ACTA 0 P H T H A L M 0 LOG I CA 72 (1994) 162-167

lnterocular differences in optic disc configuration in the unilateral exfoliation syndrome

Goji Tomita, Paivi Puska and Christina Raitta

Department of Ophthalmology, Helsinki University Central Hospital, University of Helsinki, Helsinki, Finland

Abstract. Thirty-seven patients with normal visual fields and open angles in both eyes, and lens exfoliation in only one eye, were examined for intraocular pressure, and optic disc topography was determined from stereo- photographs of the fundi with the computerized three- dimensional image analyzer. The rim-to-disc radius ra- tios at intervals of 30" were obtained with the instrument. The mean intraocular pressure of the exfoliative eyes (17.5 f 2.9 mmHg) was significantly higher than that of the nonexfoliative eyes (15.9 f 2.8 mmHg). In the exfolia- tive eyes, the mean rim-to-disc radius ratio at the infe- rior-temporal was significantly smaller, and the higher the intraocular pressure of the exfoliative eye in com- parison with that of the nonexfoliative eye, the smaller were the rim-to-disc radius ratios at the inferior-tempo- ral and inferior radii. In the exfoliative eyes, pressure-de- pendent changes are observable in the inferior and infe- rior-temporal areas of the optic disc prior to visual field loss.

Key words: unilateral exfoliation syndrome - optic nerve head - computerized image analysis - rim-to-disc ratio - glaucoma.

The exfoliation syndrome has been claimed to be a risk factor for glaucoma (Tarkkanen 1962; Hansen & Sellevold 1968; Aasved 1971). Among persons with exfoliation, glaucoma affects about 20% com- pared with 1.2% among persons without exfolia- tion (Aasved 1971). Of the remaining 80% of non- glaucomatous persons with exfoliation, some 15 to 35% later develop ocular hypertension or glau- coma (Henry et al. 1987; Klemetti 1988). However, how to distinguish those who may be at risk for de-

162

veloping glaucoma from those who will retain nor- mal vision is still not known. Puska & Raitta (1992) showed that there were no differences in the global optic disc parameters such as rim area or cup vol- ume between the two eyes of patients with unilat- eral exfoliation and no glaucoma, although the ex- foliative eyes had significantly higher intraocular pressures (IOP).

The purpose of the present study was to look for the sectional differences in optic disc contig...- tion between the two eyes of patients with unilat- eral exfoliation syndrome and no visual field loss. We evaluated the rim-to-disc radius ratio at inter- vals of 30", using a computerized image analysis system.

Material and Method

The unilateral exfoliation syndrome without vis- ual field loss was defined as normal visual fields with open angles in both eyes and lens exfoliation observed in only one eye. Initially, we prospectively identified 66 patients with unilateral exfoliation syndrome in whom exfoliation of the lens was de- tected after dilation of the pupil in one eye, and no biomicroscopically detectable exfoliation at a pu- pillary dilation of at least 6 nun in the contralateral eye. None of these patients had previous intraocu- lar operations or used ocular medication. We then further selected those patients who met the follow- ing criteria: 1) previous experience of automated perimetry and no visual field loss detectable with

the Octopus automated perimeter (Octupus 200R, Program G1, Interzeag, Schlieren, Switzerland) in both eyes with reliable measurements (false-posi- tive and negative ratios less than 15%); 2) availa- bility of well-focused and clear-imaged simulta- neous stereophotographs of the optic discs of both eyes for computerized image analysis.

We defined a visual field as normal for the cen- tral 26" when it did not meet the following criteria of an abnormal visual field (Caprioli 1991): 1) three or more adjacent points of >5 dB loss each; 2) two or more adjacent points of >10 dB; or 3) a dif- ference of > 10 dB across the nasal horizontal meri- dian at two or more adjacent points. For the outer visual field of the G1 program, we defined as ab- normal those with two or more absolute defects on the nasal side. If a patient had an abnormal visual field, we repeated the measurements at the next visit and did not define the visual field as abnormal if the field losses had altered (i.e., if they were at different locations or had disappeared).

Thirty-seven patients met all the entry criteria and were enrolled in the current study. The series comprised 32 women and 5 men aged 45 to 81 years (mean 67.7 years).

Stereophotography of the optic disc, intraocular pressure measurement with a Goldmann applana- tion tonometer, and visual field testing were done within 1 month. The simultaneous stereo-fundus camera (Topcon TRC-S52, Topcon Co., Tokyo, Japan) was used for optic disc photography with Kodak Ektachrome lOOHC film.

Topographic analysis of stereoscopic slides was performed with the computerized three-dimensio- lan image analysis system Topcon IMAGEnet (Top- con TRC-S52, Topcon Co., Tokyo, Japan) by one examiner (PP) masked to clinical information. The principles and mode of operation of the instru- ment have been described elsewhere (Varma et al. 1988, 1989a; Nanba et al. 1989; Shirakashi et al. 1992). The inner border of the whitish scleral ring is referred to as the disc edge. The optic cup cir- cumference was defined as being 100-200 pm below the surface of the optic disc. We changed the cup definition for individual discs to obtain a good correspondence with the clinical stereoscopic view of the optic cups, because it has been demon- strated that in specific types of optic discs, esti- mates of cup-to-disc ratios obtained by clinical examination can differ from those obtained by image analysis (varma et al. 1989b). In the stereo-

11 *

Superior Superior-temporal

Superior-nasal K r - 7 Ternporal- Nasal- / \ s m 'r.. / superior

Nasal-inferio

Inferior- iasal

-iTempora' Ternporal- inferior

Fig 1. Rim-to-disc radius ratios were measured at the sections divided at intervals of 30".

scopic view, we considered the cup edge to be the portion where the nerve fibers intersect the im- aginary plane established by the inner margin of the peripapillary scleral rim (Zeyen & Caprioli 1993). Ifthe examiner could not agree with the cup circumferene determined by the computer, the cup definition was changed repeatedly until the computer had shown the circumference which was almost comparable to the examiner's evaluation with the stereoscopic view. A modification of Littmann's method (Littmann 1982) adapted for use with the Topcon fundus camera (unpublished data and formula, Topcon Co., Tokyo, Japan) was used to correct the magnification of the central fundus, taking into account the anterior corneal curvature, refraction, and axial length.

We obtained the horizontal and vertical cup-to- disc ratios, rim-to-disc ratio, and cup volume (mm3) as the global parameters of optic disc topo- graphy. As parameters for evaluating sectional dif- ferences in the optic disc, we obtained the rim-to- disc radius ratios at intervals of lo", which were computed by the internal software of the instru- ment. We used measurements for 12 sections at in- tervals of 30" including the horizontal and vertical axes (i.e., the nasal, nasal-superior, superior-nasal, superior, superior-temporal, temporal-superior,

163

temporal, temporal-inferior, inferior-temporal, in- ferior, inferior-nasal, and nasal-inferior (Fig. 1). The median of the intraexaminer coefficients of variation was 2.6% for the vertical cup-to-disc ratio, 3.8% for the horizontal cup-to-disc ratio, 1.4% for the rim-to-disc 4.7% for the cup volume, and 8.9% for the rim-to-disc radius ratios, when 10 randomly selected optic disc photographs of 10 pa- tients were analyzed twice with an interval of at least 1 week.

The two-tailed paired Student's t-test was used to compare differences between the two eyes. The two-tailed two-sample Student's t-test was used to compare the independent groups. Spearman's correlation coefficient was used for correlation analysis. A p-value of less than 0.05 was considered to be statistically significant.

Sections

Results

Of the 37 patients, 3 (8.1%) had an IOP of 22 mmHg in the exfoliative eye, and 1 (2.7%) IOPs of 23 mmHg in both eyes. The remaining patients had IOPs of less than 22 mmHg in both eyes. As com- pared with the nonexfoliative eyes (range 11-23 mmHg), the exfoliative eyes (range 13-23 mmHg) had a significantly higher mean IOP (t=4.33, p = 0.0001; Table 1). The mean f SD of the mean de- fect of the visual field indices was 1.89 * 1.10 dB for the exfoliative eyes and 1.75 k 1.21 dB for the non- exfoliative eyes. No significant difference was ob- served in the mean defect between the two eyes. There were no significant differences in global

Exfoliative Nonexfoliative

(mean f SD) eye eye P

(mean f SD)

Exfoliative Nonexfoliative

(mean f SD)

eye eye (mean f SD)

IOP ( m a g ) 17.5 f 2.9 15.9 f 2.8 < 0.01

Disc parameters Cup-to-disc ratio Vertical 0.51 f0.13 0.48f0.14 NS Horizontal 0.51 f0.13 0.51 f0.16 NS

Rim-to-disc ratio 0.72 f0.13 0.73f0.14 NS

Cup volume (mm') 0.20f0.18 0.19*0.17 NS

P

IOP: intraocular pressure; SD: standard deviation; NS: not significant.

164

Table 2. Sections were divided at intervals of 30". Rim-to-disc radius ratio at each section (n = 37).

NS: not significant.

optic disc parameters between the two eyes (Table 1). No significant differences were also observed in the ratio of the horizontal to vertical cup-to-disc ratio between the two eyes (1.05 f 0.32 for the ex- foliative eyes and 0.96 f 0.20 for the nonexfolitive eyes).

Of the rim-to-disc radius ratios at intervals of 30", the ratio at the inferior-temporal radius was significantly smaller in the exfoliative eye than in the nonexfoliative eye (t = 2.04, p = 0.048; Table 2). Significant correlations were found between the difference in IOP and the interocular differences (exfoliative eye - nonexfoliative eye) between the rim-to-disc radius ratios at the inferior-temporal (r = - 0.325, p = 0.049) and the inferior (r =

- 0.384, p = 0.019) radii (i.e., the higher the IOP of the exfoliative eye in comparison with that of the nonexfoliative eye, the smaller were the rim-to- disc radius ratios at the inferior-temporal and infe- rior radii) (Table 3).

On the basis of the interocular difference (exfoli- ative eye - nonexfoliative eye) in IOP (mean dif- ference, 1.62 mmHg), patients were assigned to one or two of three groups: IOP difference < 2 mmHg (group with s m a l l differences, n=21), IOP dif- ferences < 2 mmHg (group with large differences,

Sections

Nasal-inferior Nasal Nasal-superior

Superior-nasal Superior Superior-temporal

Temporal-superior Temporal Temporal-inferior

Inferior- temporal Inferior Inferior-nasal

r P

- 0.18 0.03 0.13 0.30 0.14 0.17 0.09 0.13 0.18

- 0.33 - 0.38 - 0.24

Exfoliative eye (mean f SD)

NS NS NS NS NS NS

NS NS NS

< 0.05 < 0.05

NS

I p Nonexfoliative eye (mean f SD)

NS: not significant.

n = 16), and IOP difference > 2 mmHg and IOPs < 22 mmHg in both eyes (subgroup with large dif- ferences but normal IOPs, n = 13, all also included in the preceding group; Table 4). In the group with small differences, the mean age was significantly older than in the group with large differences and the subgroup with large differences but normal IOPs (t = 3.27 and 3.01, p = 0.002 and 0.005), and the IOP of the nonexfoliative eyes was significantly higher than in the group with large differences and the subgroup with large differences but nor- mal IOPs (t = 2.27 and 2.38, p = 0.029 and 0.024); neither the IOP nor the rim-to-disc radius ratios of the inferior-temporal and inferior differed be- tween the two eyes. In the group with large dif- ferences, the rim-to-disc radius ratios of the infe- rior-temporal and inferior radii were significantly smaller in the exfoliative eyes than in the contra- lateral eyes (t = 2.77 and 3.10, p = 0.014 and 0.007) or in the exfoliative eyes of the group with small differences (t = 2.70 and 2.54, p = 0.011 and 0.016).

Table 4. Parameters of the groups with small differences and large differences.

Group with small differences (n = 2 1) Age (mean f SD, years)

Rim-to-disc radius ratio IOP (&g)

Inferior-temporal Inferior

Group with large differences (n = 16) Age (mean k SD, years)

Rim-to-disc radius ratio IOP (mmHg)

Inferior-temporal Inferior

Subgroup with large differences but normal IOPs (n = 13) Age (mean f SD, years)

Rim-to-disc radius ratio IOP (&g)

Inferior-temporal Inferior

70.2 f 4.4 17.7 f 2.8 16.7 f 2.8

0.52 f 0.15 0.55 f 0.14

0.53 f 0.16 0.54 f 0.16

64.4 f 6.5” 18.2 Ifr 2.9 14.5 f 2.4b

0.38 f 0.15‘ 0.43 f 0.15‘

0.49 f 0.12 0.55 f 0.14

64.3 f 7.1” 17.3 f 2.9 14.5 f 2.3b

0.42 f 0.16 0.46 f 0.15

0.49 f 0.13 0.57 f 0.14

NS

NS NS

< 0.001

< 0.05 < 0.01

< 0.001

NS < 0.05

Group with small differences: IOP differences (exfoliative eye - nonexfoliative eye) <2 mmHg. Group with large dif- ferences: IOP differences > 2 mmHg. Subgroup with large differences but normal IOPs: IOP differences > 2 mmHg and IOPs < 22 mmHg in both eyes. IOP: intraocular pressure; SD: standard deviation; NS: not significant. a Significantly younger than in the group with small differences (p < 0.01).

Significantly lower than in the group with small differences (p < 0.51). Significantly smaller than in the group with small differences (p < 0.05).

165

In the subgroup with large differences but normal IOPs, the rim-to-disc radius ratio of the inferior radius was significantly smaller in the exfoliative eyes than in the contralateral eyes (t=2.37, p = 0.035); rim-to-disc radius ratios of the inferior- temporal and inferior radii in the exfoliative eyes were smaller than those of the group with small differences of borderline significance (t = 1.93 and 1.84, p = 0.063 and 0.074).

Differences

Although the exfoliation syndrome has been claimed to be a risk factor for capsular glaucoma, the relatively large number of patients with this syndrome show normal IOP level and visual field. This raises the question of whether there are any factors or parameters for distinguishing those who may develop glaucoma. To try to find an answer to the question, we made a study design of interocu- lar paired comparisons between the two eyes of pa- tients with unilateral exfoliation syndrome and no visual field loss, paying special attention to any dif- ferences in the configuration of the optic nerve head, measured with a computerized image ana- lysis system. Since the optic discs of the two eyes usually resemble one another in the normal popu- lation (Armaly 1967; Carpel & Engstrom 1981; Ca- rassa et al. 1991), any differences in optic disc con- figuration were assumed to be due to direct or in- direct innuences of the exfoliation syndrome.

In the current study, no differences in global optic disc parameters were found between the ex- foliative and nonexfoliative fellow eyes, but the ex- foliative eyes had significantly higher IOPs. The rim-to-disc radius ratios were then compared at in- tervals of 30" to evaluate any sectional differences in optic disc topography, and the raio at the infe- rior-temporal radius was found to be significantly smaller in the exfoliative eyes. Furthermore, the greater the difference in IOP between the two eyes, the smaller were the rim-to-disc ratios at the infe- rior-temporal and inferio; radii. When the series was grouped as patients with IOP differences (ex- foliative eye - nonexfoliative eye) < 2 mmHg (group with small differences), or > 2 mmHg (group with large differences, including the subgroup with large differences despite normal IOPs), no dif- ferences in optic disc parameters were observed in the group with small differences. In contrast, in the

166

group with large differences, the inferior part of the optic disc was significantly damaged in the ex- foliative eyes, and these changes were observed even though the IOPs were within the normal range (subgroup with large differences but normal

Assuming that the nonexfoliative eyes in the unilateral exfoliation syndrome represent the indi- vidual, normative IOP, our findings suggest that an IOP rise of at least 2 mmHg higher than its ba- seline affect the optic disc. Changes in the optic disc may occur even in eyes without a so-called statistically high IOP. Linner et al. (1989) believed that the exfoliative process in the posterior seg- ment including the optic nerve might explain why a Significantly large percentage of disc pallor was found in exfoliative than in nonexfoliative eyes with ocular hypertension, despite a lack of signifi- cant differences in IOP or visual field findings. However, they compared two independent groups in which the individual IOP baseline may have been different. Thus, differences in optic disc pal- lor might have been found even though the aver- age IOP was not significantly different between the groups. Hence, optic disc changes in the exfolia- tion syndrome can be considered in large part to be pressure-dependent changes. The persistently higher IOP in the exfoliative eyes in many cases of the unilateral exfoliation syndrome implies the need for vigilance even when the IOP level is within the normal range.

IOP).

Acknowledgments

Dr. Tomita was a visiting research scientist from the De- partment of Ophthalmology, Gifu University School of Medicine, Japan.

This study was supported by grants from the Silma- saatio (the Eye Foundation) and the Y j o Jahnsson Foun- dation.

The Topcon IMAGEnet was kindly provided for this study by Topcon Europe Inc, Amsterdam, The Nether- lands.

None of the authors has any financial, commercial, or proprietary interest in the instruments mentioned in the study.

References

Aasved H (1971): Intraocular pressure in eyes with and without fibrillopanthia epitheliocapsularis (so-called exfoliation or pseudoexfoliation). Acta Ophthalmol (Copenh) 49: 601-610.

Armaly M F (1967): Genetic determination of cup/disc ratio of the optic nerve. Arch Ophthalmol78: 35-43.

Carpioli J (1991): Automated perimetry in glaucoma. Am

Carassa R G, Schwartz B & Takamoto T (1991): Increased preferential optic disc asymmetry in ocular hypertens- ive patients compared with control subjects. Ophthal- mology 98: 681-691.

Carpel E F & Engstrom P F (1981): The normal cup-disk ratio. Am J Ophthalmol91: 588-597.

Hansen E & Sellevold 0 L (1968): Pseudoexfoliation of the lens capsule I. Clinical evaluation with special re- gard to the presence of glaucoma. Acta Ophthalmol

Henry J C, Krupin T, Schmitt M, Lauffer J, Miller E, Ewing M Q & Scheie H G (1987): Long-term follow-up of pseudoexfoliation and the development of elevated intraocular pressure. Ophthalmology 9 4 545-552.

Klemetti A (1988): Intraocular pressure in exfoliation syndrome. Acta Ophthalmol (Copenh) (suppl) 184:

Linner E, Schwartz B & Araujo D (1989): Optic disc pallor and visual field defect in exfoliative and non-exfolia- tive, untreated ocular hypertension. Int Ophthalmol

Littmann H (1982): Zur Bestimmung der wahren GroBe Objekts auf dem hintergrund des lebenden Auges. Klin Monatsbl Augenheilkd 180: 286-289.

Nanba K, Shirakashi M, Fukuchi T & Iwata K (1989): Stereomorphometry of optic disc cupping with a com- puter image analyzer IMAGEnet. Rinsho Ganka 43:

Puska P & Raitta C (1992): Exfoliation syndrome as a risk factor for optic disc changes in nonglaucomatous eye. Graefes Arch Clin Exp Ophthalmol230: 501-504.

J Ophthalmol 111: 235-239.

45: 1095-1104.

54-58.

13: 21-24.

535-538.

Shirakashi M, Nanba K & Iwata K (1992): Changes in reversal of cupping in experimental glaucoma. Longi- tudinal study. Ophthalmology 99: 1104-1110

Tarkkanen A (1962): Pseudoexfoliation of the lens cap- sule. A clinical study of 418 patients with special ref- erence to glaucoma, cataract and changes of the vi- treous. Acta Ophthalmol (Copenh) (suppl) 71: 1-98.

Varma R, Steinmann W C, Spaeth G I & Wilson R P (1988): Variability in digital analysis of optic disc topo- graphy. Graefes Arch Clin Exp Ophthalmol 226

Varma R, Douglas G R, Steinmann W C, Wijsman K, Mawson D & Spaeth G L (1989a): A comparative evalu- ation of three methods of analyzing optic disc topo- graphy. Ophthalmic Surg 20: 813-819.

Varma R, Spaeth G L, Steinmann W C & Katz L J (1989b): Agreement between clinicians and an image analyzer in cup-to-disc ratios. Arch Ophthalmol 107: 526-529.

Zeyen T G & Caprioli J (1993): Progression of disc and field damage in early glaucoma. Arch Ophthalmol 111:

435-442.

62-65.

Received on November llth, 1993.

Corresponding author:

Goji Tomita, MD Department of Ophthalmology Gifu University School of Medicine 40 Tsukassa-Machi Gifu-shi, Gifu 500 Japan.

167