Journal of Neuro-Ophthalmology, December 2001, Volume 21, Issue 4

Polarimetric Nerve Fiber Analysis in Patients with Visible Optic Nerve Head Drusen

OBJECTIVE: To evaluate the effect of visible optic nerve head drusen (ONHD) on retinal nerve fiber layer (RNFL) thickness retardation by using scanning laser polarimetry. METHODS: Twenty-three eyes of 13 patients with visible ONHD and 26 eyes of 13 age- and sex-matched control subjects were involved in the study. Ophthalmologic examination, scanning laser polarimetry with nerve fiber analyser (NFA) type II GDX, automated Humphrey visual field testing, and red-free fundus photography were performed. Eyes with ONHD were classified from grade 0 to III according to the amount of visible drusen. Thus, grade 0 discs had no clinically visible ONHD and grade III discs represented the presence of dense drusen. RESULTS: Measurements with NFA of RNFL thickness retardation showed significant decrease in eyes with visible ONHD compared with control eyes (P < 0.05). Although no significant difference was found between grade I and grade II discs regarding NFA measurements, grade III discs had significantly lower values, indicating the greater amount of RNFL loss with higher grade ONHD. Documentation of increased percentage of visual field defects with higher grade drusen was also in accordance with this finding. CONCLUSIONS: NFA can quantitatively detect the decrease in retardation of RNFL thickness in eyes with visible ONHD and can be used as an indicator of nerve fiber layer loss in these cases.

Journal of Neuro- Ophthalmology 21( 4): 245- 249, 2001. • 2001 Lippincott Williams & Wilkins, Inc., Philadelphia Original Contribution Polarimetric Nerve Fiber Analysis in Patients with Visible Optic Nerve Head Drusen Sinan Tatlipinar, MD, Sibel Kadayifgilar, MD, Banu Bozkurt, MD, § ansal Gedik, MD, Ergun Karaagaoglu, PhD, Mehmet Orhan, MD, and Murat Irkeg, MD Objective: To evaluate the effect of visible optic nerve head drusen ( ONHD) on retinal nerve fiber layer ( RNFL) thickness retardation by using scanning laser polarimetry. Methods: Twenty- three eyes of 13 patients with visible ONHD and 26 eyes of 13 age- and sex- matched control subjects were involved in the study. Ophthalmologic examination, scanning laser polarimetry with nerve fiber analyser ( NFA) type IIGDX, automated Humphrey visual field testing, and red- free fundus photography were performed. Eyes with ONHD were classified from grade 0 to III according to the amount of visible drusen. Thus, grade 0 discs had no clinically visible ONHD and grade III discs represented the presence of dense drusen. Results: Measurements with NFA of RNFL thickness retarda­tion showed significant decrease in eyes with visible ONHD compared with control eyes ( P < 0.05). Although no significant difference was found between grade I and grade II discs re­garding NFA measurements, grade III discs had significantly lower values, indicating the greater amount of RNFL loss with higher grade ONHD. Documentation of increased percentage of visual field defects with higher grade drusen was also in accordance with this finding. Conclusions: NFA can quantitatively detect the decrease in retardation of RNFL thickness in eyes with visible ONHD and can be used as an indicator of nerve fiber layer loss in these cases. Key Words: Optic nerve head drusen- Retinal nerve fiber layer- Nerve fiber analyser- Visual field. Optic nerve head drusen ( ONHD) were first described clinically by Liebreich in 1868 ( 1). These hyaline struc­tures appear as globular bodies protruding from optic disc and blur the disc margins. They are mostly bilateral From the Departments of Ophthalmology ( ST, SK, BB, SG, MO, MI) and Biostatistics ( EK), Hacettepe University School of Medicine, Ankara, Turkey. Address correspondence and reprint requests to Sibel Kadayifcilar, MD, Associate Professor of Ophthalmology, Hacettepe Hastanesi Goz ABD, Sihhiye 06100, Ankara, Turkey; E- mail: sibelkd@ mailcity. com The authors have no proprietary interest in the Nerve Fiber Analyser. and often familial. Lorentzen reported an incidence of 3.4 per 1000 individuals ( 2). ONHD are mostly seen in small discs ( 3). A small scleral canal is thought to cause stasis of axoplasmic flow and result in death of ganglion cell axons and formation of ONHD. This build- up of drusen, in turn, damages progressively more axons. The result is retinal nerve fiber layer ( RNFL) atrophy and associated visual field defects. These visual field defects are similar to those of glaucoma or other optic neurop­athies ( 2). Since ONHD distort the optic disc shape, it is also quite difficult to interpret the cupping. The RNFL can be evaluated with several different techniques, such as red- free ophthalmoscopy, photogra­phy, optical coherence tomography ( OCT), scanning la­ser tomography, scanning laser microperimetry, laser biomicroscopy, and scanning laser polarimetry ( 4- 6). RNFL defects associated with ONHD were previously documented by RNFL photography and, more recently, by OCT ( 5,6). In this study, the RNFL in eyes with visible ONHD is evaluated by scanning laser polarim­etry, a diagnostic technique that measures the thickness of peripapillary RNFL in vivo quantitatively ( 7). This method does not require a reference plane and provides rapid and objective data. PATIENTS AND METHODS Between November 1998 and December 1999, 26 eyes of 13 patients with visible ONHD were examined at our Ophthalmology Department. Ophthalmologic evalu­ation included visual acuity testing, intraocular pressure measurement ( Goldmann tonometry), slit- lamp examina­tion, indirect slit- lamp biomicroscopy with + 90 D lens, color and red- free photography using a digital fundus camera ( FF 450 IR; Zeiss, Oberkochen, Germany), and automated visual field examination with Humphrey cen­tral threshold program ( 30- 2 or 24- 2) ( Humphrey Aller-gan Instruments, San Leandro, CA, USA). Scanning la­ser polarimetry and visual field testing were performed 245 246 S. TATLIPINAR ET AL. on a separate day. Informed consent was obtained from all patients with ONHD and from the control subjects. The inclusion criteria were the presence of ONHD visible by indirect slit- lamp biomicroscopy in one or both eyes and the absence of any other known ocular pathologic lesion affecting the RNFL ( e. g., glaucoma and other optic neuropathies, previous retinal surgery or retinal laser treatment). Eyes with ONHD were graded from 0 to III according to the ophthalmoscopic appear­ance of drusen ( 6). Grade 0 discs had no clinically visible ONHD. Grade I discs had a few scattered drusen, and grade II discs had more numerous ONHD. Grade III discs were noted to have dense drusen with obscuration of the optic cup ( Fig. 1). A nerve fiber analyser ( NFA II GDX version 1.0.08; Laser Diagnostic Technologies Inc, San Diego, CA, USA) was used to measure the thickness of the RNFL. This noninvasive technique is described in detail else­where ( 8- 11). Briefly, NFA consists of a laser source, a polarizer, a scanning unit, a polarization modulator, a compensator, and a polarization detector. NFA projects 780 nm polarized laser light on the retina. The parallel microtubules within the retinal ganglion cell axons be­have as a birefringent medium and change the polariza­tion state of laser light. The phase shift of polarization is called " retardation" and is presumed to be linearly cor­related to RNFL thickness. The compensator corrects for the effect of lens and corneal birefringence on the RNFL retardation measurement. The RNFL thickness is calculated automatically in each of 256x256 individual retinal positions. To elimi­nate any operator- related bias, measurements were per­formed by the same trained examiner ( BB). To avoid the effect of corneal birefringence on the measurements, macular scans were performed. None of the patients or FIG. 1. A: Red- free fundus photograph of a patient with grade III superior and inferior nasal field defects. the control subjects displayed axes or amplitudes not corrected by the commercial device. The field size was set to 15x15 degrees and pupils were undilated. Three good- quality images were obtained and results were av­eraged ( Fig. 2). The measuring ellipse was positioned along a line at 1.5 disc diameters concentric with the disc margin. Two parameters, namely ellipse average ( EA) and total polar integral ( TPI), calculated by NFA were used for comparison between the patients with visible ONHD and control subjects ( 8). Briefly, EA ( in microns) is the average thickness of the RNFL beneath the ellipse surrounding the optic disc. The integral under the retar­dation curve ( polarimetric data analysis), which repre­sents the cross- sectional area of the RNFL along the measuring ellipse, is also calculated ( TPI, in mm2). Twenty- six eyes of 13 age- and sex- matched normal subjects, taken as controls, also underwent ophthalmo­logic examination, including scanning laser polarimetry with NFA and visual field testing with Humphrey 30- 2 threshold program. These subjects were free of ocular disease with 20/ 20 vision in each subject, intraocular pressures less than 21 mm Hg, and normal- appearing optic discs. Statistical analysis was performed using the Mann- Whitney U test and an independent, two- tailed Student f- test. A P value of less than 0.05 was considered statis­tically significant. RESULTS There were six male and seven female subjects in the visible ONHD group with an average age of 29.6 ( SD, 19.6) years ( range, 12- 68). Of the 13 age- matched con­trol subjects, 6 were male and 7 were female with an average age of 29.6 ( SD, 18.9) ( range, 12- 65). Except : - S - II - J" - E • U - S - t • r - C E S *• _- _• a • a & i t • ) • • i a m m i • p * • • * i * • • • • • • • MB I s m • * > • - • I B -- optic nerve head drusen ( OS). B: Visual field of the same eye shows / Neuro- Ophthalmol, Vol. 21, No. 4, 2001 NFA IN OPTIC NERVE HEAD DRUSEN 247 E m s V 1 B H B MM FIG. 2. Scanning laser polarimetric image of the eye in Figure 1. A: Graph showing the RNFL thickness in temporal, superior, nasal, inferior, and temporal quadrants, compared with the normative database. Measurements of the patient are almost at the same level in temporal and nasal quadrants but below the lower limit of normative database in some parts of inferior and superior quadrants. Note the attenuation of the typical double hump of RNFL indicating nerve fiber layer thinning. B: Fundus image of patient. C: Thickness map; showing thicker RNFL ( dark) in the superior and inferior quadrants, and thinner RNFL ( light) in the nasal and temporal area; however, in this patient there is an overall decrease in RNFL. for three unilateral cases, all other patients had bilateral visible ONHD. Among the three unilateral cases, ONHDs were visible in one OD and two OSs. Thus, a total of 23 eyes with visible ONHD were involved in the study. There were 7 grade I, 10 grade II, and 6 grade III discs. In 7 of 10 bilateral cases, ONHD grades were the same. Visual acuities ranged from 20/ 20 to 20/ 30 in eyes with visible drusen. The results of two parameters ( EA, TPI) ( mean [ SD]) of eyes with visible ONHD and control eyes are given in Table 1 and Figures 3 and 4. Comparisons of RNFL thickness measurements were made between the eyes with visible drusen ( all grades as a single group) and control eyes, between each of the three grades and con­trols, and finally between each grade of ONHD. Both parameters ( EA, TPI) indicated a significant de­crease in retardation of RNFL thickness in eyes with visible drusen ( 67.2 ( xm and 0.576 mm2, respectively) compared with those of control eyes ( 81.03 ( im and 0.697 mm2, respectively; P = 0.0001 for both param­eters, two- tailed Student f- test) ( Figs. 3 and 4). When each grade of ONHD was compared with the control eyes, only a subset of age- and sex- matched con­trol subjects rather than the whole control group was used for each comparison. Grade I discs differed from the control discs significantly in both EA and TPI with P = 0.043 and P = 0.003, respectively ( Mann- Whitney U test). When grade II discs were compared with the control discs, the difference was significant with regard to EA and TPI ( P = 0.041 and P = 0.048, respectively, Mann- Whitney U test). The most significant difference was between grade III discs and control discs, however ( P = 0.0001 for both parameters, Mann- Whitney {/ test) ( Table 1). When ONHD grades were compared with each other, no statistically significant difference was found between grade I and II discs ( P > 0.05, Mann- Whitney U test). Both parameters showed a significant decrease in retar­dation in grade III discs compared with grade I discs ( EA, P = 0.005; TPI, P = 0.048; Mann- Whitney U test). When grade II discs were compared with grade III discs, EA measurement showed a significant difference, indicating that the RNFL was thinner in grade III ONHD ( P = 0.022, Mann- Whitney U test) ( Table 1). TABLE 1. Retardation due to retinal nerve fiber layer thickness ( mean [ SD]) and percentage of visual field defects of eyes with optic nerve head drusen ( ONHD) and of control eyes Control ONHD ( Total) Grade I Grade II Grade III Eyes ( n) 26 23 7 10 6 EA ([ xm) 81.03( 12.1) 67.2 ( 10.08) 70.0 ( 5.4) 70.0( 11.1) 57.8 ( 6.7) TPI ( mm2) 0.697 ( 0.10) 0.576 ( 0.07) 0.563 ( 0.03) 0.617 ( 0.09) 0.523 ( 0.03) % with VFD 0 60.8 42.8 50 100 EA, ellipse average; TPI, total polar integral; VFD, visual field defects. J Neuro- Ophthalmol, Vol. 21, No. 4, 2001 248 S. TATLIPINAR ET AL. a T 1 T N = 7 10 6 36 GRADE I GRADE II GRADE III CONTROL FIG. 3. Box- plot diagram showing minimum, maximum, and me­dian levels of total polar integral values in nerve fiber analysis of patients with visible optic nerve head drusen and con­trol subjects. Visual field defects were present in 42.8% of patients with grade I, 50% of patients with grade II, and 100% of patients with grade III ONHD ( Table 1). Most common field defects are found to be blind spot enlargement and arcuate defects followed by inferonasal defects and gen­eralized constriction ( Fig. 1). DISCUSSION Alteration in axonal transport seems to be the basic pathogenic mechanism for the formation of ONHD ( 3). A small scleral canal is thought to cause stasis of axo-plasmic transport at the optic nerve head and result in accumulation of hyaline material ( 12). This accumula­tion of hyaline material further impedes the axoplasmic flow and causes death of more axons. Thus, ONHD is a progressive disease in which resulting RNFL atrophy is manifested as visual field defects, reported in 71% of eyes with visible drusen ( 13). Generally, these defects show a nerve fiber bundle pattern similar to that of glau­coma; however, visual field defects do not necessarily match the position of drusen in the optic disc ( 14). Two major theories proposed for the development of visual field defects can be summarized as direct compression of RNFL by drusen or vascular events caused by drusen. Several methods have been used for the evaluation of RNFL in eyes with ONHD. Mustonen and Nieminen performed photography to document atrophy of the peripapillary RNFL ( 5). However, interpretation of RNFL photography is subjective and often difficult. Re­cently, OCT was shown to document the nerve fiber layer thinning due to ONHD ( 6). OCT is a noninvasive and noncontact technique that incorporates near infrared, coherent light passing through Michaelson interferom­eter ( 15). It gives high- resolution, cross- sectional imag­ing of the retina. With OCT, superior and inferior RNFL measurements were found to be significantly thinner in eyes with visible drusen compared with in control eyes. Higher grade ONHD had more visual field defects in accordance with thinner RNFL measurements by OCT ( 6). Our data, similarly, have disclosed that higher grade ONHD was associated with visual field defects and de­creased retardation values. In this study, we used NFA to evaluate RNFL thick­ness in eyes with visible ONHD. NFA is a computerized scanning laser polarimetry device that can objectively and quantitatively measure the RNFL thickness in terms of retardation in vivo. In patients with glaucoma and ocular hypertension, a reduction of RNFL thickness re­tardation was documented by NFA ( 10,11). It works with undilated pupils, and a complete scan of 65,536 points takes only 0.7 seconds. Although placing the measuring ellipse at the disc margin for grade III discs was rela­tively difficult, the experienced operator could establish disc border by using the segments of disc margin unin-volved by ONHD. In this study, NFA parameters dealing with the whole circumference of optic disc ( EA, TPI) were used rather than sectoral ( superior, inferior, tempo­ral, nasal) thickness measurements, since it was previ­ously documented that visual field defects do not neces­sarily match the localization of drusen in the optic disc. We found that there was a significant reduction in retar­dation of RNFL thickness compared with controls. Pres­ence of visual field defects was in accordance with this finding. Grade III discs showed the most significant de­crease in retardation, and all of the patients with grade III discs had visual field defects. On the other hand, grade I and II discs had similar measurements with a similar percentage of visual defects. In conclusion, NFA can quantitatively detect the RNFL thickness retardation in eyes with visible ONHD. Eyes with visible ONHD had significantly lower RNFL thickness retardation values compared with normal eyes. With increasing grades of ONHD, the nerve fiber layer loss was found to be more prominent. NFA appears to be a sensitive indicator of RNFL loss. Further longitu­dinal studies are needed to assess the efficacy of NFA to document progressive RNFL changes in patients with visible ONHD. 1 JL T _ L T _ L N- 7 10 6 26 GRADE I GRADE II GRADE III CONTROL FIG. 4. Box- plot diagram showing minimum, maximum, and me­dian levels of ellipse average values in nerve fiber analysis of patients with visible optic nerve head drusen and con­trol subjects. / Neuro- Ophthalmol, Vol. 21, No. 4, 2001 NFA IN OPTIC NERVE HEAD DRUSEN 249 REFERENCES 1. Liebreich R. In discussion of Iwanoff A. Ueber Neuritis Optica. Klin Monatsb Augenheilkd 1868; 6: 426- 7. 2. Lorentzen SE. Drusen of the optic disc: a clinical and genetic study. Acta Ophthalmol 1966; Suppl 90: 1- 180. 3. Spencer WH. Drusen of the optic disc and aberrant axoplasmic transport. The XXXIV Edward Jackson memorial lecture. Am J Ophthalmol 1978; 85: 1- 12. 4. Assi A, Watts P, McAllister J. Estimation of the retinal nerve fiber layer thickness in the papillomacular area of long standing stage IV macular holes. Br J Ophthalmol 1999; 83: 573- 76. 5. Mustonen E, Nieminen H. Optic disc drusen: a photographic study. II. Retinal nerve fiber layer photography. Acta Ophthalmol ( Copenh) 1982; 60: 859- 72. 6. Roh S, Noecker RJ, Schuman JS, et al. Effect of optic nerve head drusen on nerve fiber layer thickness. Ophthalmology 1998; 105: 878- 885. 7. Weinreb RN, Shakiba S, Sample PA, et al. Association between quantitative nerve fiber layer measurement and visual field loss in glaucoma. Am J Ophthalmol 1995; 120: 732- 8. 8. Nerve Fiber Layer Analyser Manual. San Diego, CA: Laser Di­agnostic Technologies Inc., 1996. 9. Anton A, Zangwill L, Emdadi A, Weinreb RN. Nerve fiber layer measurements with scanning laser polarimetry in ocular hyperten­sion. Arch Ophthalmol 1997; 115: 331- 4. 10. Tjon- Fo- Sang M, de Vries J, Lemij HG. Measurement by nerve fiber analyzer of retinal nerve fiber layer thickness in normal sub­jects and patients with ocular hypertension. Am J Ophthalmol 1996; 122: 220- 7. 11. Weinreb RN, Shakiba S, Zangwill L. Scanning laser polarimetry to measure the nerve fiber layer of normal and glaucomatous eyes. Am J Ophthalmol 1995; 119: 627- 36. 12. Tso MOM. Pathology and pathogenesis of drusen of the optic nervehead. Ophthalmology 1981; 88: 1066- 80. 13. Savino PJ, Glaser JS, Rosenberg MA. A clinical analysis of pseudopapilledema. II. Visual field defects. Arch Ophthalmol 1979; 97: 71- 5. 14. Sarkies NJC, Sanders MD. Optic disc drusen and episodic visual loss. Br J Ophthalmol 1987; 71: 537- 9. 15. Pedut- Kloizman T, Pakter HM, Schuman JS, Sczwartz JC, Hee MR. Ophthalmic diagnosis using optical coherence tomography. Ophthalmol Clin North Am 1998; 11: 465- 86. J Neuro- Ophthalmol, Vol. 21, No. 4, 2001