January 2007
Volume 48, Issue 1
Free
Glaucoma  |   January 2007
Relationship between Visual Field Sensitivity and Retinal Nerve Fiber Layer Thickness as Measured by Optical Coherence Tomography
Author Affiliations
  • Csilla Ajtony
    From the Department of Ophthalmology and the
  • Zsolt Balla
    From the Department of Ophthalmology and the
  • Szabolcs Somoskeoy
    Center for Medical Information Technology, University of Pécs Medical and Health Sciences Center, Pécs, Hungary.
  • Balint Kovacs
    From the Department of Ophthalmology and the
Investigative Ophthalmology & Visual Science January 2007, Vol.48, 258-263. doi:10.1167/iovs.06-0410
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Csilla Ajtony, Zsolt Balla, Szabolcs Somoskeoy, Balint Kovacs; Relationship between Visual Field Sensitivity and Retinal Nerve Fiber Layer Thickness as Measured by Optical Coherence Tomography. Invest. Ophthalmol. Vis. Sci. 2007;48(1):258-263. doi: 10.1167/iovs.06-0410.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To evaluate the strength and pattern of the relationship between visual field (VF) sensitivity and retinal nerve fiber layer (RNFL) thickness as measured by StratusOCT (Carl Zeiss Meditec, Inc., Dublin, CA).

methods. Three hundred eleven subjects—45 normal, 102 with preperimetric glaucoma (PPG), and 164 with primary open-angle glaucoma (POAG)—were enrolled in this cross-sectional study. The relationship between RNFL thickness and VF sensitivity, expressed as mean deviation (MD) and mean sensitivity (MS), were evaluated with linear and nonlinear regression models, and the coefficient of determination (R 2) was calculated. The association between RNFL/VF was described by bivariate Pearson correlation coefficients.

results. The correlation of RNFL and the VF parameters MS and MD in normal and PPG eyes was not significant. In POAG eyes, RNFL and both MS (r = 0.733) and MD (r = 0.718) correlated significantly. Linear regression plots of MS or MD against RNFL thickness demonstrated a negligible degree of determination in normal (R 2 = 0.0378 and 0.0121, respectively) and PPG groups (R 2 = 0.0215 and 0.0151, respectively), whereas their relationship fit a curvilinear regression model (R 2 = 0.6947 and 0.723) in the POAG group. Receiver operating characteristic (ROC) curves describing the VF parameters and average RNFL thickness (AVG) were evaluated to differentiate PPG from POAG eyes. Repeated analysis with the best-performing test parameter, pattern standard deviation (PSD) (AUROC = 0.937) with a cutoff of 1.9 dB, showed that regression profiles in the POAG group with PSD >1.9 dB maintained a strong curvilinear RNFL/VF relationship, whereas those with PSD <1.9 dB exhibited a relationship almost indistinguishable from the PPG group.

conclusions. Evaluation of the structure–function relationship in normal subjects and those with PPG or POAG showed strong curvilinear regression in POAG eyes with PSD >1.9 dB and RNFL AVG thickness below 70 μm, whereas no correlation was detectable above these values.

Glaucoma is characterized by progressive degeneration of retinal ganglion cells and their axons that leads to nerve fiber layer loss, optic disc cupping, and consecutive glaucomatous visual field changes. 
Retinal nerve fiber layer (RNFL) loss is considered an early sign of glaucoma. 1 Because injury due to glaucoma is largely irreversible, early detection and prevention of glaucomatous damage is crucial. Examination of the optic nerve head (ONH) and peripapillary RNFL is essential in detecting and monitoring the disease. Recent advances in imaging technologies using the optical properties of the RNFL allow objective and quantitative assessment of the RNFL thickness. 2 3 4 5 Optical coherence tomography (OCT) is a noninvasive, noncontact technique for imaging the layered structure of the retina. 6 7 OCT computes RNFL thickness measurements from high-resolution cross-sectional images of the retina and is able to identify diffuse and focal RNFL defects reproducibly that occur in glaucoma 8 9 10 that have been shown to correlate quantitatively with visual field (VF) abnormalities. 11 12 13 14 Both macular and peripapillary NFL thicknesses, as measured by OCT, show statistically significant correlation with glaucoma; nevertheless, the peripapillary RNFL thickness turns out to be the best marker in glaucoma assessment. 15 16 17  
The purpose of this cross-sectional study was to evaluate the strength and pattern of the relationship between VF sensitivity and RNFL thickness as measured by the StratusOCT (Carl Zeiss Meditec, Inc., Dublin, CA), to find the most accurate VF test parameter in this association, and to define an approximate average peripapillary nerve fiber layer thickness where visual field abnormalities characteristic of glaucoma appear. 
Methods
We reviewed data of 266 patients who were referred to the Department of Ophthalmology, University of Pécs, Hungary, between November 2004 and September 2005. 
All subjects underwent full clinical ophthalmic evaluation, including medical, ocular, and family history; visual acuity testing; intraocular pressure measurement (IOP); slit lamp stereo biomicroscopy; and indirect ophthalmoscopy. 
We included all patients who fulfilled the following criteria: age older than 35 years; glaucomatous optic neuropathy (GON); open angles; good-quality scans obtained in peripapillary RNFL thickness evaluation by OCT, with a signal-to-noise ratio of >35; good-quality standard automated perimetry (SAP) performed at ±1 month from OCT imaging; and refractive error within a ±4-spherical diopter range, with less than ±2 cylinder diopters. 
Criteria for exclusion of a patient from the study were best corrected visual acuity on the Snellen-chart worse than 20/40, corneal or lens opacity that interfered with clinical evaluation of the optic disc and posterior pole through an undilated pupil, or significant parapapillary atrophy that caused blind spot enlargement on the visual field tests, interfered with VF readings, or causing false nerve fiber layer thickness data by OCT evaluation. Patients with tilted discs or any vitreal or retinal diseases were excluded, as well. IOPs were not considered for classification, because we focused on existing glaucomatous optic neuropathy and concomitant visual field defect, regardless of IOP value readings. 
The nature of these tests was explained to each subject, and verbal informed consent was received. All methods adhered to the Declaration of Helsinki for research involving human subjects. 
Patients had SITA Standard 30-2 perimetry (Humphrey Systems Model 750; Carl Zeiss Meditec, Inc.). A reliable VF test was defined as one with less than 20% fixation loss and false-positive and -negative responses. Glaucomatous VF defect was defined as two or more contiguous test locations demonstrating a threshold sensitivity loss on the pattern deviation plot with P < 0.01, three or more such contiguous test locations with P < 0.05 with at least one of the points depressed to P < 0.01, or a 10-dB difference across the nasal horizontal midline at two or more test locations. Data were analyzed by one grader (C.A.) who was masked from all other patient records. All patients had been subjected to automated perimetry before. For comparison, we used Humphrey global indices mean deviation (MD) and pattern SD (PSD), and we also calculated mean sensitivity (MS) by recording each point on the dB scale except the two points for the blind spot. 
OCT was performed with StratusOCT (Carl Zeiss Meditec, Inc.) with software version 3.0 without a built-in normative database. The RNFL thickness average analysis report, using the fast RNFL thickness scan with three sequential circular scans of 3.4 mm diameter over the optic disc, was performed. Before the scan the pupil size of each subject was determined under ambient conditions. Those below 4 mm were then dilated with 0.5% tropicamide. One of the authors (Z.B.) performed all image acquisitions. For statistical analysis, data of the left eye were converted into right-eye format. 
Glaucomatous optic neuropathy was defined as cupping, rim notching, or diffuse thinning of the optic disc. Grading of the optic disc was by clinical evaluation employing dilated stereoscopic ophthalmoscopy with a hand-held 78-D lens and by stereophotograph evaluation by two graders who were masked from all other patient data. If interobserver disagreement occurred, a consensus had to be achieved. 
Two hundred sixty-six subjects fulfilled our inclusion–exclusion criteria. One eye was randomly chosen if both eyes were eligible for the study. One hundred two of these were classified as preperimetric glaucomatous (PPG) and 164 with primary open-angle glaucoma (POAG) on the basis of SAP results. PPG was defined as eyes having glaucomatous optic neuropathy without any characteristic VF defects above. 18 For POAG classification, both GON and glaucomatous VF defect had to be present. 
Forty-five eyes of 45 normal subjects were consecutively enrolled. These patients were referred for disorders other than glaucoma, had no ophthalmic disease or conditions as defined by our exclusion criteria, and fulfilled all the inclusion criteria except for GON. Healthy eyes had a measured IOP of 21 mm Hg or less, and SAP results were without any glaucomatous VF defect, when graded as just described. One eye was randomly chosen if both eyes were eligible, otherwise the contralateral eye of patients having mild cataract (n = 2), posterior vitreous detachment (n = 1) or other nonglaucomatous eye disease (n = 1, asteroid hyalosis in one eye) was enrolled. 
Statistical analysis was performed on computer (SPSS ver 11.0 software; SPSS, Chicago, IL). Means comparison analysis of paired parameters between the groups was evaluated by one-way ANOVA including Levene’s homogeneity of variance test. Post hoc adjustment for multiple comparisons was performed by the Games-Howell method, if variances in groups were not equal, and by the Tukey honest significant difference (HSD) test, when equal variances were assumed. Association between RNFL thickness and visual field parameters was characterized by bivariate correlation analysis computing Pearson correlation coefficients. Data were reported as the mean ± SD. Pearson correlation coefficients with absolute values equal to or greater than 0.5 suggesting a strong association with P < 0.01 were accepted as statistically significant. Receiver operating characteristic (ROC) curves were used to describe the ability of each VF parameter and the average RNFL thickness to differentiate eyes with PPG from eyes with POAG. Sensitivity and specificity of each test parameter was determined by obtaining the highest sensitivities, with target specificity set at ≥90%. Regression analysis between RNFL and VF parameters was performed, employing linear and curvilinear (quadratic) models with corresponding scatterplots showing best-fit regression curves and regression coefficients. In all statistical analyses, P < 0.05 was considered statistically significant. 
Results
Two-hundred and sixty-six eyes of 266 patients with glaucoma and 45 healthy eyes fulfilled our inclusion and exclusion criteria. On the basis of SAP outcome, 102 eyes were clinically classified as having PPG and 164 eyes as having glaucoma. Baseline characteristics of the study population and the means comparison analysis, are summarized in Table 1 . Significant differences were detected in every test parameter in POAG versus normal and POAG versus PPG groups at P < 0.001. MD, PSD, and age were nonsignificant in PPG versus normal groups, but MS and AVG were significant. 
For purposes of comparison between groups, VF MS and RNFL AVG thickness were adjusted for age. Data for normalization were obtained from healthy eyes in our clinic (n = 45). Plotting RNFL thickness against age, we obtained a 0.36-μm decrease per year (r = −0.37, P < 0.05, R 2 = 0.136) and 0.0569-dB loss per year (r = −0.46, P < 0.01, R 2 = 0.215) for calculated VF sensitivity data (MS). 
RNFL AVG, measured by OCT, as an objective quantitative parameter, was first correlated with quantitative measurement of VF, by using calculated MS data. The same procedure was performed with the VF global indices MD and PSD. First, we calculated MS data in all groups, because we assumed that in contrast to MD, this is an individual actual parameter characteristic of the patient not attenuated by age normalization based on the built-in normative database, and it therefore could be directly compared to—and might show a more sensitive relationship with—RNFL AVG, which is also a directly measured individual actual parameter of the patient. Table 2summarizes the Pearson correlation coefficients between RNFL and all parameters of VF in normal and both POAG and PPG groups. The correlations between AVG and MS, MD, or PSD were all significant at the 0.01 level in the POAG group, with Pearson coefficients of 0.733, 0.718, and −0.689, respectively. In contrast, in eyes in the normal and PPG groups, no significant association was detected between AVG and either of the corresponding VF parameters. 
The relationship between RNFL thickness and VF parameters within groups were further characterized by regression analyses using linear and curvilinear (quadratic) models, the latter giving the better curve fit in the POAG group, with higher coefficients of determination (R 2). VF parameters were treated as dependent variables and RNFL thickness as the independent variable in all regression analyses. Scatterplots with regression curves of AVG versus MD in normal (R 2 = 0.0121), PPG (R 2 = 0.0151), and POAG (R 2 = 0.723) eyes are shown in Figure 1
Coefficient of determination in the same groups for the regression curves of AVG versus MS were 0.0378 in normal, 0.0215 in PPG, and 0.6947 in POAG eyes (data not shown). 
Scatterplots of both VF MD and VF MS against the RNFL indicate that there was an overlap in RNFL thickness distribution among the eyes in our groups. To assess a differentiation parameter other than dividing our patients with glaucoma on the basis of their SAP results as described previously, we generated ROC curves, and sensitivities at fixed specificity (≥90%) were calculated for each test parameter (VF and OCT). Results are shown in Figure 2 . The widest area under the receiver operating characteristic curve (AUROC) was obtained for PSD (0.937, P < 0.001; Table 3 ). 
The cutoff point for PSD was 1.9 dB, with 78% sensitivity and 90.2% specificity. Based on this finding, we subdivided both our clinically PPG and POAG groups by this PSD value. Descriptive statistics of resultant subgroups are detailed in Table 4 . Pearson correlation between AVG and MS, MD, or PSD in the POAG subgroup with PSD >1.9 dB (128 eyes) remained significant at P < 0.01 with values of 0.733, 0.727, and −0.674, respectively. 
Thirty-six eyes from the original 164 eyes of the POAG group with PSD <1.9 dB reverted in their Pearson correlation coefficients to values similar to those found for the PPG group (i.e., 0.165 for AVG/MS, 0.021 for AVG/MD, and −0.072 for AVG/PSD with none reaching statistical significance). Correlation data in the PPG subgroup with PSD >1.9 dB did not change (i.e., no significant correlation was found; Table 5 ). 
Scatterplots and regression analyses using the curvilinear model were repeated for groups of PPG and POAG eyes subdivided by the PSD cutoff value. As shown in Figure 3 , only the POAG group with PSD >1.9 dB exhibited a strong curvilinear relationship between RNFL AVG and VF parameters. POAG eyes with PSD <1.9 dB had a relationship almost indistinguishable from that of the PPG group shown earlier. 
Discussion
Assessing the amount of glaucomatous damage is the first step toward the correct management of glaucoma. The damage is usually estimated by observation of structures affected by glaucoma (RNFL and optic disc) and by testing visual function by perimetry. It is of great importance to know the way the damage to specific structures affects visual function. 
In this study, an objective quantitative measurement of RNFL thickness, as measured by OCT, was correlated with a quantitative measurement of VF as expressed in VF MS and with MD which, in contrast to MS, is a parameter adjusted to general normative data. Our starting hypothesis was that because of this difference, we would find a more appropriate correlation and regression examining MS versus AVG. However, we found that, in all our analyses, MS and MD showed an almost indistinguishable relationship with RNFL, therefore either one or both of them could be used for data interpretation. Because both RNFL and VF sensitivity are affected by age, for means comparison analysis between our groups, age adjustment had to be performed. We defined a declining rate of RNFL AVG with age as 0.36 μm per year, which is in agreement with previous findings in the literature. Alamouti and Funk 19 reported a 0.44-μm yearly decline, whereas Kanamori et al. 20 calculated 0.22 μm per year using OCT II. VF sensitivity loss has already been reported to be 0.06 dB per year 18 by Humphrey perimetry, and our results showed a 0.056-dB loss per year. 
The correlation of RNFL with VF abnormalities has already been reported in the literature. 21 22 In our study, we compared structural and functional parameters of glaucoma patients with and without VF abnormalities. The discriminating power of OCT has been described between normal subjects and patients who have glaucoma. 23 24 We compared our data to those from normal subjects, but focused mostly on our patients with PPG and POAG (i.e., we tried to estimate to what degree RNFL structural loss is responsible for VF changes). First we analyzed the RNFL AVG versus different VF parameters. We found no correlation in the PPG group, but a significant correlation in the POAG eyes. Based on this difference, we refined our grouping criteria with another parameter, dividing all eyes according to the best discriminating parameter, PSD, with a cutoff value of 1.9 dB. Subsequent correlation and regression test results suggested that eyes having a milder VF defect (i.e., PSD below 1.9 dB) were very similar to the eyes of the preperimetric group. Nouri-Mahdavi et al. 25 described that in early glaucoma, with their group average MD of −2.9 dB, no correlation was found between MD/PSD and RNFL AVG thickness using the OCT 2000 (Carl Zeiss Meditec, Inc.). It has been noted that in the earlier stages of glaucoma a so-called functional reserve seems to exist, but this may be only a result of the logarithmic decibel scaling of the VF or due to the effect of the logarithmic scale, which minimizes sensitivity changes at high dB levels but maximizes changes at low dB levels. 26 Our study found that this might indeed be the case, due to the overlap of both RNFL thickness and VF sensitivity distribution in normal, preperimetric and early stages of glaucoma. It is notable that almost all the eyes in the preperimetric group had an RNFL thickness of 68.0 μm or more, and separation of our patients based on PSD showed an almost identical minimum RNFL AVG (70.73 μm). Using OCT, the average RNFL thickness in normal eyes has been reported to vary from 90 to 128 μm. 9 15 16 22 24 27 28 We found an average RNFL thickness of 96.48 ± 8.24 μm in our normal group. Thus, a mean value of RFNL thickness of approximately 70 μm—even considering its wide range of interindividual variability—may represent a profound threshold value in glaucomatous structural changes. To our knowledge, there are no published data estimating RNFL loss in the progression of glaucoma in the presence of VF defects, as measured by StratusOCT. Our study suggests that all normal, preperimetric, or early glaucomatous eyes could be found with AVG above 68.0 μm, whereas RNFL/VF changes become significant in eyes with AVG below this value. This observation is also consistent with postmortem histologic measurements in patients with glaucoma, indicating that at least 25% to 40% of retinal ganglion cells were lost before abnormalities were statistically detected by automated visual field testing. 29 30  
The present study has certain limitations—that is, selection of patients with glaucoma based on their SAP results was not masked, even if only typical glaucomatous abnormalities seen on the pattern deviation plot alone were strictly taken into account. Nevertheless, the VF-test reader was able to see global indices on the whole VF printout. 
A cross-sectional study can identify structural parameters that are associated with visual function, but cannot address the temporal relationship between structure and function. More longitudinal data are needed to clarify structural change with its corresponding functional decline during the progression of glaucoma. Regression analysis of the structure–function relationship may give important information in assessing the pattern of progression. 
In summary, the strength of the relationship between structure and function detected in our study is in good agreement with previous reports in the literature. The correlation is significant in POAG eyes, but no correlation was detected between VF and RNFL parameters in normal and PPG eyes. Either the moderate discriminating power of the commercially available StratusOCT in visualizing very early structural RNFL changes or SAP’s detection of only the more pronounced VF defects in early-stage glaucoma may explain these findings. This is combined with the great interindividual variability in RNFL thickness, which causes a significant overlap between normal, PPG, and early POAG groups. 
 
Table 1.
 
Descriptive Statistics and Means Comparisons of Subject Characteristics in Study Groups
Table 1.
 
Descriptive Statistics and Means Comparisons of Subject Characteristics in Study Groups
POAG (n = 164) PPG (n = 102) Normal (n = 45) P 1 Normal-POAG P 2 Normal-PPG P 3 PPG-POAG
MD (dB) −4.33 ± 5.00 −1.18 ± 1.00 −0.81 ± 0.92 <0.001* NS* <0.001*
PSD (dB) 4.05 ± 3.31 1.60 ± 0.27 1.54 ± 0.23 <0.001* NS* <0.001*
MS (dB) 23.94 ± 4.81 27.48 ± 1.35 28.48 ± 1.04 <0.001* , † <0.001* <0.001* , †
AVG (μm) 77.99 ± 17.06 90.56 ± 10.96 96.48 ± 8.24 <0.001* , † 0.002* <0.001* , †
AGE (y) 63.27 ± 10.21 55.52 ± 9.59 55.33 ± 8.47 <0.001, ‡ NS, ‡ <0.001, ‡
Table 2.
 
Pearson Correlation between AVG and VF Test Parameters within Normal, POAG and PPG Groups
Table 2.
 
Pearson Correlation between AVG and VF Test Parameters within Normal, POAG and PPG Groups
MD MS PSD
POAG PPG Normal POAG PPG Normal POAG PPG Normal
AVG 0.718* 0.123 0.027 0.733* 0.147 0.105 −0.689* −0.034 0.004
Figure 1.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in the Normal, PPG, and POAG groups. RNFL average thickness expressed in micrometers, VF MD in decibels.
Figure 1.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in the Normal, PPG, and POAG groups. RNFL average thickness expressed in micrometers, VF MD in decibels.
Figure 2.
 
ROC curves for VF and RNFL test parameters.
Figure 2.
 
ROC curves for VF and RNFL test parameters.
Table 3.
 
Area under the ROC Curve for Test Parameters
Table 3.
 
Area under the ROC Curve for Test Parameters
Area SE 95% CI
Lower Upper
MD 0.776* 0.028 0.722 0.830
PSD 0.937* 0.014 0.909 0.964
MS 0.792* 0.027 0.740 0.844
AVG 0.716* 0.031 0.655 0.777
Table 4.
 
Descriptive Statistics of Test Variables in POAG and PPG Subgroups Based on Their PSDs
Table 4.
 
Descriptive Statistics of Test Variables in POAG and PPG Subgroups Based on Their PSDs
POAG PPG
PSD > 1.90 PSD < 1.90 PSD > 1.90 PSD < 1.90
AVG (μm) 74.80 ± 17.18 89.34 ± 10.68 89.29 ± 6.61 90.70 ± 11.35
MD (dB) −5.11 ± 5.39 −1.57 ± 1.00 −1.77 ± 1.03 −1.11 ± 0.99
MS (dB) 23.11 ± 5.10 26.89 ± 1.27 26.79 ± 1.66 27.55 ± 1.30
PSD (dB) 4.70 ± 3.49 1.74 ± 0.12 2.13 ± 0.41 1.54 ± 0.17
AGE (y) 64.10 ± 10.52 60.33 ± 8.53 59.99 ± 10.91 55.03 ± 9.37
n 128 36 10 92
Table 5.
 
Pearson Correlations between AVG and VF Test Parameters in the POAG and PPG Subgroups, Based on Their Corresponding PSDs
Table 5.
 
Pearson Correlations between AVG and VF Test Parameters in the POAG and PPG Subgroups, Based on Their Corresponding PSDs
MD MS PSD
POAG PPG POAG PPG POAG PPG
I II I II I II I II I II I II
AVG 0.727* −0.021 0.278 0.109 0.733* 0.165 −0.359 0.183 −0.674* −0.112 0.407 −0.072
Figure 3.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in POAG subgroups based on their PSD values. RNFL average thickness expressed in micrometers and VF mean deviation in decibels.
Figure 3.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in POAG subgroups based on their PSD values. RNFL average thickness expressed in micrometers and VF mean deviation in decibels.
TuulonenA, LehtolaJ, AiraksinenPJ. Nerve fiber layer defects with normal visual fields: do normal optic disc and normal visual field indicate absence of glaucomatous abnormality?. Ophthalmology. 1993;100:587–597.discussion 597–598 [CrossRef] [PubMed]
SchumanJS, WollsteinG, FarraT, et al. Comparison of optic nerve head measurements obtained by optical coherence tomography and confocal scanning laser ophthalmoscopy. Am J Ophthalmol. 2003;135:504–512.(published correction in Am J Ophthalmol. 2003;136:following 403) [CrossRef] [PubMed]
DaPozzo S, FuserM, VattovaniO, DiStefano G, RavalicoG. GDx-VCC performance in discriminating normal from glaucomatous eyes with early visual field loss. Graefes Arch Clin Exp Ophthalmol. 2005;16:1–7.
BlumenthalEZ, WilliamsJM, WeinrebRN, GirkinCA, BerryCC, ZangwillLM. Reproducibility of nerve fiber layer thickness measurements by use of optical coherence tomography. Ophthalmology. 2000;107:2278–2282. [CrossRef] [PubMed]
CarpinetoP, CiancagliniM, ZuppardiE, FalconioG, DoronzoE, MastropasquaL. Reliability of nerve fiber layer thickness measurements using optical coherence tomography in normal and glaucomatous eyes. Ophthalmology. 2003;110:190–195. [CrossRef] [PubMed]
SchumanJS, HeeMR, PuliafitoCA, et al. Quantification of nerve fiber layer thickness in normal and glaucomatous eyes using optical coherence tomography. Arch Ophthalmol. 1995;113:586–596. [CrossRef] [PubMed]
JaffeGJ, CaprioliJ. Optical coherence tomography to detect and manage retinal disease and glaucoma (review). Am J Ophthalmol. 2004;137:156–169. [CrossRef] [PubMed]
PierothL, SchumanJS, HertzmarkE, et al. Evaluation of focal defects of the nerve fiber layer using optical coherence tomography. Ophthalmology. 1999;106:570–579. [CrossRef] [PubMed]
MokKH, LeeVW, SoKF. Retinal nerve fiber loss pattern in high-tension glaucoma by optical coherence tomography. J Glaucoma. 2003;12:255–259. [CrossRef] [PubMed]
JeoungJW, ParkKH, KimTW, KhwargSI, KimDM. Diagnostic ability of optical coherence tomography with a normative database to detect localized retinal nerve fiber layer defects. Ophthalmology. 2005;112:2157–2163. [CrossRef] [PubMed]
WeinrebRN, ShakibaS, SamplePA, et al. Association between quantitative nerve fiber layer measurement and visual field loss in glaucoma. Am J Ophthalmol. 1995;120:732–738. [CrossRef] [PubMed]
El BeltagiTA, BowdC, BodenC, et al. Retinal nerve fiber layer thickness measured with optical coherence tomography is related to visual function in glaucomatous eyes. Ophthalmology. 2003;110:2185–2191. [CrossRef] [PubMed]
LeungCK, ChanWM, HuiYL, et al. Analysis of retinal nerve fiber layer and optic nerve head in glaucoma with different reference plane offsets, using optical coherence tomography. Invest Ophthalmol Vis Sci. 2005;46:891–899. [CrossRef] [PubMed]
PaunescuLA, SchumanJS, PriceLL, et al. Reproducibility of nerve fiber thickness, macular thickness, and optic nerve head measurements using StratusOCT. Invest Ophthalmol Vis Sci. 2004;45:1716–1724. [CrossRef] [PubMed]
GuedesV, SchumanJS, HertzmarkE, et al. Optical coherence tomography measurement of macular and nerve fiber layer thickness in normal and glaucomatous human eyes. Ophthalmology. 2003;110:177–189. [CrossRef] [PubMed]
LeungCK, ChanWM, YungWH, et al. Comparison of macular and peripapillary measurements for the detection of glaucoma: an optical coherence tomography study. Ophthalmology. 2005;112:391–400. [CrossRef] [PubMed]
WollsteinG, SchumanJS, PriceLL, et al. Optical coherence tomography (OCT) macular and peripapillary retinal nerve fiber layer measurements and automated visual fields. Am J Ophthalmol. 2004;138:218–225. [CrossRef] [PubMed]
European Glaucoma Society. Terminology and Guidelines for Glaucoma. 1998;44.European Glaucoma Society Savona, Italy.
AlamoutiB, FunkJ. Retinal thickness decreases with age: an OCT study. Br J Ophthalmol. 2003;87:899–901. [CrossRef] [PubMed]
KanamoriA, EscanoMF, EnoA, et al. Evaluation of the effect of aging on retinal nerve fiber layer thickness measured by optical coherence tomography. Ophthalmologica. 2003;217:273–278. [CrossRef] [PubMed]
KwonYH, HongS, HonkanenRA, AlwardWL. Correlation of automated visual field parameters and peripapillary nerve fiber layer thickness as measured by scanning laser polarimetry. J Glaucoma. 2000;9:281–288. [CrossRef] [PubMed]
LeungCK, ChongKK, ChanWM, et al. Comparative study of retinal nerve fiber layer measurement by StratusOCT and GDx VCC, II: structure/function regression analysis in glaucoma. Invest Ophthalmol Vis Sci. 2005;46:3702–3711. [CrossRef] [PubMed]
BudenzDL, MichaelA, ChangRT, McSoleyJ, KatzJ. Sensitivity and specificity of the StratusOCT for perimetric glaucoma. Ophthalmology. 2005;112:3–9. [CrossRef] [PubMed]
HohST, GreenfieldDS, MistlbergerA, LiebmannJM, IshikawaH, RitchR. Optical coherence tomography and scanning laser polarimetry in normal, ocular hypertensive, and glaucomatous eyes. Am J Ophthalmol. 2000;129:129–135. [CrossRef] [PubMed]
Nouri-MahdaviK, HoffmanD, TannenbaumDP, LawSK, CaprioliJ. Identifying early glaucoma with optical coherence tomography. Am J Ophthalmol. 2004;137:228–235. [CrossRef] [PubMed]
SchlottmannPG, De CillaS, GreenfieldDS, CaprioliJ, Garway-HeathDF. 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. [CrossRef] [PubMed]
LeungCK, ChanWM, ChongKK, et al. Comparative study of retinal nerve fiber layer measurement by StratusOCT and GDx VCC, I: correlation analysis in glaucoma. Invest Ophthalmol Vis Sci. 2005;46:3214–3220. [CrossRef] [PubMed]
KanamoriA, NakamuraM, EscanoMF, SeyaR, MaedaH, NegiA. Evaluation of the glaucomatous damage on retinal nerve fiber layer thickness measured by optical coherence tomography. Am J Ophthalmol. 2003;135:513–520. [CrossRef] [PubMed]
QuigleyHA, DunkelbergerGR, GreenWR. Retinal ganglion cell atrophy correlated with automated perimetry in human eyes with glaucoma. Am J Ophthalmol. 1989;107:453–464. [CrossRef] [PubMed]
Kerrigan-BaumrindLA, QuigleyHA, PeaseME, KerriganDF, MitchellRS. Number of ganglion cells in glaucoma eyes compared with threshold visual field tests in the same persons. Invest Ophthalmol Vis Sci. 2000;41:741–748. [PubMed]
Figure 1.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in the Normal, PPG, and POAG groups. RNFL average thickness expressed in micrometers, VF MD in decibels.
Figure 1.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in the Normal, PPG, and POAG groups. RNFL average thickness expressed in micrometers, VF MD in decibels.
Figure 2.
 
ROC curves for VF and RNFL test parameters.
Figure 2.
 
ROC curves for VF and RNFL test parameters.
Figure 3.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in POAG subgroups based on their PSD values. RNFL average thickness expressed in micrometers and VF mean deviation in decibels.
Figure 3.
 
Scatterplots of RNFL (AVG) vs. VF (MD) in POAG subgroups based on their PSD values. RNFL average thickness expressed in micrometers and VF mean deviation in decibels.
Table 1.
 
Descriptive Statistics and Means Comparisons of Subject Characteristics in Study Groups
Table 1.
 
Descriptive Statistics and Means Comparisons of Subject Characteristics in Study Groups
POAG (n = 164) PPG (n = 102) Normal (n = 45) P 1 Normal-POAG P 2 Normal-PPG P 3 PPG-POAG
MD (dB) −4.33 ± 5.00 −1.18 ± 1.00 −0.81 ± 0.92 <0.001* NS* <0.001*
PSD (dB) 4.05 ± 3.31 1.60 ± 0.27 1.54 ± 0.23 <0.001* NS* <0.001*
MS (dB) 23.94 ± 4.81 27.48 ± 1.35 28.48 ± 1.04 <0.001* , † <0.001* <0.001* , †
AVG (μm) 77.99 ± 17.06 90.56 ± 10.96 96.48 ± 8.24 <0.001* , † 0.002* <0.001* , †
AGE (y) 63.27 ± 10.21 55.52 ± 9.59 55.33 ± 8.47 <0.001, ‡ NS, ‡ <0.001, ‡
Table 2.
 
Pearson Correlation between AVG and VF Test Parameters within Normal, POAG and PPG Groups
Table 2.
 
Pearson Correlation between AVG and VF Test Parameters within Normal, POAG and PPG Groups
MD MS PSD
POAG PPG Normal POAG PPG Normal POAG PPG Normal
AVG 0.718* 0.123 0.027 0.733* 0.147 0.105 −0.689* −0.034 0.004
Table 3.
 
Area under the ROC Curve for Test Parameters
Table 3.
 
Area under the ROC Curve for Test Parameters
Area SE 95% CI
Lower Upper
MD 0.776* 0.028 0.722 0.830
PSD 0.937* 0.014 0.909 0.964
MS 0.792* 0.027 0.740 0.844
AVG 0.716* 0.031 0.655 0.777
Table 4.
 
Descriptive Statistics of Test Variables in POAG and PPG Subgroups Based on Their PSDs
Table 4.
 
Descriptive Statistics of Test Variables in POAG and PPG Subgroups Based on Their PSDs
POAG PPG
PSD > 1.90 PSD < 1.90 PSD > 1.90 PSD < 1.90
AVG (μm) 74.80 ± 17.18 89.34 ± 10.68 89.29 ± 6.61 90.70 ± 11.35
MD (dB) −5.11 ± 5.39 −1.57 ± 1.00 −1.77 ± 1.03 −1.11 ± 0.99
MS (dB) 23.11 ± 5.10 26.89 ± 1.27 26.79 ± 1.66 27.55 ± 1.30
PSD (dB) 4.70 ± 3.49 1.74 ± 0.12 2.13 ± 0.41 1.54 ± 0.17
AGE (y) 64.10 ± 10.52 60.33 ± 8.53 59.99 ± 10.91 55.03 ± 9.37
n 128 36 10 92
Table 5.
 
Pearson Correlations between AVG and VF Test Parameters in the POAG and PPG Subgroups, Based on Their Corresponding PSDs
Table 5.
 
Pearson Correlations between AVG and VF Test Parameters in the POAG and PPG Subgroups, Based on Their Corresponding PSDs
MD MS PSD
POAG PPG POAG PPG POAG PPG
I II I II I II I II I II I II
AVG 0.727* −0.021 0.278 0.109 0.733* 0.165 −0.359 0.183 −0.674* −0.112 0.407 −0.072
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×