- Email: [email protected]
Quandtadve Analysis of Optic Disc Cupping in Compressive Optic Neuropathy
Quantitative Analysis of Optic Disc Cupping in Compressive Optic Neuropathy Stefania Bianchi-Marzoli, MD/ Joseph F. Rizzo III, MD, 2 Rosario Brancato, MD/ Simmons Lessell, MD2 Purpose: Cupping of the optic disc, a characteristic sign of glaucoma, has been anecdotally described in association with compressive optic neuropathy. The aim of this study is to perform a masked, controlled, and quantitative measurement of the optic disc cup to determine if compressive lesions of the afferent visual pathway were associated with increased cupping. Methods: The ratio of cup area:disc area of 29 patients with intracranial lesions impinging on the optic nerves and the chiasm (14 with pituitary adenomas, 7 with me ningiomas, 6 with craniopharyngi!Jmas, and 2 with aneurysms) was compared with those of 20 age-matched control subjects. The areal ratios were derived planimetrically from hand-drawn images of magnified stereophotographs. Patients were divided into three groups based on the degree of laterality of visual compromise. Uninvolved eyes served as an internal control for patients with unilateral disease. Results: The median ratio of cup area:disc area was 0.37 for all eyes with visual compromise (n =51) and 0.10 for control eyes, which was statistically significant (P = 0.0001). The median intereye difference in the ratio of cup area:disc area was 0.13 for patients with unilateral lesions and 0.04 for control subjects. This difference also was statistically significant (P = 0.0001 ). Conclusions: The finding of intereye asymmetry in patients with unilateral optic nerve compression is convincing evidence that the enlarged cup is an acquired feature. Several types of compressive lesions of the anterior visual pathway can be associated with increased cupping of the optic disc in the absence of increased intraocular pressure. Ophthalmology 1995; 102:436-440
Enlargement of the optic cup, the essential topographic feature of glaucomatous optic atrophy, also has been re ported in anterior and posterior ischemic optic neurop athy, syphilitic optic neuritis, several forms of hereditary optic neuropathy, shock, and traumatic optic neuropa thy. l-ID There is also anecdotal, uncontrolled evidence that
supports the occurrence of cupping in patients with com pressive optic neuropathy.9- 17 These reports stimulated us to conduct a masked, controlled, and quantitative analysis of optic cup size in patients with compressive lesions of the optic nerves and chiasm and in an age matched, control population. The purpose is to determine if these compressive lesions were associated with increased cupping.
Originally received: March 23, 1994. Revision accepted: October 5, 1994. 1 Department of Ophthalmology and Visual Sciences, University of Mil ano, H. S. Raffaele, Milano.
Methods
2
Department of Ophthalmology, Harvard Medical School and the Mas sachusetts Eye and Ear Infirmary, Boston. Presented in part at the ARVO Annual Meeting, Sarasota, 1991. Reprint requests to Joseph Rizzo, MD, Massachusetts Eye and Ear In firmary , 243 Charles St, Boston, MA 02114.
436
From the database of the Massachusetts Eye and Ear In firmary, we identified 198 patients with tumors or aneu rysms near the optic nerves or chiasm. One hundred thir teen of the patients were excluded because stereophoto
Bianchi-Marzoli et al · Optic Disc Cupping in Compressive Optic Neuropathy graphs of the optic discs were unavailable or of poor quality. Three black subjects were excluded because of the tendency for blacks to have larger cups. 18 Subjects were excluded if they had a history ofocular disease, head trauma, neurosurgery, history of local or systemic corti costeroid therapy, refractive error greater than 4 diopters spherical equivalent, congenital disc anomaly, normal vi sual function (50 patients), family history of glaucoma, or intraocular pressure on any occasion of greater than 21 mmHg (3 subjects). The following data were abstracted from the records of the remaining 82 patients: best-cor rected Snellen visual acuity, refractive error, color vision (Ishihara), pupillary light reactions, applanation intra ocular pressure (obtained on at least 2 examinations), and Goldmann visual fields. Of the 29 study subjects, there were 17 men and 12 women, all of whom had visual dysfunction and unam biguous neuroradiologic signs of a compressive lesion of the optic nerves or chiasm. They ranged in age from 16 to 82 years (mean± standard deviation, 50.7 ± 17.9 years). There were 14 pituitary adenomas, 7 meningiomas, 6 craniopharyngiomas, and 2 aneurysms. Stereophoto graphs of the optic nerves were taken of each subject on the date of the initial visit. Patients were divided into three groups based on clin ical pattern of visual loss: symmetrical (group 1), asym metric (group 2), or strictly unilateral (group 3). Group I included 13 patients (8 with pituitary adenoma, 1 with olfactory groove meningioma, and 4 with craniophar yngioma), group 2 included nine patients (6 with pituitary adenoma, 2 with craniopharyngioma, and 1 with aneu rysm), and group 3 included seven patients (6 with me ningioma and I with aneurysm). All patients in group 3 had unequivocal evidence of an optic neuropathy on the involved side. Neuroimaging (computed tomography or magnetic resonance) supported the clinical interpreta tion of a unilateral optic neuropathy for all patients in group 3. The control subjects, who were solicited from friends and relatives, included 20 age-matched white men (n = 5) and women (n = 15) who met the exclusion criteria listed above. The mean age of the control subjects was 43.7 years (standard deviation, 22.3 years) (range, 16-76 years). All control subjects had normal findings on neuro ophthalmologic examination and intraocular pressures less than 21 mmHg. The identity of the patients was masked. Stereopho tographs were randomly mixed, projected, and magnified (X 15) onto paper. The disc and cup margins were plotted with the aid of a hand-held stereoviewer. The boundary of the cup was identified by noting, among other things, stereoscopic contours and deviations of small vessels on the surface of the optic disc. Margins ofdisc and cup were traced with a digital stylus, and ratio of cup area to disc area was calculated by a computer. Reproducibility of the method of measuring the ratios of cup area:disc area was assessed by two observers (SBM and JFR) making three sets of measurements of ten ran domly selected stereophotographs. The results were used to determine coefficients of variation for intra- and inter-
observer measurements. Thereafter, all measurements were made by one of us (SBM). The median ratio of cup area:disc area of both eyes of control subjects was compared with that of patients with the Mann-Whitney U test. A similar comparison was made after separating patients into groups according to the specific type of mass, which was based on clinical, radiologic, and, in some instances, histologic information. Data obtained from unaffected eyes of patients with strictly unilateral lesions were excluded from these anal yses (n = 7). Intereye differences in the ratio of cup area:disc area of all patients and (separately) control subjects were as sessed with the paired Student's t test. Additionally, the intereye difference in the cup:disc ratio of all patients ver sus control subjects was assessed with the Mann-Whitney U test. This latter test also was applied after the patients had been grouped according to the degree of symmetry ofvisual loss. The intereye difference for patients in group 3 were plotted in relation to duration of visual symptoms and degree of visual loss. Absolute values of intereye dif ferences were used.
Results The intra- and interobserver coefficients of variation were 0.03 and 0.02, respectively. Subsequent data are expressed as median (50th percentile) values with parenthetical no tation ofthe 25th and 75th percentiles. The median ratio ofcup area:disc area was 0.37 (0.25, 0.54) for all eyes with visual compromise (n = 51) and 0.1 (0.06, 0.17) for all control eyes (Table I), the difference of which was statis tically significant (P = 0.000 I). Significant differences also were found when separate comparisons of control subjects to patients with pituitary adenoma, meningioma, cran iopharyngioma, and intracranial aneurysm were made (P < 0.004 in each case). The intereye difference in the ratio of cup area:disc area between the controls' two eyes and (separately) the patients' two eyes was not statistically significant in either case (P = 0.84; P = 0.13, respectively; paired Student's t test). In addition, the median intereye difference in the ratio of cup area:disc area was 0.06 (0.03, 0.13) for all patients and 0.04 (0.02, 0.09) for control subjects (Table 2), and the comparison between controls and patients did Table 1. Median Ratio of Cup Area:Disc Area 25th,
75th No. of No. of Subjects Eyes Median Percentile Control subjects Patients Pituitary adenoma Meningioma Craniopharyngioma Aneurysm
20 29
14
7 6 2
40 51 28 8 12 3
0.1 0.37 0.37 0.59 0.26 0.78
0.06, 0.25, 0.27, 0.22, 0.07, 0.58,
0.17 0.54 0.45 0.81 0.35 0.79
437
Ophthalmology
Volume 102, Number 3, March 1995
Table 2. Median Intereye Difference in the Ratio of Cup Area:Disc Area
Control subjects Patients Group 1 Group 2 Group 3
No. of Subjects
No. of Eyes
20 29 13 9
58
7
40 26
18
14
Median 0.04 0.06 0.04 0.06 0.13
25th, 75th Percentile 0.02, 0.03, 0.02, 0.02, 0.11,
0.09 0.13 0.07 0.13 0.44
not show a statistically significant difference (P = 0.1; Mann-Whitney U). The median intereyedifference in the ratio ofcup area: disc area was 0.04 (0.02, 0.07) for patients in group I, 0.06 (0.02, 0.13) for those in group 2, and 0.13 (0.11, 0.44) for those in group 3. The intereye dif ference in ratio ofcup area:disc area between control sub jects and patients in group 3 was statistically significant (P = 0.0001 ), but this was not so for comparisons between control subjects and patients in groups 1 and 2 (P = 0.19 and P = 0.3, respectively) (Fig 1). There was no correlation between the magnitude ofthe intereye difference in ratio ofcup area: disc area and the duration of visual symptoms or the severity of visual loss.
Discussion Our data confirm an association between compression of the afferent visual pathway and cupping of the optic nerve. This relation had been noted previously in anecdotal reports9 - 16 and in a series of 16 patients by Kupersmith and Krohn. 17 These earlier studies, however, relied exclu sively on subjective and nonstereoscopic clinical estimates of the cup size, which is potentially subject to error. Be cause of this problem, 19•20 we studied our patients with a highly reproducible planimetric method. Furthermore, we evaluated cup:disc areas rather than linear measurement of diameters because of the increased accuracy afforded by measurements taken around a circumference rather than along a single meridian. The median ratio of cup area:disc area was larger in patients than in control subjects. This comparison, how ever, is simply one of statistical probability. To more rig orously ascertain whether cupping was associated with compression, we divided the patients into three groups based on symmetry of visual function. In this way, the unaffected eyes of patients with strictly unilateral disease served as an internal control because there is normally only a small intra-individual difference in cup:disc ratios. The most important result of this study was the statis tically significant intereye difference in patients with uni lateral optic neuropathy (Fig 1; Table 2). This finding is convincing evidence that the enlarged cup is an acquired feature. The median intereye differences were relatively small but did reach 0.27 or more in three of the seven patients with unilateral visual loss (Fig 2). Are the findings,
438
then, simply of interest for research or do they also have clinical use? In each case of unilateral visual compromise, the larger cup was associated with reduced Snellen visual acuity, dyschromatopsia, afferent pupillary defect (except in 1 patient), and visual field defect. This constellation offind ings is unambiguous evidence ofan optic neuropathy and suggests that the enlarged cup is a meaningful finding. In addition, the larger cup was always present on the involved side, and the intereye difference was smaller with asym metric, rather than unilateral, compression and smallest with bilaterally symmetric compression. This continuum provides additional supportive evidence that the cupping is related to compression; therefore, it is a clinically useful observation. The relatively small differences in ratios of cup area: disc area reported here should not suggest that the asym metries are of doubtful significance or that they would be hard to discern clinically. In this context, ratios using the formula for area of a circle (1rr) will, of necessity, be smaller than ratios using linear dimensions, which is the standard method when reporting cup:disc measurements. Consequentially, the larger squared value (of the radius) in the denominator of an areal ratio minimizes the result much more than the doubled value in the denominator of a linear ratio. This rationale accounts for the relatively small cup:disc ratios in our patients and control subjects. Our results are contrary to those of two earlier inves tigations. Schwartz21 studied 17 patients with a variety of neurologic problems, including optic neuritis, who were generally younger than patients in the other groups used for comparison. He concluded that cupping was not an important feature of neurologic disease affecting the optic nerve. Radius and Maumenee7 studied eight cases of compressive and other optic neuropathies and compared the cup:disc ratios with patients with mild diabetic reti nopathy. They concluded that cupping may be seen after trauma to the optic nerve but not after compression by a mass. Cupping of the optic nerve head is almost always as sociated with increased intraocular pressure. 22 •23 Non glaucomatous cupping occurs uncommonly but can be
.. 25th
0,5
"
~
o,4
~Q. = ."'"..' ".:.
o,3
• median •
75th
~ :;;
!
0,2
0,1
~
;
Controls
Group 1
I
Group 2
Group 3
Figure 1. Median (25th, 75th percentile) intereye difference in ratio of cup area:disc area of control subjects and the three groups of patients, which were divided based on symmetry of visual loss.
Bianchi-Marzoli et al · Optic Disc Cupping in Compressive Optic Neuropathy Figure 2. Gadolinium-enhanced Tl-weighted magnetic resonance scan of the brain in the coronal plane (top). A round, high intensity signal is adjacent to the optic nerve on the right side. This tumor compressed the optic nerve and caused progressive ipsilateral visual loss. The optic nerves of this patient (right eye, center; left eye, bottom) show pallor of the right eye and marked asymmetry in the degree of cupping between the eyes. Several vessels on the temporal side of the right optic nerve demarcate the temporal boundary of the enlarged cup. This is an extreme example of the asymmetry in cupping that may be seen.
observed in some cases of optic neuropathy secondary to arteritis, syphilis, trauma, shock, and in certain inherited disorders (autosomal dominant optic neuropathy and Le ber hereditary optic neuropathy). 1- 10 Normal-tension glaucoma is an enigmatic disease in which progressive cupping and visual loss occur without intraocular hyper tension .5 Glaucomatous enlargement of the optic nerve cup is thought to result from axonal loss and posterior bowing of the lamina cribrosa. 24 - 26 A single case of cupping as sociated with compression of the optic nerve (by an aneu 27 rysm) was studied pathologically by Portney and Roth, and loss of axons and glial tissue within the optic nerve head was thought to account for the cupping. Our study was not designed to assess the mechanism of increased cupping but it would seem likely that axonal loss is an important factor. Axonal loss alone, however, is not necessarily a suffi cient explanation for cupping. For example, significant visual loss from nonarteritic anterior ischemic optic neu ropathy or the devastating loss of axons that occurs with occlusion of the central retinal artery are not recognized causes of cupping. Loss of nonneural tissue on the optic nerve head, and very likely other unrecognized factors, is probably an important determinant that influences the potential for increased cupping. 24 How alteration in the nonneural composition of the optic nerve head might re sult from intracranial compression is not clear. Some have suggested that vascular compromise may cause cupping in situations of compression. 4 • 11 In summary, the finding of intereye asymmetry in pa tients with strictly unilateral visual loss clearly demon strates that compressive lesions can be associated with cupping of the optic nerve. The combination of increased cupping and normal intraocular pressure poses a risk for misdiagnosis of normal-tension glaucoma. Asymmetry in cupping, however, is only one factor that would be used in deciding if neuro-ophthalmologic investigation is nec essary. A detailed assessment of afferent visual function usually provides the necessary evidence to distinguish cupping caused by glaucoma versus compression.
References I. Sebag J, Thomas JV, Epstein DL, Grant WM. Optic disc cupping in arteritic anterior ischemic optic neuropathy re
439
Ophthalmology
2. 3. 4. 5. 6. 7. 8. 9. 10. II. 12. 13. 14. 15.
440
Volume 102, Number 3, March 1995
sembles glaucomatous cupping. Ophthalmology 1986;93: 357-61. Hayreh SS. Anterior ischaemic optic neuropathy. II. Fundus on ophthalmoscopy and fluorescein angiography. Br J Ophthalmol 1974;58:964-80. Sonty S, Schwartz B. Development of cupping and pallor in posterior ischemic optic neuropathy. Int Ophthalmol 1983;6:213-20. Blazar HA, Scheie HG. Pseudoglaucoma. Arch Ophthalmol 1950;44:499-513. Levene RZ. Low tension glaucoma: a critical review and new material. Surv Ophthalmol 1980;24:621-64. Kline LB, Glaser JS. Dominant optic atrophy: the clinical profile. Arch Ophthalmol 1979;97: 1680-6. Radius RL, Maumenee AE. Optic atrophy and glaucoma tous cupping. Am J Ophthalmol 1978;85: 145-53. Drance SM, Morgan RW, Sweeney VP. Shock-induced optic neuropathy: a cause of nonprogressive glaucoma. N Eng! J Med 1973;288:392-5. Miller NR. Walsh and Hoyt's Clinical Neuro-Ophthalmol ogy, 4th ed. Vol. I. Baltimore: Williams & Wilkins, 1982;334-6. Trobe JD, Glaser JS, Cassady J, et al. Nonglaucomatous excavation of the optic disc. Arch Ophthalmol 1980;98: 1046-50. Dalsgaard-Nielsen E. Glaucoma-like cupping of the optic disc and its etiology. Arch Ophthalmol 1937;15:151-78. Thiel G. Glaukom ohne Hochdruck. Klin Monatsbl Au genheilkd 1930;85:106-7. Knapp A. Course in certain cases of atrophy of the optic nerve with cupping and low tension. Arch Ophthalmol 1940;23:41-7. Hamann J. Beitrag zur Diagnose und Therapie von Hy pophysentumoren. Ztschr f Augenh Eilkd 1929;68:317-30. Koyanagi Y, Takahashi T. Kavernose Sehnervenatrophie
bei Orbitaltumoren. Graefes Arch Ophthalmol 1925;115: 596-619. 16. Kalenak JW, Kosmorsky GS, Hassenbusch SJ. Compression of the intracranial optic nerve mimicking unilateral normal pressure glaucoma. J Clin Neuro-ophthalmol 1992;12:230-5. 17. Kupersmith MJ, Krohn D. Cupping of the optic disc with compressive lesions of the anterior visual pathway. Ann Ophthalmol 1984; 16:948-53. 18. Beck RW, Messner DK, Musch DC, et al. Is there a racial difference in physiologic cup size? Ophthalmology 1985;92: 873-6. 19. Trobe JD, Glaser JS, Cassady JC. Optic atrophy: differential diagnosis by fundus observation alone. Arch Ophthalmol 1980;98: I 040-5. 20. Varma R, Steinmann WC, Scott IU. Expert agreement in evaluating the optic disc for glaucoma. Ophthalmology 1992;99:215-21. 21. Schwartz B. Differences between cupping and pallor of the optic disc in glaucomatous and neurological optic nerve disease. In: Acta XXIII Concilium Ophthalmologicum, Kyoto, 1978;450: I I08-11. 22. Schwartz B. Cupping and pallor of the optic disc. Arch Ophthalmol 1973;89:272-7. 23. Kirsch RE, Anderson DR. Clinical recognition of glauco matous cupping. Am J Ophthalmol 1973;75:442-54. 24. Hayreh SS. Pathogenesis of cupping of the optic disc. Br J Ophthalmol 1974;58:863-76. 25. Quigley HA, Green WR. The histology of human glaucoma cupping and optic nerve damage: clinicopathologic corre lation in 21 eyes. Ophthalmology 1979;86: 1803-27. 26. Quigley HA, Addicks EM. Regional differences in the structure ofthe lamina crib rosa and their relation to glaucomatous optic nerve damage. Arch Ophthalmol 1981 ;99: 137-43. 27. Portney GL, Roth AM. Optic cupping caused by an intra cranial aneurysm . Am J Ophthalmol 1977;84:98-103.
Enlargement of the optic cup, the essential topographic feature of glaucomatous optic atrophy, also has been re ported in anterior and posterior ischemic optic neurop athy, syphilitic optic neuritis, several forms of hereditary optic neuropathy, shock, and traumatic optic neuropa thy. l-ID There is also anecdotal, uncontrolled evidence that
supports the occurrence of cupping in patients with com pressive optic neuropathy.9- 17 These reports stimulated us to conduct a masked, controlled, and quantitative analysis of optic cup size in patients with compressive lesions of the optic nerves and chiasm and in an age matched, control population. The purpose is to determine if these compressive lesions were associated with increased cupping.
Originally received: March 23, 1994. Revision accepted: October 5, 1994. 1 Department of Ophthalmology and Visual Sciences, University of Mil ano, H. S. Raffaele, Milano.
Methods
2
Department of Ophthalmology, Harvard Medical School and the Mas sachusetts Eye and Ear Infirmary, Boston. Presented in part at the ARVO Annual Meeting, Sarasota, 1991. Reprint requests to Joseph Rizzo, MD, Massachusetts Eye and Ear In firmary , 243 Charles St, Boston, MA 02114.
436
From the database of the Massachusetts Eye and Ear In firmary, we identified 198 patients with tumors or aneu rysms near the optic nerves or chiasm. One hundred thir teen of the patients were excluded because stereophoto
Bianchi-Marzoli et al · Optic Disc Cupping in Compressive Optic Neuropathy graphs of the optic discs were unavailable or of poor quality. Three black subjects were excluded because of the tendency for blacks to have larger cups. 18 Subjects were excluded if they had a history ofocular disease, head trauma, neurosurgery, history of local or systemic corti costeroid therapy, refractive error greater than 4 diopters spherical equivalent, congenital disc anomaly, normal vi sual function (50 patients), family history of glaucoma, or intraocular pressure on any occasion of greater than 21 mmHg (3 subjects). The following data were abstracted from the records of the remaining 82 patients: best-cor rected Snellen visual acuity, refractive error, color vision (Ishihara), pupillary light reactions, applanation intra ocular pressure (obtained on at least 2 examinations), and Goldmann visual fields. Of the 29 study subjects, there were 17 men and 12 women, all of whom had visual dysfunction and unam biguous neuroradiologic signs of a compressive lesion of the optic nerves or chiasm. They ranged in age from 16 to 82 years (mean± standard deviation, 50.7 ± 17.9 years). There were 14 pituitary adenomas, 7 meningiomas, 6 craniopharyngiomas, and 2 aneurysms. Stereophoto graphs of the optic nerves were taken of each subject on the date of the initial visit. Patients were divided into three groups based on clin ical pattern of visual loss: symmetrical (group 1), asym metric (group 2), or strictly unilateral (group 3). Group I included 13 patients (8 with pituitary adenoma, 1 with olfactory groove meningioma, and 4 with craniophar yngioma), group 2 included nine patients (6 with pituitary adenoma, 2 with craniopharyngioma, and 1 with aneu rysm), and group 3 included seven patients (6 with me ningioma and I with aneurysm). All patients in group 3 had unequivocal evidence of an optic neuropathy on the involved side. Neuroimaging (computed tomography or magnetic resonance) supported the clinical interpreta tion of a unilateral optic neuropathy for all patients in group 3. The control subjects, who were solicited from friends and relatives, included 20 age-matched white men (n = 5) and women (n = 15) who met the exclusion criteria listed above. The mean age of the control subjects was 43.7 years (standard deviation, 22.3 years) (range, 16-76 years). All control subjects had normal findings on neuro ophthalmologic examination and intraocular pressures less than 21 mmHg. The identity of the patients was masked. Stereopho tographs were randomly mixed, projected, and magnified (X 15) onto paper. The disc and cup margins were plotted with the aid of a hand-held stereoviewer. The boundary of the cup was identified by noting, among other things, stereoscopic contours and deviations of small vessels on the surface of the optic disc. Margins ofdisc and cup were traced with a digital stylus, and ratio of cup area to disc area was calculated by a computer. Reproducibility of the method of measuring the ratios of cup area:disc area was assessed by two observers (SBM and JFR) making three sets of measurements of ten ran domly selected stereophotographs. The results were used to determine coefficients of variation for intra- and inter-
observer measurements. Thereafter, all measurements were made by one of us (SBM). The median ratio of cup area:disc area of both eyes of control subjects was compared with that of patients with the Mann-Whitney U test. A similar comparison was made after separating patients into groups according to the specific type of mass, which was based on clinical, radiologic, and, in some instances, histologic information. Data obtained from unaffected eyes of patients with strictly unilateral lesions were excluded from these anal yses (n = 7). Intereye differences in the ratio of cup area:disc area of all patients and (separately) control subjects were as sessed with the paired Student's t test. Additionally, the intereye difference in the cup:disc ratio of all patients ver sus control subjects was assessed with the Mann-Whitney U test. This latter test also was applied after the patients had been grouped according to the degree of symmetry ofvisual loss. The intereye difference for patients in group 3 were plotted in relation to duration of visual symptoms and degree of visual loss. Absolute values of intereye dif ferences were used.
Results The intra- and interobserver coefficients of variation were 0.03 and 0.02, respectively. Subsequent data are expressed as median (50th percentile) values with parenthetical no tation ofthe 25th and 75th percentiles. The median ratio ofcup area:disc area was 0.37 (0.25, 0.54) for all eyes with visual compromise (n = 51) and 0.1 (0.06, 0.17) for all control eyes (Table I), the difference of which was statis tically significant (P = 0.000 I). Significant differences also were found when separate comparisons of control subjects to patients with pituitary adenoma, meningioma, cran iopharyngioma, and intracranial aneurysm were made (P < 0.004 in each case). The intereye difference in the ratio of cup area:disc area between the controls' two eyes and (separately) the patients' two eyes was not statistically significant in either case (P = 0.84; P = 0.13, respectively; paired Student's t test). In addition, the median intereye difference in the ratio of cup area:disc area was 0.06 (0.03, 0.13) for all patients and 0.04 (0.02, 0.09) for control subjects (Table 2), and the comparison between controls and patients did Table 1. Median Ratio of Cup Area:Disc Area 25th,
75th No. of No. of Subjects Eyes Median Percentile Control subjects Patients Pituitary adenoma Meningioma Craniopharyngioma Aneurysm
20 29
14
7 6 2
40 51 28 8 12 3
0.1 0.37 0.37 0.59 0.26 0.78
0.06, 0.25, 0.27, 0.22, 0.07, 0.58,
0.17 0.54 0.45 0.81 0.35 0.79
437
Ophthalmology
Volume 102, Number 3, March 1995
Table 2. Median Intereye Difference in the Ratio of Cup Area:Disc Area
Control subjects Patients Group 1 Group 2 Group 3
No. of Subjects
No. of Eyes
20 29 13 9
58
7
40 26
18
14
Median 0.04 0.06 0.04 0.06 0.13
25th, 75th Percentile 0.02, 0.03, 0.02, 0.02, 0.11,
0.09 0.13 0.07 0.13 0.44
not show a statistically significant difference (P = 0.1; Mann-Whitney U). The median intereyedifference in the ratio ofcup area: disc area was 0.04 (0.02, 0.07) for patients in group I, 0.06 (0.02, 0.13) for those in group 2, and 0.13 (0.11, 0.44) for those in group 3. The intereye dif ference in ratio ofcup area:disc area between control sub jects and patients in group 3 was statistically significant (P = 0.0001 ), but this was not so for comparisons between control subjects and patients in groups 1 and 2 (P = 0.19 and P = 0.3, respectively) (Fig 1). There was no correlation between the magnitude ofthe intereye difference in ratio ofcup area: disc area and the duration of visual symptoms or the severity of visual loss.
Discussion Our data confirm an association between compression of the afferent visual pathway and cupping of the optic nerve. This relation had been noted previously in anecdotal reports9 - 16 and in a series of 16 patients by Kupersmith and Krohn. 17 These earlier studies, however, relied exclu sively on subjective and nonstereoscopic clinical estimates of the cup size, which is potentially subject to error. Be cause of this problem, 19•20 we studied our patients with a highly reproducible planimetric method. Furthermore, we evaluated cup:disc areas rather than linear measurement of diameters because of the increased accuracy afforded by measurements taken around a circumference rather than along a single meridian. The median ratio of cup area:disc area was larger in patients than in control subjects. This comparison, how ever, is simply one of statistical probability. To more rig orously ascertain whether cupping was associated with compression, we divided the patients into three groups based on symmetry of visual function. In this way, the unaffected eyes of patients with strictly unilateral disease served as an internal control because there is normally only a small intra-individual difference in cup:disc ratios. The most important result of this study was the statis tically significant intereye difference in patients with uni lateral optic neuropathy (Fig 1; Table 2). This finding is convincing evidence that the enlarged cup is an acquired feature. The median intereye differences were relatively small but did reach 0.27 or more in three of the seven patients with unilateral visual loss (Fig 2). Are the findings,
438
then, simply of interest for research or do they also have clinical use? In each case of unilateral visual compromise, the larger cup was associated with reduced Snellen visual acuity, dyschromatopsia, afferent pupillary defect (except in 1 patient), and visual field defect. This constellation offind ings is unambiguous evidence ofan optic neuropathy and suggests that the enlarged cup is a meaningful finding. In addition, the larger cup was always present on the involved side, and the intereye difference was smaller with asym metric, rather than unilateral, compression and smallest with bilaterally symmetric compression. This continuum provides additional supportive evidence that the cupping is related to compression; therefore, it is a clinically useful observation. The relatively small differences in ratios of cup area: disc area reported here should not suggest that the asym metries are of doubtful significance or that they would be hard to discern clinically. In this context, ratios using the formula for area of a circle (1rr) will, of necessity, be smaller than ratios using linear dimensions, which is the standard method when reporting cup:disc measurements. Consequentially, the larger squared value (of the radius) in the denominator of an areal ratio minimizes the result much more than the doubled value in the denominator of a linear ratio. This rationale accounts for the relatively small cup:disc ratios in our patients and control subjects. Our results are contrary to those of two earlier inves tigations. Schwartz21 studied 17 patients with a variety of neurologic problems, including optic neuritis, who were generally younger than patients in the other groups used for comparison. He concluded that cupping was not an important feature of neurologic disease affecting the optic nerve. Radius and Maumenee7 studied eight cases of compressive and other optic neuropathies and compared the cup:disc ratios with patients with mild diabetic reti nopathy. They concluded that cupping may be seen after trauma to the optic nerve but not after compression by a mass. Cupping of the optic nerve head is almost always as sociated with increased intraocular pressure. 22 •23 Non glaucomatous cupping occurs uncommonly but can be
.. 25th
0,5
"
~
o,4
~Q. = ."'"..' ".:.
o,3
• median •
75th
~ :;;
!
0,2
0,1
~
;
Controls
Group 1
I
Group 2
Group 3
Figure 1. Median (25th, 75th percentile) intereye difference in ratio of cup area:disc area of control subjects and the three groups of patients, which were divided based on symmetry of visual loss.
Bianchi-Marzoli et al · Optic Disc Cupping in Compressive Optic Neuropathy Figure 2. Gadolinium-enhanced Tl-weighted magnetic resonance scan of the brain in the coronal plane (top). A round, high intensity signal is adjacent to the optic nerve on the right side. This tumor compressed the optic nerve and caused progressive ipsilateral visual loss. The optic nerves of this patient (right eye, center; left eye, bottom) show pallor of the right eye and marked asymmetry in the degree of cupping between the eyes. Several vessels on the temporal side of the right optic nerve demarcate the temporal boundary of the enlarged cup. This is an extreme example of the asymmetry in cupping that may be seen.
observed in some cases of optic neuropathy secondary to arteritis, syphilis, trauma, shock, and in certain inherited disorders (autosomal dominant optic neuropathy and Le ber hereditary optic neuropathy). 1- 10 Normal-tension glaucoma is an enigmatic disease in which progressive cupping and visual loss occur without intraocular hyper tension .5 Glaucomatous enlargement of the optic nerve cup is thought to result from axonal loss and posterior bowing of the lamina cribrosa. 24 - 26 A single case of cupping as sociated with compression of the optic nerve (by an aneu 27 rysm) was studied pathologically by Portney and Roth, and loss of axons and glial tissue within the optic nerve head was thought to account for the cupping. Our study was not designed to assess the mechanism of increased cupping but it would seem likely that axonal loss is an important factor. Axonal loss alone, however, is not necessarily a suffi cient explanation for cupping. For example, significant visual loss from nonarteritic anterior ischemic optic neu ropathy or the devastating loss of axons that occurs with occlusion of the central retinal artery are not recognized causes of cupping. Loss of nonneural tissue on the optic nerve head, and very likely other unrecognized factors, is probably an important determinant that influences the potential for increased cupping. 24 How alteration in the nonneural composition of the optic nerve head might re sult from intracranial compression is not clear. Some have suggested that vascular compromise may cause cupping in situations of compression. 4 • 11 In summary, the finding of intereye asymmetry in pa tients with strictly unilateral visual loss clearly demon strates that compressive lesions can be associated with cupping of the optic nerve. The combination of increased cupping and normal intraocular pressure poses a risk for misdiagnosis of normal-tension glaucoma. Asymmetry in cupping, however, is only one factor that would be used in deciding if neuro-ophthalmologic investigation is nec essary. A detailed assessment of afferent visual function usually provides the necessary evidence to distinguish cupping caused by glaucoma versus compression.
References I. Sebag J, Thomas JV, Epstein DL, Grant WM. Optic disc cupping in arteritic anterior ischemic optic neuropathy re
439
Ophthalmology
2. 3. 4. 5. 6. 7. 8. 9. 10. II. 12. 13. 14. 15.
440
Volume 102, Number 3, March 1995
sembles glaucomatous cupping. Ophthalmology 1986;93: 357-61. Hayreh SS. Anterior ischaemic optic neuropathy. II. Fundus on ophthalmoscopy and fluorescein angiography. Br J Ophthalmol 1974;58:964-80. Sonty S, Schwartz B. Development of cupping and pallor in posterior ischemic optic neuropathy. Int Ophthalmol 1983;6:213-20. Blazar HA, Scheie HG. Pseudoglaucoma. Arch Ophthalmol 1950;44:499-513. Levene RZ. Low tension glaucoma: a critical review and new material. Surv Ophthalmol 1980;24:621-64. Kline LB, Glaser JS. Dominant optic atrophy: the clinical profile. Arch Ophthalmol 1979;97: 1680-6. Radius RL, Maumenee AE. Optic atrophy and glaucoma tous cupping. Am J Ophthalmol 1978;85: 145-53. Drance SM, Morgan RW, Sweeney VP. Shock-induced optic neuropathy: a cause of nonprogressive glaucoma. N Eng! J Med 1973;288:392-5. Miller NR. Walsh and Hoyt's Clinical Neuro-Ophthalmol ogy, 4th ed. Vol. I. Baltimore: Williams & Wilkins, 1982;334-6. Trobe JD, Glaser JS, Cassady J, et al. Nonglaucomatous excavation of the optic disc. Arch Ophthalmol 1980;98: 1046-50. Dalsgaard-Nielsen E. Glaucoma-like cupping of the optic disc and its etiology. Arch Ophthalmol 1937;15:151-78. Thiel G. Glaukom ohne Hochdruck. Klin Monatsbl Au genheilkd 1930;85:106-7. Knapp A. Course in certain cases of atrophy of the optic nerve with cupping and low tension. Arch Ophthalmol 1940;23:41-7. Hamann J. Beitrag zur Diagnose und Therapie von Hy pophysentumoren. Ztschr f Augenh Eilkd 1929;68:317-30. Koyanagi Y, Takahashi T. Kavernose Sehnervenatrophie
bei Orbitaltumoren. Graefes Arch Ophthalmol 1925;115: 596-619. 16. Kalenak JW, Kosmorsky GS, Hassenbusch SJ. Compression of the intracranial optic nerve mimicking unilateral normal pressure glaucoma. J Clin Neuro-ophthalmol 1992;12:230-5. 17. Kupersmith MJ, Krohn D. Cupping of the optic disc with compressive lesions of the anterior visual pathway. Ann Ophthalmol 1984; 16:948-53. 18. Beck RW, Messner DK, Musch DC, et al. Is there a racial difference in physiologic cup size? Ophthalmology 1985;92: 873-6. 19. Trobe JD, Glaser JS, Cassady JC. Optic atrophy: differential diagnosis by fundus observation alone. Arch Ophthalmol 1980;98: I 040-5. 20. Varma R, Steinmann WC, Scott IU. Expert agreement in evaluating the optic disc for glaucoma. Ophthalmology 1992;99:215-21. 21. Schwartz B. Differences between cupping and pallor of the optic disc in glaucomatous and neurological optic nerve disease. In: Acta XXIII Concilium Ophthalmologicum, Kyoto, 1978;450: I I08-11. 22. Schwartz B. Cupping and pallor of the optic disc. Arch Ophthalmol 1973;89:272-7. 23. Kirsch RE, Anderson DR. Clinical recognition of glauco matous cupping. Am J Ophthalmol 1973;75:442-54. 24. Hayreh SS. Pathogenesis of cupping of the optic disc. Br J Ophthalmol 1974;58:863-76. 25. Quigley HA, Green WR. The histology of human glaucoma cupping and optic nerve damage: clinicopathologic corre lation in 21 eyes. Ophthalmology 1979;86: 1803-27. 26. Quigley HA, Addicks EM. Regional differences in the structure ofthe lamina crib rosa and their relation to glaucomatous optic nerve damage. Arch Ophthalmol 1981 ;99: 137-43. 27. Portney GL, Roth AM. Optic cupping caused by an intra cranial aneurysm . Am J Ophthalmol 1977;84:98-103.