Ophthalmoscopic assessment of the optic nerve head is of utmost importance for diagnosis of optic nerve anomalies and diseases including the glaucomas. The parameters used for the ophthalmoscopic evaluation can be classified into qualitative parameters, such as presence of an optic disc haemorrhage or unusual shape of the neuroretinal rim, which can be assessed subjectively by the examiner ; and quantitative parameters, such as area of the optic disc and neuroretinal rim, which can be measured objectively.1 The measurements have conventionally been performed by confocal laser scanning tomography of the optic nerve head or planimetry of optic disc photographs.234
Optical coherence tomography (OCT) technique has been introduced in ophthalmology as a means to assess the superficial and deep retinal layers including measurement of the retinal nerve fibre layer thickness.5678910111213 Furthermore it allows cross-sectional imaging and measurements of the optic disc. In contrast to all optical techniques available so far, it retrieves signals not only from the surface of the retina, but also from a depth of several mm.1415 The OCT is a computer assisted precision optical instrument that generates cross sectional images (tomograms) of the retina with< 10 microns axial resolution.16 It works on the principle of low coherence interferometry, the detailed descriptions of which have been published.567891011121314
OCT displays the tomograms in real time using a false colour scale that represents the degree of light backscattering from tissues at different depths in the retina.17 The recent release of an OCT machine, the OCT3 (Carl Zeiss Ophthalmic Systems-Humphrey Division, Dublin, CA, USA; following software release 2.0, called Stratus-OCT) enables the computation of two-dimensional and three-dimensional morphometric data of the optic disc using multiple cross-sectional scans of the disc.14 The first purpose of the study was to compare morphometric data of the optic disc obtained by OCT with those obtained by an established method, planimetry of optic disc photographs.
The key measure of all disc parameters obtained by the OCT is the optic disc size, which depends on the correct detection of the disc margin in six radial OCT scans. The disc margin in OCT measurements is determined at the level of the pigment epithelium or choriocapillaries layer. Therefore, differing properties in respect of light reflection and transmission of the pigment epithelium and the layers overlaying the pigment epithelium may be a source of error. The second purpose of the study was to determine whether OCT measurements of the optic disc are influenced by the amount of parapapillary atrophy or pigmentary conus at the disc margin.
Materials and Methods
With informed consent 94 eyes of 47 Asian-Indian were subjected to digital optic disc photography and OCT optic disc scanning by an experienced member of the hospital staff. All tested eyes had visual acuity of 6/9 or better with refraction ranging from ± 3.50 to -5.0 dioptres. The subjects were either normal or glaucoma suspects. Normal subjects had no family history of glaucoma; no history of retinal pathology, laser or intraocular surgical treatment and IOP less than 21mmHg. Suspicious glaucomatous subjects had no retinal pathology, laser or surgical therapy, IOP of 22 to 30mmHg with cup/disc ratio less than 0.7, IOP of less than 22mmHg with asymmetrical cupping (cup/disc ratio difference >0.2 between two eyes) or cup/disc ratio >0.6; all in the presence of a normal visual fields.
Photographs and OCT images were selected on the basis of image quality rather than on the presence of any particular optic disc features. In 10 eyes of 6 patients either the image quality of optic nerve head or the OCT scan was not favourable for the study. Five eyes of 3 patients were excluded due to the media haziness producing inability to identify the margins of the cup or disc on planimetry. Three eyes of 2 patients were excluded due to poor resolution produced by the OCT machine (the OCT image poor in red and yellow colour). In the remaining 2 eyes of 2 patients, the OCT machine was unable to recollect the data from the storage device. Both eyes of these six patients were excluded and the final analysis was done on 82 eyes of 41 patients.
Optic disc photography
Subjects' pupils were dilated with 1% tropicamide. Digital 20-degree colour optic disc images were taken with a telecentric fundus camera (FF450 plus, Carl Zeiss Meditech AG, Jena, Germany), equipped with a digital 3-CCD camera (JVC, Japan). The borders of the optic cup and optic disc were manually outlined and measurements of the disc area and cup area were taken. The evaluation gave absolute measurements in the dimensions of the fundus structures taking into account the magnification of the camera and the digitising equipment (Visupac 3.2.1. Digital Image Archiving System, Carl Zeiss Meditech AG, Jena, Germany). Because the path of light in the camera follows the telecentric principle the magnification of the camera is fixed and not dependent on the dimensions of the subject's eye. The optic cup was defined on the basis of contour and not of pallor. The border of the disc was identical with the inner side of the peripapillary scleral ring. The latter is a thin white band encircling the optic disc. Parapapillary atrophy was differentiated into a peripheral zone alpha with irregularities of the retinal pigment epithelium and, if present, into zone beta with denuded pigment epithelium and visible large choroidal vessels as described earlier.15 For the purpose of this study, presence and if present, the amount of clock hours adjacent to the disc of parapapillary atrophy zone beta and additionally of congenital pigmented crescent were noted. One experienced unmasked observer (WMB) performed all measurements.
Imaging with the OCT
On the same day, all eyes underwent optic disc imaging with the OCT3 (software release A1.1) using the "fast optical disc" protocol (Carl Zeiss Ophthalmic Systems-Humphrey Division, Dublin, CA, USA). Six equally spaced cross-sectional scans of the optic disc in different meridians (spaced 30 degrees) were obtained, all running through a common central axis. This resulted in a spoke like pattern centered on the disc. With each scan length of 4 mm and depth of 2 mm is measured. In the OCT3 evaluation protocol "optic nerve head", the software identifies the anterior and innermost borders of the second echo-rich structure (thought to represent the retinal pigment epithelium/choriocapillaries layer) automatically at each side of the optic disc. All six scans of each eye were reviewed on screen by one experienced observer (WMB). The detected position of the end of the retinal pigment epithelium/choriocapillaries reflection was corrected manually if necessary. The correct identification was checked and the sensitivity to detect surface border was corrected if necessary. At a fixed offset of 150 microns anterior to the plane of the retinal pigment epithelium/choriocapillaries reflection, the system constructs the disc area and cup area from the diameter of the disc and rim and cup in each of the six scans. Rim area is represented by the difference of disc area and cup area. Cup area was assessed by a second approach taking into account a possible movement of the patient's eye during the 1.9 sec of measurement (“topographic cup area”). Rim area was computed by multiplying the average of the rim width of the six cross-sectional scans times the circumference of the disc (“horizontal integrated rim width”).
Previous reports estimate the mean morphometric optic disc size of 2.58 ± 0.65 mm2 in Indian18 population, 2.94 ± 0.74mm2 in blacks and 2.63 ± 0.46mm2 in whites19 and 2.15 ± 0.36 mm2 by OCT analysis.14 Expecting to get optic disc area results with significant difference (d) of 0.5mm2; power of 80% (Zb = 0.842) and significance level of 0.01 (Za = 2.576) was used to determine the sample size. The sample size was determined using the formula n = 2(Za ± Zb)2S2/d2. Considering the standard deviation (S) of 0.65 the minimum sample size calculated was 40.
Statistical analysis was performed using SPSS version 11.5.1. Systematic differences were assessed by the t-test for paired samples and by comparing the differences with the means. The influence of the presence of parapapillary atrophy and pigmentary conus was tested comparing the differences of the measurement modalities between optic discs with and optic discs without parapapillary atrophy and pigmentary conus respectively with the t-test for unpaired samples. The Pearson correlation test was used to assess the relation between the amount of parapapillary atrophy and pigmentary conus respectively with the differences between the measurement modalities. Because multiple comparisons were carried out, the level of significance was set to an error probability of 0.01 (two sided comparisons).
The characteristics of the study population are summarised in Table 1.
Mean disc size in OCT measurements was significantly smaller than in photographic planimetry (P < 0.001; Table 2). Figure 1 shows the significant systematic difference between the values determined by the two techniques plotted against the average.17 The differences tended to be higher with increasing disc area (r = -0.37, P < 0.001; Figure 1).
Similar to disc area, cup area, rim area and cup/disc area ratio measurements have smaller values for OCT measurements compared to planimetry measurements (p < 0.001, Table 2) and there was no inter-eye difference between the two groups (p > 0.5, paired students t-test). The data results were reanalysed with randomly selected one eye of the patient (41 eyes of 41 patients) and similar results for all comparisons were obtained.
Optic discs with parapapillary atrophy (n = 38) and discs without parapapillary atrophy (n = 44) differed slightly, but not significantly, in the relative difference between disc size values obtained by the two measurement modalities (relation of the difference between measurement modalities to the mean of the measurement modalities: 19.5% ± 8.5% vs. 15.9% ± 7.5%, P = 0.2). This means that optic discs with parapapillary atrophy on OCT were smaller by 19.5% as opposed to eyes without parapapillary atrophy, in which OCT disc was smaller by 15.9% in comparison to the mean of both measurement modalities. The relative difference did not correlate with the amount of parapapillary atrophy (r = -0.17,P = 0.29). No significant difference was noted on imaging with OCT and planimetry between eyes with parapapillary atrophy and eyes without parapapillary atrophy for disc area, cup area, cup disc ratio and rim area on (Table 3).
Optic discs with pigmentary conus (n = 12) and discs without pigmentary conus (n = 70) did not differ in the relative difference between OCT disc size and photographic disc size (18.9% ± 9.1 % vs. 17.3% ± 8.7%; P = 0.65). There was no significant correlation of the relative difference and the amount of pigmentary conus (r = -0.04, P = 0.9). When comparing with unpaired t test it was noted that there was no significant difference on imaging with OCT and Planimetry, between the eyes with pigmentary conus and eyes without pigmentary conus (Table 4). The data were reanalysed with a randomly selected eye of the patient (41 eyes of 41 patients) and similar results for all comparisons were obtained.
This study compared measurements of the optic disc obtained by OCT and planimetry of fundus photographs. Disc size, cup area, rim area and cup disc ratios obtained by OCT were smaller than planimetry measure. The difference between the parameters measured with two imaging technique showed an increase with increasing disc size (Figure 1). This trend of increasing differences with increasing disc size point to the conclusion that the measurement algorithms present in the two imaging techniques vary proportionally with the changing disc sizes.
The cup/disc area ratio, which differed significantly between OCT and planimetry measurements, cannot be explained by differing magnifications alone. Obviously, the reference plane of the OCT with the pre-set offset of 150 µm anterior to the pigment epithelial layer defines different proportions of the disc belonging to rim and cup than the results of photographic planimetry suggest.
Furthermore, the OCT software uses several other algorithms to construct rim area (e.g. rim area and horizontal integrated rim area) and cup area (cup area and topographical cup area), which give differing results among themselves. Further studies are needed to determine an optimal method of rim/cup separation and measurement in the OCT. The same holds true for several volume parameters, present in the analysis protocol, constructed by the OCT software.
The key structure for the detection of the disc margin in OCT measurements, the end of the retinal pigment epithelium/choriocapillaries layer, is buried beneath other layers. Differing optical properties of these layers, namely parapapillary atrophy and pigmentary conus did not lead to a systematic increase or decrease in the differences between the two measurement modalities. Furthermore, there was no association of the amount of parapapillary atrophy and pigmentary conus with differences in disc size obtained by OCT measurement and by photographic planimetry. Either both measurement modalities are affected in the same way by the presence or amount of parapapillary atrophy and pigmentary conus or, parapapillary atrophy and pigmentary conus do not lead to a systematic measurement error of OCT derived disc size and do not lead to an increase in unsystematic scatter with increasing amount of pigmentary conus and parapapillary atrophy.The latter seems to be more likely.
So far, optic disc measurements by the OCT have been compared with results of confocal scanning-laser tomography (CSLO).14 Schuman and colleagues examined OCT3 data obtained by the automatic determination of optic disc margin without manual correction. They found significantly larger figures for the area measurements obtained by OCT in comparison to CSLO measurements. The results of both modalities correlated significantly (for area measures: range of r: 0.64 - 0.75, for volume measures: range of r: 0.34 - 0.72). The investigation by Schuman and colleagues suggests that fully automated optical coherent tomography of the optic nerve head may be a technique for examinations of the optic nerve head without the need of interactive corrections by trained technicians. The cup/disc area ratio differed significantly between OCT and CSLO measurements hinting at a difference in the separation of the disc into rim and cup by the two methods. In Schuman's study, OCT3 measurements performed similar to CSLO measurements in separating glaucoma patients from normal subjects. The areas under the receiver operator characteristic curves were found to be similar and in the range of 0.61 to 0.79 for both devices.
There are limitations of the present study. K-readings of the patients' corneas or measurements of the axial lengths were not available. Therefore an individual correction for the magnification of the optics of the eye, e.g. the method of Littmann,16 was not possible. The magnifications of the subjects' eyes were assumed according to the dimensions of Gulstrand's eye for the planimetry of the fundus photographs. Standard values of K-readings and refractive error were set for the OCT measurements. There was, however, no relationship with refractive error and the difference in disc size between the two measurement modalities (r = 0.11,P = 0.351).
To summarise, there are substantial differences in the way of assessing the data by two imaging modalities. Irregularities at the disc margin such as parapapillary atrophy zone beta and pigmentary conus do not seem to be a systematic source of error for measurements of disc size with the OCT. Since there are differences in optic disc parameters values obtained on OCT when compared to established technique of planimetry, further studies to evaluate the usefulness of optic disc measurements obtained by the OCT seem to be worthwhile.
1. Jonas JB, Budde WM, Panda-Jonas S. Ophthalmoscopic evaluation of the optic nerve head Surv Ophthalmol. 1999;43:293–320
2. Caprioli J, Miller JM, Sears M. Quantitative evaluation of the optic nerve head in patients with unilateral visual field loss from primary open-angle glaucoma Ophthalmology. 1987;94:1484–7
3. Garway-Heath DF, Hitchings RA. Quantitative evaluation of the optic nerve head in early glaucoma Br J Ophthalmol. 1998;82:352–61
4. Chauhan BC, McCormick TA, Nicolela MT, LeBlanc RP. Optic disc and visual field changes in a prospective longitudinal study of patients with glaucoma: comparison of scanning laser tomography with conventional perimetry and optic disc photography Arch Ophthalmol. 2001;119:1492–99
5. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson Wg, Chang W, et al Optical coherence tomography Science. 1991;254:1178–81
6. Hee MR, Izatt JA, Swanson EA, Huang D, Schuman JS, Lin CP, et al Optical coherence tomography of the human retina Arch Ophthalmol. 1995;113:325–32
7. Wilkins JR, Puliafito CA, Hee MR, Duker JS, Reichel E, Coker JG, et al Characterization of epiretinal membranes using optical coherence tomography Ophthalmology. 1996;103:2142–51
8. Schuman JS, Hee MR, Puliafito CA, Wong C, Pedut-Kloizman T, Lin CP, et al Quantification of nerve fibre layer thickness in normal and glaucomatous eyes using optical coherence tomography Arch Ophthalmol. 1995;113:586–96
9. Schuman JS, Pedut-Kloizman T, Hertzmark E, Hee MR, Wilkins JR, Coker JG, et al Reproducibility of nerve fiber layer thickness measurements using optical coherence tomography Ophthalmology. 1996;103:1889–98
10. Huang D, Stinson WG, Schuman JS. High-resolution measurement of retinal thickness using optical coherence domain reflectometry Invest Ophthalmol Vis Sci. 1991;32:1019
11. Mistlberger A, Liebmann JM, Greenfield DS, Pons ME, Hoh ST, Ishikawa H, et al Heidelberg retina tomography and optical coherence tomography in normal, ocular hypertensive, and glaucomatous eyes Ophthalmology. 1999;106:2027–32
12. Blumenthal EZ, Williams JM, Weinreb RN, Girkin CA, Berry CC, Zangwill LM. Reproducibility of nerve fiber layer thickness measurements by use of optical coherence tomography Ophthalmology. 2000;107:2278–82
13. Jones AL, Sheen NJL, North RV, Morgan JE. The Humphrey optical coherence tomography scanner: quantitative analysis and reproducibility study of the normal human retinal nerve fibre layer Br J Ophthalmol. 2001;85:673–7
14. Schuman JS, Wollstein G, Farra T, Hertzmark E, Aydin A, Fujimoto JG, et al Comparison of optic nerve head measurements obtained by optical coherence tomography and confocal scanning laser ophthalmoscopy Am J Ophthalmo l. 2003;135:504–12
15. Jonas JB, Fernandez MC, Naumann GO. Glaucomatous parapapillary atrophy Occurrence and correlations Arch Ophthalmol. 1992;110:214–22
16. Littmann H. Determination of the true size of an object on the fundus of the living eye (original article), "Zur Bestimmung der wahren Gr φsseeines Objektes auf dem Hintergrund des lebenden Auges," which originally appeared in Klinische Monatsblδtter für Augenheilkunde 1982; 180:286-9.Translated by TD Williams Optom Vis Sci. 1992;69:717–20
17. Hagimura N. Optical coherence tomographic features of normal ocular fundus Jpn J Clin Ophthalmol. 1998;52:1459–62
18. Jonas JB, Thomas R, George R, Berenshtein E, Muliyil J. optic disc morphology in south Indian: the vellore eye study Br J Ophthalmol. 2003;87:189–96
19. Varma R, Tielsch JM, Quigley HA, Hilton SC, Katz J, Spaeth GL, et al Race-, age-, gender-, and refractive error-related differences in the normal optic disc Arch Ophthalmol. 1994;112:1068–76