Copyright 2000 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2000
Glaucoma destroys the ganglion cell axons in the retina and may produce localized and/or diffuse damage of the retinal nerve fiber layer (RNFL). Unlike diffuse damage, localized defects are easy to detect. At the location of the RNFL defect the normal striated pattern of RNFL disappears and the localized defect is well outlined against the surrounding healthy nerve fiber bundles.1 Although single or multiple slitlike defects are probably not a sign of abnormality, they may sometimes be misinterpreted as a localized RNFL defect. This study demonstrates another RNFL finding to be differentiated from a true localized RNFL defect, ie, what we call a "pseudodefect" of the RNFL.
When routine RNFL photographs were examined, pseudodefects were detected in a total of 13 patients (Figure 1). Four patients were found during glaucoma screening and no information was available from the Finnish Glaucoma Patient Association. The remaining 9 patients (7 female and 2 male) had a mean age of 62 years (age range, 41-75 years). We also had the opportunity to measure 6 patients with RNFL pseudodefects using Optical Coherence Tomography (Zeiss-Humphrey, San Leandro, Calif).2 Several optical coherence tomographic scans were taken across the pseudodefect. Using a prototype software version, an estimate of the mean thickness of the RNFL at the site of the pseudodefect was compared with the mean thickness of the adjacent normal retina on both sides of the pseudodefect (Figure 2). The mean of 3 measurements was used for each location. The optic disc size of the 6 eyes was measured using a Heidelberg Retina Tomograph (Heidelberg Engineering, Heidelberg, Germany).3 Informed consent was obtained from all 6 patients who were examined.
Within 1 to 4 disc diameters from the optic disc margin the pseudodefects of the retinal nerve fiber layer look like oblong or braid-shaped chinks or cracks in the normal retinal nerve fiber layer structure (small arrows). Further in the periphery, the pseudodefect has a flat, beaten appearance (large arrow).
An Optical Coherence Tomographic (Zeiss-Humphrey, San Leandro, Calif) image of a retinal nerve fiber layer pseudodefect (arrow) adjacent to the blood vessel.
Pseudodefects were found in a total of 13 patients during routine RFNL photography performed because of suspected glaucoma, a positive family history of glaucoma (9 patients), or in screening for glaucoma by the Finnish Glaucoma Patient Association (4 patients). In 4 eyes found during glaucoma screening, all patient data were unknown, as only the screening photographs were available. Five of the remaining 9 patients were high myopes (spherical equivalent of −7.0, −9.25, −10.0, −10.0, and −16.5 diopters [D]), 2 were mild myopes (−1.25 and −2.25 D), and 2 patients were hyperopes (+6.5 and +2.0 D). Seven of 9 patients had ocular hypertension (range, 23-28 mm Hg). Except for the pseudodefects, the patients with ocular hypertension had otherwise normal RNFL, normal optic disc appearance, and normal Humphrey 30-2 visual field results. Five of these 7 patients also had follow-up ranging from 4 to 10 years. One of these 5 patients developed glaucoma during the follow-up period that subsequently resulted in enlargement of optic disc cupping, diffuse thinning of the RNFL, and a glaucomatous visual field defect. In this progressive case, the shape of the pseudodefect did not change; however, it became less easily visible with overall thinning of the RNFL.
The pseudodefects were found along the major retinal blood vessels. Within 1 to 4 disc diameters from the optic disc margin, they looked like oblong or braid-shaped chinks or cracks within the normal RNFL structure (Figure 1). In 3 eyes these pseudodefects could be followed far into the periphery (up to 4 disc diameters temporal to the macula) where the pseudodefect no longer looked like a chink but had a flat, beaten appearance. In 1 patient the pseudodefect was seen in the inferior nasal retina; all others were located temporal to the optic disc and/or the macula. In 2 patients pseudodefects were found in both eyes.
In 5 of 6 patients the optical coherence tomographic scan showed a thinning of the RNFL at the pseudodefect. The mean ± SD thickness of the RNFL was 132 ± 49 µm at the pseudodefect, 220 ± 49 µm on the temporal side of the pseudodefect, and 203 ± 70 µm on the nasal side of pseudodefect. The differences of the retinal thickness values at the pseudodefects compared with the average retinal thickness values on either side of the pseudodefect were 212, 136, 111, 38, 32, and −53 µm (mean difference, 79 µm). The respective differences were 70%, 56%, 49%, 22%, 21%, and −30% (mean difference, 37%). The mean ± SD optic disc area of the 6 eyes using the Heidelberg Retina Tomograph measurements was 1.40 ± 0.2 mm2 (reference range, 1.1-1.6 mm2), which is smaller than the mean of our patients with glaucoma (2.1 mm2).4
We found 1 previous report on pseudodefects in the literature. In 1992 Chihara and Chihara5 described 3 patients with what they called a cleavage of the RNFL. They reported that the cleavage was associated with high myopia (−14, −14, and −15 D) and postulated that the number of axons might be insufficient to cover the enlarged posterior segment of these myopic eyes. All of their patients were younger than 35 years. They did not find a cleavage in any of 144 emmetropic or hyperopic eyes. The cleavage was unassociated with any visual dysfunction measured with automated perimetry.
In our larger number of patients we found several differences and new findings compared with the previously published Japanese cases, ie, (1) although most frequent in high myopes, pseudodefects were found also in hyperopes and mild myopes, (2) our patients were all older than 40 years, (3) in 2 patients the changes were seen bilaterally, (4) the mean optic disc size of 1.4 mm2 of the eyes with pseudodefects was smaller compared with our previously reported mean disc size of 2.1 mm2,4 and (5) the mean decrease in retinal thickness was 79 µm (37%) at the pseudodefect, which is larger than the mean difference of 39 µm reported between glaucomatous and normal eyes.2
Chihara and Chihara5 examined the 3 eyes with pseudodefect using dark-field imaging of the scanning laser ophthalmoscopy and did not find any lacquer crack lesions, neovascularization, or clumping of the retinal pigment epithelium associated with the pseudodefect. Their results of fluorescein angiography confirmed the integrity of Bruch's membrane. They suggested that the pseudodefect may be associated with a breakdown of intraretinal neural connections and may later lead to an RNFL defect. In the 1 progressive case of our series the shape of the pseudodefect itself did not change; however, it became less easily visible with advancing glaucomatous damage.
It is probable that we need histopathological studies to learn more about the so far unknown etiology of the pseudodefects. However, it is important clinically to recognize the existence of pseudodefects and differentiate them from true RNFL defects in the diagnosis and follow-up of subjects with glaucoma.
The authors have no commercial or proprietary interest in the devices or companies mentioned in this article.
Presented in part at the 30th Turkish National Meeting of Ophthalmology, Antalya, Turkey, September 15, 1996, and the Association of Vision Research in Ophthalmology meeting, Fort Lauderdale, Fla, May 15, 1998.
Reprints: Anja Tuulonen, MD, PhD, Department of Ophthalmology, University of Oulu, FIN-90220 Oulu, Finland (e-mail: firstname.lastname@example.org).
Tuulonen A, Yalvac IS. Pseudodefects of the Retinal Nerve Fiber Layer Examined Using Optical Coherence Tomography. Arch Ophthalmol. 2000;118(4):572-575. doi: