Transcranial radiographic and tomographic analysis of the lateral and midpoint inclined planes of the articular eminence Wayne Ichikawa, DDS, MS,” Daniel M. Laskin, DDS, IMY,~ and Henry M. Rosenberg, DDS,” Chicago, Ill., and Richmond, Va. DEPARTMENT OF ORAL AND MAXILLOFACIAL SURGERY, MEDICAL COLLEGE OF VIRGINIA SCHOOL OF DENTISTRY, AND DEPARTMENT OF ORAL AND MAXILLOFACIAL SURGERY AND RADIOLOGY, UNIVERSITY OF ILLINOIS COLLEGE OF DENTISTRY A radiologic study of the articular eminence was performed on 10 skulls by means of tomograms and transcranial radiographs. Angulation of the articular eminence at the lateral and midpoint inclined planes was measured from the radiographs and compared with direct anatomic measurements. With both radiographic methods, the results showed no significant difference between left and right side measurements. The mean angulation of the midpoint inclined plane was significantly greater than that of the lateral inclined plane with both techniques. With transcranial radiographs, the angulation values were the same as the direct anatomic measurements despite a 25-degree change in skull orientation required for the radiographic method. In the tomographic study, all measurements corresponded with the anatomic and transcranial measurements except that angulation of the lateral inclined plane differed slightly from the direct anatomic measurement. The results show that both tomography and transcranial radiography are adequate for visualization of the midpoint region. However, tomography is preferable for viewing the lateral aspect of the articular eminence. (ORAL SURC ORAL MED ORAL PATHOL 1990;70:516-22)
T
he lateral aspect of the articular eminence can be easily seenon visual examination of the human temporomandibular joint (TMJ) from the side. The midpoint is hidden from view; however, because the articular eminence is concavein its mediolateral axis, as a result of this anatomic configuration, radiographic examination of the TMJ may not depict a true representation of the articulating surface. For the clinician, an accurate appreciation of the shape and slope of the articular eminence is an important factor in diagnosis and treatment planning.’ The morphology of the articular eminence has been studied in medial to lateral sections both radiographSupported in part by USPHS research grant DE-06946 from the National Institute of Dental Research, National Institutes of Health, Bethesda, Md. aFormer resident, Department of Oral and Maxillofacial Surgery, University of Illinois College of Dentistry. Currently in private practice, San Jose, Calif. bProfessorand Chairman, Department of Oral and Maxillofacial Surgery, Medical College of Virginia, Richmond. cProfessorEmeritus, Department of Radiology, University of Illinois College of Dentistry, Chicago. 7/16/24958
516
ically and in cadaver specimens.*However, there is a lack of information in the literature about the relationship of the actual slope of the eminence and what can be seen and measured by various radiographic techniques. In a previous study, detailed anatomic information about the relationship between the lateral and midpoint inclined planes of the articular eminence was reported.3 The purpose of this study was to determine whether the same anatomic landmarks could be visualized and accurately measuredon either transcranial radiographs or tomograms. MATERIAL
AND METHODS
Ten skulls (20 TMJs) were selected for this study. Direct angulation measurements of the lateral and midpoint inclined planes of each articular eminence were available from a previous study.3 The transcranial radiographic study consisted of three views per TMJ: (I) a plain view serving as a control (Fig. l), (2) a view with the lateral inclined plane indicated by a radiopaque marker (Fig. 2), and (3) a view with the midpoint inclined plane indicated by a radiopaque marker (Fig. 3). The radiopaque markers consisted of 2%gauge stainless steel wire
Volume 70 Number 4
Transcranial
analysis of lateral and midpoint inclined planes
517
Fig. 1. Control transcranial radiograph without marker. Condyle (C) is positioned loosely for reference only. AE, Articular eminence; GF, glenoid fossa.
Fig. 2. Transcranial radiograph showing lateral inclined plane (LIP) outlined with radiopaque marker. Condyle (C) is positioned loosely for reference only. AE, Articular eminence.
luted anteroposteriorly on the surface of the glenoid fossa in the appropriate parasaggital plane with soft wax in a manner similar to that described by Weinberg.4 Articular eminence angles were measured directly on the radiographs and compared with the direct skull measurements adjusted for a 25degree rotation. Right and left side measurements were compared by the Student’s t test. One-way analysis of variance with repeated measures and the Newman-Keuls test were applied to determine the relationship between the various radiographic views. A Philips Polytome U machine (Philips Dental Systems, Stanford, Conn.) was used to expose the transcranial views of each TMJ (kVp = 60; mA = 40; 6 X 8 inch diaphragm; reciprocating grid; Kodak
Lanex regular screenswith Kodak OM- 1 film [Eastman Kodak Co., Rochester, N.Y.]; automatic processing at 2% minutes with Kodak R-P X-Omat; source to film = 115 cm). Skulls were stabilized by a special cephalometric headholder describedby Rosenberg and Graczyk.5 All skulls were horizontally rotated 25 degreesto prevent superimposition of the condyles. The skulls were oriented so that the Frankfort horizontal plane was perpendicular to the Polytome table. The resulting radiographs therefore provided reference points that included porion and the Frankfort horizontal plane, which corresponded to the film edge. This facilitated orientation for direct articular eminence angulation measurements from the radiographs. Magnification error was not calcu-
518
Ichikawa, Laskin, and Rosenberg
ORAL SURG ORAL
MED ORAL PATHOL. October 1990
Fig. 3. Transcranial radiograph showing midpoint inclined plane (MIP) outlined with radiopaque marker. Condyle (C) is positioned loosely for reference only. AE, Articular eminence; GF, glenoid fossa.
Fig. 4. Control tomogram showing lateral inclined plane (LIP) without radiopaque marker. Headholder ear rod (ER) is located in external acoustic meatus. Glenoid fossa is blocked from view.
lated, becausethe slopes of the inclined planes were all relative to each other. As a check on the reproducibility of the transcranial radiographic method, five skulls were randomly selected and plain transcranial projections were exposedon one side only at two different times. The inclined plane of the articular eminence was measured on each projection and correlation analysis was applied. Becausethe transcranial radiographs in this study were exposed with the skulls rotated 25 degrees,angulation measurementsmade in a parasagittal plane were not true values. To correct for this factor, direct measurementsat a 25-degree angle to the parasagit-
tal plane were made on the 10 skulls selected for radiographic study. These measurements were then compared with those obtained at the true parasagittal plane with the use of the Student’s t test. The tomographic study involved the same 10 skulls selected for use in the transcranial study. A Philips Polytome U machine was used to expose the tomograms (kVp = 40 to 50; mA = 25; time = 6 seconds; 60 mm diaphragm; no grid; Kodak Lanex regular screenswith Kodak TMG-1 film; automatic processing at 2% minutes with Kodak R-P X-Omat; source to film = 145 cm). The headholder was the same as that used for the transcranial projections. The Frankfort horizontal plane was again used as a reference to
Volume70 Number 4
Transcranial
analysis of lateral and midpoint inclined planes
519
5. Tomogram showing lateral inclined plane outlined with radiopaque marker. Headholder ear rod (EZ?)is located in the external acoustic meatus. Glenoid fossa is blocked from view.
Fig.
Fig. 6. Control tomogram showing midpoint inclined plane (MZP] without radiopaque marker. Headholder ear rod (ER) is located in external acoustic meatus. Glenoid fossa is fully visible.
facilitate orientation for direct measurement of angulations from the radiographs. Magnification factors
were not calculated becauseangulations were all relative to each other.
Four views per TMJ were exposed: (1) a control view of the lateral inclined plane without a radiopaque marker (Fig. 4), (2) a view of the lateral inclined plane with a radiopaque marker (Fig. 5), (3) a control view of the midpoint inclined plane without a radiopaque marker (Fig. 6), and (4) a view of the midpoint inclined plane with a radiopaque marker (Fig. 7). The radiopaque marker consisted of 1.5 mm diameter solder.* This material was easily shaped *Litton, Chicago,III. (60%lead and 40%tin solderalloy).
and luted on the surface of the glenoid fossa with soft wax. Angles of inclination were measured directly from the tomograms and compared with direct skull measurements.3Right versus left measurementswere also compared. One-way analysis of variance with repeated measures and the Newman-Keuls test were used to examine relationships between the four tomographic projections. The reproducibility of the tomographic method was tested by randomly selecting five skulls and measuring the lateral inclined plane with a radiopaque marker on one side at two different times. Correlation analyses was applied by means of the Pearson product-moment coefficient of correlations.
Ichikawa,
520
Laskin. and Rosenberg
ORAL SURC ORAL MED ORAL PATHOL October 1990
Fig. 7. Tomogram showing midpoint inclined plane outlined with radiopaque marker. Headholder ear rod {IX) is located in external acoustic meatus. Glenoid fossa is fully visible.
Table
I. Comparison of direct articular eminence measurements with those made on transcranial radiographs Measuremem
Side
Plain view control
Lateral direct measure
Lateral marked
Midpoint direct measure
58.0” 58.7" 58.3"
39.6" 38.7" 39.1"
35.1" 38.6" 37.1"
58.4”
55.1”
57.5" 51.9"
58.9" 57.0"
Right
Left Mean
Table
Midpoint marked
II. Comparison of direct articular eminence measurements with those made on tomograms Measurement
Side
Right Left Mean
Lateral control
Lateral direcl measure
Lateral marked
Midpoint control
Midpoint direct measure
Midpoint marked
31.4" 29.2" 30.3"
37.3" 38.0" 37.6"
32.0" 32.1" 32.0"
55.8" 58.5" 57.1"
58.0" 58.6" 58.3"
53.9" 55.1" 54.5"
RESULTS
Repeated measurements on five transcranial and five tomographic projections yielded correlation coefficients with very high significance (r = 0.9723, p < 0.005; and r = 0.9797, p < 0.0025; respectively), confirming the reliability of the techniques used in this study. Moreover, a comparison of direct anatomic measurements at the parasagittal plane with those made on the radiographs of the skulls rotated 25 degrees revealed no significant difference. This finding was important in showing that corrections for angulation factors were not necessarywhen studying tran-
scranial radiographs with a change in skull orientation of 25 degrees. The mean angulation measurementsfrom the three transcranial views are shown in Table I. No signiiicant difference was found between the right and left sidesin any of the views. One-way analysis of variance with repeated measures and multiple comparison testing with the Newman-Keuls test also showed no significant difference betweendirect measurementsat the lateral and midpoint inclined planes and those made on the marked radiographs (right: F = 33.82; df= 4,36; p < 0.01) (left: F = 26.76; df = 4,36;
Volume 70 Number 4
Transcranial
p < 0.01). However, the lateral inclined planes were significantly less steep than the midpoint inclined planes. The plain view control measurements corresponded to the midpoint direct values. Results of the tomographic study are shown in Table II. No significant difference was found between measurementsmade on the right and left sides. However, one-way analysis of variance with repeated measuresshowed a significant difference between the lateral andmidpoint inclined planes (right: F = 48.26; df = $45; p < 0.01) (left: F = 79.04; df = 5,45; p < 0.01). Multiple comparison analysis with the Newman-Keuls test showed that the lateral control view measurementswere similar to those made on the lateral tomographic view with a radiopaque marker. The midpoint direct measuresand the measurements made on the midpoint control view and on the midpoint tomogram with a radiopaque marker were also similar. All lateral measurements were significantly different from midpoint values. An unexpected finding was that the lateral direct measure differed significantly from that on the lateral control view and on the lateral view with radiopaque marker. The difference was small in relation to the midpoint values, but it was statistically significant. DISCUSSION
The same relationships found by direct anatomic study3 were demonstrated in the transcranial radiographic study. The lateral and midpoint slopes were shown to be significantly different, with the angulation of the lateral slope being less than that of the midpoint slope (Table I). However, a comparison of these values with those from other published studies is difficult becauseeach study differs with respect to method of analysis and radiographic technique.6-8 This applies to the tomographic studies as well63 9,lo In the tomographic study, the relationships between the lateral and midpoint inclined planes were also similar to those found by direct anatomic measurement (Table II). What was seen on a midpoint tomogram corresponded to the true value. It also corresponded to the midpoint tagged by the radiopaque marker. However, when the measurements from the plain lateral tomograms were compared with the direct anatomic measuresand with those on the marked tomograms, the angulation of the lateral inclined plane on the plain tomogram corresponded to that on the radiograph with a lateral radiopaque tag, but it was not equal to the true value obtained by direct anatomic measurement (Table II). Becausethe magnitude of this difference was not large, the discrepancy can possibly be explained by experimental error. The midpoint region may be easier to target by tomography because of a flatter configuration in the medio-
analysis of lateral and midpoint inclined planes
521
lateral axis. Additionally, the existence of stability at the midpoint is corroborated by Hjortsjo and coworkers,’ ’ who found that the midsagittal region was smoother, with a more regular outline than lateral sections, which were more irregularly shaped. A significant finding from this study was that the lateral slope of the articular eminence was not readily visible on the plain-view transcranial radiograph; what was seenwas the midpoint inclined plane. This is different than what was reported by Lewis12 and Weinberg.4 Lewis12studied conventional transcranial radiographs and demonstrated that the lateral portions of the joint were accessible to radiographic inspection but that the central and medial parts were obscured by superimpositions. Weinberg4 drew similar conclusions using 0.01O-inch wire luted to various positions in the glenoid fossa.Wire luted to the lateral border corresponded to the fossa image, wire luted medially was projected inferiorly, and wire at the medial border was projected inferiorly out of the image. It is important to note, however, that the radiographic techniques of Lewis’* and Weinberg4 and their x-ray angulations were different from those used in this study. Because the incline of the articular eminence can sometimeshave important clinical implications,’ it is necessary to be able to make this assessmentaccurately. Radiographs are generally used for this purpose.As shown in this study, however, the information obtained will vary with the technique used. Because the incline at the midpoint is the steepestpart of the eminence, and therefore most representative, it is preferable to view this region. This can be accomplished by either transcranial or tomographic radiography. If information about the lateral aspect of the eminence is required, however, the useof tomography is preferable, even though this study showed that the actual slope was slightly greater than that of the radiographic image. Comparison of transcranial and tomographic methods has shown that both possesssome degree of inherent error and distortion.13-16Becausethey each provide different types of information, one method should not be favored over the other; rather, the two techniques should be used to complement one another. REFERENCES
1. Hall MB, Brown RW, Sclar AG. Anatomy of the TMJ articular eminence before and after surgical reduction. J Craniomandibular Pratt 1984;2:135-40. 2. Eckerdal 0. Tomography of the temporomandibular joint. Correlation between tomographic image and histologic sections in a three-dimensional system. Acta Radio1 [Diag] [Suppl] (Stockh) 1973;329:1-107. 3. Ichikawa W, Laskin DM. Anatomic study of the angulation of
522
4. 5.
6.
7.
8.
9. 10.
11.
Ichikawa,
Laskin, and Rosenberg
the lateral and mid-point inclined planes of the articular eminence. J Craniomandibular Pratt 1989;7:22-6. Weinberg LA. What we really see in a TMJ radiograph. J Prosthet Dent 1973;30:898-913. Rosenberg HM, Graczyk RJ. Temporomandibular articulation tomography: a corrected anteroposterior and lateral cephalometiic technique. ORAL SURG ORAL MED ORAL PATHOL 1986;62:198-204. Lindblom G. On the anatomy and function of the temporomandibular joint. Studies on clinical bite-rehabilitation material including arthrosis cases, with special references to radiographic findings. Acta Odontol Stand [Suppl] 1960;28: l-287. Lawther WL. A roentgenographic study of the temporomandibular joint using a special head positioner. Angle Orthod 1956;26:22-33. Quirch JS, Carraro JJ, Itoiz ME. Correlation between articular eminentia and the depth of glenoid fossa. J Periodont Res 1966;36:227-32. Ricketts RM. Variations of the temporomandibular joint as revealed by cephalometric laminagraphy. Am J Orthod 1950;36:877-98. Goldman SM, Taylor R. Retrospective radiographic evaluation of 100 temporomandibular joint patients. J Prosthet Dent 1985;53:566-9. Hjortsjo C-H, Persson P-I, Sonesson B. Studies on the shape of the articular eminence with its relation to the mechanism in the temporomandibular joint. Odontol Rev 1953;4:187-202.
ORAL SURG ORAL MED ORAL PATHOL October 1990 12. Lewis GR. Temporomandibular joint radiographic technics. Comparison and evaluation of results. Dent Radiogr Photogr 1964;37:8-20. 13. Klein IE, Blatterfein L, Miglino JC. Comparison of the fidelity of radiographs of the mandibular condyles made by different techniques. J Prosthet Dent 1970;24:419-52. 14. Larheim TA, Tveito L. Reproducibility of temporomandibular joint radiographs using oblique lateral transcranial projection and lateral tomographic technique. Dentomaxillofac Radiol 1980;9:85-90. 15. Lundberg M, Welander U. The articular cavity in the temporomandibular joint. A comparison between the oblique-lateral and the tomographic image. Medicamundi 1970;15:27-9. 16. Eckerdal 0, Lundberg M. Temporomandibular joint relations as revealed by conventional radiographic techniques. A comparison with the morphology and tomographic images. Dentomaxillofac Radio1 1979;8:65-70. Reprint requests to: Dr. Daniel M. Laskin Department of Oral and Maxillofacial Box 566, MCV Station Richmond, VA 23298-0566
Surgery
T
he lateral aspect of the articular eminence can be easily seenon visual examination of the human temporomandibular joint (TMJ) from the side. The midpoint is hidden from view; however, because the articular eminence is concavein its mediolateral axis, as a result of this anatomic configuration, radiographic examination of the TMJ may not depict a true representation of the articulating surface. For the clinician, an accurate appreciation of the shape and slope of the articular eminence is an important factor in diagnosis and treatment planning.’ The morphology of the articular eminence has been studied in medial to lateral sections both radiographSupported in part by USPHS research grant DE-06946 from the National Institute of Dental Research, National Institutes of Health, Bethesda, Md. aFormer resident, Department of Oral and Maxillofacial Surgery, University of Illinois College of Dentistry. Currently in private practice, San Jose, Calif. bProfessorand Chairman, Department of Oral and Maxillofacial Surgery, Medical College of Virginia, Richmond. cProfessorEmeritus, Department of Radiology, University of Illinois College of Dentistry, Chicago. 7/16/24958
516
ically and in cadaver specimens.*However, there is a lack of information in the literature about the relationship of the actual slope of the eminence and what can be seen and measured by various radiographic techniques. In a previous study, detailed anatomic information about the relationship between the lateral and midpoint inclined planes of the articular eminence was reported.3 The purpose of this study was to determine whether the same anatomic landmarks could be visualized and accurately measuredon either transcranial radiographs or tomograms. MATERIAL
AND METHODS
Ten skulls (20 TMJs) were selected for this study. Direct angulation measurements of the lateral and midpoint inclined planes of each articular eminence were available from a previous study.3 The transcranial radiographic study consisted of three views per TMJ: (I) a plain view serving as a control (Fig. l), (2) a view with the lateral inclined plane indicated by a radiopaque marker (Fig. 2), and (3) a view with the midpoint inclined plane indicated by a radiopaque marker (Fig. 3). The radiopaque markers consisted of 2%gauge stainless steel wire
Volume 70 Number 4
Transcranial
analysis of lateral and midpoint inclined planes
517
Fig. 1. Control transcranial radiograph without marker. Condyle (C) is positioned loosely for reference only. AE, Articular eminence; GF, glenoid fossa.
Fig. 2. Transcranial radiograph showing lateral inclined plane (LIP) outlined with radiopaque marker. Condyle (C) is positioned loosely for reference only. AE, Articular eminence.
luted anteroposteriorly on the surface of the glenoid fossa in the appropriate parasaggital plane with soft wax in a manner similar to that described by Weinberg.4 Articular eminence angles were measured directly on the radiographs and compared with the direct skull measurements adjusted for a 25degree rotation. Right and left side measurements were compared by the Student’s t test. One-way analysis of variance with repeated measures and the Newman-Keuls test were applied to determine the relationship between the various radiographic views. A Philips Polytome U machine (Philips Dental Systems, Stanford, Conn.) was used to expose the transcranial views of each TMJ (kVp = 60; mA = 40; 6 X 8 inch diaphragm; reciprocating grid; Kodak
Lanex regular screenswith Kodak OM- 1 film [Eastman Kodak Co., Rochester, N.Y.]; automatic processing at 2% minutes with Kodak R-P X-Omat; source to film = 115 cm). Skulls were stabilized by a special cephalometric headholder describedby Rosenberg and Graczyk.5 All skulls were horizontally rotated 25 degreesto prevent superimposition of the condyles. The skulls were oriented so that the Frankfort horizontal plane was perpendicular to the Polytome table. The resulting radiographs therefore provided reference points that included porion and the Frankfort horizontal plane, which corresponded to the film edge. This facilitated orientation for direct articular eminence angulation measurements from the radiographs. Magnification error was not calcu-
518
Ichikawa, Laskin, and Rosenberg
ORAL SURG ORAL
MED ORAL PATHOL. October 1990
Fig. 3. Transcranial radiograph showing midpoint inclined plane (MIP) outlined with radiopaque marker. Condyle (C) is positioned loosely for reference only. AE, Articular eminence; GF, glenoid fossa.
Fig. 4. Control tomogram showing lateral inclined plane (LIP) without radiopaque marker. Headholder ear rod (ER) is located in external acoustic meatus. Glenoid fossa is blocked from view.
lated, becausethe slopes of the inclined planes were all relative to each other. As a check on the reproducibility of the transcranial radiographic method, five skulls were randomly selected and plain transcranial projections were exposedon one side only at two different times. The inclined plane of the articular eminence was measured on each projection and correlation analysis was applied. Becausethe transcranial radiographs in this study were exposed with the skulls rotated 25 degrees,angulation measurementsmade in a parasagittal plane were not true values. To correct for this factor, direct measurementsat a 25-degree angle to the parasagit-
tal plane were made on the 10 skulls selected for radiographic study. These measurements were then compared with those obtained at the true parasagittal plane with the use of the Student’s t test. The tomographic study involved the same 10 skulls selected for use in the transcranial study. A Philips Polytome U machine was used to expose the tomograms (kVp = 40 to 50; mA = 25; time = 6 seconds; 60 mm diaphragm; no grid; Kodak Lanex regular screenswith Kodak TMG-1 film; automatic processing at 2% minutes with Kodak R-P X-Omat; source to film = 145 cm). The headholder was the same as that used for the transcranial projections. The Frankfort horizontal plane was again used as a reference to
Volume70 Number 4
Transcranial
analysis of lateral and midpoint inclined planes
519
5. Tomogram showing lateral inclined plane outlined with radiopaque marker. Headholder ear rod (EZ?)is located in the external acoustic meatus. Glenoid fossa is blocked from view.
Fig.
Fig. 6. Control tomogram showing midpoint inclined plane (MZP] without radiopaque marker. Headholder ear rod (ER) is located in external acoustic meatus. Glenoid fossa is fully visible.
facilitate orientation for direct measurement of angulations from the radiographs. Magnification factors
were not calculated becauseangulations were all relative to each other.
Four views per TMJ were exposed: (1) a control view of the lateral inclined plane without a radiopaque marker (Fig. 4), (2) a view of the lateral inclined plane with a radiopaque marker (Fig. 5), (3) a control view of the midpoint inclined plane without a radiopaque marker (Fig. 6), and (4) a view of the midpoint inclined plane with a radiopaque marker (Fig. 7). The radiopaque marker consisted of 1.5 mm diameter solder.* This material was easily shaped *Litton, Chicago,III. (60%lead and 40%tin solderalloy).
and luted on the surface of the glenoid fossa with soft wax. Angles of inclination were measured directly from the tomograms and compared with direct skull measurements.3Right versus left measurementswere also compared. One-way analysis of variance with repeated measures and the Newman-Keuls test were used to examine relationships between the four tomographic projections. The reproducibility of the tomographic method was tested by randomly selecting five skulls and measuring the lateral inclined plane with a radiopaque marker on one side at two different times. Correlation analyses was applied by means of the Pearson product-moment coefficient of correlations.
Ichikawa,
520
Laskin. and Rosenberg
ORAL SURC ORAL MED ORAL PATHOL October 1990
Fig. 7. Tomogram showing midpoint inclined plane outlined with radiopaque marker. Headholder ear rod {IX) is located in external acoustic meatus. Glenoid fossa is fully visible.
Table
I. Comparison of direct articular eminence measurements with those made on transcranial radiographs Measuremem
Side
Plain view control
Lateral direct measure
Lateral marked
Midpoint direct measure
58.0” 58.7" 58.3"
39.6" 38.7" 39.1"
35.1" 38.6" 37.1"
58.4”
55.1”
57.5" 51.9"
58.9" 57.0"
Right
Left Mean
Table
Midpoint marked
II. Comparison of direct articular eminence measurements with those made on tomograms Measurement
Side
Right Left Mean
Lateral control
Lateral direcl measure
Lateral marked
Midpoint control
Midpoint direct measure
Midpoint marked
31.4" 29.2" 30.3"
37.3" 38.0" 37.6"
32.0" 32.1" 32.0"
55.8" 58.5" 57.1"
58.0" 58.6" 58.3"
53.9" 55.1" 54.5"
RESULTS
Repeated measurements on five transcranial and five tomographic projections yielded correlation coefficients with very high significance (r = 0.9723, p < 0.005; and r = 0.9797, p < 0.0025; respectively), confirming the reliability of the techniques used in this study. Moreover, a comparison of direct anatomic measurements at the parasagittal plane with those made on the radiographs of the skulls rotated 25 degrees revealed no significant difference. This finding was important in showing that corrections for angulation factors were not necessarywhen studying tran-
scranial radiographs with a change in skull orientation of 25 degrees. The mean angulation measurementsfrom the three transcranial views are shown in Table I. No signiiicant difference was found between the right and left sidesin any of the views. One-way analysis of variance with repeated measures and multiple comparison testing with the Newman-Keuls test also showed no significant difference betweendirect measurementsat the lateral and midpoint inclined planes and those made on the marked radiographs (right: F = 33.82; df= 4,36; p < 0.01) (left: F = 26.76; df = 4,36;
Volume 70 Number 4
Transcranial
p < 0.01). However, the lateral inclined planes were significantly less steep than the midpoint inclined planes. The plain view control measurements corresponded to the midpoint direct values. Results of the tomographic study are shown in Table II. No significant difference was found between measurementsmade on the right and left sides. However, one-way analysis of variance with repeated measuresshowed a significant difference between the lateral andmidpoint inclined planes (right: F = 48.26; df = $45; p < 0.01) (left: F = 79.04; df = 5,45; p < 0.01). Multiple comparison analysis with the Newman-Keuls test showed that the lateral control view measurementswere similar to those made on the lateral tomographic view with a radiopaque marker. The midpoint direct measuresand the measurements made on the midpoint control view and on the midpoint tomogram with a radiopaque marker were also similar. All lateral measurements were significantly different from midpoint values. An unexpected finding was that the lateral direct measure differed significantly from that on the lateral control view and on the lateral view with radiopaque marker. The difference was small in relation to the midpoint values, but it was statistically significant. DISCUSSION
The same relationships found by direct anatomic study3 were demonstrated in the transcranial radiographic study. The lateral and midpoint slopes were shown to be significantly different, with the angulation of the lateral slope being less than that of the midpoint slope (Table I). However, a comparison of these values with those from other published studies is difficult becauseeach study differs with respect to method of analysis and radiographic technique.6-8 This applies to the tomographic studies as well63 9,lo In the tomographic study, the relationships between the lateral and midpoint inclined planes were also similar to those found by direct anatomic measurement (Table II). What was seen on a midpoint tomogram corresponded to the true value. It also corresponded to the midpoint tagged by the radiopaque marker. However, when the measurements from the plain lateral tomograms were compared with the direct anatomic measuresand with those on the marked tomograms, the angulation of the lateral inclined plane on the plain tomogram corresponded to that on the radiograph with a lateral radiopaque tag, but it was not equal to the true value obtained by direct anatomic measurement (Table II). Becausethe magnitude of this difference was not large, the discrepancy can possibly be explained by experimental error. The midpoint region may be easier to target by tomography because of a flatter configuration in the medio-
analysis of lateral and midpoint inclined planes
521
lateral axis. Additionally, the existence of stability at the midpoint is corroborated by Hjortsjo and coworkers,’ ’ who found that the midsagittal region was smoother, with a more regular outline than lateral sections, which were more irregularly shaped. A significant finding from this study was that the lateral slope of the articular eminence was not readily visible on the plain-view transcranial radiograph; what was seenwas the midpoint inclined plane. This is different than what was reported by Lewis12 and Weinberg.4 Lewis12studied conventional transcranial radiographs and demonstrated that the lateral portions of the joint were accessible to radiographic inspection but that the central and medial parts were obscured by superimpositions. Weinberg4 drew similar conclusions using 0.01O-inch wire luted to various positions in the glenoid fossa.Wire luted to the lateral border corresponded to the fossa image, wire luted medially was projected inferiorly, and wire at the medial border was projected inferiorly out of the image. It is important to note, however, that the radiographic techniques of Lewis’* and Weinberg4 and their x-ray angulations were different from those used in this study. Because the incline of the articular eminence can sometimeshave important clinical implications,’ it is necessary to be able to make this assessmentaccurately. Radiographs are generally used for this purpose.As shown in this study, however, the information obtained will vary with the technique used. Because the incline at the midpoint is the steepestpart of the eminence, and therefore most representative, it is preferable to view this region. This can be accomplished by either transcranial or tomographic radiography. If information about the lateral aspect of the eminence is required, however, the useof tomography is preferable, even though this study showed that the actual slope was slightly greater than that of the radiographic image. Comparison of transcranial and tomographic methods has shown that both possesssome degree of inherent error and distortion.13-16Becausethey each provide different types of information, one method should not be favored over the other; rather, the two techniques should be used to complement one another. REFERENCES
1. Hall MB, Brown RW, Sclar AG. Anatomy of the TMJ articular eminence before and after surgical reduction. J Craniomandibular Pratt 1984;2:135-40. 2. Eckerdal 0. Tomography of the temporomandibular joint. Correlation between tomographic image and histologic sections in a three-dimensional system. Acta Radio1 [Diag] [Suppl] (Stockh) 1973;329:1-107. 3. Ichikawa W, Laskin DM. Anatomic study of the angulation of
522
4. 5.
6.
7.
8.
9. 10.
11.
Ichikawa,
Laskin, and Rosenberg
the lateral and mid-point inclined planes of the articular eminence. J Craniomandibular Pratt 1989;7:22-6. Weinberg LA. What we really see in a TMJ radiograph. J Prosthet Dent 1973;30:898-913. Rosenberg HM, Graczyk RJ. Temporomandibular articulation tomography: a corrected anteroposterior and lateral cephalometiic technique. ORAL SURG ORAL MED ORAL PATHOL 1986;62:198-204. Lindblom G. On the anatomy and function of the temporomandibular joint. Studies on clinical bite-rehabilitation material including arthrosis cases, with special references to radiographic findings. Acta Odontol Stand [Suppl] 1960;28: l-287. Lawther WL. A roentgenographic study of the temporomandibular joint using a special head positioner. Angle Orthod 1956;26:22-33. Quirch JS, Carraro JJ, Itoiz ME. Correlation between articular eminentia and the depth of glenoid fossa. J Periodont Res 1966;36:227-32. Ricketts RM. Variations of the temporomandibular joint as revealed by cephalometric laminagraphy. Am J Orthod 1950;36:877-98. Goldman SM, Taylor R. Retrospective radiographic evaluation of 100 temporomandibular joint patients. J Prosthet Dent 1985;53:566-9. Hjortsjo C-H, Persson P-I, Sonesson B. Studies on the shape of the articular eminence with its relation to the mechanism in the temporomandibular joint. Odontol Rev 1953;4:187-202.
ORAL SURG ORAL MED ORAL PATHOL October 1990 12. Lewis GR. Temporomandibular joint radiographic technics. Comparison and evaluation of results. Dent Radiogr Photogr 1964;37:8-20. 13. Klein IE, Blatterfein L, Miglino JC. Comparison of the fidelity of radiographs of the mandibular condyles made by different techniques. J Prosthet Dent 1970;24:419-52. 14. Larheim TA, Tveito L. Reproducibility of temporomandibular joint radiographs using oblique lateral transcranial projection and lateral tomographic technique. Dentomaxillofac Radiol 1980;9:85-90. 15. Lundberg M, Welander U. The articular cavity in the temporomandibular joint. A comparison between the oblique-lateral and the tomographic image. Medicamundi 1970;15:27-9. 16. Eckerdal 0, Lundberg M. Temporomandibular joint relations as revealed by conventional radiographic techniques. A comparison with the morphology and tomographic images. Dentomaxillofac Radio1 1979;8:65-70. Reprint requests to: Dr. Daniel M. Laskin Department of Oral and Maxillofacial Box 566, MCV Station Richmond, VA 23298-0566
Surgery
