Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Aug 23.
Published in final edited form as: Ophthalmic Surg Lasers Imaging. 2003;34(6):498–504.

Optical Coherence Tomography Disc Assessment in Optic Nerves With Peripapillary Atrophy

Edward Lai 1, Gadi Wollstein 1, Lori Lyn Price 1, Lelia A Paunescu 1, Paul C Stark 1, James G Fujimoto 1, Joel S Schuman 1
PMCID: PMC1950848  NIHMSID: NIHMS24573  PMID: 14620759

Abstract

BACKGROUND AND OBJECTIVE

Optical coherence tomography (OCT) is able to determine the optic disc margin automatically. The aim of this study was to investigate the accuracy of the automatic OCT optic nerve head measurements in the presence of peripapillary atrophy.

PATIENTS AND METHODS

This was a cross-sectional, retrospective study. Thirty-one subjects with peripapillary atrophy underwent optic nerve head scanning with OCT version 3. Nineteen of the eyes were classified clinically as having glaucoma, nine had suspected glaucoma, and three were normal. Automatic OCT results were compared with manual tracing results.

RESULTS

Significant differences were found between most OCT optic nerve head automated and manual disc assessment parameters; however, good agreement was found between the two methods for all parameters (intraclass correlation, 0.71 to 0.94). Areas under receiver operator characteristics curves for clinical status were similar for all parameters with both methods.

CONCLUSION

Automated OCT optic nerve head analysis may be used in the clinical setting in the presence of peripapillary atrophy; however, caution should be used when comparing individual results with population-derived optic nerve head results.

INTRODUCTION

Glaucoma is a group of optic neuropathies in which there is damage to retinal ganglion cells and the nerve fiber layer with a characteristic appearance of the optic nerve head. This damage may precede visual field defects, which occur later in the course of disease.1–5 Early diagnosis and the ability to detect progression of glaucoma are of great significance because the damage caused is largely irreversible. Several technologies to assist in early detection and following the progression of glaucoma are now available and can provide objective quantitative analysis. These include several imaging modalities that are capable of scanning the optic nerve head. Confocal scanning laser ophthalmoscopy and scanning laser biomicroscopy both require manual tracing of the optic disc margin. On the other hand, optical coherence tomography (OCT) is able to determine the optic disc margin automatically. A previous cross-sectional study found high correlation between results of automated and manual tracing of the optic nerve margin as measured by OCT in normal subjects and patients with glaucoma.6 The automated definition of the optic disc margin as determined by OCT is dependent on the analysis algorithm’s recognition of the termination of the retinal pigment epithelium/choriocapillaris.7–9 However, the accuracy of the OCT optic nerve head automated measurements in the presence of peripapillary atrophy, where the retinal pigment epithelium/choriocapillaris edge does not overlay the disc margin, is questionable. Moreover, the frequency and size of peripapillary atrophy has been shown to be larger in eyes with glaucoma than in normal eyes.10–16 The aim of this study was to investigate the effect of peripapillary atrophy on OCT optic nerve head measurements.

PATIENTS AND METHODS

Participants of the study were consecutively recruited from the glaucoma service at New England Eye Center, Tufts-New England Medical Center, Boston, Massachusetts, between January and July 2002. All subjects underwent comprehensive ophthalmic examination including visual acuity testing, slit-lamp biomicroscopy, indirect ophthalmoscopy, visual field testing, confocal scanning laser ophthalmoscopy, and OCT examinations. All tests were completed within 6 months of each other. Eyes with a refractive error of more than −6.0 diopters were excluded from the study to eliminate confounding findings due to elongated axial length. The inclusion criteria required the presence of overt peripapillary beta zone atrophy, characterized as a central zone of chorioretinal atrophy with visible large choroidal vessels and sclera, with the widest diameter being no less than one-quarter of the disc diameter. Institutional Review Board/Ethics Committee approval was obtained for the study and all participants gave their approval to participate in the study. This study followed the principles of the Declaration of Helsinki.

The study group consisted of 31 eyes, of which 19 eyes were classified clinically as having glaucoma, 9 as having suspected glaucoma, and 3 as normal. The mean age of the subjects was 63.5 ± 12.1 years, the mean refractive error was −1.52 ± 2.66 diopters, and the mean horizontal and vertical cup-to-disc ratio as determined clinically was 0.7 ± 0.2. Visual field defects, as described below, were present in 14 of 31 eyes.

Visual Field Testing

All subjects underwent Humphrey full-threshold 24-2 achromatic perimetry (Carl Zeiss Meditech, Dublin, CA), Swedish Interactive Thresholding Algorithm standard 24-2 perimetry (Carl Zeiss Meditech), or Frequency Doubling Technique N-30 perimetry (Welch Allyn, Skaneateles Falls, NY, and Carl Zeiss Meditech). A reliable visual field test was defined as one with fewer than 30% fixation losses, false-positive responses, or false-negative responses. Normal visual field test results were defined as having no cluster of three or more adjacent points depressed more than 5 dB or two adjacent points depressed more than 10 dB. Glaucomatous visual field defects were defined as typical arcuate defects, paracentral scotomata, or a nasal step consisting of clusters of depressed points as defined earlier.

Confocal Scanning Laser Ophthalmoscopy

All confocal scanning laser ophthalmoscopy scans were conducted using the Heidelberg Retina Tomograph (HRT) (software version 2.01; Heidelberg Engineering GmbH, Heidelberg, Germany). HRT scans were used only for preliminary identification of eyes with peripapillary atrophy. Three scans of 15° × 15° were done in one sitting. One of the investigators (GW), experienced with this device, evaluated all HRT reflectance maps to identify eyes that may have qualified for the study. All eyes with peripapillary atrophy chosen from the HRT screening were confirmed with clinical findings. No further analysis was conducted using the HRT results.

Optical Coherence Tomography

OCT is a noninvasive technology that provides high-resolution measurements and cross-sectional imaging of the retina and nerve fiber layer. This technique is similar to an ultrasound B-scan but gathers data from optical differences using scattered light instead of sound waves, the details of which are discussed elsewhere.7–9,17,18 All subjects’ eyes were scanned with Stratus OCT (software version A1.0; Carl Zeiss Meditech) after dilating the pupils with 1% tropicamide and 2.5% phenylephrine. Optic nerve head analysis was constructed from six consecutive, equally spaced scans in a radial spoke pattern centered on the optic nerve head. The optic nerve head margin was determined automatically at the termination of the retinal pigment epithelium/choriocapillaris.

Three observers independently analyzed the optic nerve head scans to determine subjectively the optic nerve head margin. This was done by manual placement of the cursor to the retinal pigment epithelium/choriocapillaris edge for each of the six radial scans for each eye. Each observer analyzed the optic nerve head scans twice at least 3 days apart using the method described earlier. The observers were masked to each other and between assessments. The following OCT optic nerve head parameters were used for the analysis: disc area, cup area, rim volume, and cup volume.

Analysis

Shrout–Fleiss reliability of a random set of intra-class correlations (ICC) was performed to compare measurements obtained for each optic nerve head parameter within a given observer. The average of the two measurements from each observer was used to analyze the agreement between observers. The mean of all six observer measurements was used to analyze the agreement between observer and automatic measurements. In both cases, the Shrout–Fleiss reliability of a random set of mean scores’ ICC was performed to compare measurements for each optic nerve head parameter between observers and between observers and automatic measurements. Analysis of variance (ANOVA) was used to compare manual values obtained between individual raters. Receiver operator characteristics (ROC) curve analysis was used to compare manual and automatic disc results to clinical diagnoses and the presence of visual field deficits.

RESULTS

Intra-observer and Inter-observer Agreements

All three observers demonstrated good intra-observer agreement with an ICC of greater than 0.75 for all parameters except rim volume, where in one instance the agreement was as low as 0.43 (Table 1). Statistically significant inter-observer differences were found when comparing the mean of the two measurements of all parameters between the observers (P < .05), but good agreement was found between observers (ICC > 0.86) (Table 2).

TABLE 1.

Intra-observer Intraclass Correlation of Optical Coherence Tomography Optic Nerve Head Measurements as Determined by Manual Detection of Optic Disc Margin

Measurement Observer 1 Observer 2 Observer 3
Disc area, mm2 0.91 0.93 0.96
Cup area, mm2 0.85 0.84 0.85
Rim volume, mm3 0.76 0.52 0.43
Cup volume, mm3 0.96 0.93 0.96
Open in a new tab

TABLE 2.

Mean Optical Coherence Tomography Optic Nerve Head Parameters for Each Observer, Inter-observer Analysis of Variance (ANOVA), and Intraclass Correlations (ICC) Between Observers

Measurement Observer 1 Observer 2 Observer 3 ANOVA ICC
Disc area, mm2 2.23 1.89 1.86 < .0001 0.90
Cup area, mm2 1.29 1.07 0.97 < .0001 0.91
Rim volume, mm3 0.15 0.17 0.19 .046 0.86
Cup volume, mm3 0.34 0.28 0.24 < .0001 0.97
Open in a new tab

Automated and Manual Agreements

Figure 1 demonstrates an example of failed automated detection of the optic nerve head margin where manual correction imposed a substantial change in the measured parameters.

Figure 1.

Figure 1

Open in a new tab

(A) Failure of automatic optical coherence tomography recognition of the termination of the retinal pigment epithelium/chorio-capillaris (left cursor) followed by (B) manual correction in each of six radial scans.

Statistically significant differences were found comparing the mean of all six manual measurements with the mean of the automated measurements for all of the parameters (P < .05), with the exception of rim volume (P = .84). The calculated power to detect a difference of 0.5 mm2 in disc area in 31 eyes was 89%. Good agreement was found between the two methods for all parameters (ICC > 0.71) (Table 3, Fig. 2).

TABLE 3.

Student’s t Test and Intraclass Correlation (ICC) Comparison of Mean Automated and Manual Optical Coherence Tomography Optic Nerve Head Parameters

Measurement Automated Manual P ICC
Disc area, mm2 2.45 1.99 < .0001 0.71
Cup area, mm2 1.35 1.11 .002 0.85
Rim volume, mm3 0.17 0.17 .84 0.82
Cup volume, mm3 0.37 0.29 < .0001 0.94
Open in a new tab

Figure 2.

Figure 2

Open in a new tab

Bland–Altman plot comparing manual and automated cup volumes.

ROC Analysis

Areas under ROC curves were calculated to determine the ability of the automated and the manual determination of the optic disc margin to detect glaucomatous abnormalities based on either the clinical classification or the presence of visual field defects (Table 4, Fig. 3). Due to the small number of normal subjects (n = 3), the clinical comparison was done between glaucoma and nonglaucoma, including suspected glaucoma in the nonglaucoma group. On the basis of this clinical classification, the area under ROC curves ranged between 0.53 and 0.64 using the automated determination and 0.49 and 0.70 using the manual determination. Based on visual field defect, the ranges were 0.56 to 0.70 and 0.61 to 0.73, respectively.

TABLE 4.

Area Under the Receiver Operator Curves Using Automated or Manual Determination of the Optic Disc Margin for Distinguishing Between Groups Based on Clinical Diagnosis and Presence of Visual Field Defect

Clinical Classification Visual Field Defect
Measurement Automated Manual Automated Manual
Disc area, mm2 0.64 0.67 0.56 0.61
Cup area, mm2 0.53 0.52 0.68 0.69
Rim volume, mm3 0.63 0.70 0.63 0.62
Cup volume, mm3 0.57 0.49 0.70 0.73
Open in a new tab

Figure 3.

Figure 3

Open in a new tab

Receiver operator characteristics curve analysis comparing manual and automated cup areas for the presence of visual field defects.

DISCUSSION

OCT is unique in its ability to detect the optic nerve head margin in an automated fashion. A previous cross-sectional study reported high correlation between OCT optic nerve head measurements obtained by automatic and manual determination of the optic disc margin.6 Given that the definition of the disc margin is based on the detection of the retinal pigment epithelium/choriocapillaris border, the presence of peripapillary atrophy may induce a significant misalignment of the disc margin read by OCT. Although the previous study may have contained subjects with peripapillary atrophy, the percentage was small in comparison with those without this pathology. Our study aimed to investigate the automatic disc assessment of OCT in discs with peripapillary atrophy.

Peripapillary atrophy is more frequently observed and is more extensive in patients with glaucoma compared with normal individuals.10–16 Other studies reported high correlation between peripapillary atrophy, optic disc damage, and mean visual field loss.19–24 Although peripapillary atrophy is known to be prevalent in highly myopic eyes,25–28 these eyes were not qualified for this study to eliminate confounding findings due to the elongated axial length. This study was designed to include only those patients with peripapillary atrophy related to glaucoma.

Despite the difficulty of defining the optic disc margin in the presence of peripapillary atrophy, good intra-observer agreements were shown for the manually corrected OCT optic nerve head measurements (Table 1). Although statistically significant differences were detected in the mean measurements between observers, they did demonstrate good inter-observer agreement (ICC = 0.86 to 0.97) (Table 2). The similar intra-observer and inter-observer agreement is contradictory to previous studies, where higher agreement was found within observers than between observers in optic nerve head assessment.29,30

Statistically significant differences were found when comparing the mean measurements of the three observers with the automated determinations of the disc margin, whereas good correlations (ICC = 0.71 to 0.94) were found between the two methods (Table 3, Fig. 2). This finding suggests that the automatic optic nerve head measurements are in agreement with the results obtained by manual detection of the disc margin, but that the actual measures are not interchangeable. This is due to the fact that the automated determination of the disc margin identified a larger disc area than the manual determination and thereafter all parameters related to disc size were subsequently larger than those measured manually.

ROC curve analysis was performed to assess whether manual OCT adjustments lead to better discrimination between groups as determined by clinical evaluation and visual field findings. The manual determination of the disc margin did not improve the ability to distinguish between groups for either the clinical assessment or the visual field evaluation (Table 4, Fig. 3). Similar areas under ROC curves were reported in a previous study where OCT optic nerve head parameters were evaluated in 159 eyes.6 The low ROC values emphasize the need for multiple parameters for glaucoma evaluation.

A significant difference between OCT optic nerve head results obtained by automated and manual disc margin determination was found in patients with peripapillary atrophy. Good agreement between the two methods was achieved and there was no difference in the area under ROC curves for distinguishing between groups as defined by clinical diagnosis and presence of visual field defect. Thus, the automated feature of the OCT optic nerve head analysis can be used in the clinical setting in the presence of peripapillary atrophy but the measurements should be evaluated with caution when compared with population study results.

Footnotes

Supported in part by NIH R01-EY13178, RO1-EY11289, P30-EY13078, by a grant from the Massachusetts Lions Eye Research Fund, Inc., by Research to Prevent Blindness, Inc., and by the American Physician Fellowship for Medicine in Israel.

References

  • 1.Sommer A, Pollack I, Maumenee A. Optic disc parameters and onset of glaucomatous field loss: I. Methods and progressive changes in disc morphology. Arch Ophthalmol. 1979;97:1444–1448. doi: 10.1001/archopht.1979.01020020106002. [DOI] [PubMed] [Google Scholar]
  • 2.Pederson JE, Anderson DR. The mode of progressive disc cupping in ocular hypertension and glaucoma. Arch Ophthalmol. 1980;98:490–495. doi: 10.1001/archopht.1980.01020030486010. [DOI] [PubMed] [Google Scholar]
  • 3.Quigley HA, Addicks EM, Green WR. Optic nerve damage in human glaucoma: III. Quantitative correlation of nerve fiber loss and visual field defect in glaucoma, ischemic neuropathy, papilledema, and toxic neuropathy. Arch Ophthalmol. 1982;100:135–146. doi: 10.1001/archopht.1982.01030030137016. [DOI] [PubMed] [Google Scholar]
  • 4.Sommer A, Katz J, Quigley HA, et al. Clinically detectable nerve fiber atrophy precedes the onset of glaucomatous field loss. Arch Ophthalmol. 1991;109:77–83. doi: 10.1001/archopht.1991.01080010079037. [DOI] [PubMed] [Google Scholar]
  • 5.Quigley HA, Katz J, Derick RJ, Gilbert D, Sommer A. An evaluation of optic disc and nerve fiber layer examinations in monitoring progression of early glaucoma damage. Ophthalmology. 1992;99:19–28. doi: 10.1016/s0161-6420(92)32018-4. [DOI] [PubMed] [Google Scholar]
  • 6.Schuman JS, Wollstein G, Farra T, et al. Comparison of optic nerve head measurements obtained by optical coherence tomography and confocal scanning laser ophthalmoscopy. Am J Ophthalmol. 2003;135:504–512. doi: 10.1016/s0002-9394(02)02093-7. [DOI] [PubMed] [Google Scholar]
  • 7.Hee MR, Izatt JA, Swanson EA, et al. Optical coherence tomography of the human retina. Arch Ophthalmol. 1995;113:325–332. doi: 10.1001/archopht.1995.01100030081025. [DOI] [PubMed] [Google Scholar]
  • 8.Schuman JS, Hee MR, Puliafito CA, et al. Quantification of nerve fiber layer thickness in normal and glaucomatous eyes using optical coherence tomography. Arch Ophthalmol. 1995;113:586–596. doi: 10.1001/archopht.1995.01100050054031. [DOI] [PubMed] [Google Scholar]
  • 9.Pieroth L, Schuman JS, Hertzmark E, et al. Evaluation of focal defects of the nerve fiber layer using optical coherence tomography. Ophthalmology. 1999;106:570–579. doi: 10.1016/s0161-6420(99)90118-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Primrose J. Early signs of the glaucomatous disc. Br J Ophthalmol. 1971;55:820–825. doi: 10.1136/bjo.55.12.820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Wilensky JT, Kolker AE. Peripapillary changes in glaucoma. Am J Ophthalmol. 1976;81:341–345. doi: 10.1016/0002-9394(76)90251-8. [DOI] [PubMed] [Google Scholar]
  • 12.Jonas JB, Nguyen NX, Gusek GC, Naumann GOH. Parapapillary chorioretinal atrophy in normal and glaucoma eyes: I. Morphometric data. Invest Ophthalmol Vis Sci. 1989;30:908–918. [PubMed] [Google Scholar]
  • 13.Jonas JB, Naumann GO. Parapapillary chorioretinal atrophy in normal and glaucoma eyes: II. Correlations. Invest Ophthalmol Vis Sci. 1989;30:919–926. [PubMed] [Google Scholar]
  • 14.Jonas JB, Konigsreuther KA, Naumann GOH. Optic disc histomorphometry in normal eyes and eyes with secondary angle-closure glaucoma: II. Parapapillary region. Graefes Arch Clin Exp Ophthalmol. 1992;230:134–139. doi: 10.1007/BF00164651. [DOI] [PubMed] [Google Scholar]
  • 15.Jonas JB, Fernandez MC, Naumann GOH. Glaucomatous parapapillary atrophy: occurrence and correlations. Arch Ophthalmol. 1992;110:214–222. doi: 10.1001/archopht.1992.01080140070030. [DOI] [PubMed] [Google Scholar]
  • 16.Jonas JB, Xu L. Parapapillary chorioretinal atrophy in normal-pressure glaucoma. Am J Ophthalmol. 1993;115:501–505. doi: 10.1016/s0002-9394(14)74453-8. [DOI] [PubMed] [Google Scholar]
  • 17.Pedut-Kloizman T, Pakter HM, Schuman JS, et al. Ophthalmic diagnosis using optical coherence tomography. Ophthalmol Clin North Am. 1998;11:1178–1181. [Google Scholar]
  • 18.Schuman JS, Kim J. Imaging of the optic nerve head and nerve fiber layer in glaucoma. Ophthalmol Clin North Am. 2000;13:383–406. [Google Scholar]
  • 19.Jonas JB, Gusek GC, Fernandez MC. Correlation of the blind spot size to the area of the optic disk and parapapillary atrophy. Am J Ophthalmol. 1991;111:559–565. doi: 10.1016/s0002-9394(14)73698-0. [DOI] [PubMed] [Google Scholar]
  • 20.Araie M, Sekine M, Suzuki Y, Koseki N. Factors contributing to the progression of visual field damage in eyes with normal-tension glaucoma. Ophthalmology. 1994;101:1440–1444. doi: 10.1016/s0161-6420(94)31153-5. [DOI] [PubMed] [Google Scholar]
  • 21.Park KH, Tomita G, Liou SY, Kitazawa Y. Correlation between peripapillary atrophy and optic nerve damage in normal-tension glaucoma. Ophthalmology. 1996;103:1899–1906. doi: 10.1016/s0161-6420(96)30409-0. [DOI] [PubMed] [Google Scholar]
  • 22.Uchida H, Ugurlu S, Caprioli J. Increasing peripapillary atrophy is associated with progressive glaucoma. Ophthalmology. 1998;105:1541–1545. doi: 10.1016/S0161-6420(98)98044-7. [DOI] [PubMed] [Google Scholar]
  • 23.Daugeliene L, Yamamoto T, Kitazawa Y. Risk factors for visual field damage progression in normal-tension glaucoma eyes. Graefes Arch Clin Exp Ophthalmol. 1999;237:105–108. doi: 10.1007/s004170050203. [DOI] [PubMed] [Google Scholar]
  • 24.Emdadi A, Kono Y, Sample PA, Maskaleris G, Weinreb RN. Parapapillary atrophy in patients with focal visual field loss. Am J Ophthalmol. 1999;128:595–600. doi: 10.1016/s0002-9394(99)00189-0. [DOI] [PubMed] [Google Scholar]
  • 25.Jonas JB, Gusek GC, Naumann GOH. Optic disk morphometry in high myopia. Graefes Arch Clin Exp Ophthalmol. 1988;226:587–590. doi: 10.1007/BF02169209. [DOI] [PubMed] [Google Scholar]
  • 26.Nevarez J, Rockwood EJ, Anderson DR. The configuration of peripapillary tissue in unilateral glaucoma. Arch Ophthalmol. 1988;106:901–903. doi: 10.1001/archopht.1988.01060140047021. [DOI] [PubMed] [Google Scholar]
  • 27.Kasner O, Fever WJ, Anderson DR. Possibly reduced prevalence of peripapillary crescents in ocular hypertension. Can J Ophthalmol. 1989;24:211–215. [PubMed] [Google Scholar]
  • 28.Vongphanit J, Mitchell P, Wang JJ. Population prevalence of tilted optic disks and the relationship of this sign to refractive error. Am J Ophthalmol. 2002;133:679–685. doi: 10.1016/s0002-9394(02)01339-9. [DOI] [PubMed] [Google Scholar]
  • 29.Tuulonen A, Jonas JB, Valimaki S, Alanko HI, Airaksinen PJ. Interobserver variation in the measurements of peripapillary atrophy in glaucoma. Ophthalmology. 1996;103:535–541. doi: 10.1016/s0161-6420(96)30661-1. [DOI] [PubMed] [Google Scholar]
  • 30.Garway-Heath DF, Poinoosawmy D, Wollstein G, et al. Inter- and intraobserver variation in the analysis of optic disc images: comparison of the Heidelberg retina tomograph and computer assisted planimetry. Br J Ophthalmol. 1999;83:664–669. doi: 10.1136/bjo.83.6.664. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES