Chapter 216 – Disc Analysis
ZINARIA Y. WILLIAMS
JOEL S. SCHUMAN
Stereoscopic Optic Nerve Head Photography
• Objective, three-dimensional, photographic representation of optic nerve head appearance.
Optic Nerve Head Morphometry-Planimetry
• Formerly used to quantify optic nerve head parameters, now supplanted by more recent technologies.
Optic Nerve Head Analyzers
• First generation of automated, quantitative measurements of optic nerve head structure (represented in this chapter by the Glaucoma Scope).
Confocal Scanning Laser Ophthalmoscopy
• Technology for optic nerve head measurements, which produces and integrates 32 coronal scans at increasing tissue depths.
Optical Coherence Tomography
• Technology for high-resolution, cross-sectional tissue imaging, well suited to nerve fiber layer and retinal measurements and becoming more useful for optic nerve head assessment.
For nearly 150 years, since Helmholtz first viewed the optic nerve in living humans, disc appearance has been used to evaluate glaucoma status. Although clinicians may disagree about the causes of optic nerve head (ONH) damage, most accept that ONH cupping, or thinning of the neuroretinal rim, is a reliable indicator of the disease.
In the evaluation of the disc, the clinician must determine whether the ONH appearance is normal or pathological, whether an anomalous appearance is the result of glaucoma or a different pathological process, and how the ONH differs from that found in previous examinations. Given the considerable variation in ONH appearance among normal subjects, the various patterns of glaucomatous cupping, and the wide variation in the assessment of ONH appearance among examiners or even from visit to visit with the same examiner, these evaluations may be exceedingly difficult to make consistently. Commonly, several variables of the clinical examination have to be considered to conclude that progression of glaucomatous damage has occurred and, accordingly, to decide whether intraocular pressure control is adequate.
As a result of the variable nature of subjective ONH assessments and because a relatively large change is necessary to conclude reliably that an actual progression of disease has occurred, techniques and instrumentation have been developed to quantify objectively ONH structures.
The first ophthalmoscope was introduced by Helmholtz in 1850, and soon thereafter the first reports of ONH changes in glaucoma appeared. It quickly became apparent that a more objective recording of ONH appearance was necessary, but no convenient technology was available for this purpose at the time. The first photograph was created on a glass plate by Nicéphore Nièpce in 1826. In 1888 George Eastman developed his “everyman’s camera,” the Kodak. However, it was not until 1889, when Eastman developed flexible film, that photography became both popular and a profession. Early fundus photography (by Howe in 1887) was limited by reflexes from both the light source and ocular tissues, improper illumination, and patient eye movement during the long exposure time of several seconds.
Tremendous technical advances in light sources, flashes, film and filter quality, techniques of development, and reduction of exposure time (to milliseconds) improved the quality of fundus photography dramatically. Stereoscopic fundus photography was first described by Thorner in 1909. Stereoscopy was attempted by movement of the patient’s eye very slightly between exposures or by movement of the camera in front of the constantly fixated eye.
The first to introduce simultaneous stereoscopic ONH photography was Nordenson in 1930, using a pair of small prisms in front of the lenses of a conventional Zeiss camera, based on the principles of a modified reflex-free Gullstrand ophthalmoscope. Other attempts were made by Norton in 1953 and Drews in 1957, but the image quality was poor. A new camera was developed in 1964 by Donaldson and revised in 1976; the design integrated two fundus cameras that used the same primary objective to obtain simultaneous photographs, employed the indirect ophthalmoscope principle (separation of the illuminating light rays that entered and reflected out of the pupil), and took advantage of a new electronic flash that provided sufficient illumination for a 35?mm color film. This camera was fast and relatively easy to operate. The depth effect was highly reproducible, and in the revised prototype, distortion was reduced to a minimum.
ONH analyzers were introduced in the early 1980s to offer automated, quantitative measurement of the disc, cup, neuroretinal rim area, and related parameters; these analyzers are based on standard fundus camera optics.
PURPOSE OF OPTIC NERVE HEAD IMAGING AND ANALYSIS
Devices described in this chapter are designed to image and/or measure ONH parameters, such as cup-to-disc ratio, disc and cup area and volume, neuroretinal rim width and area, area of pallor, and cup slope, in a simple, fast, cost-effective, objective, accurate, and reproducible way. Additional features in some of the technologies include quantitative or semiquantitative analysis of results and correction for highly variable patient-related factors such as refractive error, axial length, and anterior corneal curvature. A major source of variation is disc size, and correction for this factor is still a future goal. Disc area may vary between individuals by as much as seven times among normal eyes. It is still not clear which of the ONH parameters mentioned has the
greatest predictive and/or prognostic value for glaucomatous damage. The goals for clinicians in the evaluation of this type of information are to differentiate between normal and glaucomatous eyes and to detect mild to moderate progression of glaucomatous damage with a high level of certainty.
Stereoscopic Optic Nerve Head Photography
Stereoscopic ONH photography is currently the most widely used technology that enables clinicians to document ONH appearance objectively. Stereoscopic photography offers several advantages—its product shows an image most clinicians are familiar and comfortable with (a real picture of the disc with its natural color), fundus cameras are widely available, and the process requires only a modest degree of technical skill from the ophthalmic photographer. Moreover, fundus cameras are probably the least expensive permanent ophthalmic imaging systems. The main disadvantages of ONH photography are subjective and qualitative assessment, the need for clear media and a dilated pupil, and the lack of an immediate result (for film systems).
Currently, two methods are used to obtain a stereoscopic image of the ONH—sequential (consecutive) and simultaneous. Sequential stereoscopic photography involves a shift of the camera joystick to opposite sides of a fully dilated pupil either manually or via a mechanical sliding carriage adapter, such as the Allen stereo separator (introduced in 1964). Several conditions must exist to obtain a consistent stereogram pair—the patient must maintain a constant head position and fixation, and camera-dependent factors such as focus depth, light intensity, stereo base, and exact camera angle must be identical for each photographic session (an almost impossible goal to achieve).
The Donaldson camera has long been considered the standard in stereoscopic ONH photography, but unfortunately it is no longer commercially available. The successor to the Donaldson camera, the Nidek 3Dx camera, was introduced in 1990. It is a simultaneous stereocamera that produces stereoscopic images in either a 35?mm split-frame format, reviewed stereoscopically by a viewing system, or a 3.5 × 5 inch, single-sheet, three-dimensional transparency. Minckler et al. found the Nidek camera to be superior to the Zeiss system in addition to being easier to use and faster. Greenfield et al. found the Nidek 3Dx system to be superior to the Donaldson camera in that it produced better-quality stereoscopic images and significantly greater reproducibility of ONH assessment ( Fig. 216-1 ).
Optic Disc Morphometry (Planimetry)
Morphometry of the ONH is an intermediate stage between freehand, qualitative drawing, based on a slit-lamp examination,
Figure 216-1 Simultaneous stereoscopic photographs of the optic nerve head taken with the 3Dx camera. Disc of an advanced primary open-angle glaucoma patient.
and quantitative, digital imaging. Described extensively by Jonas and colleagues in Germany, this method is based on stereophotographs of the ONH taken using a telecentric Zeiss fundus camera with an Allen stereo separator. The 15° color slides or transparencies are projected on a scale of 1 to 15. The outline margins of the optic disc, optic cup, peripapillary scleral ring, and peripapillary chorioretinal atrophy are plotted by hand on paper and analyzed morphometrically (using a planimetry computer, which calculates areas according to their coordinates).
The optic cup is outlined on the basis of contour, when viewed stereoscopically, and not on the basis of pallor. Helpful clues in the determination of the cup margin are bent vessels and variation in tissue heights and planes. The optic disc margin is judged to be the inner margin of the peripapillary scleral ring, which is a thin, white band that encircles the ONH (see Fig. 216-1 ). This is easier to detect on the temporal than on the nasal disc border. In the calculation of absolute values and areas in units (i.e., mm or mm2 ), a correction for ocular and photographic magnification is made using Littman’s method, which compensates for the constant twofold magnification of the telecentric fundus camera and for ocular magnification that arises from refractive error and anterior corneal curvature. Littman’s method may produce false values in cases of abnormally high refractive power of the lens (e.g., from cataracts that cause myopia of more than 1D) and so far is not applicable for cases of aphakia or pseudophakia.
The disc surface, which includes the peripapillary region, is divided into four sectors on the basis of the anatomic organization of the four main retinal vessel branches and the nerve fiber layer (NFL) bundles ( Fig. 216-2 ). Sector A (temporal) covers 64° and represents the papillomacular fiber-bundle area; sector B (superior temporal) and sector C (inferior temporal) represent the arcuate nerve fiber-bundle areas; and sector D covers 116° and represents the remaining nasal area. Sectors B and C each occupy 90°, and their midlines reside 13° temporal to the vertical optic disc axis (according to a previous study that showed this to be the most common location for neuroretinal rim notches). Neuroretinal rim and optic disc areas are measured separately in each of the four sectors. Total neuroretinal rim area is calculated as the difference between the disc and cup areas.
In a prospective, masked, longitudinal study, Caprioli et al. followed 193 eyes of primary open-angle glaucoma and ocular hypertensive patients for 3.3 ± 1.0 years and compared different
Figure 216-2 Optic disc area divided into four sectors. The green and red sectors represent superior temporal and inferior temporal retina—note that their middle axis is tilted temporally from the midline (angle ß). The blue sector represents 64° of retina on the temporal side and the yellow sector covers the remaining 116° on the nasal side.
methods to evaluate the ONH and NFL for glaucomatous change. The detection rate of structural glaucomatous change was 15% using qualitative evaluation of ONH stereophotographs, 7.2% using qualitative evaluation of NFL stereophotographs, 3.6% using manual stereoplanimetry of the disc rim area, and 13.2% using relative NFL height measured by the Rodenstock analyzer (see later). A low degree of overlap between the various techniques may indicate that qualitative techniques have a higher sensitivity in the detection of small focal changes and quantitative measurements are better for the detection of local diffuse loss. Manual planimetry of the neural rim area showed changes in eyes that demonstrated no progression using qualitative ONH and NFL evaluation.
The Discam (Marcher Enterprises Ltd, Hereford, UK), is a new computer-assisted planimetric system that uses a digitized semiautomatic charge-coupled device (CCD) camera to capture 20° × 20° ONH photographs and display them stereoscopically. The Discam acquires sequential disc photographs in a semi-automatic fashion, and the optic disc and the cup margins are outlined using the technique previously described to obtain quantitative parameters of the ONH. Optic disc measurement using the Discam was found to have high interobserver and intraobserver repeatability. 7a In a more recent study, 386 eyes underwent ONH analysis using the Discam digital optic disc stereo camera. The following parameters were used for this study: cup-to-disc area ratio, vertical cup-to-disc ratio, and horizontal cup-to-disc ratio. The results were compared with ONH analysis using confocal scanning laser ophthalmoscopy (see later) and stereoscopic disc photography (stereography) to evaluate the ability of these three methods to detect glaucomatous changes. Although the results of all three techniques are not interchangeable, good agreement was found between ONH measurements measured by the Discam and the other technologies. 7b
Digital, Quantitative Imaging of the Optic Nerve Head
Multiple computerized systems have emerged since the early 1980s; each attempts to give objective, accurate, reproducible, quantitative, detailed topographic maps of the ONH and the peripapillary NFL and the potential for longitudinal comparison. Most systems share several basic features—they are all noncontact and noninvasive, a charge-coupled device (CCD) camera enables the operator to see an image of the fundus and a real-time tomographic image, and the data are stored digitally. Despite these external similarities, fundamental differences exist in the imaging technologies in regard to basic concept, light source, operator-dependent variables, and reproducibility of measurements.
Optic Nerve Head Analyzers
PAR IS 2000 (IMAGENET).
The original PAR IS 2000 digitally captures simultaneous, monochromatic, stereoscopic ONH images using the TRC-SS fundus camera or digitizes images from 35?mm slides using a color video camera. The operator defines the ONH margin by the selection of four coincident points (control points) on both images of one stereo pair, from which an elliptical disc contour is fitted automatically; the cup is defined as the area located 120?mm posterior to the disc margin. After image registration, cross-correlation, and enhancement, the analysis system generates stereometric parameters, three-dimensional topographic plots of the ONH, and vessel-shift analysis. The complete process takes 15–30 minutes, depending primarily on the ease of image acquirement. Both inter- and intraobserver variability in the calculation of cup-to-disc ratio is in the range 25–30%. The primary source of variability is the determination of the disc margin by the operator.  The currently available version has improved software and also may be used for fluorescein and indocyanine green angiography.
RODENSTOCK OPTIC NERVE HEAD ANALYZER.
The Rodenstock ONH analyzer (no longer commercially available)
Figure 216-3 Glaucoma Scope technique based on computed raster stereography principle. About 25 parallel horizontal lines are projected onto the optic nerve head at an oblique angle. The lines are deflected relative to a reference plane close to the disc margin and proportionate to the depth of excavation. (Adapted with permission from Netland PA. The Glaucoma-Scope: principles, techniques, and applications. In: Schuman JS, ed. Imaging in glaucoma. Thorofare, NJ: Slack; 1997:17–32.)
utilizes a simultaneous stereoscopic video camera and is used to capture an image of the ONH. Two sets of seven evenly spaced lines are projected onto the captured image, and from their horizontal displacement the computer generates a contour map of the ONH surface and calculates the depth in each of the 1600 points on the optic disc. The margin of the ONH is defined interactively by the operator in four cardinal locations and an ellipse is fitted around those points. The cup margin is defined by a curve that connects all points 150?µm below the disc margin.
THE GLAUCOMA SCOPE.
The Glaucoma Scope performs ONH analysis based on a technique of computer raster stereography described by Holm and Krakau in 1970. The system consists of an optical head, used for image acquisition, a monitor that shows a video image of the ONH, and a computerized image analysis component. A halogen lamp using near-infrared light (750?nm) produces a series of approximately 25 parallel, horizontal, and equally spaced lines, which are projected onto the ONH at an angle of 9°. The projected lines are deflected proportionately to the depth of excavation—a shallow cup creates a small deflection whereas a deep cup causes a large deflection of the lines ( Fig. 216-3 ).
Several variables of the image are determined by the operator, for example, a reference point and the disc margin. The reference point selected by the operator, most commonly at a major blood vessel branch, defines the center of a 128 × 128 pixel area and is used for future image registration. For future images of the same ONH, the operator chooses a reference point that ideally is within 5–10 pixels of the original reference point. At the initial visit the operator outlines the disc margin—at least eight points are marked around the disc and the major blood vessels are identified. The outline of the disc margin and blood vessels appears on the printout as a landmark but has no effect on the algorithm calculations of depth values. The refractive error of the eye is recorded for correction of magnification in the disc measurements.
The ONH and projected lines are viewed in real time on the video monitor and the operator may optimize focus and illumination. Multiple images are captured and stored in digital form
Figure 216-4 Glaucoma Scope analysis. The data shown are from a 9-year-old girl who had corticosteroid-induced unilateral glaucoma. A, Optic nerve head demonstrates marked cupping of the right eye with extensive loss of the neuroretinal rim temporally. B, Glaucoma Scope analysis correlates with the clinical photographs. Left image shows quantitative data; gray-scale image to the right. (With permission from Netland PA. The Glaucoma-Scope: principles, techniques, and applications. In: Schuman JS, ed. Imaging in glaucoma. Thorofare, NJ: Slack, 1997:17–32.)
on optical discs. The quality (or focus) of the captured image is assessed on a logarithmic scale and depends on the ratio of horizontal line data to nonhorizontal line data (nonhorizontal line data are created mainly by blood vessels) and on the degree of overexposure (to eliminate images that are too bright to be analyzed). The analysis system of the Glaucoma Scope automatically analyzes the image with the highest quality scale unless other images are chosen by the operator. A computer algorithm converts horizontal line data from the captured image into corresponding numerical depth values. A reference plane for the depth measurements is calculated as the mean depth of two 50?µm columns placed about 350?µm nasal and temporal to the ONH margin. The optic cup is defined as an area inside the ONH that lies 140?µm or more below the reference plane.
The depths or elevations of over 8750 real data points are calculated in an area of approximately 350 × 280 pixels (which corresponds roughly to a 20° area on the retina). The printout contains a gray-scale map with up to 722 numerical values, a three-dimensional grid map, a battery of ONH parameters, a disc and cup outline, and a progression report that shows points that have changed by more than 50?µm on follow-up tests (see Fig. 216-4 ). Pendergast and Shields found the standard deviation of measurements to be less than 50?µm.
The Glaucoma Scope is limited by the need for a pupil size of at least 4.5–5?mm, clear media, and an experienced operator. In addition, because of the nature of this technology, only surface topography can be analyzed using the Glaucoma Scope.
Gundersen et al. examined 138 normal subjects and 102 glaucoma patients for vertical and horizontal cup-to-disc ratio, minimum rim width within the 60° and 90° sectors across the vertical meridian, and rim and cup area. The most sensitive parameter for differentiation of normal from glaucomatous eyes was minimum rim width at 60° and 90° (which represents localized changes of the optic disc), followed by vertical cup-to-disc ratio. Global indices, such as rim and cup area and horizontal cup-to-disc ratio, were the least sensitive. These
Figure 216-5 Confocal scanning laser ophthalmoscopy. The tissue is scanned in 16–64 coronal sections at consecutive focal planes in the HRT I. The three-dimensional image acquired consists of 256 × 256 pixels; acquisition time is approximately 1.4?sec. In the HRT II, the three-dimensional image acquired consists of 384 × 384 pixels with an acquisition time of approximately 1.0?sec. (Adapted from Schuman JS, Noeker RJ. Imaging of the optic nerve head and nerve fiber layer in glaucoma. Ophthalmol Clin North Am. 1995;8:259–79.)
differences were most obvious in eyes affected by mild to moderate glaucomatous field defects.
Confocal Scanning Laser Ophthalmoscopy
Confocal scanning laser ophthalmoscopy (CSLO) offers real-time, three-dimensional imaging of the ONH and NFL, with reduced need for pupillary dilation or clear media. A confocal optical system is designed to allow only a “thin” slice of the target tissue to be in focus on the image plane—light rays reflected from higher or lower focal planes are blocked, which creates high-resolution tomographic images. The tissue is illuminated and imaged point by point through a pinhole. The system is confocal because both the illumination pinhole and the imaging pinhole correspond to the same focal point on the tissue. A three-dimensional image may be obtained by varying the x, y, and z coordinates. A pair of galvanometer-controlled mirrors facilitates rapid horizontal and vertical scanning across the object.
The Heidelberg Retina Tomograph (HRT, Heidelberg Engineering GmbH, Heidelberg, Germany) is the only commercially available confocal scanning laser ophthalmoscope. There are two systems. The HRT I is a scanning system for acquisition and analysis of the posterior pole. The typical use of the HRT is for the assessment of the ONH. It uses a diode laser of 670?nm to scan a three-dimensional image from a series of optical sections at 32 consecutive focal planes ( Fig. 216-5 ). Because laser tomographic scanning occurs one point at a time, the confocal scanning optical microscope reconstructs an ONH image by bringing a series of two-dimensional digitized images into registration. The registration process corrects for microsaccades that occur during image acquisition. The topography image consists of 256 × 256 pixel elements, each of which is a measurement of height at its corresponding location. The optical transverse resolution is approximately 10?µm, whereas the axial resolution is about 300?µm. The transverse field of view can be 10° × 10°, 15° × 15°, or 20° × 20°. In current clinical practice, three scans of each eye are taken and then averaged to create a mean topography image. The printed report shows a topographic image and a reflectivity image of the ONH and its contour line, ONH stereometric parameters, and a mean-height contour of the peripapillary retina ( Figs. 216-6 and 216-7 ).
Figure 216-6 Confocal scanning laser ophthalmoscopy printed report—HRT I. The report is from a subject with glaucoma. A, The topographic (left) and reflectivity image (right) illustrates the ONH. In the contour graph (below), the white line represents the reference plane at which there is a height of zero. The red line represents the height of the reference line between the cup and disk. The green line is the retinal height of the subject eye at the contour line showing the typical double hump feature at the superior and inferior poles. B, The topographic image is shown with the cup represented in red, the sloping neural tissue in blue, and the rim in green. The ONH parameters and subject classification are shown on the right. The classification number for the HRT I is determined by an automated algorithm devised by Frederick Mikelberg based on the ONH and retinal parameters.
Figure 216-7 Optic nerve head. The standard reference plane at 50?µm below the retinal surface. (Adapted with permission from Zangwill L, Horn SV, Lima MDS, et al. Optic nerve head topography in ocular hypertensive eyes using confocal scanning laser ophthalmology. Am J Ophthalmol. 1996;122:520–5.)
In a comparative study, expert clinicians evaluated 72 normal patients and 51 patients with early glaucoma using qualitative assessment of stereoscopic optic disc photographs and CSLO imaging. The Heidelberg Retina Tomograph was reported to be more sensitive than clinical assessment in detecting early glaucomatous disc changes. In another study, investigators analyzed 13 ocular hypertensive eyes that subsequently developed reproducible visual field defects and 13 normal eyes that had undergone sequential optic disc images. HRT has also been found to detect glaucomatous changes in the optic disc before visual field changes occurred.
The recently developed HRT II is designed for topographic ONH analysis. The HRT II is small, lightweight, portable, and almost completely automatic. All image acquisition parameters are either fixed or predetermined. The system automatically acquires 16–64 image planes covering a field of view fixed at 15° × 15° using 384 × 384 pixels per plane. Utilizing an internal fixation target, the HRT II automatically acquires three images with the use of a quality control system that will acquire additional images if one or more of the images acquired cannot be used (e.g., fixation loss). The printed report shows a topographic and reflectivity image of the ONH and its contour line, details of the classification, and ONH stereometric parameters ( Fig. 216-8 ).
The original HRT is a research-oriented tool with a wide range of applications, including measuring retinal circulation when combined with the Heidelberg Retina Flowmeter; the HRT II is restricted to ONH analysis.
An inherent limitation of this technology lies in the reference plane. A reference plane is required to calculate cup area, cup-to-disc ratio, cup volume, rim area, rim volume, retinal NFL thickness, and retinal NFL cross-sectional area. The reference plane used by the current software may change over time, especially in patients with glaucoma who have changing topography. Another obstacle is the manual delineation of the optic disc margin by the operator and the influence of this on ONH parameters.
The most recent published data regarding analysis of ONH parameters using CSLO indicate that the slope of the cup (“the third central moment of the depth distribution”) is the most significant parameter in the prediction of glaucoma status (see Fig. 216-8 ).   There appears to be some ability to discriminate between normal and glaucomatous eyes using CSLO, with a sensitivity and specificity of about 85%; however, considerable overlap exists between normal, ocular hypertensive, and glaucomatous eyes.  Finally, some authors have claimed the ability to determine NFL thickness, or cross-sectional area, using CSLOs—by using a reference point in the nasal retina or in the macula, a given thickness is determined (calculated) to represent NFL. This indirect measurement of NFL thickness is probably not the best method for NFL analysis, nor is it particularly accurate given the low axial resolution of CSLO (approximately 300?mm) and the fact that superior technologies exist for the direct assessment of NFL thickness (see Chapter 217 ). Software developed by Chauhan and colleagues to detect topographic changes in the optic disc and peripapillary retina appears to provide the best longitudinal data analysis to date using HRT II. Seventy-seven subjects with early glaucomatous visual field damage were followed
Figure 216-8 Confocal scanning laser ophthalmoscopy printed report— HRT II. This report is from a different subject with glaucoma. The topographic image (left) is shown with the cup represented in red, the sloping neural tissue in blue, and the rim in green. The reflectivity image (right) illustrates the classification of the six ONH sectors. Each sector is marked with a green check mark, a red cross, or a yellow exclamation mark to illustrate being within normal limits, outside normal limits, or borderline, respectively. A bar graph represents this further in the right middle panel. The stereometric parameters are displayed in the left middle panel. The classification number for HRT II is derived from an algorithm developed by Wollstein et al. at Moorfields Eye Hospital. (Courtesy of Heidelberg Engineering, Inc., Carlsbad, CA.)
with scanning laser tomography and with conventional perimetry. Investigators found glaucomatous disc changes determined with scanning laser tomography to occur more frequently than visual field changes. This result suggests that glaucomatous damage and progression may be detected earlier using scanning laser polarimetry. 
Optical Coherence Tomography
Optical coherence tomography (OCT) is a diagnostic imaging technology that utilizes interferometry and low-coherence light in the near-infrared range (approximately 840?nm) to achieve high-resolution (about 10?µm for OCT 1 and 2, about 7–8?µm for OCT 3), cross-sectional imaging of the eye ( Fig. 216-9 ). In OCT, a beam of low-coherence light is split to the tissue of interest (probe beam) and to a reference mirror at a known variable position (reference beam). Multiple echoes are reflected or backscattered from the eye, but for the two beams to recombine and produce positive interference on a photodetector (and hence to create an image) their pulses must arrive simultaneously or within the short coherence length of the light source. Thus, it is the low-coherence light source that primarily determines the longitudinal resolution. Transverse resolution is determined by the probe beam diameter (20?mm) and effective transverse resolution is affected by transverse pixel spacing. In OCT 1 and 2, a sequence of 100 successive, longitudinal measurements (i.e., A-scans) is used to construct a false color topographic image of tissue microsections that appears remarkably similar to histological sections; OCT 3 uses 500 axial scans acquired in 1 second.
OCT 2 and 3 are equipped with ONH analysis software for the clinical assessment of the ONH. The ONH analysis evaluates the disc and cup anatomy and quantifies the amount of nerve tissue in the optic nerve at the disc. Quantification of the nerve tissue is done by calculating the cross-sectional area of the nerve tissue and then calculating the minimum distance from the disk to the retina surface. The analysis program detects the anterior surface of the NFL and retinal pigment epithelium (RPE) by searching each A-scan axially for the highest rates of change in reflectivity. Once these boundaries have been determined, the algorithm identifies and measures all features of disc anatomy based on the anatomical markers (disc reference points) on each side of the disc where the RPE/choriocapillaris/choroid reflection ends ( Fig. 216-10 ).
Although it is a very valuable tool for glaucoma assessment, the principal application of OCT in this disease prior to the new ONH analysis software has been in NFL thickness measurement
Figure 216-9 Optical coherence tomography principle. Axial profiles of backscattered light within the target tissue are measured by translation of the reference mirror, and the interferometric signal is recorded. Constructive interference creates an image that is seen by a photodetector only when the time paths to both the reference mirror and the tissue scanned are within the coherence length of the light source, which thus predicts a resolution of 10?µm in the eye. (Adapted with permission from Huang D, Swanson EA, Lin C, et al. Optical coherence tomography. Science. 1991;254:1178–81.)
Figure 216-10 Optical coherence tomography. Disc topography scan pattern for ONH analysis. A, Normal—note the thicker nerve fiber layer superiorly and inferiorly. Borders of the retina and nerve fiber layer are designed by a computer algorithm that calculates threshold reflectivity. Retinal thickness equals the distance between the posterior (posteriormost blue line) and anterior retinal border (white line). Nerve fiber layer thickness is equal to the distance between the white line and anterior blue line. B, A patient who has advanced primary open-angle glaucoma. Note severe and generalized thinning of nerve fiber layer.
(see Chapter 217 ). It has tremendous utility in the diagnosis and monitoring of retinal, particularly macular, pathologies (see Section 8 ). ONH analysis using the OCT is being investigated more extensively. Preliminary results suggest that OCT ONH assessment correlates highly with HRT ONH measurements, and both perform similarly in discriminating between normal and glaucomatous eyes. Recently, investigators recruited 236 eyes to evaluate the relationship between OCT-generated ONH parameters using OCT 2 and 3 and HRT-measured parameters to compare the association between the two technologies with glaucoma status as determined by perimetry and clinical evaluation. HRT and OCT 2 and 3 ONH measurements were found to have a statistically significantly high correlation with each other and with disease status.
The technologies described here represent the efforts of many investigators to quantify and objectively analyze ONH structure to differentiate between normal and glaucomatous eyes as soon as possible in the disease process and to detect the progression of the disease with the least possible additional neural damage. Although these devices offer some increase in the sensitivity to early or progressive damage, and measurement of ONH is a critical aspect of the care of patients who have glaucoma, many factors must be considered in the evaluation of the glaucomatous eye. These techniques add to, but do not replace, a careful clinical examination, functional testing such as perimetry, and other measures. These instruments help to add objectivity and quantitative data for clinical analysis but do not provide information that is otherwise unobtainable (unlike technologies for NFL thickness measurement described in Chapter 217 ); if the choice is between a measurement of ONH carried out using any of the devices described here and the experienced clinician’s examination of the ONH and NFL, the clinician’s result must be selected.
1. Duke-Elder S. Anomalies of the intra-ocular pressure. In: Duke-Elder WS, ed. Textbook of ophthalmology, Vol 3. St Louis: Mosby; 1941:3280–429.
2. Hurtes R, ed. Evolution of ophthalmic photography. Boston: Little, Brown; 1976.
3. Jonas JB, Gusek GC, Naumann GO. Optic disc, cup and neuroretinal rim size, configuration and correlations in normal eyes. Invest Ophthalmol Vis Sci. 1988; 29:1151–8.
4. Greenfield DS, Zacharia P, Schuman JS. Comparison of Nidek 3Dx and Donaldson simultaneous stereoscopic disk photography. Am J Ophthalmol. 1993; 116:741–7.
5. Minckler D, Nichols T, Morales R. Preliminary clinical experience with the Nidek 3Dx camera and lenticular stereo disk images. J Glaucoma. 1992;1:184–6.
6. Jonas J, Konigsreuther KA. Optic disk appearance in ocular hypertensive eyes. Am J Ophthalmol. 1994;117:732–40.
7. Caprioli J, Prum B, Zeyen T. Comparison of methods to evaluate the optic nerve head and nerve fiber layer for glaucomatous change. Am J Ophthalmol. 1996; 121:659–67.
7A. Shuttleworth GN, Khong CH, Diamond JP. A new digital optic disc stereo camera: intraobserver and interobserver repeatability of optic disc measurements. Br J Ophthalmol. 2000;84:403–7.
7B. Correnti AJ, Wollstein G, Price LL, Schuman JS. Comparison of Optic Nerve Head Assessment with a Digital Stereoscopic Camera (Discam), Scanning Laser Ophthalmoscopy, and Stereography. Ophthalmology (in press).
8. Varma R, Spaeth GL. The PAR IS 2000: a new system for retinal digital image analysis. Ophthalmic Surg. 1988;19:183–92.
9. Varma R, Steinmann WC, Spaeth GL, Wilson RP. Variability in digital analysis of optic disc topography. Graefes Arch Clin Exp Ophthalmol. 1988;226:435–42.
10. Bishop KI, Werner EB, Krupin T, et al. Variability and reproducibility of optic disk topographic measurements with the Rodenstock Optic Nerve Head Analyzers. Am J Ophthalmol. 1988;106:696–702.
11. Holm O, Krakau C. A photographic method for measuring the volume of papillary excavations. Ann Ophthalmol. 1970;1:327–32.
12. Netland PA. The Glaucoma-Scope: principles, techniques, and applications. In: Schuman JS, ed. Imaging in glaucoma. Thorofare, NJ: Slack; 1997:17–32.
13. Pendergast S, Shields MB. Reproducibility of optic nerve head topographic measurements with the Glaucoma-Scope. J Glaucoma. 1995;4:170–6.
14. Gundersen KG, Heijl A, Bengtsson B. Sensitivity and specificity of structural optic disc parameters in chronic glaucoma. Acta Ophthalmol Scand. 1995;Separatum:1–6.
15. Schuman JS, Noecker RJ. Imaging of the optic nerve head and nerve fiber layer in glaucoma. Ophthalmol Clin North Am. 1995;8:259–79.
16. Echelman DA, Shields MB. Optic nerve imaging. In: Albert DM, Jakobiec FA, eds. Principles and practice in ophthalmology, Vol 3. Philadelphia: WB Saunders; 1994:1310–29.
17. Zangwill L, de Souza Lima M, Weinreb RN. Confocal scanning laser ophthalmoscopy to detect glaucomatous optic neuropathy. In Schuman JS, ed. Imaging in glaucoma. Thorofare, NJ: Slack; 1997.
18. Zangwill L, Horn SV, Lima MDS, et al. Optic nerve head topography in ocular hypertensive eyes using confocal scanning laser ophthalmoscopy. Am J Ophthalmol. 1996;122:520–5.
19. Zangwill L, Shakiba S, Caprioli J, Weinreb RN. Agreement between clinicians and a confocal scanning laser ophthalmoscope in estimating cup/disk ratios. Am J Ophthalmol. 1994;119:415–21.
20. Weinreb RN, Luski M, Bartsch D-U, Morsman D. Effect of repetitive imaging on topographic measurements of the optic nerve head. Arch Ophthalmol. 1993;111: 636–8.
21. Mikelberg F, Wijsman K, Schulzer M. Reproducibility of topographic parameters obtained with the Heidelberg retina tomograph. J Glaucoma. 1993;2:101–3. Weinreb RN. Diagnosing and monitoring glaucoma with confocal scanning laser ophthalmoscopy. J Glaucoma. 1995;4:225–7.
22. Brigatti L, Caprioli J. Correlation of visual field with scanning confocal laser optic disc measurements in glaucoma. Arch Ophthalmol. 1995;113:1191–4.
23. Mikelberg FS, Prafitt CM, Swindale NV, et al. Ability of the Heidelberg retina tomograph to detect early glaucomatous visual field loss. J Glaucoma. 1995;4: 242–7.
24. 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.
25. Chauhan BC, Blanchard JW, Hamilton DC, LeBlanc RP. Technique for detecting serial topographic changes in the optic disc and peripapillary retina using scanning laser tomography. Invest Ophthalmol Vis Sci. 2001;41:775–82.
26. 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;119:1492–9.
27. Huang D, Swanson EA, Lin C, et al. Optical coherence tomography. Science. 1991;254:1178–81.
28. Schuman JS, Farra T, Mattox CG, et al. OCT optic nerve head assessment: comparison with HRT. Puerto Rico: American Glaucoma Society; March 2002.
29. 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(4):504–12.