonographic results did differ significantly from the MRI results in ␹2 testing, p ⫽ 0.003

Positive predictive value ⫽ TP/(TP ⫹ FP) 16/(16 ⫹ 20) 0.44 Negative predictive value ⫽ TN/(FN ⫹ TN) 63/(7 ⫹ 63) 0.9

altogether, p ⫽ 0.008 for osteoarthroses, p ⫽ 0.008 for disk degeneration, at a level of significance of ␣ ⫽ 0.05. DISCUSSION Close association of TMD with condylar and temporal fossa degeneration, i.e., joint osteoarthrosis and disk degeneration has been repeatedly emphasized (Stegenga et al. 1993). Mejersjö and Hollender (1984) and Wiberg and Wänman (1998) reported that, in the TMD patient population, between 30% and 39% of TMJ were affected. However, our study found merely 22% (23 of 106) of TMJ to be affected by degeneration. Splint therapy has been reported to be successful (Okeson et al. 1988; Truelove et al. 1992) for treatment of clinical symptoms; however, in cases of progressing arthritis, arthroscopic lavage, lysis (Clark et al. 1991) and arthroplasty (Ericsson and Westesson 1992) are performed today. MRI, conventional tomography and computed tomography (CT) of the soft and hard tissue components of the TMJ are not available in most TMD clinics today. Conventional tomography and CT, moreover, include radiation exposure and are therefore unsuitable for re-

Table 2. Cross-table of 3-D sonographic vs MRI findings regarding disk degeneration Disk degeneration ⫹ 3-D sonography

–

⫹

16 True positives (TP)

22 False-positives (FP)

All with positive test 38

⫺

9 False-negatives (FN)

59 True negatives (TN)

All with negative test 68

All with disk degeneration 25 Sensitivity ⫽ TP/(TP ⫹ FN) ⫽ 0.64

All without disk degeneration 81 Specificity ⫽ TN/(FP ⫹ TN) ⫽ 0.73

Everyone TP ⫹ FP ⫹ FN ⫹ TN Pre-test probability ⫽ (TP ⫹ FN)/(TP ⫹ FP ⫹ FN ⫹ TN) ⫽ 0.24

Positive predictive value ⫽ TP/(TP ⫹ FP) 16/(16 ⫹ 22) 0.42 Negative predictive value ⫽ TN/(FN ⫹ TN) 59/(9 ⫹ 59) 0.87

630

Ultrasound in Medicine and Biology

Volume 32, Number 5, 2006

Table 3. Diagram of true and false-positives-negatives for osteoarthrosis and disk degeneration osteoarthrosis

disk degeneration 63

16

20

16

59

22 7

true positive

false positive

peated follow-up examinations in a chronic disease such as TMJ arthritis. Cost and availability limit the use of MRI in many countries; most authors therefore manage TMD based on the clinical symptoms. Several studies have been conducted to determine the accuracy of clinical and adjunctive diagnostic tests; the accuracy rates obtained for TMJ disorders ranged from 43% to 90% (Anderson et al. 1989; Roberts et al. 1991; Brady et al. 1993). From this relative clinical imprecision in the literature, a screening sonography with the low positive predictive value and high negative predictive value as found in this study seems a significant advantage to exclude osteoarthrosis and disk degeneration in a fast screening examination. Thus, an MRI could prospectively become scheduled if a chairside 3-D sonograph cannot exclude degenerative joint alteration. 3-D sonography may therefore eventually be performed at the first

Table 4. Diagram of 3-D sonographic accuracy, sensitivity, specificity, positive predictive value and negative predictive value in diagnosing osteoarthrosis and disk degeneration versus MRI 100% 90%

osteoarthrosis

disk degeneration

80% 70% 60% 50% 40% 30% 20% 10% 0% accuracy

sensitivity

specificity

positive predictive value

negative predictive value

true negative

9

false negative

examination. On the other hand, the 3-D sonography does not yet yield the expected results and reliability of the MRI examination, and therefore the differences within the results are significant. TMJ sonography is not yet ready to replace MRI diagnosis but may become applied as a screening examination to exclude osteoarthrosis and disk degeneration and therewith to reduce patient strain and economize the treatment of TMD. From the high percentages of osteoarthrosis found (Brooks et al. 1992; Westesson et al. 1987; Wiberg and Wänman 1998) in asymptomatic volunteers, it seems probable that only a low percentage of the general population either has symptoms or seeks treatment for symptoms of osteoarthrosis. Asymptomatic patients incorrectly diagnosed with osteoarthrosis may incur unnecessary and potentially damaging treatment. Average sensitivity levels and high specificity should therefore avoid false-positive diagnoses and treatment. 2-D ultrasonography has been advocated for diagnosis of osteoarthrosis (Emshoff et al. 2003). Sonography was reported to have good delineation of anatomic detail in 2-D application. The examination is also fast and comfortable to the patient; disregarding the initial outlay for the sonographic equipment, it is an economic alternative, although only advanced equipment provides 3-D reconstruction today. In this study, prospective interpretation of 3-D sonography to the absence of osteoarthrosis was rated at 90%, and 87% for disk degeneration compared with 79% and 83% in 2-D sonography (Emshoff et al. 2003; Landes et al. 2000). These accurate results for diagnosis of absence of joint or disk degeneration make 3-D sonography appear to be valuable in the exclusion of osteoarthrosis and disk degeneration in routine screening in a TMD clinic. The data further may become more accurate for diagnosing the presence of osteoarthrosis and disk degeneration once the 5- to 12-MHz

3-D TMJ-sonography for osteoarthrosis ● C. A. LANDES et al.

631

entire length of the condylar surface both transversely and especially longitudinally to look for the morphology of the condyle. Future refined equipment will yield even more convincing images. The more recent equipment (see mpeg file at http://www.constantinlandes.net) allowed continuous depiction. However this equipment was not yet available throughout this evaluation and will be prospectively compared with rapid MRI sequence-reconstructed motion cycles. Arthroscopic or surgical confirmation of imaging results was not available except in three patients, when

Fig. 2. The lateral condylar pole and articular capsule are less regular in this example of osteoarthrosis rotated for 3-D sagittal viewing; the native picture is shown left; the outlined structures of the disk, condyle and capsule are shown on the right: (a) closed-mouth; and (b) open-mouth position.

probe becomes replaced by a 12- to 14-MHz probe. The higher frequency should permit higher diagnostic efficacy, because of defined tissue differentiation and near-field clarity than lower frequency ranges. Moreover, real-time “4D”–3-D sonography as the next step is currently under evaluation (mpeg movie files can be downloaded at http:// www.constantinlandes.net). There are reported pitfalls (Emshoff et al. 2003) in diagnosing osteoarthrosis on the basis of findings from the static 2-D sonographic evaluation. If the transducer is not parallel (Landes et al. 2000) or perpendicular (Emshoff et al. 2003) to the TMJ, the condyle becomes hypoechogenic, resulting in false-negative results. A suboptimal angulation of the transducer makes the correctly positioned disk disappear from the sonographic picture in 2-D linear B-scan. If the beam does not intersect the disk at its lateral portion in the axial plane, higher echoes result from artifactual echoes between the osseous surfaces (e.g., echoes that were reflected repeatedly between osseous surfaces: Malzer 1992). In contrast the tissue block obtained by 3-D sonography could freely be cut in levels. This made the diagnosis of the entire condylar head for complex deformation easier and 3-D in-depth view of the hypoechogenic disk was possible. The main reasons for false-positive interpretations are found in still low or limited sonographic resolution to surface irregularities and normal condylar morphology present. Figures 2 and 3 are therefore realistic pictures that were evaluated in this study. It is important to scan the

Fig. 3. Shows a degenerated condyle. (a) 3-D sagittal; (b) 3-D frontal with only moderate disk alterations, but a marked condylar osteophyte; (c) shows the identical joint in open position in 3-D sagittal; and (d) the 3-D frontal view.

632

Ultrasound in Medicine and Biology

all explorations supported the sonographic findings. False-negative and false-positive imaging findings should prospectively become definitely assessed by surgical exploration. However, postmortem histologic cryosections have shown a high correlation of MRI findings and therefore it seems justified to correlate to MRI as reference standard (Tasaki et al. 1993). When osteoarthrosis, according to above criteria, has been observed in normal asymptomatic volunteers, this questions what should be considered to be abnormal condyle morphology. In terms of “disease classification,” a more accurate diagnostic operational criterion to define TMJ condylar osteoarthrosis that is closely related to the clinical signs and symptoms as pain and discomfort is needed, to avoid over- and undertreatment and to obtain a more cost-effective outcome. The observers have been involved with TMJ sonography and MRI throughout 5 y, and the learning curve has entered a steady slope; however, individual observer variation can influence the results. Retest examinations have shown a 96% concordance. MRI stays partly artificial in examination setting, and an anxious patient may exhibit stronger dislocation as a consequence of bruxism while in the prone position. The principal advantage of 3-D sonography was to obtain a complete overview of the condyle, not a single transection scanning the condyle perpendicular. This made the interpretation much more feasible because deliberate cuts could be placed within the scan volume. The angulation did not have to be kept precisely. Yet the upper margin of the 3-D volume selected from the 2-D pilot picture can only be adjusted as a straight line. It cannot yet become modified in nonlinear fashion to follow the slope of the condyle in real-time mode. This could prospectively reproduce the translation of an identical point on the condyle in motion. Alternatively, the condyle and disk interaction could become real-time visualized at a constant anatomical height relative to the condyle. Sonography, despite its limitations, is an ideal method for evaluating the condyle and its abnormalities. Prospectively, real-time 3-D sonography will be compared with higher frequency and resolution with MRI rapid-sequence motion reconstructions. REFERENCES Anderson GC, Schiffman EL, Schellhas KP, Fricton JR. Clinical vs. arthrographic diagnosis of TMJ internal derangement. J Dent Res 1989;68:826 – 829. Brady AP, McDevitt L, Stack JP, Downey D. A technique for magnetic resonance imaging of the temporomandibular joint. Clin Radiol 1993;47:127–133. Brooks SL, Westesson PL, Eriksson L, Hansson LG, Barsotti JB. Prevalence of osseous changes in the temporomandibular joint of asymptomatic persons without internal derangements. Oral Surg Oral Med Oral Pathol 1992;73:118 –122.

Volume 32, Number 5, 2006 Clark GT, Moody DG, Sanders B. Arthroscopic treatment of temporomandibular joint locking resulting from disc derangement: Twoyear results. J Oral Maxillofac Surg 1991;49:157–164. De Leeuw R, Boering G, Stegenga B, de Bont LGM. Radiographic signs of temporomandibular joint osteoarthrosis and internal derangement 30 years after nonsurgical treatment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;79:382–392. De Leeuw R, Boering G, van der Kuijl B, Stegenga B. Hard and soft tissue imaging of the temporomandibular joint 30 years after diagnosis of osteoarthrosis and internal derangement. J Oral Maxillofac Surg 1996;54:1270 –1280. Eberhard D, Bantleon HP, Steger W. Functional magnetic resonance imaging of temporomandibular joint disorders. Eur J Orthod 2000; 22:489 – 497. Emshoff R, Bertram S, Rudisch A, Bodner G, Ga␤ner R. Temporomandibular joint osteoarthrosis diagnosed with high resolution ultrasonography versus magnetic resonance imaging: How reliable is high resolution ultrasonography? J Oral Rehabil 2003;30:812– 817. Eriksson L, Westesson PL. Temporomandibular joint discectomy. Oral Surg Oral Med Oral Pathol 1992;74(3):259 –272. Helkimo M. Epidemiological surveys of dysfunction of the masticatory system. Oral Sci Rev 1976;7:54 – 69. Kircos LT, Ortendahl DA, Mark AS, Arakawa M. Magnetic resonance imaging of the TMJ disc in asymptomatic volunteers. J Oral Maxillofac Surg 1987;45:852– 854. Landes C, Walendzik W, Klein C. Sonography of the temporomandibular joint from 60 examinations and comparison with MRI and axiography. J Craniomaxillofac Surg 2000;28:352–361. Landes CA, Sterz M. Evaluation of condylar translation by sonography versus axiography in orthognathic surgery patients. J Oral Maxillofac Surg 2003;61:1410 –1417. Landes CA. Proximal segment positioning in bilateral sagittal split osteotomy: Intraoperative dynamic positioning and monitoring by sonography. J Oral Maxillofac Surg 2004;62:22–28. Landes CA, Goral WA, Sader R, Mack MG. 3-D sonography for diagnosis of disk dislocation of the temporomandibular joint compared to MRI. Ultrasound Med Biol 2006;32:633– 639. Malzer U. Schlierenoptische Untersuchungen am isolierten Kniegelenksmodell. Marburg:Germany, 1992 Videoproduktion Fa. Siemens. Mejersjö C, Hollender L. TMJ pain and dysfunction: Relation between clinical and radiographic findings in the short and long-term. Scand J Dent Res 1984;92:241–248. Okeson JP. Long-term treatment of disc-interference disorders of the temporomandibular joint with anterior repositioning occlusal splints. J Prosthet Dent 1988;60:611– 615. Roberts C, Katzberg RW, Tallents RH, Espeland MA, Handelman SL. The clinical predictability of internal derangements of the temporomandibular joint. Oral Surg Oral Med Oral Pathol 1991;71:412– 414. Siegert R. Ultrasonic imaging of the head and neck region. Dtsch Z Mund KieferGesichtschir 1987;11:67– 69. Sokolow L. Pathology and pathogenesis of osteoarthrosis. In: McCarthy DJ, ed. Arthritis and Allied Conditions: A Textbook of Rheumatology. Philadelphia, PA: Lea & Febiger, 1979;1135–1142. Stegenga B, de Bont LG, de Leeuw R, Boering G. Assessment of mandibular function impairment associated with temporomandibular joint osteoarthrosis and internal derangement. J Orofac Pain 1993;7:183–195. Tasaki MM, Westesson PL. Temporomandibular joint: Diagnostic accuracy with sagittal and coronal MR imaging. Radiology 1993;186: 723–729. Truelove EL, Sommers EE, LeResche L, Dworkin SF, Von Korff M. Clinical diagnostic criteria for TMD: New classification permits multiple diagnosis. J Am Dent Assoc 1992;123:47–54. Westesson PL, Katzberg RW, Tallents RH, Sanchez-Woodworth RE, Svensson SA. CT and MR of the temporomandibular joint: Comparison with autopsy specimens. AJR Am J Roentgenol 1987;148: 1165–1171. Wiberg B, Wänman A. Signs of osteoarthrosis of the temporomandibular joints in young patients. A clinical and radiographic study. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;86:158 –164.