Trick, Gary L. PhD; Bhatt, Sonali S. BS; Dahl, David MS; Skarf, Barry PhD, MD
Department of Eye Care Services, Henry Ford Health Sciences Center, Detroit, Michigan, and Chicago Medical School, Chicago, Illinois.
Manuscript received February 28, 2001; accepted September 4, 2001.
Address correspondence and reprint requests to Gary L. Trick, PhD, Department of Eye Care Services, Henry Ford, Health Sciences Center, 2799 W. Grand Blvd., Detroit, Michigan, 48202; E-mail: GLTrick1@aol.com
Grant support: none
Commercial relationship: none (N)
Papilledema is a diagnosis that describes true swelling (edema) of the optic disc resulting from elevated intracranial pressure (ICP). Pseudopapilledema refers to an optic nerve anomaly in which there is elevation of the optic disc surface and blurring of its margins that can resemble papilledema or the disc edema associated with other optic neuropathies. Unlike true edema of the optic disc, pseudopapilledema (PP) is considered benign and is frequently related to the presence of optic disc drusen, which are calcific bodies within the optic nerve head. Papilledema, on the other hand, can be caused by a space-occupying intracranial lesion such as a tumor or hemorrhage. Frequently, however, papilledema is associated with an idiopathic elevation of ICP, a condition known as pseudotumor cerebri or as benign intracranial hypertension. Since the optic disc can appear outwardly similar in papilledema and pseudopapilledema, careful evaluation of the disc morphology is often critical in diagnosing these conditions. Clinically, differentiating papilledema from PP can be a challenge for most physicians, neurologists, neurosurgeons, optometrists and for many ophthalmologists. When the diagnosis is uncertain, patients are referred for neuroimaging studies to rule intracranial pathology. More reliable and simple methods to distinguish PP from true disc swelling would help eliminate much uncertainty, unnecessary hardship, expense, and worry.
Confocal scanning laser (CSL) tomography provides a direct, noninvasive, quantitative evaluation of optic disc morphology. The accuracy, reliability, and reproducibility of this technique when used to measure optic disc excavation in glaucomatous optic neuropathy have been demonstrated (1–3). Recently, CSL tomography has also been shown to be useful as a means of measuring and monitoring the optic disc elevation due to papilledema in patients with pseudotumor cerebri (PTC) (4–6). CSL tomography also has been demonstrated to have a high sensitivity for detecting small changes in disc volume in PTC patients (5). Consequently, it has been suggested that CSL tomography might be useful for the follow-up of these patients (6). We are unaware of any published studies in which CSL tomography was used to systematically evaluate optic disc topography in PP. Therefore, this study was designed to determine if CSL tomography could be used to accurately quantify optic disc topography in PP and, if so, to determine whether the optic disc morphology of PP is quantitatively similar to or different from the topography of the edematous disc in PTC.
The Heidelberg Retinal Tomograph (HRT, Heidelberg Engineering, Heidelberg, Germany) was used to image the optic disc. The HRT is a confocal microscope that uses a 650-nm diode laser to scan the retinal surface in three dimensions. To generate three-dimensional topographic images, the HRT acquires a series of transverse optical sections taken at 32 consecutive equally spaced focal planes over a scan depth that ranges from .5 to 4.0 mm. Each image is generated from a 256 by 256 pixel matrix (65,536 pixels) in which each pixel represents retinal surface height at a specific location. The section images are automatically aligned for horizontal and vertical shifts due to any fixation instability during acquisition of the image. By combining the images in each series, the software generates a topographic map (also containing 256 by 256 pixels) in which each pixel has a value describing surface elevation at that point. These images are rapidly obtained. (Total acquisition time is approximately 1.6 seconds for the 32 images). The elevation measures are expressed relative to a reference plane that, in this study, was chosen to be the focal plane of the eye (see later discussion).
A 15° by 15° image, centered relative to the optic disc, was chosen for all images obtained in this study. Images were acquired using the standard HRT protocol (version 2.01) in which the elevation of the retinal surface is calculated relative to a reference plane placed 50 μm posterior to the mean retinal elevation along an arc concentric with the optic disc margin in the temporal segment of the optic disc (350° to 356°). Three topographic images were obtained through the undilated pupil for each eye and every image was corrected for tilt using a reference ring placed along the margin of the topographic image with an outer diameter of 94% of image size. A mean image, created by averaging three images from each eye, was used for the analysis.
The (pre-programmed) conventional HRT analysis measures two parameters that have been shown to be associated with the degree of optic disc swelling. These parameters are the volume above reference plane and the volume above surface. The volume above the reference plane is defined as the volume within the contour line that is above the reference plane. The HRT software permits the definition of reference planes parallel to the (x, y) plane of the coordinate system. A reference plane is specified by its absolute or relative z coordinate (height). In the version 2.01 software the reference plane is automatically defined at a location 50 μm posterior to the mean height of the retinal surface, that is at z = –50 μm in the relative coordinate system. Therefore, volume above the reference plane is a measure of the tissue volume inside the contour line that is anterior to this reference plane.
The volume above the surface is the volume within the contour line that is above the curved surface. The curved surface lies on the contour line at every boundary point, following the height variation of the retinal surface along the contour line. The height in its center is specified by the mean height of the retinal surface along the disc contour line; all connecting lines from the center point to a boundary point are straight lines. For purposes of this study the contour line that was drawn at 1800 μm from the center of the optic disc.
To further refine the analysis and evaluate the regional topography of the papilla and peripapillary retina, the HRT data were transformed as follows. With 0° defined as temporal, eight meridians were selected for analysis. As illustrated in Figure 1, the eight meridians ranged from 0° to 315° in 45° increments. Mean retinal elevation was then calculated for each of the 17 positions along each meridian, determined in 100 μm steps from 100 to 1700 μm from the center of the optic disc. In this procedure the center of the optic disc is defined as the center of gravity of all points along the contour line and automatically determined. At each of these positions the retinal surface height was calculated as the mean surface height along an arc extending ± 22.5° from each meridian on the circumference of a ring one pixel wide and having the same radius as the location. The mean retinal surface height for each location is then expressed relative to the mean surface height determined for a ± 5° arc centered at a radius of 1700 μm from the center of the optic disc along the 0° meridian (i.e., normalized relative to this reference value in a flat retina).
All study procedures were conducted in accordance with the tenets of the Declaration of Helsinki. Optic nerve head images were collected from 10 patients with clinically diagnosed pseudopapilledema (PP) ranging from 14 to 57 years of age (mean = 32.2 ± 14.0 years). The results for the PP patients were compared with data from 17 patients with clinically diagnosed pseudotumor cerebri who were 10 to 39 years of age (mean = 31.5 ± 7.0 years). The diagnosis of PTC was based on the observation of clinical papilledema on funduscopic examination in the presence of elevated CSF pressure and a negative CT scan of the head. Patients with pseudopapilledema were selected from among those referred with a diagnosis of possible papilledema and from among patients who were diagnosed with pseudopapilledema on routine examination, only when they met the following criteria: 1) abnormal elevation of the optic disc; 2) indistinct disc margin; 3) retinal vessels of normal caliber; 4) spontaneous venous pulsations present; 5) no obscuration of any vessels crossing the disc margin; 6) no retinal or disc hemorrhage or exudate; 7) appearance unchanged on six month follow-up exam.
Individuals with more than 6.0 diopters of either myopia or hyperopia or more than 1.0 diopter of astigmatic error were excluded. Individuals with a prior history of ophthalmic disease or any systemic disease with ocular manifestations were excluded.
For comparison with previously published reports (5–7), the mean “volume above the reference” and the mean “volume above the surface” were determined and for each of these parameters the difference between groups was assessed using a Student t test.
For the regional analysis statistical comparisons were based upon the normalized retinal elevations determined for each of the 17 positions along each meridian. A descriptive analysis of the variation in retinal elevation along each meridian was generated based upon the mean elevation at each distance from the center of the optic disc in both patient groups. The surface height values for each of the 17 locations along the eight meridians in the PP group were compared with those of the PTC group utilizing Student t test.
The “volume above the reference” was determined for all patients in each group (Fig. 2). The mean “volume above the reference” was lower for the PP patients (2.821 ± 1.215 mm3) than for the PTC patients (4.556 ± 2.379 mm3). The between-group difference was statistically significant:P = 0.0038. For each participant the “volume above the surface” also was determined (Fig. 3). The mean “volume above the surface” was less for the PP patients (1.730 ± 1.062 mm3) than for the PTC patients (2.793 ± 1.498 mm3) and this difference also was statistically significant (P = 0.0075)
In order to identify the local factors contributing to these global differences between groups, the mean surface height as a function of distance along each meridian was calculated for all eight meridians. In the PP patients, maximal surface height was greatest in the superonasal region where it exceeded 530 μm. Maximum surface height also exceeded 500 μm in the nasal (513 μm) and inferonasal (512 μm) meridians (Fig. 4). Maximum surface height was least in the temporal (154 μm) and inferotemporal (280 μm) meridians (Fig. 4). Other meridians were intermediate. Overall surface topography in the PP group was symmetrical along the vertical axis and asymmetric along the horizontal axis. Surface height in the nasal aspect peaked at 400 μm from the center of the optic disc while in the temporal aspect surface height peaked at 600 μm from the center of the optic disc. Along the vertical axis, surface height peaked at 600 μm from the center of the disk while inferiorly surface height peaked at 700 μm from the center. Along the oblique axes the symmetry was intermediate between the vertical and horizontal axes with peak surface height between 500 and 700 μm from the center of the optic disc.
A similar trend was observed in the PTC group with retinal surface height greatest in the superonasal and least in the temporal aspects of the disc (Fig. 5). Surface height was asymmetric along the horizontal axis with little elevation temporally compared with nasally. Surface height reached a maximum at 500 μm from the center of the disc nasally and at 700 μm temporally. Along the vertical axis, elevation was essentially symmetric, reaching a peak at 800 μm from the center of the disc both superiorly and inferiorly. As was the case for the PP patients symmetry along the oblique axes was intermediate between the vertical and horizontal axes (Fig. 5).
As suggested by the global analysis, mean surface height was generally greater in the PTC patients than in the PP patients (Fig. 6). For all eight meridians, the difference between the two groups was essentially zero in the center of the papilla and increased to reach a maximum at 900 to 1200 μm from the center of the optic disc. The difference in surface height decreased more peripherally. For each of the meridians, the surface height differences were generally symmetrical although there was a somewhat greater asymmetry in the vertical than in the horizontal axis. Along the horizontal axis the difference in retinal surface height between the two groups is symmetrical with the difference in surface height reaching 240 μm nasally and 237 μm temporally. In both cases, the maximum between group differences was found at 900 μm from the center of the disc. For the superior meridian, the between group difference in surface height difference peaked (270 μm) at 1200 μm from the center of the disc while inferiorly the greatest difference (222 μm) was detected at 1100 μm from the center of the disc (Fig. 7).
To describe the difference in surface height between the two groups more precisely, Student t tests were used to compare measurements for each of the locations along all eight meridians. No statistically significant differences were observed within a circle of radius 500 μm from the center of the papilla. The most significant between group differences were detected in the peripapillary retina: 1) Superiorly at 1300 μm (P = 0.000916) and 1400 μm (P = 0.000721) 2) Superonasally at 1300 μm (P = 0.000906) and 1400 μm (P = 0.000833) and 3) Inferotemporally at 1100 μm (P = 0.00094).
In pseudopapilledema there is an elevation of the optic disc that appears similar to the papilledema associated with pseudotumor cerebri. However, the optic disc elevation in pseudopapilledema is usually secondary to an underlying benign process such as optic disc drusen, whereas the papilledema in PTC is a direct result of elevated intracranial pressure. Given the distinct mechanisms mediating the optic disc changes in these two conditions, we speculated that careful quantitative evaluation of the optic disc morphology could reveal differences in disc topography reflecting the diverse pathophysiology. Reliable topographic differences between the two conditions could also be useful for differential diagnosis.
We found that SLO tomography could be used to quantify optic disc topography in both PP and PTC. Several previous reports have documented the utility of SLO tomography in PTC but we believe this is the first report on the use of SLO tomography to quantify optic disc topography in PP. We observed a significant anterior displacement of the papillary and peripapillary surfaces in PTC patients. The magnitude and position of this anterior displacement was similar to what has been reported previously. In particular, surface topography in the PTC group was symmetrical along the vertical axis and asymmetric along the horizontal axis. In the PTC group surface height peaked at between 500 and 800 μm from the center of the optic disc (depending upon the meridian) and was greatest in the superonasal, nasal, and superior regions and least in the temporal region.
In the PP patients the gross optic disc morphology appeared similar to the PTC patients. Surface height peaked at between 400 and 700 μm from the center of the optic disc depending upon the meridian and was greatest in the superonasal, nasal, and inferonasal regions and least in the temporal region.
Despite these similarities, quantitative analysis revealed distinct differences between the two groups. Significant differences in “volume above the reference” and “volume above the surface” were detected indicating that surface height was generally lower for PP patients than for PTC patients. However, local analysis revealed that the surface height differences were not uniform. Comparison of retinal elevation values for each location along all meridians demonstrated that PP and PTC have differences that are statistically significant particularly at the superior, superonasal, and infratemporal aspects of the disc. Furthermore, in the central 500 μm of the papilla no statistically significant differences between PP and PTC were detected, while near the neuroretinal rim and in the peripapillary retina PP patients exhibited significantly lower surface height than PTC patients in all meridians. This finding indicates that the disc swelling of papilledema extends well beyond the disc margin elevating peripapillary retina, as well as the optic disc. In contrast, the elevation due to PP is more closely confined to the papilla itself. Thus, the pathophysiologies underlying the two diseases appear distinct, producing a larger area as well as a greater magnitude of elevation in the PTC patients.
This study also demonstrates that measures of regional optic disc topography localize similarities and differences between PP and PTC that are not evident in the global measures of disc volume. This finding implies that quantitative regional topography may be more sensitive to subtle differences optic disc morphology than the conventional HRT analysis. However, further investigation is needed to determine if the HRT can be used to monitor the progression or remission of regional differences.
In conclusion, this study demonstrated that the HRT could quantify optic disc topography in pseudopapilledema as well as in pseudotumor cerebri. Therefore, the HRT can be used as a reliable method of measuring the optic disc elevation as well as excavation. Although the standard HRT procedures requires delineation of a contour line at the margin of the optic disk, the technique we present here allows a less subjective measure of the degree of elevation in the optic disc. The surface elevation measurements at various selected meridians allow for an accurate composite picture of the regional topography in both conditions. Recognition of regional differences in the degree of elevation may be beneficial to the clinician detecting an apparently swollen disc but unsure of whether it is the result of a benign process (such as pseudopapilledema) or an indication of elevated intracranial pressure.
1. Mikelberg FS, Wijsman K, Schulzer M. Reproducibility of topographic parameters obtained with the Heidelberg Retina Tomograph. J Glaucoma 1993; 2:101–3.
2. Rohrschneider K, Burk ROW, Kruse FE, Volcker HE. Reproducibility of the optic nerve head topography with a new laser tomographic scanning device. Ophthalmology 1994; 101:1044–9.
3. Weinreb RN, Lusky M, Bartsch DU, Morsman D. Effect of repetitive imaging on the topographic measurements of the optic nerve head. Arch Ophthalmol 1993; 111:636–8.
4. Trick GL, Vesti E, Tawansy K, Skarf B, Gartner J. Quantitative evaluation of papilledema in pseudotumor cerebri. Invest Ophthalmol Vis Sci 1998 Sep;39(10):1964–71.
5. Mulholland DA, Craig JJ, Rankin SJA. Use of scanning laser ophthalmoscopy to monitor papilloedema in idiopathic intracranial hypertension. Br J Ophthalmol 1998; 82 (11):1301–5.
6. Gobel W, Lieb WE, Grein HJ. Quantitative and objective follow-up of papilledema with the Heidelberg Retina Tomograph. [German]Ophthalmologe 1999 Sep; 94(9):673–7.
7. Janknecht P, Funk J, Hansen LL. Reproducibility of measurements with the Heidelberg retina tomograph in fundus elevation. Ophthalmologe 1995 Dec;92(6):862–5.
© 2001 Lippincott Williams & Wilkins, Inc.