Glaucoma is one of the leading causes of blindness. Chronic open-angle and angle closure glaucoma are progressive optic neuropathies leading to accelerated, progressive, and irreversible visual field damage, loss of retinal ganglion cells, and decrease of peripapillary retinal nerve fiber layer thickness (RNFLT). Clinical assessment of the severity and progression of glaucomatous damage has traditionally been made with automated threshold perimetry.1,2 In the recent years, imaging technologies [time-domain and Fourier-domain optical coherence tomography (OCT), Heidelberg retina tomography, and scanning laser polarimetry] gained a role in the detection of the structural damage, and in future they may potentially be used for quantitative measurement of disease progression.3–9 However, the relationship between functional (visual field) deterioration and structural damage has not yet been completely understood. Therefore, investigation of structure-function relationship remains in the focus of glaucoma research.3–22
Using the Humphrey Field Analyzer (Carl Zeiss Meditec Inc., Dublin, CA), it has been shown that the relationship between the corresponding structural and visual field parameters is curvilinear for populations comprising normal and glaucoma eyes, and different when a linear or curvilinear regression model is used.13–15 Relationship is also different when a logarithmic visual field parameter (eg, sensitivity expressed in dB) or an unlogged measure of luminance is used.15,16 It has also been shown that the relationship between retinal sensitivity and RNFLT is stronger with Fourier-domain OCT technology than scanning laser polarimetry,5 and similar to or stronger than that with time-domain OCT technology.4,11 No structure-function relationship has been established for healthy eyes.13
Surprisingly, research on RNFLT-based structure-function relationship has been on the basis of almost exclusively on visual field tests made with the Humphrey Field Analyzer system.3–6,7–23 Limited detailed research has been published on the Octopus perimeters (Haag-Streit AG, Koeniz-Berne, Switzerland), a different advanced visual field analyzing system that is used worldwide.1 Previously we investigated the correlation between scanning laser polarimetric quadrant RNFLT and mean sensitivity (MS) of the corresponding hemifield measured with the Octopus G2 program,24 but we did not evaluate structure-function relationship for narrow RNFLT sectors. Recently, 2 different groups studied structure-function relationship between the global indices of Octopus perimetry and the 360-degree RNFLT measured with the Topcon-100 OCT or the Heidelberg Retina Tomograph II, respectively, in different severity stages of glaucoma.7,25 However, both groups used the tendency-oriented perimetry (TOP) strategy of the Octopus system, which performs regional averaging and therefore makes it impossible to determine the exact relationship between measures smaller than global parameters.
Several core technical features of the Octopus perimeters1 and the Humphrey Field Analyzers2 are different. Interestingly, there were almost no investigations to find evidence which of these might be more advantageous for the detection and follow-up of glaucoma. Most importantly, in the Octopus system, the test points follow the nerve fiber bundle distribution and their density is higher around the macula,1,26 which may potentially represent an advantage for detailed structure-function research. In contrast, in the Humphrey system, the test points are equally distributed, with 6 degrees distance between the test points.2 In the Octopus 101 perimeter, the background is calibrated to a mesopic value (4 asb) and the 0 dB stimulus intensity is calibrated to 1000 asb luminance. In the Octopus G (General) programs, there are 59 test points in the central 30-degree field, whereas the Humphrey Field Analyzer has 54 test points (including 2 points for the blind spot) in the 24-2 program, which is most commonly used for evaluation of the central visual field. In addition, the stimulus duration in the Octopus 101 perimeter is 100 ms, but it is 200 ms in the Humphrey Field Analyzer. These differences limit application of structure-function research results gained with the Humphrey Field Analyzer system on the Octopus perimeters.
In the current investigation, our goal was to establish and compare the relationship between each of the 16 visual field clusters of the Octopus perimeter and the corresponding 16 RNFLT sectors determined with the RTVue-100 Fourier-domain OCT (RTVue OCT) and scanning laser polarimetry with variable corneal compensation (GDx-VCC) and enhanced corneal compensation (GDx-ECC) respectively.
Participants and Patient Groups
The research protocol was approved by the Institutional Review Board for Human Research of Semmelweis University, Budapest and all procedures adhered to the tenets of the Helsinki Declaration. Informed consent was obtained from all participants before enrollment. For inclusion, all participants had to have, in the study eye, refractive error within ±8.0 D, sufficient central vision for optimal fixation, image quality sufficient for optimal evaluation, and no macular or other pathology except for glaucoma on stereoscopic evaluation. The participants were whites, who were participating in a long-term imaging study in the Glaucoma Center of the Semmelweis University in Budapest. All participants had at least 3 years of experience of Octopus normal G2 visual field threshold testing and scanning laser polarimetry examination when data collection for the current study was started. Data collection was made prospectively for 1.5 to 3 years with 6-month intervals between June 2008 and June 2011. Altogether, data for 568 visits were analyzed. In all visits, the participants underwent determination of the best corrected visual acuity, Octopus G2 threshold perimetry of the central 30 degrees visual field using phases 1 and 2 of the normal strategy, Fourier-domain OCT using the Glaucoma Protocol of the RTvue-100 OCT, scanning laser polarimetry with GDx-VCC and GDx-ECC, and a detailed ophthalmological examination for glaucoma or progression of glaucoma. The healthy control subjects received no treatment. Patients with ocular hypertension and glaucoma were on intraocular pressure-lowering medication. No study eyes underwent cataract surgery or filtration surgery during the data collection period. All measurements were made by the same trained PhD student (A.G.).
The patient population comprised 15 healthy subjects with no optic nerve head damage, reliable and reproducible normal Octopus G2 visual field tests with mean defect (MD) not worse than 2 dB, and intraocular pressure consistently <21 mm Hg; 20 ocular hypertensive subjects with normal optic nerve head and visual field (with MD not worse than 2 dB) and untreated intraocular pressure consistently >21 mm Hg; 75 perimetric glaucoma patients (55 primary open-angle glaucoma, 7 juvenile open-angle glaucoma, 8 normal pressure glaucoma, 4 chronic angle closure glaucoma, and 1 pseudoexfoliative glaucoma patients) characterized by glaucomatous neuroretinal rim loss, and reliable and reproducible visual field defect typical for glaucoma (inferior and/or superior paracentral or arcuate scotomas, nasal step, hemifield defect or generalized depression with MD worse than 2 dB). The demographics of the participants are shown in Table 1.
Visual Field Testing and Determination of the Visual Field Clusters
The same Octopus 101 perimeter (Haag-Streit AG, formerly Interzeag AG, Schlieren, Switzerland) was used for all tests. The G 2 test of the central 30 degrees visual field (phases 1 and 2, which provide doubled threshold determination) with normal (bracketing) strategy were applied for all examinations. Current ametropia was corrected for according to the manufacturer’s recommendation. If necessary, artificial tear drops were given before the 5-minute dark adaptation period. The right eye was tested first in all cases. Only reproducible tests with <20% false-positive and 20% false-negative response rates were used for evaluation. For the current study, 16 visual field clusters were set (Fig. 1). Grouping of the test points into visual field clusters was determined on the basis of their spatial relationship to the optic nerve head and anatomy of nerve fiber bundles,1,26 and correction was made for the distance of the peripapillary RNFLT measuring ellipse from the edge of the optic nerve head. To ensure uniform handling of all data, the clusters of the left eyes were mirrored and numbered as those of right eyes. For structure-function investigations, MS values were calculated for each visual field cluster, in dB. In Octopus perimetry, pathologic MD is a positive value and the normal range for global MD is ±2.00 dB. Corrected loss variance (CLV) characterizes the focal defects.1 It is the square of the SD of the test point sensitivity values corrected for short-term fluctuation and is expressed in dB.2 In the current investigation, global MD and CLV were used to characterize the visual field damage for each patient group.
OCT was performed with the RTVue-100 OCT instrument (Optovue Inc., Fremont, CA) with software version 4.0. The working principle of the device has been described in detail elsewhere.6,12,16,27 In brief, the RTVue OCT uses a near-infrared light source centered at 840 nm, with a 50 nm bandwidth. For RNFLT measurements, the standard glaucoma protocol was used.27 This includes a 3D optic disc scan for the definition of the disc margin based on computer-assisted determination of retinal pigment epithelium endpoints, an optic nerve head scan to measure the optic disc parameters and RNFLT within an area of diameter 4 mm, centered on the predefined disc, and the standard ganglion cell complex scan. Each optic nerve head scan consists of 12 radial lines and 6 concentric rings, which are used to create an RNFLT map. The measuring circle (920 points) of 3.5 mm diameter is derived from this map after the sample circle is adjusted so as to be centered on the optic disc. The measured RNFLT is automatically given for the total circle, the superior and inferior sectors, and each of the sixteen 22.5 degree-sized sectors of the measuring circle (Fig. 1). The 16 sectors are numbered in sequence from the temporal side of the horizontal meridian (clockwise for the right eye, and anticlockwise for the left eye). In the current investigation, structure-function relationship was investigated for all 16 separate peripapillary RNFLT sectors. No pupil dilatation was made. Image quality was carefully checked after each image acquisition, and all images of insufficient quality or with any artifact were rejected and reacquired. Only images with signal strength index >40 were used. Optic nerve head was defined at the first visit. This definition was automatically applied for all later measurements using the instrument software.
Scanning Laser Polarimetry
Scanning laser polarimetry was performed using the GDx-VCC instrument (Carl Zeiss Meditec Inc.) with software version 5.5.1. This instrument can work either with VCC or (in a special research mode) with ECC. These variants of the noninvasive technique have been described in detail elsewhere.5,8,9,14,18,28 In brief, the instrument projects a beam of 780 nm polarized laser light onto the retina through the pupil. The birefringent ocular structures cause retardation in the polarized light passing through them, and this retardation is measured automatically. In the VCC mode, as a first step, the corneal polarimetric axis and magnitude values are calculated and then this information is used to correct the measured retardation around the optic nerve head. In the ECC mode, the compensator is adjusted so that it combines with the corneal retardation to produce a bias retardation of approximately 55 nm and a slow axis close to vertical. The software then measures a higher total retardation than the RNFL retardation alone, and the signal-to-noise ratio is improved as a result. The bias retardation and axis in each image are measured from the macular region and the actual RNFL retardation is derived mathematically. The actual bias is determined from each image and removed from the final RNFL image. Before each image acquisition, the current ametropia was corrected for, using the software. In the first visit, 1 corneal image per eye was obtained for all subsequent VCC and ECC measurements. All RNFLT measurements were made through undilated pupil, and were based on the same fixed-size measurement circle 3.2 mm in diameter centered on the operator-placed ellipse. Image quality was checked after each measurement, and images of inadequate quality (quality score <8) were reacquired. In both GDx-VCC and GDx-ECC, the measured RNFLT is automatically given for the total circle, the superior and inferior sectors, and 64 equally distributed plots along the 360 degrees measurement ellipse. In order to make the RTVue OCT sectors and the GDx sectors comparable, RNFLT values automatically determined for each set of 4 neighboring plots were averaged into 1 RNFLT sector. This provided the same sixteen 22.5 degree-sized RNFLT sectors of the measuring circle as the software of the RTVue OCT (Fig. 1). The resulting 16 RNFLT sectors were numbered in sequence from the temporal side of the horizontal meridian (clockwise for the right eye, and anticlockwise for the left eye). Structure-function relationship was investigated for the 16 peripapillary RNFLT sectors.
Structure-function relationship was analyzed between each of the 16 visual field cluster MS values and RNFLT of the corresponding 16 peripapillary sectors measured with RTVue OCT, GDx-VCC, and GDx-ECC, respectively. For each eye, the average of all values measured during follow-up was calculated for each parameter, and these individual mean values were applied for determination of structure-function relationship. The Stata 6.0 program package was used for statistical analysis. Linear (y=a+bx) and parabolic (y=a+bx+cx2) relationships, and first-order inverse [y=a+b×ln(x)] relationship (logarithmically transformed RNFLT) were investigated both with logarithmic (dB) and linear (1/Lambert) scales of cluster MS. The following formula was used to convert the individual dB values to 1/Lambert scale: dB=10×log10 (1/Lambert). Cluster MS was the dependent variable and RNFLT was the independent variable. The best fitting model was selected using either the F-test (nested models) or Akaike information criterion (AIC) and adjusted R2, as necessary. Strength of the relationship was characterized with coefficient of determination (R2). The Wilcoxon signed rank test of absolute residuals was used to compare the goodness of fit of the corresponding curves between the methods. P values of <0.01 were considered statistically significant.
On the basis of the results of the F-test, significantly better fit was found for parabolic than linear curves. The AIC values for parabolic fit were lower than those for first-order fit with logarithmically transformed RNFLT, and the adjusted R2 values were higher for logarithmic than linear (1/Lambert) cluster MS scale. Altogether, parabolic relationship with logarithmic cluster MS and linear sector RNFLT values provided the best fit. For the total study population (110 eyes), the relationship between the corresponding sector RNFLT and cluster MS values is shown in Figures 2 and 3 for RTVue OCT and GDx-ECC, respectively. The R2 values found for the corresponding visual field cluster MS and sector RNFLT values are shown in Table 2. For the RTVue OCT, structure-function relationship was significant (P<0.0001) for all sectors. The R2 values were highest for the temporal, superotemporal, and inferotemporal RNFLT sectors. For both GDx-VCC and GDx-ECC, the highest R2 values were seen for superotemporal RNFLT sectors. For both GDx methods, significant (P<0.01) parabolic relationship was seen for all but the temporal and nasal RNFLT sectors. The overall highest R2 value (0.6943) was found for the upper superotemporal RNFLT sector (ST 1 or 4) with GDx-ECC.
For several RNFLT sectors, the goodness of fit differed significantly across the 3 imaging methods (Table 2). Significantly better fit was seen for the RTVue OCT than either GDx method for 1 temporal lower RNFLT sector (TL 2 or 15). In contrast, significantly better fit was seen for 1 superotemporal RNFLT sector (ST 1 or 4) with both GDx methods than with RTVue OCT. For 2 RNFLT sectors (SN 2 or 6 and IN 1 or 12), the relationship with GDx-ECC was significantly stronger than that with GDx-VCC.
When the glaucoma eyes (n=75) were evaluated separately, significant (P<0.001) parabolic structure-function relationship was found for all RNFLT sectors with RTVue OCT, and at P<0.01 level for the superotemporal and superonasal (ST 2 or 3, ST 1 or 4, SN 1 or 5, and SN 2 or 6) and inferotemporal and inferonasal (IN 2 or 11, IN 1 or 12, IT 1 or 13, and IT 2 or 14) RNFLT sectors with GDx-VCC. For GDx-ECC, significant (P<0.01) parabolic structure-function relationship was found for all but the temporal (TU 1 or 1, TL 2 or 15, and TL 1 or 16) and nasal (NU 2 or 8 and NL 1 or 9) RNFLT sectors (detailed data are not shown). When the healthy eyes (n=15) were investigated separately, no significant R2 values were seen for any RNFLT sector with any imaging method.
In the current study, we investigated the structure-function relationship between MS of 16 Octopus visual field clusters and the corresponding sixteen 22.5-degree-sized peripapillary RNFLT sectors measured with RTVue OCT, GDx-VCC, and GDx-ECC, respectively. As far as we know, this is the first study in which structure-function relationship was investigated for the Octopus visual field clusters and narrow RNFLT sectors. Most of the previous research was based on the Humphrey Field Analyzer system (central 24 or 30 degrees visual field), which is essentially different compared with the Octopus G program used by us. Of the many differences, location of the test points seems to be the most important for structure-function investigations. In contrast to the Humphrey system, where the test points are equally distributed,2 in the G programs of Octopus perimetry, the positions of test points are based on the nerve fiber bundles and the distance from the optic nerve head.1 This makes it easier to group functionally related test points into clusters, and couple defined areas of the optic nerve head contour line with the corresponding visual field clusters.29–31 In the current study, we used the normal (bracketing) strategy for the 30 degrees G program. As RNFLT measurement is made along a measuring ellipse, which is wider than the optic nerve head contour line, correction was made on the clusters for the distance between the position of the ellipse and the edge of the disc, based on the peripapillary location, as shown in Figure 1. Previously, RNFLT-based structure-function relationship was analyzed only for global Octopus visual field parameters.7,24,25 In 2 of the previous 3 investigations, the TOP strategy was used. In the TOP strategy, sensitivity is tested only once for each test point and the final threshold is influenced by the results of the neighboring points.1 Although this approach saves time and is useful in clinical practice, it is not suitable to establish exact relationship between clusters and RNFLT sectors.
In the current investigation, cluster MS values were averaged for all visits, for each cluster. The use of averaged cluster MS values instead of individual measurement values certainly reduced the variability and helped to recognize tendencies. As using averages influenced all models equally, the results of the comparisons between the models were not influenced.
In our study, 3 methods (2 different working principles) were used to measure RNFLT for the same 16 peripapillary sectors. The RTVue OCT is a Fourier-domain OCT, which defines RNFLT based on the instrument’s segmentation algorithm,27 whereas scanning laser polarimetry (GDx-VCC and GDx-ECC) determines RNFLT on the basis of birefringence, which is related to the intracellular structure of the ganglion cell axons.8,9,32 As 29 of our 110 eyes had a typical scan score <80 with GDx-VCC (which represents an atypical retardation pattern), we also investigated the potential benefit of neutralization of this pattern with GDx-ECC on structure-function relationship.
We found similar structure-function relationship for the total study population and the glaucoma subgroup, both of which showed a wide range of global MD and RNFLT values. The best relationship between sector RNFLT and cluster MS was parabolic using logarithmic cluster MS and linear sector RNFLT values. The R2 values were highest for the temporal, superotemporal, and inferotemporal RNFLT sectors with RTVue OCT and the superotemporal RNFLT sectors with both GDx methods. Although the overall highest R2 value was found with GDx-ECC for the upper superotemporal RNFLT sector, the relationship was more general with RTVue OCT (highly significant for all sectors) than either GDx method (R2 was not significant for 2 temporal and 2 nasal RNFLT sectors with both GDx-VCC and GDx-ECC). The relatively poor performance of GDx-VCC and GDx-ECC on the temporal and nasal areas has previously been published,8,9,14,18 thus this result was not unexpected. This phenomenon is at least partly caused by the variation of nerve fiber layer-related birefringence around the optic nerve head, and the poorer signal-to-noise ratio in these areas.33 As a consequence, significantly better fit was seen for RTVue OCT than GDx-VCC or GDx-ECC for 1 inferotemporal RNFLT sector. Our results suggest that, of the structural methods tested in the current study, RTVue OCT is the most suitable for structure-function investigation with the Octopus visual field clusters.
For 2 of the 16 RNFLT sectors, the goodness of fit with GDx-ECC was significantly stronger than that with GDx-VCC. This finding supports the previously published results on the benefits of ECC over VCC for structure-function investigation.8,9 As in clinical practice, patient populations usually comprise eyes with atypical retardation pattern, use of GDx-ECC is generally recommended when scanning laser polarimetry is applied for evaluation of structure-function relationship.
When the healthy eyes were investigated separately, no significant structure-function relationship was found. This supports the findings of the previous structure-function studies made with the Humphrey Field Analyzer, in which no relationship was found for eyes with any visual field damage.3,4,6,13
It is difficult to compare our findings with previous results. As discussed above, no similar study using Octopus visual field clusters has been published. Thus, due to the population-related and disease severity-related and technical differences between the current and previous investigations, direct comparison of our R2 data to those reported by others is not possible. In addition, due to the test point distribution of the Humphrey Field Analyzer, in most studies ≤8 sectors were investigated for structure-function relationship.3–9,11–16,18–20,22,23,25 Despite these limitations, some tendencies seem to be similar to our results and the findings previously reported using the Humphrey system.3–5,8,9,13,14 Structure-function relationship between RNFLT and MS was curvilinear, Fourier-domain OCT performed better than GDx-VCC and ECC, and GDx-ECC performed better than GDx-VCC.
In conclusion, a significant parabolic relationship was found between the 16 Octopus visual field cluster MS values, determined with normal test strategy and the G program, and the corresponding 16 RNFLT sectors measured with RTVue OCT, for a population comprising normal, ocular hypertensive, and glaucoma eyes, and for the glaucoma subgroup. A similar but less general relationship was seen with GDx-VCC and GDx-ECC, of which GDx-ECC showed a stronger relationship. Our results show that Octopus visual field clusters can be successfully applied for detailed structure-function research.
The authors thank Matthias Monhart for his help with the definition of the visual field clusters.
1. Weijland A, Fankhauser F, Bebie H, et al..Automated Perimetry.2004:5th ed.Schlieren:Haag Streit AG.
2. Heijl A, Patella VM.Essential perimetry.2002:3rd ed.Dublin:Carl Zeiss Meditec Inc..
3. Miglior S, Riva I, Guareschi M, et al..Retinal sensitivity and retinal nerve fiber layer thickness measured by optical coherence tomography in glaucoma.Am J Ophthalmol.2007;144:733–740.
4. Lee JR, Jeoung JW, Choi J, et al..Structure-function relationships in normal and glaucomatous eyes determined by time- and spectral-domain optical coherence tomography.Invest Ophthalmol Vis Sci.2010;51:6424–6430.
5. Aptel F, Sayous R, Fortoul V, et al..Structure-function relationships using spectral-domain optical coherence tomography: comparison with scanning laser polarimetry.Am J Ophthalmol.2010;150:825–833.
6. Rao HL, Zangwill LM, Weinreb RN, et al..Structure-function relationship in glaucoma using spectral-domain optical coherence tomography.Arch Ophthalmol.2011;129:864–871.
7. Gonzalez-Hernandez M, Pablo LE, Armas-Dominguez K, et al..Structure-function relationship depends on glaucoma severity.Br J Ophthalmol.2009;93:1195–1199.
8. Bowd C, Tavares IM, Medeiros FA, et al..Retinal nerve fiber layer thickness and visual sensitivity using scanning laser polarimetry with variable and enhanced corneal compensation.Ophthalmology.2007;114:1259–1265.
9. Mai TA, Reus NJ, Lemij G.Structure-function relationship is stronger with enhanced corneal compensation than with variable corneal compensation in scanning laser polarimetry.Invest Ophthalmol Vis Sci.2007;48:1651–1658.
10. Kanamori A, Naka M, Nagai-Kusuhara A, et al..Regional relationship between retinal nerve fiber layer thickness and corresponding visual field sensitivity in glaucomatous eyes.Arch Ophthalmol.2008;126:1500–1506.
11. Takagishi M, Hirooka K, Baba T, et al..Comparison of retinal nerve fiber layer thickness measurements using time domain and spectral domain optical coherence tomography, and visual field sensitivity.J Glaucoma.2011;20:383–387.
12. Kim NR, Lee ES, Seong GJ, et al..Structure-function relationship and diagnostic value of macular ganglion cell complex measurement using Fourier-domain OCT in glaucoma.Invest Ophthalmol Vis Sci.2010;51:4646–4651.
13. Ajtony C, Balla Z, Somoskeoy S, et al..Relationship between visual field sensitivity and retinal nerve fiber layer thickness as measured by optical coherence tomography.Invest Ophthalmol Vis Sci.2007;48:258–263.
14. Schlottmann PG, De Cilla S, Greenfield DS, et al..Relationship between visual field sensitivity and retinal nerve fiber layer thickness as measured by scanning laser polarimetry.Invest Ophthalmol Vis Sci.2004;45:1823–1829.
15. Leung CK-S, Chong KK-L, Chan W-M, et al..Comparative study of retinal nerve fiber layer measurement by stratus OCT and GDx VCC, II: structure/function regression analysis in glaucoma.Invest Ophthalmol Vis Sci.2005;46:3702–3711.
16. Cho JW, Sung KR, Lee S, et al..Relationship between visual field sensitivity and macular ganglion cell complex thickness as measured by spectral-domain optical coherence tomography.Invest Ophthalmol Vis Sci.2010;51:6401–6407.
17. Ferreras A, Pablo LE, Garway-Heath DF, et al..Mapping standard automated perimetry to the peripapillary retinal nerve fiber layer in glaucoma.Invest Ophthalmol Vis Sci.2008;49:3018–3025.
18. Bowd C, Zangwill LM, Medeiros FA, et al..Structure-function relationships using confocal scanning laser ophthalmoscopy, optical coherence tomography, and scanning laser polarimetry.Invest Ophthalmol Vis Sci.2006;47:2889–2895.
19. Lee P-J, Liu CJ-L, Wojciechowski R, et al..Structure-function correlations using scanning laser polarimetry in primary angle-closure glaucoma and primary open-angle glaucoma.Am J Ophthalmol.2010;149:817–825.
20. Leung CK-S, Cheung CY-L, Weinreb RN, et al..Retinal nerve fiber layer imaging with spectral-domain optical coherence tomography: a variability and diagnostic performance study.Ophthalmology.2009;116:1257–1263.
21. Lopez-Peña MJ, Ferreras A, Larrosa JM, et al..Relationship between standard automated perimetry and retinal nerve fiber layer parameters obtained with optical coherence tomography.J Glaucoma.2011;20:422–432.
22. Leite MT, Zangwill LM, Weinreb RN, et al..Structure-function relationships using the Cirrus spectral domain optical coherence tomography and standard automated perimetry.J Glaucoma
23. Garway-Heath DF, Poinoosawmy D, Fitzke FW, et al..Mapping the visual field to the optic disc in normal tension glaucoma eyes.Ophthalmology.2000;107:1809–1815.
24. Katsanos A, Kóthy P, Holló G.Comparison between fixed-angle and customized corneal-polarization compensation methods in scanning laser polarimetric measurement of the retinal nerve fiber layer in glaucoma.Eye.2005;19:152–158.
25. Cvenkel B, Kontestabile AŠ.Correlation between nerve fiber layer thickness measured with spectral domain OCT and visual field in patients with different stages of glaucoma.Graefes Arch Clin Exp Ophthalmol.2011;249:575–584.
26. Hogan MJ, Alvarado JA, Weddel JE.Histology of the Human Eye.1971.Philadelphia:W.B. Saunders;491–498526–538.
27. Garas A, Vargha P, Holló G.Reproducibility of retinal nerve fiber layer and macular thickness measurement with the RTVue-100 optical coherence tomography.Ophthalmology.2010;117:738–746.
28. Tóth M, Holló G.Enhanced corneal compensation for scanning laser polarimetry on eyes with atypical polarization pattern.Br J Ophthalmol.2005;89:1139–1142.
29. Weber J, Ulrich H.A perimetric nerve fiber bundle map.Int Ophthalmol.1991;15:193–200.
30. Weber J, Dannheim F, Dannheim D.The topographical relationship between optic disc and visual fields in glaucoma.Acta Ophthalmol.1990;68:568–574.
31. Lefrançois A, Valtot F, Barrault O.New diagnosis approaches: our experience with Octopus Field Analysis (OFA V2.2), the new software for analysis of visual field.J français d’ophtalmol.2009;32:3–14.
32. Garas A, Tóth M, Vargha P, et al..Comparison of repeatability of retinal nerve fiber layer thickness measurement made using the RTVue Fourier-domain optical coherence tomography and the GDx scanning laser polarimeter with variable or enhanced corneal compensation.J Glaucoma.2010;19:412–417.
33. Huang XR, Bagga H, Greenfield DS, et al..Variation of peripapillary nerve fiber layer birefringence in normal human subjects.Invest Ophthalmol Vis Sci.2004;45:3073–3080.
© 2014 by Lippincott Williams & Wilkins.