Purpose: The purpose of this work was to establish criteria for the diameters of normal extraocular muscles and superior ophthalmic vein and width of the optic.
Journal of Computer Assisted Tomography 24(3):493-496 © 2000 Lippincott Williams & Wilkins, Inc., Philadelphia
Normative Measurements of Orbital Structures Using MRI ..
..
Ali Ozgen and Usttin Aydingoz
Purpose: The purpose of this work was to establish criteria for the diameters of normal extraocular muscles and superior ophthalmic vein and width of the optic nerve-sheath complex, to determine normal globe position as seen on MRI, and to investigate the effects of age and sex on these structures. Method: Diameters of the extraocular muscles and superior ophthalmic vein, width of the optic nerve-sheath complex, and distance from the interzygomatic line to the posterior margin of the globe were calculated for 200 normal orbits of 100 patients on Tl-weighted MR scans. Results: Normal ranges for diameters of the extraocular muscles were as follows (mean ± 2 SD): medial rectus, 3.2-4.9 mm; lateral rectus, 2.6-4.8 mm; inferior rectus, 3.7-6.0 mm; superior group, 3.1-5.6 mm; superior oblique, 2.4-4.1 mm. The normal position of the posterior pole of the globe was 8.9 mm behind the interzygomatic line (range 5.0-12.7 mm). The mean diameters of the extraocular muscles in male patients were significantly larger than those in female patients (p < 0.001). Conclusion: These data may be of value in quantitatively evaluating MRI of the orbit. Index Terms: Magnetic resonance imaging-Orbit, oculomotor muscles-Orbit, anatomy-Nerves, optic.
MRI is widely used to image the orbit. The normal MR and CT anatomy of the orbit has been described in many references. Although normative measurements of orbital structures as seen on CT have been reported ( 1) and attempts made to establish the diameters of normal extraocular (oculomotor) muscles with ultrasound (2-4), to our knowledge, no normative MR data exist regarding these structures. In this study, we present data for the diameters of normal extraocular muscles and superior ophthalmic vein, width of the optic nerve-sheath complex, and the normal anterior protuberance of the globe from the orbit as seen on MRI.
patients (4 7 men and 53 women) were evaluated in this study. No contrast medium was given to any patient. The age range of the patients was 16-77 (mean 41 years). Informed consent was obtained from all patients.
Methods Axial and coronal 3-mm-thick sections with 0.3 mm intersection gap were obtained in all patients with a 0.5 T superconducting MR system by the use of a 13 cm diameter circular surface coil. Calibration of the scanner is periodically confirmed by imaging a standard. Measured images of the standard are required to be within ±0.33% of the actual value. The matrix was 256 x 256 corresponding to a field of view of 13 cm. The study consisted of spin echo Tl-weighted imaging (TRITE = 400130 ms) performed in axial and coronal planes with three excitations. Although the scanner has its own software to optimize contrast and brightness of the images, these settings were manually optimized to best define the orbital structures where necessary. Patients were asked to maintain primary gaze and gentle eye closure during the scans to prevent asymmetrical extraocular muscle contraction. Scans of the patients that could not maintain a forward gaze and scans with motion artifacts were not included in the study.
MATERIALS AND METHODS Patients All MR images in this study were obtained prospectively from patients who were referred to our department for MRI of the brain. All patients were free of clinical evidence or history of thyroid disease and any orbital or muscular disorder. Two hundred normal orbits of 100 From the Department of Radiology, School of Medicine, Hacettepe University, Sihhiye, 06100, Ankara, Turkey. Address correspondence and reprint requests to Dr. A. Ozgen.
493
A. OZGEN AND
494
Imaging of the orbit was performed immediately after imaging of the brain. Axial scans were obtained almost parallel to the optic nerve. Coronal scans were obtained perpendicular to the axial plane. To determine the normal globe position, the interzygomatic line at the midglobe section was used as a reference line. The length of the interzygomatic line and the perpendicular distance between the interzygomatic line and the posterior margin of the globe at the midglobe section were measured on axial scans. The width of the optic nerve-sheath complex was also measured perpendicular to its course in axial sections where the middle portion of the nerve was visualized. Horizontal diameters of the lateral and medial rectus muscles were measured on axial scans (Fig. lA-C). The diameter of the superior ophthalmic vein was also measured on axial scans at its proximal part (Fig. lD). The superior rectus and the levator palpebra superior muscles were measured together as a single muscle group (the superior muscle group) before they become separate with a fat plane in between. Diameters of the superior muscle group, the inferior rectus muscle, and the superior oblique muscle were measured on coronal scans (Fig. lE). The diameter of each muscle was measured at its maximum. Measurements were done on magnified hard copy images with the same magnification factor for all patients and then were converted to true size in millimeters.
Statistics Data for 200 orbits of 100 patients were evaluated. Mean and SD values of the data for each orbit were computed. The paired samples t test was used to compare
FIG. 1. MR scans of 42-year-old man. A-0: Axial images. The interzygomatic line (A and B) and the perpendicular distance from the interzygomatic line to the posterior margin of globe (C and D) are shown. White lines indicate maximum diameters of the medial rectus muscle, lateral rectus muscle, optic nerve-sheath complex, and superior ophthalmic vein, where measurements were made. E: Coronal image. White lines indicate maximum diameters of superior muscle group, inferior rectus muscle, and superior oblique muscle, where measurements were made.
J Comput Assist Tomogr, Vol. 24, No. 3, 2000
U.
AYDINGOZ
data obtained from the right and left orbits. The independent samples t test was used to compare data obtained from male and female patients. We calculated correlations using the Pearson correlation. Ninety-five percent normal cutoff values were computed by adding and subtracting 2 SD from the mean.
RESULTS Mean values and normal ranges for the diameters of extraocular muscles and the superior ophthalmic vein, the distance between the interzygomatic line and posterior margin of the globe, the width of the optic nervesheath complex, the length of the interzygomatic line, and the difference ranges between right and left orbit data are given in Table 1. No statistically significant difference was found between data for the right and left orbits in any orbital structure (p < 0.001). The mean ages of male and female patients were not statistically different (p = 0.16). The mean diameters of the extraocular muscles and the mean length of the interzygomatic line in male patients were significantly larger than those in female patients (p < 0.001). These comparative data are given in Table 2. The length of the interzygomatic line was positively correlated with the diameters of all extraocular muscles (p 2 0.25, r :::; 0.015) but the inferior rectus muscle (p = 0.16, r = 0.107). The width of the optic nerve-sheath complex, the diameter of the superior ophthalmic vein, and the globe position showed no statistically significant difference between male and female patients (p = 0.312, p = 0.316, and p = 0.179, respectively). We found no consistent correlation between age and
MEASUREMENTS OF ORBITAL STRUCTURES USING MRI
495
TABLE 1. Normative orbital measurements as seen on MRI Measurement
Mean (mm)
Normal range (mm) (mean± 2 SD)
Difference rangea (mm)
Muscle Medial rectus Lateral rectus Superior group Inferior rectus Superior oblique Superior ophthalmic vein Globe positionh Optic nerve-sheath complex Interzygomatic line
4.0 3.7 4.4 4.8 3.2 1.9 8.9 4.4 99
3.2--4.9 2.6-4.8 3.1-5.6 3.7-6.0 2.4--4.1 1.0-2.9 5.0-12.7 3.4-5.5 92-107
0.9 0.9 0.8 1.1 1.2 0.9 1.2 0.7
The medial and lateral rectus muscles and the superior ophthalmic vein were measured on axial plane; the superior group, inferior rectus, and superior oblique muscles were measured on coronal plane. The diameter of each muscle was measured at its maximum. a Difference range between right and left orbit data in 95th percentile limits. h Perpendicular distance between the interzygomatic line and the posterior margin of the globe.
Extraocular muscle enlargement may be due to primary neoplasm, metastatic tumor, nonspecific inflammation, vascular malformation, infection, acromegaly, and trauma as well as Graves ophthalmopathy, which is the most common cause (5-8). Enlargement of the superior ophthalmic vein may be due to vascular malformation, thrombosis, nonspecific inflammation, and compression. Optic nerve-sheath complex enlargement may arise from primary neoplasm, infection, nonspecific inflammation, metastatic tumor, hemorrhage, and increased intracranial pressure (9,10). Suggested MR protocols of the orbit include 3- to 5-mm-thick sections, 15 to 24 cm field of view with 256 x 256 matrix or 12 to 16 cm field of view with 256 x 128 to 192 matrix (9-11 ). Our imaging protocol, 3-mm-thick sections with 256 x 256 matrix corresponding to a field of view of 13 cm with three excitations, resulted in sufficient resolution and soft tissue contrast. Scans acquired with one or two excitations obtained with 1 or 1.5 T MR scanners would probably result in comparable images in a shorter scan time. This study consisted of spin echo Tl-weighted images as this sequence is widely used in imaging of the orbit and provides good contrast between
the width of the optic nerve-sheath complex, the diameter of the superior ophthalmic vein, or the globe position (p 2:: 0.195). Statistically significant correlation was found only between age and the diameter of the inferior and lateral rectus muscles (r = 0.20 , p = 0.047, and r = 0.24, p = 0.015, respectively). That is, these muscles tend to be slightly larger in older patients. No correlation was found between age and the diameters of the medial rectus muscle, the superior muscle group, and the superior oblique muscle.
DISCUSSION
MRI and CT are widely used techniques to image the orbit. High soft tissue contrast, multiplanar imaging capability, and lack of ionizing radiation make MRI superior to CT for imaging of the orbit. The normal MR and CT anatomy of the orbit has been described in many references. Normative measurements of orbital structures as seen on CT have also been described (1). However, we could not find normative MR data regarding these orbital structures.
TABLE 2. Diameters of normal extraocular muscles and length of the interzygomatic line in male and female patients as seen on MRI Normal range (mm) (mean± SD)
Meana (mm) Measurement
Men
Women
Men
Women
Muscle Medial rectus Lateral rectus Superior group Inferior rectus Superior oblique Interzygomatic line
4.2 3.9 4.5 5.1 3.7 101
3.9 3.5 4.2 4.6 3.5 97
3.4-5.0 2.8-5.0 3.3-5.9 3.9-6.2 2.9--4.5 94-108
3.0-4.8 2.5--4.5 3.0-5.4 3.5-5.8 2.7--4.3 92-103
The medial and lateral rectus muscles were measured on axial plane; the superior group, inferior rectus, and superior oblique muscles were measured on coronal plane. The diameter of each muscle was measured at its maximum. a Mean values for male and female patients are statistically different (p :::; 0.009).
J Comput Assist Tomogr, Vol. 24, No. 3, 2000
496
A. OZGEN AND
the retroorbital fat and the extraocular muscles and the optic nerve-sheath complex. Visualization of an eye muscle is optimal if the plane of section is parallel to the course of the muscle (10,12). The courses of the medial and lateral rectus muscles are almost parallel to each other as they are parallel to the axial plane. So, we preferred to use data obtained on axial sections to calculate the diameters of these extraocular muscles. However, the courses of the inferior rectus muscle and superior muscle group are not parallel to each other. Besides, none of them has a parallel course to the sagittal plane. Coronal scans may visualize these muscles close to their true cross-sections, although a small angulation may occur between the true crosssection of the muscle and the plane of section due to the slant angulation of these muscles. Measuring diameters of the extraocular muscles in the axis perpendicular to the orbital wall is used in CT and MRI worldwide (1,8,13-18). Therefore, we used coronal sections to measure the diameters of the inferior rectus muscle and the superior muscle group. Although there is no ideal section to visualize the superior oblique muscle due to its oblique course besides its slant angulation parallel to the orbital wall, we believe that this muscle is best visualized on coronal sections. The normative MR data established in this study are close to the normative CT data for orbital structures (1 ). Differences for upper normal limits for the diameters of extraocular muscles, optic nerve-sheath complex, and length of the interzygomatic line were :::;8.5%. A difference of 0.9 mm was encountered in normal limits for the anterior protuberance of the globe from the orbit, 5.9 mm in MR and 5.0 mm in CT. As normative data for the superior ophthalmic vein and the superior oblique muscle were not established before, no comparison is possible. Measurements of the same structure using CT and MR may result in different values depending on the imaging planes. Besides, every change in the window settings may result in different measurements in the same scan. We believe that both factors as well as the use of different patient populations may be responsible for differences between normative CT and MR data. Attempts were also made to establish diameters of normal extraocular muscles with sonography (2-4). As sonography is a real-time imaging method and sonographic measurements are highly observer dependent, these studies resulted in different norms for diameters of healthy extraocular muscles as they differ from our results. In conclusion, we present normative measurements for some orbital structures as seen on MRI. We believe that these data may help observers to accurately assess the enlargement of the extraocular muscles, the optic nerve-
J Comput Assist Tomogr, Vol. 24, No. 3, 2000
U.
AYDINGOZ
sheath complex, and the superior ophthalmic vein and to determine whether exophthalmos (or enophthalmos) is present in a practical quantitative method. These values may particularly be useful in evaluating the patients with Graves disease, which generally causes diffuse extraocular muscle enlargement and exophthalmos.
REFERENCES 1. Ozgen A, Ariyurek M. Normative measurements of orbital structures using CT. AIR 1998;170:1093-6. 2. Demer JL, Kerman BM. Comparison of standardized echography with magnetic resonance imaging to measure extraocular muscle size. Am J Ophthalmol 1994;118:351-61. 3. Byrne SF, Gendron EK, Glaser JS, Feuer W, Atta H. Diameter of normal extraocular muscles with echography. Am J Ophthalmol 1991;112:706-13. 4. McNutt LC, Kaefring SL, Ossoinig KC. Echographic measurement of extraocular muscles. In: White D, Brown RE, eds. Ultrasound in medicine. New York: Plenum Press, 1977:927-32. 5. Trokel SL, Hilal SK. Recognition and differential diagnosis of enlarged extraocular muscles in computed tomography. Am J Ophthalmol 1979;87:503-12. 6. Rothfus WE, Curtin HD. Extraocular muscle enlargement: CT review. Radiology 1983;151:677-81. 7. Patrinely JR, Osborn AG, Anderson RL, Whiting AS. Computed tomographic features of nonthyroid extraocular muscle enlargement. Ophthalmology 1989;96:1038-47. 8. Ozgen A, Alp MN, Ariyiirek, M, Tiitiincii NB, Can I, Giinalp I. Quantitative CT of the orbit in Graves disease. Br J Radio! 1999; 72:757-62. 9. Grossman CB. Magnetic resonance imaging and computed tomography of the head and spine. 2nd ed. Baltimore: Williams & Wilkins, 1995:605-53. 10. Mafee MF. Eye and orbit. In: Som PM, Curtin HD, eds. Head and neck imaging. 3rd ed. St. Louis: Mosby, 1996:1009-128. 11. Hesselink JR, Karampekios S. Normal computed tomography and magnetic resonance imaging anatomy of the globe, orbit, and visual pathways. Neuroimag Clin North Am 1996;6:15-27. 12. UnsOld R, Newton TH, De Groot J. CT evaluation of extraocular muscles-anatomic-CT correlations. Albrecht von Graefes Arch Klin Ophthalmol 1980;214:155-80. 13. Nugent RA, Belkin RI, Neigel JM, et al. Graves orbitopathy: correlation of CT and clinical findings. Radiology 1990;177:675-82. 14. Hallin ES, Feldon SE. Graves ophthalmopathy: 1. Simple CT estimates of extraocular muscle volume. Br J Ophthalmol 1988;72: 674-77. 15. Chen YL, Chang TC, Huang KM, Tzeng SS, Kao SCS. Relationship of eye movement to computed tomographic findings in patients with Graves ophthalmopathy. Acta Ophthalmol 1994;72: 472-7. 16. Wilson RG, Pope RM, Michell MJ, Cannon R, McGregor AM. The use of real-time ultrasound in Graves ophthalmopathy: a comparison with computed tomography. Br J Radial 1989;62:705-9. 17. Hiromatsu Y, Kojima K, Ishisaka N, et al. Role of magnetic resonance imaging in thyroid-associated ophthalmopathy: its predictive value for therapeutic outcome of immunosupressive therapy. Thyroid 1992;4:299-305. 18. Nishikawa M, Yoshimura M, Toyoda N, et al. Correlation of orbital muscle changes evaluated by magnetic resonance imaging and thyroid-stimulating antibody in patients with Graves ophthalmopathy. Acta Endocrinol 1993;129:213-9.
