Glaucoma is the leading cause of irreversible blindness

Transcription

1 Glaucoma Characteristics of Optic Disc Morphology in Glaucoma Patients with Parafoveal Scotoma Compared to Peripheral Scotoma Kyoung In Jung, Hae-Young L. Park, and Chan Kee Park PURPOSE. We investigated the amount and location of optic nerve head (ONH) abnormal points and the retinal nerve fiber layer (RNFL) thickness with paracentral scotoma compared to peripheral scotoma. METHODS. Totals of 35 normal tension glaucoma (NTG) patients with isolated parafoveal scotoma (PFS) within a 108 radius in one hemifield, and 35 patients with isolated peripheral nasal step (PNS) within the nasal periphery outside 108 of fixation in one hemifield were enrolled if their mean deviation was greater than 10 decibels (db). Global and sector optic disc stereometric parameters obtained by Heidelberg retina tomography and analyzed by Moorfields regression analysis (MRA), and retinal RNFL thickness measured using Cirrus spectral domain-optical coherence tomography were compared between the two groups. The percentages of superior and inferior field defects were evaluated. RESULTS. In PFS, superior field defects (82.9%) were found to be dominant, whereas PNS showed a predominance of inferior field defects (80.0%). The PFS group revealed smaller rim area, more glaucomatous cup shape than the PNS group (P ¼ 0.036, 0.012, respectively). In MRA classification, the percentage outside of normal limits (ONL) was greater in the PFS group (P ¼ 0.006). Compared to the PNS group, the PFS group exhibited more glaucomatous ONH morphology in the temporal and inferotemporal sectors in a sector analysis of optic disc parameters, and had thinner RNFLs in the inferior quadrant, and at 7 and 8 o clock (P ¼ 0.007, 0.003, 0.005, respectively). CONCLUSIONS. In early NTG, paracentral scotoma may be more significant than peripheral scotoma because of narrower optic disc rim and larger cup, especially inferotemporally. (Invest Ophthalmol Vis Sci. 2012;53: ) DOI: / iovs Glaucoma is the leading cause of irreversible blindness worldwide. 1 A wide range of racial variation in the prevalence of normal tension glaucoma (NTG) has been From the Department of Ophthalmology and Visual Science, Seoul St. Mary s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea. Presented at a meeting of the European Association for Vision and Eye Research on October 6, Submitted for publication March 25, 2012; revised May 21, 2012; accepted June 9, Disclosure: K.I. Jung, None;H.-Y.L. Park, None;C.K. Park, None Corresponding author: Chan Kee Park, Department of Ophthalmology and Visual Science, Seoul St. Mary s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea, 505 Banpo-dong, Seocho-ku, Seoul , Korea: ckpark@catholic.ac.kr. reported. 2 An epidemiologic study in Korea revealed a high prevalence (3.5%) of primary open-angle glaucoma, with a high percentage (77%) of NTG. 3 Many studies have reported that visual field defects are more likely to be deeper, steeper, and closer to fixation in NTG compared to HTG, although no significant difference has been noted in the pattern of visual field defects in a number of reports. 4 9 The characteristics of visual field change in NTG have been reported as the occurrence of dense scotomas close to fixation and often at low levels of total loss, whereas in primary open-angle glaucoma (POAG), similar paracentral scotomas tend to occur in association with more advanced field loss. 6 Recently, it has been reported that NTG patients with parafoveal scotoma have different risk factor profiles; for example, systemic hypotension was higher in patients with parafoveal scotoma compared to those with nasal step. 10 Preserving central vision is of paramount importance in treating glaucoma. A parafoveal visual field defect is significant because patients with central visual disturbance are at risk for losing visual acuity, leading to decreased driving performance The central and paracentral retina is highly populated by ganglion cells. Area of peak ganglion cell density, which averages 35,000 cells/mm 2, are found approximately 1 mm from the foveal center, with a higher concentration of retinal ganglion cells in the central/paracentral retina compared to the periphery. 14 We assumed that glaucoma subjects with isolated paracentral visual field defects may have more structural difference apparent in the optic nerve head (ONH). Therefore, we compared the optic disc parameters measured using confocal scanning laser ophthalmoscopy and the retinal nerve fiber layer (RNFL) thickness profile determined by Cirrus spectral domain-optical coherence tomography (SD-OCT) between subjects with parafoveal scotoma (PFS) and peripheral nasal step (PNS) in early NTG. METHODS Our retrospective observational study protocol was approved by the Institutional Review Board of the Catholic University of Korea, Seoul, Korea. The study design followed the tenets of the Declaration of Helsinki for biomedical research. Eyes with NTG and meeting the eligibility criteria were enrolled consecutively from a database of patients who were examined for glaucoma between March 1, 2009 and December 31, 2009 in the Department of Ophthalmology, Seoul St. Mary s Hospital. Inclusion criteria for all patients were best-corrected visual acuity of 20/40 or better, refractive error of greater than 6 spherical diopters (D) and 2 D cylinder, and transparent ocular medium (nuclear color or opalescence, cortical or posterior subcapsular lens opacity <1) according to the Lens Opacities Classification System III. 15 The inclusion criteria for early NTG were eyes with a normal open angle on gonioscopy, initial Investigative Ophthalmology & Visual Science, July 2012, Vol. 53, No. 8 Copyright 2012 The Association for Research in Vision and Ophthalmology, Inc. 4813

2 4814 Jung et al. IOVS, July 2012, Vol. 53, No. 8 untreated intraocular pressure (IOP) of <21 mm Hg, and the presence of a glaucomatous optic neuropathy (diffuse or focal rim thinning, cupping, notching, hemorrhage, RNFL defect) with a corresponding reproducibly reliable visual field defect coincident with our criteria for PFS or PNS. Subjects with a history of previous intraocular surgery, diseases that may affect the peripapillary area, or systemic disease or medication that may affect visual acuity were excluded. Each patient had undergone comprehensive ophthalmic examinations, including a best corrected visual acuity test, slit-lamp biomicroscopy, IOP measurement by Goldmann applanation tonometry, gonioscopy, central corneal thickness (CCT) measurements using ultrasonic pachymetry (SP-3000; Tomey Corp., Nagoya, Japan), and dilated funduscopic examination. Stereoscopic optic disc photography and monoscopic red-free digital fundus photography (Canon Cf-60 UW with Canon EOS D-6 CCD camera; Canon, Tokyo, Japan) were performed. Patients were re-examined usually at 6- to 12-month intervals; the same tests were performed with the exception of CCT measurements. Data regarding age at the time of study entry, visual acuity, maximum known untreated IOP from records, spherical equivalent, and disc hemorrhage detection during follow-up were collected. Confocal Scanning Laser Ophthalmoscopy Topographic analysis of the ONH was performed by confocal scanning laser ophthalmoscopy using a Heidelberg Retina Tomograph II (HRT II; Heidelberg Engineering, Heidelberg, Germany). The spherical equivalent refractive error of each eye was adjusted in the dioptric ring of the HRT. The topographic images were obtained through dilated pupils and were analyzed using Advanced Glaucoma Analysis 3.0 software (HRT II; Heidelberg Engineering). All scans were required to have an intrascan SD of <30 lm. In cases of poor centration (more than onequarter of the disc outside the target circle), the scans were excluded from the study. The margin of the optic disc was traced manually by experienced users while viewing the stereo photographs under a stereoscopic viewer, and the inner edge of Elschnig s ring was defined with at least a four-point contour line. This was reviewed by a glaucoma specialist (CKP) who was blinded to patient identity and clinical history. The HRT II software displays several windows in which the topographic results are detailed. Eleven global optic disc stereometric parameters (disc area, cup area, rim area, cup/disc area ratio, rim/disc area ratio, cup volume, rim volume, mean cup depth, maximum cup depth, height variation contour, and cup shape measure) and six sector parameters were obtained with the HRT. Following the convention of measuring angles around the ONH, temporal was defined as 08, superior as 908, nasal as 1808, and inferior as 2708 (regardless of whether a left or right eye was analyzed). Therefore, the angles increased clockwise for right eyes and counterclockwise for left eyes. The ONH was then divided into six sectors: temporal (T) with angular extension from 458 to þ458, temporal-superior (TS) from þ458 to þ908, temporal-inferior (TI) from þ2708 to þ3158, nasal (N) from þ1358 to þ2258, nasal-superior (NS) from þ908 to þ1358, and nasal-inferior (NI) from þ2258 to þ The global and six sector (T, TS, TI, N, NS, and NI) ONH classifications also were obtained by Moorfields regression analysis (MRA). ONHs were classified as outside of normal limits (ONL) when a parameter was outside of the 99.9% confidence interval (CI), borderline when a parameter was between the 95% and 99.9% CIs, and within normal limits (WNL) when a parameter was within the 95% CI. These classifications were made for the entire optic disc considering the global rim area, and for each separate sector considering the rim area of the relative segment. Optical Coherence Tomography (OCT) Scanning The Cirrus SD-OCT (software version ; Carl Zeiss Meditec, Dublin, CA) uses the spectral domain technology of an optic disc cube obtained from a three-dimensional dataset composed of 200 A-scans from each of 200 B-scans. The software determines automatically the center of the disc and then extracts a circumpapillary circle (radius 1.73 mm) from the cube data set for RNFL thickness measurements. All subjects were measured through dilated pupils, using the Optic Disc Cube program (Carl Zeiss Meditec). Poor-quality scans were defined as those with a signal strength of <6, the presence of overt misalignment of the surface detection algorithm on 15% of consecutive A-scans or 20% of cumulative A-scans, or overt decentration of the measurement circle location as assessed subjectively. These were excluded. The RNFL thickness parameters evaluated in our study were temporal, superior, nasal, and inferior quadrant thicknesses, and thicknesses for all of the 12 clock-hour positions, with the 9 o clock (3 o clock for left eyes) position being temporal, 12 o clock position being superior, 3 o clock position being nasal, and 6 o clock position being inferior. Visual Field Testing Standard automated perimetry (SAP) was performed with a Humphrey field analyzer (Carl Zeiss Meditec), applying the Swedish interactive threshold algorithm (SITA) standard and program 24-2 test. Optical lens correction was placed in front of the tested eye, while the other eye was occluded with a patch. A reliable test was defined as <30% fixation losses, false-positives or false-negatives. When the test results did not meet the reliability criteria, the test was repeated. The visual field indices, expressed as mean deviation (MD) and pattern standard deviation (PSD), and the visual field index (VFI) from SAP were evaluated. All ophthalmic examinations, two consecutive perimetry tests performed on different days, and all topographic analyses were performed within 1 month from the subject s date of enrollment. Visual Field Criteria for PFS and PNS One experienced glaucoma specialist divided the patients into two groups based on pattern deviation probability plots. The PFS group included patients with isolated PFS within 12 points of a central 108 radius in one hemifield on the SITA 24-2 test (Fig. 1). The PNS group included those with isolated PNS within the nasal periphery outside 108 of fixation in one hemifield. Criteria for a PFS or PNS visual field defect were the presence of three or more points with a P value of <5%, one of which had a P value of <1%, among 12 points in each group on the pattern deviation plot. Patients with visual field defects in the central 108 and peripheral nasal visual fields, with visual field defects other than the central or nasal peripheral fields, or with visual field defects in the superior and inferior hemifields were excluded. When both eyes fulfilled the inclusion criteria, one eye per subject was chosen randomly for the study. Statistical Analysis Statistical analyses were performed with SPSS software (ver. 17.0; SPSS, Chicago, IL). Differences between the two groups were assessed by Student s t-test for continuous parameters and the v 2 test for numeric parameters. Each group was divided into superior and inferior field defect groups, and the percentage of these groups were compared between the PFS and PNS groups by the v 2 test. Global and sector optic disc stereometric parameters were compared between the two groups by Student s t-test. Global and sector MRA classification was compared between the two groups by the v 2 test. We considered this study as an exploratory analysis. Therefore, we did not adjust P value to control overall significance level to 5% in all statistical tests. RESULTS There was little difference in age, sex, IOP, visual acuity, or disc hemorrhage between the two groups (P ¼ 0.497, 0.236, 0.187,

3 IOVS, July 2012, Vol. 53, No. 8 Optic Disc Morphology in Parafoveal Scotoma 4815 Comparison of Topographic Optic Disc Parameters and MRA Classification between the PFS and PNS Groups Rim area measured using HRT was smaller in the PFS group than in the PNS group (P ¼ 0.036). Compared to the PNS group, the PFS group displayed more positive values for cup shape measurements (P ¼ 0.012, Table 2). There was a tendency that the cup/disc area ratio and cup volume was higher, and rim/disc area ratio was lower in the PFS group compared to the PNS group (P ¼ 0.052, 0.052, and 0.079, respectively). In the sector analysis of optic disc parameters, the cup area, rim area, cup/disc area ratio, rim/disc area ratio, cup volume, rim volume, and height variation contour revealed more glaucomatous ONH morphology in the temporal and inferotemporal quadrants in the PFS group compared to the PNS group, but there were no great differences in the other quadrants (Table 3). In terms of MRA classification, the percentage ONL was greater in the PFS group (85.7%) than in the PNS group (51.4%, P ¼ 0.006, Table 4). The PFS group also showed a greater percentage ONL (74.3%) compared to the PNS group (34.3%, P ¼ 0.003) in the temporal-inferior MRA classification. FIGURE 1. Pattern deviation plot divided into two subfields of the Humphrey visual field. The parafoveal scotoma group includes abnormal points within 12 points of a central 108 radius (dashed-line). The peripheral nasal step group has abnormal points within 12 nasal peripheral points (dotted-line) in one hemifield , and 0.617, respectively; Table 1). Spherical equivalent tended to be more myopic in the PFS group ( D) than in the PNS group ( D, P ¼ 0.035). Comparison of Visual Field Features between the PFS and PNS Groups Both groups had similar MD and PSD (MD P ¼ and PSD P ¼ 0.248, Table 1). VFI was 89.47% in the PFS group and 93.2% in the PNS group. Superior field defects were dominant (82.9%) in the PFS group, whereas inferior field defects were dominant (80.0%) in the PNS group (Fig. 2). Comparison of RNFL Thickness between the PFS and PNS Groups The RNFL thickness measurements are shown in Table 5 and Figure 3. The 3608 average RNFL showed no great difference between the PFS ( lm) and the PNS ( lm, P ¼ 0.625) groups. The inferior quadrant RNFL was thinner in the PFS group (P ¼ 0.007), while the superior quadrant RNFL was thinner in the PNS group (P ¼ 0.006). The RNFL thicknesses in the 7 and 8 o clock sectors were less in the PFS group (P ¼ and 0.005, respectively), and the RNFL thickness in the 11 o clock sector was less in the PNS group (P ¼ 0.008). DISCUSSION In our study, paracentral scotomas were associated with marked ONH abnormality in locations important for reading and seeing detail in glaucomatous optic neuropathy. A VFI that places greater weight on the location in the paracentral visual TABLE 1. Characteristics of Patients with Parafoveal Scotoma (PFS) and Peripheral Nasal Step (PNS) Parameter PFS Group (N ¼ 35) PNS Group (N ¼ 35) P Value* Age (y) Male/Female 14/21 18/ Right/left laterality 15/20 14/ Maximum untreated IOP (mm Hg) CCT (lm) Visual acuity (logmar) Spherical equivalent (D) Mean deviation (db) PSD (db) Visual field index 89.47% 93.26% Disc hemorrhage 7 (20.0) 7 (20.0) Follow-up period (mo) Interval between visits (mo) Continuous variables are expressed as N (percentage), mean 6 SD, or percentage. * Statistically significant differences between the PFS and PNS groups (P < 0.05) by Student s t-test for continuous variables or v 2 test for categorical data are indicated in bold.

4 4816 Jung et al. IOVS, July 2012, Vol. 53, No. 8 FIGURE 2. The parafoveal scotoma group showed a predominance of superior field defects (82.9%), whereas the peripheral nasal step group had a higher percentage of inferior field defects (80.0%). TABLE 2. Heidelberg Retina Tomography Parameters of Global and Sector (T, N) Parameters in the PFS and PNS Groups Parameter PFS Group (N ¼ 35) PNS Group (N ¼ 35) P Value* Global Disc area (mm 2 ) Cup area (mm 2 ) Rim area (mm 2 ) Cup/disc area ratio Rim/disc area ratio Cup volume (mm 3 ) Rim volume (mm 3 ) Mean cup depth (mm) Maximum cup depth (mm) Height variation contour Cup shape measure Temporal Disc area (mm 2 ) Cup area (mm 2 ) Rim area (mm 2 ) Cup/disc area ratio Rim/disc area ratio Cup volume (mm 3 ) Rim volume (mm 3 ) Mean cup depth (mm) Maximum cup depth (mm) Height variation contour Cup shape measure Nasal Disc area (mm 2 ) Cup area (mm 2 ) Rim area (mm 2 ) Cup/disc area ratio Rim/disc area ratio Cup volume (mm 3 ) Rim volume (mm 3 ) Mean cup depth (mm) Maximum cup depth (mm) Height variation contour Cup shape measure Continuous variables are expressed as mean 6 SD. * Statistically significant differences between the PFS and PNS groups (P < 0.05) by Student s t-test are indicated in bold.

5 IOVS, July 2012, Vol. 53, No. 8 Optic Disc Morphology in Parafoveal Scotoma 4817 TABLE 3. Heidelberg Retina Tomography Optic Disc Sector Stereometric Parameters (TS, TI, NS, NI sectors) in the PFS and PNS Groups Sector Optic Disc Stereometric Parameters PFS Group (N ¼ 35) PNS Group (N ¼ 35) P Value* Sector Optic Disc Stereometric Parameters PFS Group (N ¼ 35) PNS Group (N ¼ 35) P Value* Temporalsuperior sector Temporalinferior sector Disc area (mm 2 ) Nasalsuperior Disc area (mm 2 ) Cup area (mm 2 ) Cup area (mm 2 ) Rim area (mm 2 ) sector Rim area (mm 2 ) Cup/disc area ratio Cup/disc area ratio Rim/disc area ratio Rim/disc area ratio Cup volume (mm 3 ) Cup volume (mm 3 ) Rim volume (mm 3 ) Rim volume (mm 3 ) Mean cup depth (mm) Mean cup depth (mm) Maximum cup depth (mm) Maximum cup depth (mm) Height variation contour Height variation contour Cup shape measure Cup shape measure Disc area (mm 2 ) Nasalinferior Disc area (mm 2 ) Cup area (mm 2 ) Cup area (mm 2 ) Rim area (mm 2 ) sector Rim area (mm 2 ) Cup/disc area ratio Cup/disc area ratio Rim/disc area ratio Rim/disc area ratio Cup volume (mm 3 ) Cup volume (mm 3 ) Rim volume (mm 3 ) Rim volume (mm 3 ) Mean cup depth (mm) Mean cup depth (mm) Maximum cup depth (mm) Maximum cup depth (mm) Height variation contour Height variation contour Cup shape measure Cup shape measure * Statistically significant differences between the PFS and PNS groups (P < 0.05) by Student s t-test are indicated in bold.

6 4818 Jung et al. IOVS, July 2012, Vol. 53, No. 8 TABLE 4. MRA of HRT: Global and Sector Analysis in the PFS and PNS Groups Sector Analysis MRA Classification PFS Group (N ¼ 35) PNS Group (N ¼ 35) P Value* Global WNL 3 (8.6%) 6 (17.1%) Borderline 2 (5.7%) 11 (31.4%) ONL 30 (85.7%) 18 (51.4%) Temporal WNL 18 (51.4%) 22 (62.9%) Borderline 7 (20.0%) 8 (22.9%) ONL 10 (28.6%) 5 (14.3%) Temporal-superior WNL 15 (42.9%) 18 (51.4%) Borderline 8 (22.9%) 6 (17.1%) ONL 12 (34.3%) 11 (31.4%) Temporal-inferior WNL 7 (20.0%) 16 (45.7%) Borderline 2 (5.7%) 7 (20.0%) ONL 26 (74.3%) 12 (34.3%) Nasal WNL 15 (42.9%) 22 (62.9%) Borderline 6 (17.1%) 7 (20.0%) ONL 14 (40.0%) 6 (17.1%) Nasal-superior WNL 16 (45.7%) 18 (51.4%) Borderline 7 (20.0%) 11 (31.4%) ONL 12 (34.3%) 6 (17.1%) Nasal-inferior WNL 13 (37.1%) 18 (51.4%) Borderline 8 (22.9%) 7 (20.0%) ONL 14 (40.0%) 10 (28.6%) * Statistically significant differences between the PFS and PNS groups (P < 0.05) by the v 2 test are indicated in bold. field arguably gives a more appropriate indication of small, but clinically important, losses in this area. The VFI clearly was lower in the PFS group than in the PNS group (P ¼ 0.008). Therefore, the PFS is located critically, although a paracentral scotoma does not necessarily mean a loss of functional vision. The most important finding of our study was the narrower optic disc rim and the larger cup, on a quantitative basis, in eyes with paracentral scotoma compared to peripheral scotoma. Before initiating the study, we predicted that given the large number of ganglion cells in the central and paracentral retina, subjects with PFS would have more glaucomatous optic disc morphology compared to subjects with PNS. 14 We observed the difference in global and sector optic disc parameters, and local RNFL thickness between eyes with PFS and eyes with PNS. The narrower neural rim and larger cup of the ONH in the PFS group compared to the PNS group also may be explained by the ratio of human midget to parasol ganglion cells, which changes from approximately 30:1 in the central retina to approximately 3:1 in the retinal periphery. 16 Midget ganglion cells, which are predominant in the central retina, generally are known to have very small receptive fields. 17 Even though one report suggests that the receptive field size shows no big difference between midget and parasol ganglion cells, receptive field radii become larger as the degree of eccentricity from fixation increases in small and large ganglion cells. 18 Therefore, glaucoma with PFS may involve a greater number of damaged ganglion cells than with PNS, at a similar level of visual field defect. The PFS group showed no big difference in average peripapillary RNFL thickness compared to the PNS group (P ¼ 0.625). We speculate that this may be related to the inconsistency of the mean axon diameter along the intraretinal axon length. Ganglion cell axon diameters increase by 20 40% over the first lm from the soma and then decrease with increasing distance from the soma. 19 Furthermore, the mean axon diameter at a location 50 lm from the soma was smaller in midget ganglion cells than in parasol ganglion cells, and it was similar between parasol and midget ganglion cells, within lm, closer to the optic disc. 19 Therefore, a marked loss of ganglion cells in the optic disc with PFS may not affect the RNFL thickness prominently because of the smaller mean axon diameter. Histologic studies investigating the distribution and function of ganglion cells in the human retina are required to determine the reasons for the structural differences in optic discs between the PFS and PNS groups. The PFS group revealed a predominance of superior field defects, which is in agreement with previous study results, and the frequent involvement of the superior hemifield our patients with PFS is consistent with previous reports. In one study, 10 of 11 eyes with arcuate glaucomatous visual field defects confined largely to the macula showed superior hemifield defects. 10 Another study found initial PFS predominantly in the superior hemifield. 10,20 However, to our TABLE 5. RNFL Thickness Measured Using HD-OCT in the PFS Group and PNS Groups Parameter PFS Group (N ¼ 35) PNS Group (N ¼ 35) P Value* Average thickness Superior thickness Nasal thickness Inferior thickness Temporal thickness Values are expressed in lm as mean 6 SD. * Statistically significant differences between the PFS and PNS groups (P < 0.05) by Student s t-test are indicated in bold.

7 IOVS, July 2012, Vol. 53, No. 8 Optic Disc Morphology in Parafoveal Scotoma 4819 FIGURE 3. The global and six sector (T, TS, TI, N, NS, and NI) classifications of the optic nerve head by MRA. Left: the eye with superior parafoveal scotoma was classified as outside normal limits by MRA. Sector analysis showed the TI sector to be outside normal limits and the T sector to be borderline. Right: the eye with inferior peripheral nasal step was classified as within normal limits by global and sector analyses of MRA. knowledge no studies have investigated the structural characteristics of ONH or the RNFL thickness to explain the superior dominance of visual field defects in PFS. Intriguingly, compared to the PNS group in our study, the PFS group showed generally smaller rim area and more glaucomatous cup shape in ONH parameters, as well as more glaucomatous ONH morphology in the TI sector, corresponding to superior hemifield defects. In addition, the PFS group had a thinner RNFL in the temporal sector, and in the 7 and 8 o clock sectors, compared to the PNS group. This may explain, at least in part, the predominance of superior visual field defects in the PFS group. Global and TI sector MRA classifications indicated more abnormal optic disc shape in the PFS group compared to the PNS group, with a comparable MD and higher PSD. Figure 4 illustrates representative cases: an eye (MD 1.68 decibels [db], PSD 2.05 db) with PFS classified as ONL by MRA, and another eye (MD 2.21 db, PSD 7.12 db) with PNS classified as WNL in global and sector MRA. The optic disc was more analogous to normal ONH form in eyes with PNS compared to PFS; thus, the MRA algorithm did not reveal optic disc abnormality in early NTG patients with PNS. In our study, we excluded patients with highly myopic eyes, as this condition frequently is associated with a tilted disc and peripapillary atrophy, which could have confounding effects on the optic disc analysis. Our result of a more myopic tendency in PFS differs from a previous finding of no significant difference in refractive error between PFS and PNS. 10 However, that study included highly myopic eyes, and the PNS group comprised only 15 patients. Limitations of our current study include its retrospective design and the relatively small number of patients. Nevertheless, we tried to elucidate the characteristics of glaucoma with PFS because to our knowledge there have been no reports comparing structural difference of ONH between PFS and PNS. Parafoveal visual field defects are important in the aspect of their location functionally because they have a greater influence on visual acuity and reading compared to peripheral visual field defects Therefore, fundamental information about the structural features of the ONH and RNFL in PFS is valuable. We tried to make as thorough an investigation of the diagnosis of glaucoma in the PFS group as possible to exclude other optic neuropathy or retinal disease that could cause central scotoma. To avoid the diagnostic bias in patients with FIGURE 4. Comparison of the RNFL thickness between the parafoveal scotoma and peripheral nasal step groups. The thicknesses at the 7, 8, and 11 o clock positions differed between the two groups (P < 0.05, by Student s t-test).

8 4820 Jung et al. IOVS, July 2012, Vol. 53, No. 8 parafoveal scotoma, we confirmed that there was no eye with decreased visual acuity and retinal diseases through fundus examination. In summary, patients with PFS, even in the early stage of NTG, appear to have more structural abnormality in the optic disc, especially in the TI sector, with thinner RNFL in the inferior quadrant. Therefore, the optic disc should be examined carefully in patients with PFS. Further studies will focus on the pattern of progression in the optic disc and visual field. References 1. Resnikoff S, Pascolini D, Etyáale D, et al. Global data on visual impairment in the year Bull World Health Organ. 2004; 82: Shields MB. Normal-tension glaucoma: is it different from primary open-angle glaucoma? Curr Opin Ophthalmol. 2008; 19: Kim CS, Seong GJ, Lee NH, Song KC. Prevalence of primary open-angle glaucoma in central South Korea the Namil study. Ophthalmology. 2011;118: Araie M. Pattern of visual field defects in normal-tension and high-tension glaucoma. Curr Opin Ophthalmol. 1995;6: Araie M, Yamagami J, Suziki Y. Visual field defects in normaltension and high-tension glaucoma. Ophthalmology. 1993; 100: Caprioli J, Spaeth GL. Comparison of visual field defects in the low-tension glaucomas with those in the high-tension glaucomas. Am J Ophthalmol. 1984;97: Chauhan BC, Drance SM, Douglas GR, Johnson CA. Visual field damage in normal-tension and high-tension glaucoma. Am J Ophthalmol. 1989;108: Hitchings RA, Anderton SA. A comparative study of visual field defects seen in patients with low-tension glaucoma and chronic simple glaucoma. Br J Ophthalmol. 1983;67: Thonginnetra O, Greenstein VC, Chu D, Liebmann JM, Ritch R, Hood DC. Normal versus high tension glaucoma: a comparison of functional and structural defects. J Glaucoma. 2010;19: Park SC, De Moraes CG, Teng CC, Tello C, Liebmann JM, Ritch R. Initial parafoveal versus peripheral scotomas in glaucoma: risk factors and visual field characteristics. Ophthalmology. 2011;118: Coeckelbergh TR, Brouwer WH, Cornelissen FW, Van Wolffelaar P, Kooijman AC. The effect of visual field defects on driving performance: a driving simulator study. Arch Ophthalmol. 2002;120: Fujita K, Yasuda N, Oda K, Yuzawa M. Reading performance in patients with central visual field disturbance due to glaucoma. Nippon Ganka Gakkai Zasshi. 2006;110: Kolker AE. Visual prognosis in advanced glaucoma: a comparison of medical and surgical therapy for retention of vision in 101 eyes with advanced glaucoma. Trans Am Ophthalmol Soc. 1977;75: Curcio CA, Allen KA. Topography of ganglion cells in human retina. J Comp Neurol. 1990;300: Iester M, Perdicchi A, Capris E, Siniscalco A, Calabria G, Recupero SM. Comparison between discriminant analysis models and glaucoma probability score for the detection of glaucomatous optic nerve head changes. J Glaucoma. 2008;17: Dacey DM, Petersen MR. Dendritic field size and morphology of midget and parasol ganglion cells of the human retina. Proc Natl Acad Sci U S A. 1992;89: De Monasterio FM, Gouras P. Functional properties of ganglion cells of the rhesus monkey retina. J Physiol. 1975;251: Lee BB. Paths to colour in the retina. Clin Exp Optom. 2004; 87: Walsh N, Fitzgibbon T, Ghosh KK. Intraretinal axon diameter: a single cell analysis in the marmoset (Callithrix jacchus). J Neurocytol. 1999;28: Hood DC, Raza AS, de Moraes CG, et al. Initial arcuate defects within the central 10 degrees in glaucoma. Invest Ophthalmol Vis Sci. 2011;52: