Comparison of computed with conventional tomography in the evaluation of temporomandibular joint disease: a study of autopsy specimens K. Tanimoto,· A. Petersson,t M. Rohlin,t L.G. Hansson'" and C.C. Johansen'" "Department of Oral Radiology, Hiroshima University, School of Dentistry, Hiroshima, Japan, tDepartment of Oral Radiology, Lund University, School of Dentistry, Malmo, Sweden and :j:Central Department of Diagnostic Radiology, University Hospital of Lund, Lund, Sweden

Received 13 March 1989 and in final form 20 May 1989 Autopsy specimens were examined both radiographically and macroscopically to compare direct computed tomography (CT) with conventional tomography (Cv"T) for their diagnostic yield of the structural bone changes in the temporomandibular joint (TMJ). Fifteen specimens were examined with corrected sagittal and 12 with corrected frontal tomography. Five joints showed bone exposure and/or disc perforation macroscopically and all of them were correctly diagnosed by both CT and cvr (sensitivity 1.0). However, false-positive diagnoses were also made, resulting in values for specificity which were somewhat lower for CT than for cvr. Frontal images did not reveal any additional 'true' findings compared with sagittal CT or cvr. For single areas of the joints the average sensitivity was low, 0.28 for CT and 0.47 for cvr, but the specificity for the joint as a whole was high with both methods (0.91 and 0.94 respectively). We conclude that cvr is superior to cr in the diagnosis of single structural bone changes but comparable for comprehensive diagnosis of TMJ disease. Keywords: Osteoarthritis; temporomandibular joint diseases; tomography, X-ray; tomography, X-ray computed

Introduction

Materials and methods

Emphasis on the possible use of computed tomography (CT) in the examination of the temporomandibular joint (TMJ) has centred mainly on the disc l - 22 and Helms et al.9 concluded that it could replace arthrography in the diagnosis of disc displacement. On the other hand, the advantage of CT is that it provides information on both the bony and soft tissue components in a single examination, and several studies have reported its application for degenerative changes 4 ,'10 , 13,20,23,24. Conventional TMJ radiography has an established role in the detection of structural bone changes and sagittal tomography has been shown to yield the most information (see, for example, Omnell and Peterssorr"). It would therefore be of interest to compare cr and conventional tomography (Cv'T). However, Maue-Dickson et al. 26 state that tomographic images obtained by these two techniques are difficult to compare because of the difference in contrast levels and section thickness. An alternative method is to compare them in a situation similar to clinical practice but with access to proof either of normality or pathology. This study was therefore undertaken, first, to compare cr with cvr for their diagnostic yield of structural bone changes in TMJ autopsy specimens, and, second, by relating the radiographic findings to the macroscopic evidence, to determine the sensitivity and specificity of the two radiographic methods from actual knowledge of the joint status.

Material Fifteen right TMJs were removed as 6 em" blocks from fresh cadavers. Nine were females (mean age 76; range 63-86 years) and six males (mean age 67; range 38-82 years). No information about their TMJ status prior to death was available. Computed tomography (CT) An acrylic holder was used to fix the specimens and direct sagittal and frontal CT performed. The TMJ was oriented with the tomographic plane perpendicular to the long axis of the condyle, i.e. corrected sagittal CT. Corrected frontal CT images parallel to the condylar long axis were then obtained by rotating the positioning device through 90° without displacing the specimen. The CT examinations were carried out with a wholebody scanner (TCT-80A, Toshiba Medical Systems, Tokyo, Japan) using 120 kV, 350 mA, a scan time of 9 s and a 320x320 reconstruction matrix. The slice thickness was 2 mm, with the slices 3 mm apart. Approximately 7-10 slices were obtained from the sagittal cr and 5-7 from the frontal. Images of at least every second slice were made with bone-detail target algorithms. Target scans with a 2x geometric enlargement and a scanning circle with a diameter of 120 mm were performed. The scans were evaluated on an off-line computer console with a 4x geometric enlargement and the optimal method of viewing the scan was determined by

© 1990 IADMFR 02S0-832x 90/01 002I-m

Dentomaxillofac. Radiol., 1990, Vol. 19, February

21

Computed and conventional tomography of TMJ disease: K. Tanimoto et al. two observers (K.T. and L.G.H.). The frontal CT images of three joints were inadequate for evaluation because of technical errors. The findings were recorded on a special form for the lateral, central and medial third of the areas listed in Table IV: thus, nine areas of the condyle and 15 areas of the temporal component of every joint were assessed. In addition, each of these scans was recorded on film for a third observer (A.P.) to evaluate.

Conventional tomography (CVT) The specimens were attached to the positioning device with the TMJ oriented with the tomographic plane perpendicular to the long axis of the condyle, i.e. corrected sagittal tomography". Corrected frontal tomographic images were made parallel to the condylar long axis by rotating it through 90°27. Tomography was performed with a Polytome U (Massiot/Philips, Paris, France) with a Bi 125/30/50R tube (Siemens, Erlangen, FRG), nominal focus size 0.6 mmxO.6 mm, focus-film distance 1.5 mm and a magnification factor of 1.3. A

hypocydoidal movement was used with a tomographic angle of 48° without a grid. The circular collimator was 1.5 em diameter and the field size diameter in the film plane 6.5 cm. Exposure factors were 60 kVp, 20-30 rnA and 6 s. A multi-film cassette with five pairs of CaW04 screens (the last pair Saphir and the others Rubin, Siemens-Elema, Sweden) and five films (CEA, Strangnas, Sweden) was used. The interspace between each tomographic section was 2.5 mm, and the cassette had an overall thickness of 10.0 mm. Four exposures were made for each TMJ specimen, two for sagittal and two for frontal tomography. The tomographic unit was moved 10 mm between the exposures for each projection so that the images covered the full mediolateral and anteroposterior extent of the TMJ. Tomograms were selected with the same anatomical appearance of the specimens as in the CT. The tomograms were evaluated by two of the observers independently (K.T. and A.P.) and, in cases of disagreement, by a third observer (M.R.) to reach a consensus.

Figure 1 An example of a true-positive diagnosis of an erosion (arrow) in the superior area of the condyle in sagittal radiological finding was confirmed by macroscopic examination (c)

22 Dentomaxillofac. Radiol., 1990, Vol. 19, February

cr (a) and CVT (b). The

Computed and conventional tomography of TMJ disease: K. Tanimoto et al. Interpretation The CT images and conventional tomograms were assessed for the following features: Concavity - a hollowed-out area on the bony surface of the joint with a well-defined cortical outline. Cyst - a well-defined, localized area of bone destruction beneath an intact cortical outline of the joint surface. Erosion - a localized area of decreased density of the joint surface and adjacent subcortical bone (Figure la, b). Flattening - a flat bony contour deviating from the convex form. Osteophyte - a marginal bony outgrowth. Sclerosis - a localized area of increased density of the cortical bony joint surface extending into the subcortical bone.

A comprehensive diagnosis, based on a subjective evaluation of the entire joint, was also made as follows: Normal - joint without any of the above features or only concavities.

Remodelled - joint with only flattening, sclerosis or a small osteophyte. Pathological - joint with erosion, cyst or a large osteophyte (Figure Ia, b).

The correspondence rate (Table III) was calculated for 24 areas (nine for the condyle and 15 for the temporal component) of each joint and the six radiographic findings (concavity, cyst, erosion, flattening, osteophyte, sclerosis). For sagittal radiography it was based on 15 and for frontal on 12 joints. It was defined from the following equation: Correspondence rate= positive results of both methods

+

negative results of both methods total number of areas Macroscopic examination The joint components were separated by opening the lower and upper joint compartment laterally, anteriorly and medially. The articular surface of the components

Figure 2 An example of a false-positive diagnosis. In both sagittal cr (a) and CVT (b), an erosion was diagnosed in the superior area of the condyle (arrow). Macroscopically, the lateral half of the condyle was assessed to have no changes and the medial half to have superficial flaking (c)

Dentomaxillofac. Radiol., 1990, Vol. 19, February 23

Computed and conventional tomography of TMJ disease: K. Tanimoto et al. was examined for any deviation in form and degenerative changes of sufficient severity to be readily visible to the naked eye and their position and extent were recorded by two of the authors (A.P. and M.R.) on a standardized map. Degenerative changes in the condyle, disc and temporal component were classified into four stages, modified from Byers et al.28 : O=no visible changes (Figure 2c). l=superficial flaking or fraying (Figure 2c). 2=localized reduction in the soft tissue thickness. 3=total loss of soft tissue with bone exposure and/or perforation of the disc (Figure lc).

Determination of sensitivity and specificity and CVT were evaluated for the ability to demonstrate macroscopic degenerative changes of Stages 2 and 3, on the basis of both the comprehensive diagnosis of the whole joint and for the different areas. The sensitivity and specificity were defined as detailed in Table 9 •

cr

e

cvr

Table I Sensitivity and specificity of IT and

Macroscopic examination Radiographic examination

Joint or area with macroscopic changes

Joint or area without macroscopic changes

Positive Negative

a (true-positive) c (false-negative)

b (false-positive) d (true-negative)

Sensitivity: the true-positive rate is the proportion of joints/areas with disease which had a positive test result a/(a+c). Specificity: the true-negative rate is the proportion of joints/areas without disease which had a negative test result d/(d+b).

Results Radiograpic examination The number of radiograpic abnormalities is shown in Table II. Thirty-seven were detected by sagittal CT and 36 by sagittal CVT. Twenty-two of the former were in the condyle and 15 in the temporal component, whereas those detected by the latter were as frequent in

the condyle as in the temporal component (19 compared with 17). In the frontal tomograms, the only findings were on the superior surface of the condyle; 16 were seen on CT and five by CVT. Out of the 16 findings on frontal cr, only six were also seen in the sagittal CT images, while only one out of five features found in frontal CVT images was also detected by sagittal CVT. The different types of radiographic pathology are also presented in Table II. Erosion was the most frequent finding, whereas only one cyst and a few concavities were observed.

Comparison between CT and CVT Table III compares the findings from cr with CVT. The average correspondence rate for sagittal cr and CVT was 98.8%. In the condyle, only four positive findings from CVT were not detected by cr and seven from cr not by CVT. With regard to the temporal component, nine positive findings from CVT were not detected by CT and seven from CT not by CVT. In frontal radiography, the average correspondence rate between frontal cr and CVT was 98.2% (Table III). In the condyle, four out of five findings from CVT were detected by cr but 12 cr findings were not revealed by CVT. Comparison of macroscopic and radiographic examinations In the condyle, 10% of the area showed macroscopic changes of Stage 3 and 24% of Stages 2+3 (Table IV). The corresponding figures for the temporal component were 8% for Stage 3 and'12% for Stages 2+3. One third of the joints had bone exposure and/or perforation of the disc. The most frequent radiographic finding consistent with a macroscopic Stage 3 change was erosion (Table II). The topographic distribution of areas with macroscopic changes of Stages 2 and 3 and the correlation with the radiographic findings are shown in Table IV. In the condyle, there were 32 areas with macroscopic changes of Stages 2 and 3, out of which 11 were detected by CT and nine by CVT. Considering Stage 3

Table II Numbers and types of radiographic findings on IT and

cvr

Type of radiographic finding Radiographic method Sagittal IT

cvr

Frontal IT

cvr

Concavity

Cyst

Erosion

Flattening

Osteophyte

Sclerosis

Total

0 0

0 1 (0)

19 (6) 13 (6)

10 (2) 8 (2)

4 (1) 6 (1)

4 (0) 8 (6)

37 (9) 36 (15)

2 (0) 2 (0)

0 0

10 (4) 3 (1)

2 (0) 0

2 (0) 0

16 (4) 5 (1)

0 0

Figures within parentheses are macroscopic changes corresponding to stage 3.

Table III Comparison between IT and cvr for examining areas of the TM joints (15 joints were examined by sagittal, 12 by frontal radiography)

Sagittal radiograpy Condyle IT

Total

CVT

Frontal radiography Temporal component

CVT

+

Total

+

+15 7 -4 784 19 791

22 788 810

+8 -9 17

7 1326 1333

+ =Ihe presence of a radiographic finding. - =00

radiographic finding.

24 Dentomaxillofac. Radiol., 1990, Vol. 19, February

Condyle

CVT

Total

+

Total

15 1335 1350

+4 -1 5

12 253 265

16 254 270

Computed and conventional tomography of TMJ disease: K. Tanimoto et al. Table IV Topographic distribution of macroscopic changes of Stages 2 and 3

Condyle Anterior Superior Posterior Total Temporal component Anterior slope Inferior part Posterior slope Roof of fossa Posterior wall of fossa Total TOTAL

Stage 2

Stage 3

3 (2/2) 6 (0/1) 9 (3/0) 18 (5/3)

4 (111) 7 (4/4) 3 (111) 14 (6/6)

3 (0/0) 3 (0/0) 2 (0/0) 0 0 8 (0/0)

4 (0/0) 4 (211) 7 (115) 3 (0/3) 0 18 (319)

26 (5/3)

32 (9/15)

Discussion

Figures within parentheses are corresponding radiographic findings (computed/conventional tomography). The total number of areas examined was 135 for the condyle and 225 for the temporal component.

changes alone six out of 14 areas were detected by both methods; five of these were found both with cr and with CVT (Figure 1). Frontal radiography added no further information. With regard to the temporal component, 26 areas were found to have macroscopic Stage 2 and 3 changes. The ability of cr to detect the changes was less than that of CVT. Only three were found by cr whereas nine were seen by CVT (two of the three true-positive cr findings were found by CVT). The correlation between the findings in the sagittal radiographs and the macroscopic Stage 3 changes is presented in Table V. The sensitivity and the specificity for the condyle were the same for the two methods, but the sensitivity of CVT for the temporal component was higher (0.50) than that of cr (0.17). The specificity was similar for both methods. Table VI presents the comprehensive radiological Table V Sensitivity and specificity of

cr and

diagnosis of the joint and its correlation with the macroscopic examination. For Stage 3, 10 out of 15 joints (66%) were correctly diagnosed by CT and 12 (80%) by CVT. In both radiographic methods, the false diagnoses were false-positive and included the radiographic finding of an erosion (Figures 2,3). However, the number of false-positive diagnoses was somewhat higher with cr. There were no false-negative diagnoses, resulting in high value of the sensitivity for both methods. However, when Stages 2 and 3 were combined, the sensitivity decreased, particularly for CVT, but the specificity remained unchanged.

Although the number of joints examined was small, the prevalence of macroscopic changes is comparable with other autopsy studies on larger samples. Thus, onethird had bone exposure and/or perforation, comparable to the figures of 24% reported by Oberg et al. 3o and 39% by Rohlin et at. 31 The prevalence of areas with the severe macroscopic changes of Stage 3, as well as localized soft tissue thinning, was also similar to that found by Rohlin et al. 31 We chose to compare corrected direct cr with corrected CVT for two reasons. First, direct cr images are of higher quality than reconstructed images 13 , 14 , 32 and, second, the technique has already been described in several clinical studies2 ,5 , 13-16 ,22 . The images were evaluated for six different structural changes previously applied in similar studies of conventional radiograph y3 1,33 ,34 . Decreased joint space, condylar angulation and superior position of the coronoid process, findings which were reported by Christiansen et al. 10 in a CT clinical study to be associated with degenerative joint disease, could not be assessed from our autopsy specimens. The correspondence rate for the two techniques was

CVT for the diagnosis of Stage 3 macroscopic changes

Diagnosis Radiographic exammatlon

True-positive

False-negative

False-positive

True-negative

Sensitivity

Specificity

Condyle"

cr

CVT Temporal component"

cr

CVT

6 6

8 S

16 13

105 lOS

0.43 0.43

0.87 0.S9

3 9

15 9

12 S

195 199

0.17 0.50

0.94 0.96

'There were 135 areas of the condyle and 225 of the temporal component.

Table VI Number of TM joints with different stages of macroscopic changes and their comprehensive radiological diagnosis: sensitivity and specificity of radiological diagnosis in detecting joints with macroscopic changes of stages 3 and 2+3 Macroscopic examination

Radiographic exammatton

True-positive

False-negative

Stage 3

cr CVT

+ +5 -5

0 0

False-positive

Stages 2+3

cr CVT

+ +S -6

True-negative

Stages 0+ 1+ 2 + -5 +5 -3 -7

Sensitivity

Specificity

1.0 1.0

0.5 0.7

0.73 0.55

0.5 0.5

Stages 0+ 1 -3 -5

+ +2 -2

-2 -2

+=pathological joints, -=normal or remodelled. A total of 15 joints were examined.

Dentomaxillofac. Radiol., 1990, Vol. 19, February 25

Computed and conventional tomography of TMJ disease: K. Tanimoto et al.

Figure 3 An example of an uncertain finding (false-positive) in the frontal cr. The cort ical outline of the medial half of the condyle is not readily visible (a) . After a change in positioning. the medial cortical outline could be seen (arrow) . However. the lateral part of the cortical outline is not revealed (b). On re-examination. the tilt of the condyle was changed and the cortical outline of both medial and lateral halves is depicted . This joint was judged to have no changes macroscopically (c)

very high, since not only were all six radiographic features considered collectively but also many areas were normal. Both sagittal CT and sagittal CVT were equally efficient, whereas frontal CT revealed more pathology than frontal CVT. However, this way of comparing the two methods overstates the capability of cr. There were several false-positive findings, resulting in a very low positive predictive value (0.24) for CT. Some of the CT images were recorded as uncertain. On the basis of our previous experience with CVT, we interpreted these as erosions, but on re-examination it was found that some disappeared when the frontal tilt of the condyle was altered (Figure 3). The efficiency in detecting joints with bone exposure and/or disc perforation was very high for both methods. However, this high sensitivity might have been achieved at the expense of more false-positive diagnoses, resulting in low values for the specificity, more 26 Dentomaxillofac. Radiol., 1990, Vol. 19, February

so for cr than for CVT. For those joints with localized soft tissue thinning (Stage 2), the sensitivity was lower but the specificity similar compared with Stage 3 joints. Since most of the 'true' findings also were seen on CVT, CT did not supplement the information from CVT. Nor, as was found in our previous study of CVT3 1 , did frontal images add to the 'true findings', as was also the case with CT. Thus, sagittal imaging would seem more valid for the diagnosis of degenerative changes. Although the sensitivity and specificity were the same for the two methods for the condyle, there was a marked difference in the former for the temporal component, due to the numerous false-negative reports with CT. The ability of both CT and CVT to reveal a single area with macroscopic bone changes was much lower than that for making a comprehensive diagnosis of the entire joint, suggesting that, to reach a diagnosis,

Computed and conventional tomography of TMJ disease: K. Tanimoto et al. the use of several parameters in combination is more valid. The level of clinical judgement, that is the process whereby an observer uses his/her training and experience to mentally weigh all the available information to arrive at a diagnosis, might have been somewhat different for the two methods. With increasing experience in interpreting the cr images and the use of a larger range of radiological parameters, such as those suggested by Christiansen et al. 10, the results of the examination might be improved. However, from this study, it appears that cr and CVT are comparable except for bony lesions in the temporal com~ment which, as was also found by Virapongse et al. ,are better visualized by CVT.

c.r

Acknowledgement The authors are grateful to Jorgen Murath for valuable assistance with the cr examinations.

References 1. Helms CA, Morrish RB, Kircos LT, Katzberg RW, Dolwick MF. Computed tomography of the meniscus of the temporomandibular joint. Preliminary observations. Radiology 1982; 145: 719-22. 2. Manzione JV, Seltzer SE, Katzberg RW, Hammerschlag SB, Chiango BF. Direct sagittal computed tomography of the temporomandibular joint. AJNR 1982; 3: 677-9. 3. Helms CA, Katzberg RW, Morrish RB, Dolwick MF. Computed tomography of the temporomandibular joint meniscus J. Oral Maxillofac Surg 1983; 41: 512-7. 4. Thompson JR, Christiansen EL, Hasso AN, Hinshaw DB Jr. Temporomandibular joints. High-resolution computed tomo. graphic evaluation. Radiology 1984; 150: 105-10. 5. Manzione N, Katzberg RW, Brodsky GL, Seltzer SE, Melhns HZ. Internal derangements of the temporomandibular joint: diagnosis by direct sagittal computed tomography. Radiology 1984; 150: 111-5. 6. Sartoris DJ, Neumann CH, Riley RW. The temporomandibular joint: true sagittal computed tomography with meniscus visualization. Radiology 1984; 150: 250-4. 7. Helms CA, Vogler JB, Morrish RB, Goldman SM, Capra RE, Proctor E. Temporomandibular joint internal derangement. CT diagnosis. Radiology 1984; 152: 459-62. 8. Thompson JR, Christiansen EL, Sauser D, Hasso AN, Hinshaw DB Jr. Dislocation of the temporomandibular joint meniscus. Contrast arthrography vs. computed tomography. AJNR 1984; 5: 747-50. 9. Helms CA, Volger JB, Morrish RB. Diagnosis by computed tomography of temporomandibular jont meniscus displacement. J Prosthet Dent 1984; 51: 544-7. 10. Christiansen EL, Thompson JR, Kopp SFO, Hasso AN, Hinshaw DB Jr. Radiographic signs of temporomandibular joint disease: an investigation utilizing X-ray computed tomography. Dentomaxillofac Radioll985; 14: 83--92. 11. Fjellstroem C-A, Olofsson O. Computed tomography.o! the temporomandibular joint meniscus. A report of preliminary tests. J Maxillofac Surg 1985; 13: 24-7. 12. Helms CA, Richardson ML, Vogler JB, Hoddick WK. Computed tomography for diagnosing temporomandibular joint disk displacement. J Craniomand Practice 1985; 3: 23-6. 13. Hiils A, Walter E, Suss Ch, Kalender WA, Luckenbach A, Schulte W. High-resolution computed tomography of the temporomandibular joint. Electromedica 1985; 53: 143--53", 14. Manco LG, Messing SG, Busino U, Fasulo CP, Sordill WC. Internal derangements of the temporomandibular joint evaluated with direct sagittal CT: a prospective study. Radiology 1985; 157: 407-12.

15. Raustia AM, Pyhtinen J, Virtanen KK. Examination of the temporomandibular joint by direct sagittal computed tomography. Clin Radioll985; 36: 291-6. 16. Schatz SL, Cohen HR, Ryvicker MJ, Deutsch AM, Manzione N. Overview of computed tomography of the temporomandibular joint. J. Computed Tomogr 1985; 9: 351-8. 17. Christiansen EL, Thompson JR, Zimmerman Get al. Computed tomography of condylar and articular disk positions within the temporomandibular joint. Oral Surg Oral Med Oral Patholl987; 64: 757-67 .. 18. Christiansen EL, Thompson JR, Hasso AN et al. CT number characteristics of malpositioned TMJ menisci. Diagnosis with CT number highlighting (blinkmode). Invest Radiol 1987; 22: 31521. 19. Jend-Rossmann I, Jend H-H, Siegert R. Significance of arthrography and computed tomography in the assessm.ent of inte~al derangement of the temporomandibular joint. J. Craniomaxillofac Surg 1987; 15: 265-9. 20. Westesson poL, Katzberg RW, Tallents RH, SanchezWoodworth RE, Svensson SA. CT and MR of the temporomandibular joint: comparison with autopsy specimens. AJR 1987; 148: 1165-71. 21. Fava C, Gatti G, Cardesi E, Parchetti R, Rocca G, Preti G. Possibilities and limits in identifying the TMJ articular meniscus with the CT scanner. A comparative anatomoradiological study. J Craniomandib Disord Facial Oral Pain 1988; 2: 141-7. 22. Heffez L, Maffee MF, Langer B. Double-contrast arthrography of the temporomandibular joint. Role of direct sagittal CT imaging. Oral Surg Oral Med Oral Patholl988; 65: 511-4. 23. Virapongse C, Sarwar M, Kier EL, Sasak C, Pillsbury H. Temporal bone disease: a comparison between high resolution computed tomography and pluridirectional tomography. Radiology 1983; 147: 743-8. 24. Larheim TA, Kolbenstvedt A. High-resolution computed tomography of the osseous temporomandibular joint. Some normal and abnormal appearances. Acta Radiol Diagn 1984; 25: 465-9. 25. Omnell K-A, Petersson A. Radiography of the temporomandibular joint utilizing oblique lateral transcranial projections. Comparison of information obtained with standardized technique and individualized technique. Odontol Revy 1976; 27: 77-92. 26. Mane-Dickson W, Trefler M, Dickson DR. Comparison of dosimetry and image quality in computed and conventional tomography. Radiology 1979; 131: 509-14. 27. Omnell K-A. Radiography of the TMJ. In: Irby WB, ed. Current advances in oral surgery, vol III. St Louis: CV Mosby 1980: 196226. 28. Byers PD, Contepomi CA, Farkas TA. A post mortem study of the hip joint. Including the prevalence of the features of the right side. Ann Rheum Dis 1970; 29: 15-31. 29. Weinstein MC, Fineberg HV. Clinical decision analysis. Philadelphia: WB Saunders 1980: 75-92. 30. Oberg T, Carlsson GE, Fajers CM. The temporomandibular joint. A morphological study on a human autopsy material. Acta Odontol Scand 1971; 29: 349-84. 31. Rohlin M, Akerman S, Kopp S. Tomography as an aid to detect macroscopic changes of the temporomandibular joint. An autopsy study of the aged. Acta Odontol Scand 1986; 44: 131-40. 32. Wilkinson T, Maryniuk G. The correlation between sagittal anatomic sections and computerized tomography of the TMJ. J Craniomandib Practice 1983; 1: 38-45. 33. Petersson A, Nanthaviroj S. Radiography of the temporomandibular joint utilizing the transmaxillary projection. A comparison of the information obtained with the oblique lateral transcranial projection versus the transmaxillary projection. Dentomaxillofac Radioll975; 4: 76-83. 34. Lindvall AM, Helkimo E, Hollender L, Carlsson GE. Radiographic examination of the temporomandibular joint. Dentomaxillofac Radioll976; 5: 24-32. Address: Dr M. Rohlin, Department of Oral Radiology, School of Dentistry, Carl Gustafs Vag 34, S-214 21 Malmo, Sweden.

Dentomaxillofac. Radiol., 1990, Vol. 19, February 27

Comparison of computed with conventional tomography in the evaluation of temporomandibular joint disease: a study of autopsy specimens.

Autopsy specimens were examined both radiographically and macroscopically to compare direct computed tomography (CT) with conventional tomography (CVT...
600KB Sizes 0 Downloads 0 Views