September 2015
Volume 56, Issue 10
Free
Glaucoma  |   September 2015
The Fast Component of Visual Field Decay Rate Correlates With Disc Rim Area Change Throughout the Entire Range of Glaucomatous Damage
Author Affiliations & Notes
  • Ji-Woong Lee
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
    Department of Ophthalmology, Pusan National University School of Medicine, Busan, South Korea
  • Eun-Ah Kim
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
  • Francisco Otarola
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
  • Esteban Morales
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
  • Fei Yu
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
    Department of Biostatistics, Jonathan and Karin Fielding School of Public Health at UCLA, Los Angeles, California, United States
  • Abdelmonem A. Afifi
    Department of Biostatistics, Jonathan and Karin Fielding School of Public Health at UCLA, Los Angeles, California, United States
  • Kouros Nouri-Mahdavi
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
  • Joseph Caprioli
    The Jules Stein Eye Institute David Geffen School of Medicine at University of California, Los Angeles (UCLA), Los Angeles, California, United States
  • Correspondence: Joseph Caprioli, Jules Stein Eye Institute, David Geffen School of Medicine at UCLA, 100 Stein Plaza, Los Angeles, CA 90095, USA; Caprioli@jsei.ucla.edu
Investigative Ophthalmology & Visual Science September 2015, Vol.56, 5997-6006. doi:10.1167/iovs.15-17006
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Ji-Woong Lee, Eun-Ah Kim, Francisco Otarola, Esteban Morales, Fei Yu, Abdelmonem A. Afifi, Kouros Nouri-Mahdavi, Joseph Caprioli; The Fast Component of Visual Field Decay Rate Correlates With Disc Rim Area Change Throughout the Entire Range of Glaucomatous Damage. Invest. Ophthalmol. Vis. Sci. 2015;56(10):5997-6006. doi: 10.1167/iovs.15-17006.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: We investigated the longitudinal relationships between the changes in neuroretinal rim area (RA) and the slow (SC) and fast (FC) components of visual field (VF) decay at various stages of glaucoma.

Methods: We divided 465 eyes of 338 patients into glaucoma suspect, and preperimetric, early, and moderate/advanced glaucoma. All patients had a minimum of 3 confocal scanning laser ophthalmoscopic examinations and 4 VF tests with follow-up of 4 or more years. A pointwise exponential regression was used to perform trend analyses on thresholds at each VF test location, which was partitioned into SC and FC. A mixed effects linear model was used to explore the associations of RA change with mean deviation (MD), visual field index (VFI), SC, and FC.

Results: Decreased RA was associated with lower mean threshold sensitivities of FC regardless of baseline severity of glaucoma (P ≤ 0.03). The mean threshold sensitivities in SC were not correlated with RA change at any stage. Decreased RA was correlated with worse MD in preperimetric, early, and moderate/advanced glaucoma (P < 0.05). Decreased RA was correlated with worse VFI in preperimetric and early glaucoma only (P ≤ 0.04).

Conclusions: A decrease in rim area was significantly correlated with the fast VF component regardless of the baseline severity of glaucoma. Mean deviation and VFI correlated with change of rim area only in certain stages of glaucoma. The identification of the fast component seems a more robust and useful measure of glaucomatous change than MD or VFI.

Glaucoma is characterized by progressive optic disc and visual field abnormalities.1 Estimating the rates of structural and functional loss is fundamental to the long-term goal of preservation of vision in patients with glaucoma.2 A better understanding of the precise relationships between structural and functional changes over time would help clinicians better measure and predict the rate of glaucoma worsening.3 
Discordance between structural and functional measures should be expected in patients with glaucoma.1,4,5 Many patients may show evidence of optic disc damage before functional loss is detected by standard automated perimetry.4,611 However, some patients may show evidence of visual field change first without clinically apparent optic disc abnormalities.5 Previous studies found a severity-dependent association between structure and function in patients with glaucoma.8,1216 
Few longitudinal studies have evaluated the nature of the structural and functional rates of change in patients with glaucoma as a function of glaucoma severity.17,18 While numerous cross-sectional studies have reported on statistically proven correlations between structure and function,1921 the use of cross-sectional data to estimate longitudinal behavior in individual patients is not the best approach.1 A technique based on trend analysis that uses a pointwise exponential model to fit the behavior of individual test locations in visual field series has been reported, and separates a slow component (SC) and a fast component (FC) of visual field decay for each eye, and operates successfully across a wide range of disease severity.2224 
The purpose of the present study was to investigate the longitudinal relationships between change in global rim area (RA) as measured with confocal scanning laser ophthalmoscopy (CSLO) with the SC and FC of visual field decay in patients at different stages of glaucoma severity, and to compare these against the standard visual field indices mean deviation (MD) and visual field index (VFI). 
Methods
Participants
In this retrospective, longitudinal, observational study, patients with suspected or established open-angle glaucoma from the Jules Stein Eye Institute's clinical database were included. This study was conducted in accord with the tenets set forth in the Declaration of Helsinki and was approved by the UCLA Institutional Review Board. The institutional review board waived consent from patients because this is a retrospective, anonymous study. Inclusion criteria were as follows: age > 30 years, baseline best-corrected visual acuity 20/80 or better, follow-up time ≥ 4 years, spherical equivalent ≥ −8 diopters and astigmatism < 3 diopters, ≥ 3 CSLO (Heidelberg Retina Tomograph II [HRT]; Heidelberg Engineering, Heidelberg, Germany) examinations with global pixel standard deviation less than 40 μm, and ≥ 4 reliable visual field examinations. The intervals between the first and last HRT examinations and visual field examinations were within 6 months. Eyes with neurologic or retinal diseases that can affect the visual field were excluded. 
Eyes were classified as having suspected glaucoma if they had a history of elevated IOP (>21 mm Hg) and/or suspicious appearance of optic nerve based on stereoscopic photographs reviewed by 2 experienced graders, but normal and reliable visual field results on at least two baseline visits.25 Disagreements between the two graders were resolved by a third experienced grader. Eyes were classified as having preperimetric glaucoma if they had glaucomatous optic neuropathy without evidence of repeatable glaucomatous visual field defect at baseline.26 Glaucomatous optic neuropathy was defined as having more than a 0.2 cup-to-disc ratio asymmetry between the 2 eyes, neuroretinal rim thinning, notching, or characteristic retinal nerve fiber layer defects indicative of glaucoma.27 Eyes were classified glaucomatous if they had a glaucomatous optic disc and two consecutive abnormal visual field test results on the baseline visits. An abnormal visual field was defined as P < 0.05 for the pattern standard deviation or a Glaucoma Hemifield Test result outside normal limits. Eyes with established glaucoma were further classified as having early (MD ≥ −6 dB) or moderate/advanced (MD < −6 dB) glaucoma.28 
Data collected from the medical record were age, sex, ethnicity, comorbidity, lens status, best-corrected visual acuity, history of glaucoma surgery and cataract surgery, and number of glaucoma medications and IOP measurements performed by Goldmann applanation tonometry. 
Perimetric Tests
All visual field tests were performed with an automated visual field analyzer (Humphrey Field Analyzer; Carl Zeiss Meditec, Inc., Dublin, CA, USA) with a 24-2 test pattern, size III white stimulus with the Swedish Interactive Threshold Algorithm standard strategy. Adequate reliability was defined as less than 15% fixation losses, less than 15% false-positive rates, and less than 30% false-negative rates. The technique of measuring rates of visual field decay has been reported in detail,22,23 and is summarized here. Rates of visual field decay were calculated with a pointwise exponential regression analysis of threshold sensitivities at 52 test locations, excluding the 2 locations corresponding to the blind spot. The association between the response variable (threshold sensitivity) and the explanatory variable (follow-up duration) was characterized by the following exponential regression models: Image not available or, equivalently, Image not available where a is the intercept, x is time, b is the mean annual rate of change in ln y, and eb represents the ratio of y in a given year to y in the year before. The decay rate is defined as Image not available To facilitate an intuitive clinical understanding of the magnitude of decay rates, the coefficients of the exponential regressions were converted into percentage per year deterioration rates, where the percentage per year decay rate is Image not available the more negative decay rate indicates a faster deterioration of perimetric sensitivities. The 52 visual field test locations were ranked according to their decay rates and were clustered into 2 subgroups (SC and FC) based on the P value for the difference in the mean rates between 2 clusters. For each possible partition, starting with a minimum number of 5 locations in a cluster, we computed a t-test statistic, and the corresponding P values were adjusted for multiple testing. Because multiple simultaneous t-tests were performed, it is desirable to correct the P values to control for false-positive results. Accordingly, the Benjamini-Hochberg correction was used to adjust the P values for multiple testing.29 The SC and FC values are unique for each eye; the mean slow and fast decay rates were calculated for the partitioned components during the follow-up period. To obtain the mean threshold sensitivities in SC and FC, the threshold sensitivity in dB unit at each of 52 visual field locations was first converted to linear (1/Lambert) scale with the following formula; Image not available Then, values from all test points within SC or FC were averaged for each eye. The average visual sensitivity per SC or FC was converted back to the dB scale for the analysis. The MD and VFI values also were analyzed with univariate linear regression analyses against time. Eyes with a significant negative trend (slope ≤ −0.3 and P < 0.05) on regression analysis of MD against time were considered to have progressive visual fields. A MD slope of −0.3 dB/y was chosen as a conservative estimate of cutoff for visual field worsening because a previous study reported that mean MD rate ranged from −0.21 to −0.35 dB/y in patients with glaucoma.30 Also, this rate represents a value that is approximately 10 times the magnitude of change due to age alone, a quantity that has been used in many prior studies.3034  
CSLO Imaging
Confocal scanning laser ophthalmoscopy images were acquired with the HRT II device and analyzed with HRT III software. Three sets of 15° field-of view scans centered on the optic disc were automatically captured at each acquisition. Subsequently, the three sets were used to create a final composite image for analysis. To construct a single examination, the software aligned a stack of individual scans. Each scan was examined for image quality and if any were outside of acceptable quality indices, the scan was discarded from the image set. The optic disc contour line was drawn on the mean topography image by an experienced observer while viewing stereoscopic photographs of the optic disc. The global and sectoral stereometric parameters generated by the HRT software were exported for further analysis. Previous studies showed that global RA was better correlated with functional indices than other HRT parameters,35 and was significantly associated with the development of primary open-angle glaucoma.36 Therefore, global RA was used as the main HRT parameter in subsequent statistical analyses. The difference of global RA from baseline at each visit (global RA at each visit minus baseline global RA) was the main structural measure that was considered. The global RA was also analyzed with univariate linear regression analyses of these values against time. 
Statistical Analysis
All statistical analyses were performed with Stata software version 13.0 (StataCorp, College Station, TX, USA). The normality of numerical data distribution was checked with the Kolmogorov-Smirnov test. Clinical characteristics of the study population were compared with ANOVA or the Kruskal-Wallis test for continuous variables and χ2 test or Fisher's exact test for categorical variables. 
Mixed effects linear regression models of MD, VFI, and mean threshold sensitivities of SC and FC were constructed to detect the adjusted association of longitudinal structural and functional changes with fixed and random effects included. Models were constructed with MD, VFI, and mean sensitivities of SC and FC set as each individual outcome in separate models. All models were fitted with fixed coefficients (fixed effect) of participants baseline age (years), follow-up time (years), baseline global RA (square millimeters), and difference of global RA from baseline at each visit (global RA at each visit minus baseline RA, in square millimeters), with random coefficients (random effects) of subject (each eye nested within subject) to detect the effect of time. The inclusion of random eye effects accounts for the nonindependence of the 2 eyes from the same subject. Because the number of HRT examinations generally was less than that of visual field, the visual fields performed within 6 months of the HRT were chosen for mixed effect linear regression model. All P values are 2-sided and differences at a level of P < 0.05 were considered statistically significant. 
Results
We included 465 eyes of 338 patients in the study. Among the 465 eyes, 88 eyes were classified as glaucoma suspect, 128 eyes as preperimetric glaucoma, 147 eyes as early glaucoma, and 102 eyes as moderate/advanced glaucoma. 
The mean (SD) follow-up times of the total study samples were 6.54 (2.13) and 6.64 (2.14) years for HRT and visual field examinations, respectively, The mean numbers of available HRT and visual field examinations of total study samples were 3.84 (1.09) and 8.86 (4.05), respectively. 
The baseline demographic, clinical, and ocular characteristics of the study sample are summarized in Tables 1 through 3. There were no statistically significant differences among study groups for baseline age, sex, eye laterality, lens status, ethnicity, comorbidity, HRT and visual field examination follow-up times, and number of HRT examinations. The number of visual field examinations increased significantly in proportion to the severity of glaucoma (P < 0.001, Table 1). 
Table 1
 
Demographic and Clinical Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 1
 
Demographic and Clinical Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Baseline HRT parameters, including the global RA, were significantly worse in the more advanced glaucoma group (P < 0.001 for all, Table 2). The proportion of eyes with evidence of progression based on MD regression analysis and that of eyes that received more than one glaucoma surgery during follow-up significantly increased in association with increasing severity of glaucoma (P = 0.01 and P < 0.001, respectively, Table 3). 
Table 2
 
Baseline HRT Stereometric Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 2
 
Baseline HRT Stereometric Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 3
 
Changes in IOP, Number of Glaucoma Medications, and Global Visual Field Parameters Among the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 3
 
Changes in IOP, Number of Glaucoma Medications, and Global Visual Field Parameters Among the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Figure 1 shows the overall distribution of rates of change for global RA, MD, VFI, SC, and FC for the entire study sample. There was a statistically significant difference in the global RA rate among four glaucoma groups (P = 0.002), and the early glaucoma group had the worst rate. Mean deviation rate, VFI rate, and FC rate were significantly worse in the more advanced groups (P = 0.038, P < 0.001, and P < 0.001, respectively; Table 4). We examined the effect of baseline global RA on global RA rate of change; a greater baseline global RA was correlated with a faster rate of global RA decay in the entire study sample (Spearman's ρ = −0.30, P < 0.001, Fig. 2). Global RA rate was positively correlated with the MD rate, VFI rate, and FC rate in the overall group (Spearman's ρ = 0.15, 0.15, and 0.14, respectively; P = 0.001, 0.002, and 0.002, respectively), while global RA rate was not correlated with SC rate (Spearman's ρ = 0.048; P = 0.30). 
Figure 1
 
Distributions of rates of change for global rim area (A), MD (B), VFI (C), SC (D), and FC (E) visual field components in entire study sample.
Figure 1
 
Distributions of rates of change for global rim area (A), MD (B), VFI (C), SC (D), and FC (E) visual field components in entire study sample.
Table 4
 
The Rates of Structural and Functional Changes Among 465 Eyes of 338 Subjects With Different Severity of Glaucoma
Table 4
 
The Rates of Structural and Functional Changes Among 465 Eyes of 338 Subjects With Different Severity of Glaucoma
Figure 2
 
Scatterplot showing the influence of baseline global rim area on the rate of global rim area change in entire study samples. The rate of global rim area change was negatively correlated with baseline global rim area (r = −0.30, P < 0.001).
Figure 2
 
Scatterplot showing the influence of baseline global rim area on the rate of global rim area change in entire study samples. The rate of global rim area change was negatively correlated with baseline global rim area (r = −0.30, P < 0.001).
With the mixed effects linear regression model, we explored the association between structural change (global RA at each visit minus baseline global RA) and longitudinal measurements of functional parameters (MD, VFI, mean threshold sensitivities of SC and FC) after adjusting for baseline age, baseline RA, and follow-up time. The greater decrease in RA was associated with worse MD, VFI, and the mean threshold sensitivities of FC in entire study sample (P < 0.001 for all, Table 5). In the overall group of patients, older baseline age was significantly associated with lower mean sensitivities of FC and SC (P < 0.05 for both). Longer follow-up time was significantly associated with lower MD, VFI, and mean sensitivities of FC (P < 0.001 for all). Smaller baseline global rim area was associated with lower MD, VFI, and mean sensitivities of FC and SC (P < 0.05 for all). In the glaucoma suspect group, only the mean threshold sensitivities of FC correlated with change in RA (P = 0.003). In preperimetric and early glaucoma, decreased RA correlated with worse MD, VFI, and the mean threshold sensitivities of FC (P < 0.05 for all). In moderate/advanced glaucoma, the decreased RA correlated with worse MD and the mean threshold sensitivities of FC (P = 0.046 and P = 0.015, respectively). There was some evidence suggesting that VFI may be associated with structural change in the moderate/advanced glaucoma group even though it did not reach statistical significance (P = 0.075). However, the mean threshold sensitivities of SC were not significantly correlated with change of RA in any of the severity groups or in the entire study sample (P = 0.352–0.878). 
Table 5
 
Results from Mixed Effects Linear Regression Model for Association Between Structural Change as a Predictor and Functional Change as Each Individual Outcome in Separate Regression Models After Adjusting for Baseline Age, Baseline Global Rim Area, and Follow-Up Time
Table 5
 
Results from Mixed Effects Linear Regression Model for Association Between Structural Change as a Predictor and Functional Change as Each Individual Outcome in Separate Regression Models After Adjusting for Baseline Age, Baseline Global Rim Area, and Follow-Up Time
Discussion
The objective of this study was to investigate the relationship between global RA decline as measured with CSLO, and the slow and fast visual field components as measured with a pointwise exponential regression model in patients with different baseline glaucoma severities. When we explored the relationship between structural and functional change with a mixed effects linear regression model after adjusting for baseline age and baseline RA, we found that the decline in threshold sensitivities of FC was significantly associated with the decline of RA in all severity groups. The greater decrease in global RA correlated with a worsening of mean threshold sensitivities of FC. The mean threshold sensitivities of SC were not correlated with this structural change in any of the severity groups. Mean deviation was not correlated with RA change in the glaucoma suspect group. Visual field index was correlated with structural change only in preperimetric and early glaucoma. 
These findings might be explained by the typically localized characteristic of glaucoma progression.22,23,37 Fast component of visual field decay better characterizes the structure and function relationship over a wide range of glaucoma severities, since it is where we believe most of the glaucoma “signal” resides. It is likely that FC is better associated with structural change than MD or VFI, since we extract and remove the noncorrelating portion (SC) of visual field deterioration. 
Shigeeda et al.38 reported that the MD rate of change would be expected to be insensitive to localized components of visual field deterioration. The VFI provides no spatial information about the visual field and seems too insensitive to regional changes or focal components of damage in the same manner.39 Therefore, point-by-point analyses may be more effective in detecting localized components of visual field progression that are closely related with structural change.2224,39 Mean deviation and VFI may lack sensitivity for detecting early glaucoma progression, because they are global indices that summarize mean damage across the entire visual field.39,40 Alasil et al.14 reported that substantial retinal nerve fiber layer thinning is necessary before MD loss is detected. The reasons for this discordance are likely, because (1) standard automated perimetry may absorb damage without showing abnormalities early on due to the redundancy of visual field sensitivity,41 (2) higher variability is found in the measurements of standard automated perimetry than those of optical coherence tomography in early glaucoma,13 and (3) MD is insensitive to localized loss.38 In addition, VFI may be less able to detect early glaucomatous change due to the ceiling effect derived from its reliance on pattern deviation probability maps.39 The fact that VFI is heavily weighted towards the central visual field, which typically is damaged later in disease, perhaps also compromises its ability to detect early glaucomatous change.39 
As opposed to the global indices MD and VFI, the method used here enables the identification of the SC and FC components separately, without loss of spatial or pattern information.22,23 Because the FC represents more focal, rapidly deteriorating test locations from glaucoma,2224 our results suggested that identification of FC may reduce the lag time between clinically detectable structural and functional changes, and enable detection of structure and function relationship in glaucoma suspect eyes that are in the early phase of developing glaucomatous damage. In contrast, structural change was not associated with SC at any glaucoma severity, because the SC component reflects more diffuse and nonspecific deterioration, more reflective of media opacity and aging.2224 While we believe that most of the glaucomatous change resides in the FC component of visual field, it certainly is possible that some slow, relatively diffuse glaucomatous visual field loss resides in SC. Henson et al.42 reported that early glaucomatous visual field loss frequently involves a diffuse component. 
In the preperimetric and early glaucoma groups, we found that a greater decrease in global RA was correlated with worse MD, VFI, and the mean threshold sensitivities of FC. Mean deviation and VFI as well as mean threshold sensitivities of FC also were found to be functional parameters that are correlated well with structural changes in preperimetric and early glaucoma groups. It may be that only after redundancy of retinal ganglion cell (RGC) function is exhausted that a strong clinical relationship between structure and function can be detected.18,41 However, Gardiner et al.43 reported that the intertest variability is sufficiently large that the observed discrepancy between the rates of structural and functional change could be a result of noise in early glaucoma. The variability could affect structural as well as functional classification. 
Although the mixed effects linear regression model revealed that decreasing RA was associated with worsening MD and the mean threshold sensitivities of FC in the moderate/advanced glaucoma group, a floor effect ultimately will influence the relationship between structure and function in advanced stages of glaucoma. The relationship between structural and functional change also can be influenced by the variability of the measurements,13,43 different measurement scales,44 different statistical methods to estimate change,17 and the particular structural or functional tests used.21,45 Bowd et al.21 reported that global associations between structure and function were stronger with optical coherence tomography than with HRT. Lamparter et al.45 reported that the correlation between structure and function was stronger in flicker defined form perimetry and frequency doubling technology perimetry than with standard automated perimetry. The correlation between structure and function might have been improved with different measurement tests in this study. 
There is a residual retinal nerve fiber layer thickness even in patients with severe glaucoma, likely because of displaced amacrine cells and nonneuronal tissues including blood vessels, glial cells.13 Likewise, a proportion of the neuroretinal RA may consist of nonneural rim tissue, including blood vessels and glia.46 Nassiri et al.18 reported that a larger baseline RA was associated with a faster rate of progression. We also found that a greater baseline RA was associated with faster rate of decay in RA with the correlation varying from −0.17 (Spearman's ρ, P = 0.05) in the preperimetric glaucoma group to −0.47 (Spearman's ρ, P < 0.001) in the moderate/advanced glaucoma group. There might be less room for decay in eyes with a smaller baseline RA, and the variability of RA measurements increases as RA decreases.47 We found that some evidence suggested that VFI may be associated with structural change in the moderate/advanced glaucoma group. However, Lee et al.48 reported that the VFI becomes highly variable in advanced glaucoma, because of the change from the use of the pattern deviation probability value to the total deviation probability value in calculations of VFI at or below −20 dB of MD. Mean deviation rate, VFI rate, and FC rate were significantly worse in the more advanced groups. Previously, we reported that worse baseline MD was associated with rapid deterioration of visual field loss.49 Chauhan et al.30 also reported that the proportion of fast (MD rate, < −1 to −2 dB/y) and catastrophic (< −2 dB/y) progressors was higher in worst tertile (median baseline MD, −7.79 dB) compared to the other two tertiles. Ernest et al.50 reviewed 85 articles and reported that more baseline visual field loss was probably associated with more visual field progression in patients with open-angle glaucoma. 
The estimated rate of global RA change found in our study (−0.018 mm2/y for the entire study sample) was similar to those reported in previous studies (range, −0.013 to −0.015 mm2/y).51,52 The correlations between global RA rate and MD, VFI, and FC rate are all fairly weak in this study. Our results are consistent with previous longitudinal studies that have explored the relationship of structure and function with HRT.17,18 Medeiros et al.17 reported a significant, but weak correlation between slopes of MD and global RA change calculated with an ordinary least squares method (Spearman's ρ = 0.19; P < 0.001), which is similar to the correlation found in our study (Spearman's ρ = 0.15; P < 0.001). Nassiri et al.18 reported that global structural and functional rates of progression were weakly correlated in eyes with suspected or established glaucoma (Spearman's ρ = 0.14; P = 0.20). These findings support previous evidence of low correspondence between structural and functional measures.18,53,54 Differences in the sensitivity–neural relationships for early and advanced visual field defects should be expected. Kerrigan-Baumrind et al.53 reported that at least 25% to 35% RGC loss was associated with statistically significant abnormalities in automated visual field testing in humans. However, Harwerth et al.54 reported that there is a loss of 6 to 8 dB in visual sensitivity over the range of smaller losses of ganglion cells where sensitivity losses appear to be uncorrelated with neural loss in monkeys. They suggested the possibility of an alteration in RGC function before the pathological loss of the cell body. In more advanced glaucoma, functional outcomes often are used to detect glaucoma progression because a structural change becomes difficult to detect.18 
The main issue with respect to the relationship between structure and function is whether or not structural changes are clinically relevant to functional outcomes.1 When glaucomatous structural change is suspicious, the clinical evidence of progression might be better corroborated with corresponding changes in FC than with MD or VFI. In addition, FC can be integrated with structural information to improve detection of glaucoma progression in future studies. Caprioli55 reported that the combination of structural and functional measurements performed better than structural or functional measurements alone to discriminate between normal and glaucoma and to detect glaucoma worsening. Medeiros et al.56 reported that a Bayesian hierarchical model for combining longitudinal structural and functional information was more sensitive than ordinary least squares regression in detecting glaucoma progression. 
The findings of this study are subject to certain limitations. The sample sizes of the each of the severity subgroups were modest. Given the retrospective and clinical nature of the study, there were a limited number of HRT and visual field examinations available that met our inclusion criteria. The number of usable examinations was determined by an acceptable frequency of making these measurements in a glaucoma practice. Owen et al.57 reported that sensitivity of detection of disease progression improves with more frequent testing of HRT examination. A smaller number of HRT scans may cause greater noise.57 The average number of HRTs used in this analysis is 3.8, and mean number of visual fields 8.9; these are reasonable numbers for a clinical practice. More HRT measurements would certainly have improved our signal-to-noise ratio. However, there is a cost related to acquiring additional measurements, in time and expense. While the findings reported here are subject to this limitation, it does not decrease the relevance or utility of our significant findings. The measurement and interpretation of the visual field FC may enhance the detection of glaucoma progression by providing functional correlates of structural change, which are insensitive to glaucoma severity. We did not consider the severity of lens opacities in the interpretation of visual fields and HRT examinations. A previous study reported that lens opacity affects image quality of HRT scans, which influenced rim area measurement variability.58 However, only eyes with global pixel standard deviation less than 40 μm were included. Previously, we reported that cataract formation minimally affects the FC rate as determined by pointwise exponential regression analysis.24 Of patients in this study, 71% (239 of 338) were of white race/ethnicity; the relationship between structure and function may be different in other ethnic populations. 
In conclusion, we found that by excluding the noncorrelating SC component of visual field decay, the FC component of visual field decay correlated with longitudinally measured structural changes in rim area at all stages of baseline glaucoma severity. Mean deviation and VFI seemed to be good functional indices in cases where glaucomatous damage was well established. Identification of the FC component of visual field progression that is related to structural change across all stages of glaucoma severity might be more helpful to detect and interpret the worsening of glaucomatous deterioration than are the global visual field indices. 
Acknowledgments
Supported by unrestricted grants from Research to Prevent Blindness, Inc., the Simms/Mann Family Foundation, and the Payden Glaucoma Research Fund. The authors alone are responsible for the content and writing of the paper. 
Disclosure: J.-W. Lee, None; E.-A. Kim, None; F. Otarola, None; E. Morales, None; F. Yu, None; A.A. Afifi, None; K. Nouri-Mahdavi, None; J. Caprioli, None 
References
Artes PH, Chauhan BC. Longitudinal changes in the visual field and optic disc in glaucoma. Prog Retin Eye Res. 2005 ; 24: 333–354.
Caprioli J. The importance of rates in glaucoma. Am J Ophthalmol. 2008 ; 145: 191–192.
Wong TY, Hyman L. Population-based studies in ophthalmology. Am J Ophthalmol. 2008 ; 146: 656–663.
Kass MA, Heuer DK, Higginbotham EJ, et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical ocular hypotensive medication delays or prevents the onset of primary open-angle glaucoma. Arch Ophthalmol. 2002; 120: 701–713, discussion 829–830.
Miglior S, Zeyen T, Pfeiffer N, et al. Results of the European Glaucoma Prevention Study. Ophthalmology. 2005 ; 112: 366–375.
Chauhan BC, Nicolela MT, Artes PH. Incidence and rates of visual field progression after longitudinally measured optic disc change in glaucoma. Ophthalmology. 2009 ; 116: 2110–2118.
Medeiros FA, Alencar LM, Zangwill LM, Bowd C, Sample PA, Weinreb RN. Prediction of functional loss in glaucoma from progressive optic disc damage. Arch Ophthalmol. 2009 ; 127: 1250–1256.
Caprioli J. Clinical evaluation of the optic nerve in glaucoma. Trans Am Ophthalmol Soc. 1994 ; 92: 589–641.
Zeyen TG, Caprioli J. Progression of disc and field damage in early glaucoma. Arch Ophthalmol. 1993 ; 111: 62–65.
Sommer A, Katz J, Quigley HA, et al. Clinically detectable nerve fiber atrophy precedes the onset of glaucomatous field loss. Arch Ophthalmol. 1991 ; 109: 77–83.
Odberg T, Riise D. Early diagnosis of glaucoma. The value of successive stereophotography of the optic disc. Acta Ophthalmol (Copenh). 1985 ; 63: 257–263.
Ohkubo S, Higashide T, Udagawa S, et al. Focal relationship between structure and function within the central 10 degrees in glaucoma. Invest Ophthalmol Vis Sci. 2014 ; 55: 5269–5277.
Hood DC, Kardon RH. A framework for comparing structural and functional measures of glaucomatous damage. Prog Retin Eye Res. 2007 ; 26: 688–710.
Alasil T, Wang K, Yu F, et al. Correlation of retinal nerve fiber layer thickness and visual fields in glaucoma: a broken stick model. Am J Ophthalmol. 2014 ; 157: 953–959.
Wollstein G, Kagemann L, Bilonick RA, et al. Retinal nerve fibre layer and visual function loss in glaucoma: the tipping point. Br J Ophthalmol. 2012 ; 96: 47–52.
Naka M, Kanamori A, Tatsumi Y, et al. Comparison of mean deviation with AGIS and CIGTS scores in association with structural parameters in glaucomatous eyes. J Glaucoma. 2009 ; 18: 379–384.
Medeiros FA, Zangwill LM, Girkin CA, Liebmann JM, Weinreb RN. Combining structural and functional measurements to improve estimates of rates of glaucomatous progression. Am J Ophthalmol. 2012 ; 153: 1197–1205.e1.
Nassiri N, Nilforushan N, Coleman AL, Law SK, Caprioli J, Nouri-Mahdavi K. Longitudinal structure-function relationships with scanning laser ophthalmoscopy and standard achromatic perimetry. Arch Ophthalmol. 2012 ; 130: 826–832.
Lee P-J, Liu CJ-L, Wojciechowski R, Bailey-Wilson JE, Cheng C-Y. Structure-function correlations using scanning laser polarimetry in primary angle-closure glaucoma and primary open-angle glaucoma. Am J Ophthalmol. 2010 ; 149: 817–825.e1.
Pinto LM, Costa EF, Melo LAS, et al. Structure-function correlations in glaucoma using matrix and standard automated perimetry versus time-domain and spectral-domain OCT devices. Invest Ophthalmol Vis Sci. 2014 ; 55: 3074–3080.
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.
Caprioli J, Mock D, Bitrian E, et al. A method to measure and predict rates of regional visual field decay in glaucoma. Invest Ophthalmol Vis Sci. 2011 ; 52: 4765–4773.
Azarbod P, Mock D, Bitrian E, et al. Validation of point-wise exponential regression to measure the decay rates of glaucomatous visual fields. Invest Ophthalmol Vis Sci. 2012 ; 53: 5403–5409.
Lee J-W, Morales E, Yu F, et al. Effect of cataract extraction on the visual field decay rate in patients with glaucoma. JAMA Ophthalmol. 2014 ; 132: 1296–1302.
Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The definition and classification of glaucoma in prevalence surveys. Br J Ophthalmol. 2002 ; 86: 238–242.
Jeong JH, Park KH, Jeoung JW, Kim DM. Preperimetric normal tension glaucoma study: long-term clinical course and effect of therapeutic lowering of intraocular pressure. Acta Ophthalmol (Copenh). 2014 ; 92: e185–e193.
Jonas JB, Budde WM, Panda-Jonas S. Ophthalmoscopic evaluation of the optic nerve head. Surv Ophthalmol. 1999 ; 43: 293–320.
Anderson DR, Patella VM. Interpretation of a single field. In: Anderson DR, ed. Automated Static Perimetry, 2nd ed. St. Louis: Mosby; 1999: 121–190.
Thissen D, Steinberg L, Kuang D. Quick and easy implementation of the Benjamini-Hochberg procedure for controlling the false positive rate in multiple comparisons. J Educ Behav Stat. 2002 ; 27: 77–83.
Chauhan BC, Malik R, Shuba LM, Rafuse PE, Nicolela MT, Artes PH. Rates of glaucomatous visual field change in a large clinical population. Invest Ophthalmol Vis Sci. 2014 ; 55: 4135–4143.
Chauhan BC, Mikelberg FS, Artes PH, et al. Canadian Glaucoma Study: 3. Impact of risk factors and intraocular pressure reduction on the rates of visual field change. Arch Ophthalmol. 2010 ; 128: 1249–1255.
Fukuchi T, Yoshino T, Sawada H, et al. The relationship between the mean deviation slope and follow-up intraocular pressure in open-angle glaucoma patients. J Glaucoma. 2013 ; 22: 689–697.
Johnson CA, Adams AJ, Lewis RA. Evidence for a neural basis of age-related visual field loss in normal observers. Invest Ophthalmol Vis Sci. 1989 ; 30: 2056–2064.
Spry PG, Johnson CA. Senescent changes of the normal visual field: an age-old problem. Optom Vis Sci. 2001 ; 78: 436–441.
Gardiner SK, Johnson CA, Demirel S. Cup size predicts subsequent functional change in early glaucoma. Optom Vis Sci. 2011 ; 88: 1470–1476.
Zangwill LM, Weinreb RN, Beiser JA, et al. Baseline topographic optic disc measurements are associated with the development of primary open-angle glaucoma: the Confocal Scanning Laser Ophthalmoscopy Ancillary Study to the Ocular Hypertension Treatment Study. Arch Ophthalmol. 2005 ; 123: 1188–1197.
Rahman R, Casson RJ, Gouveia SM, Salmon JF. Optic disc morphology on presentation of chronic glaucoma. Eye (Lond). 2002 ; 16: 665–667.
Shigeeda T, Tomidokoro A, Araie M, Koseki N, Yamamoto S. Long-term follow-up of visual field progression after trabeculectomy in progressive normal-tension glaucoma. Ophthalmology. 2002 ; 109: 766–770.
Artes PH, O'Leary N, Hutchison DM, et al. Properties of the statpac visual field index. Invest Ophthalmol Vis Sci. 2011 ; 52: 4030–4038.
Smith SD, Katz J, Quigley HA. Analysis of progressive change in automated visual fields in glaucoma. Invest Ophthalmol Vis Sci. 1996 ; 37: 1419–1428.
Harwerth RS, Quigley HA. Visual field defects and retinal ganglion cell losses in patients with glaucoma. Arch Ophthalmol. 2006 ; 124: 853–859.
Henson DB, Artes PH, Chauhan BC. Diffuse loss of sensitivity in early glaucoma. Invest Ophthalmol Vis Sci. 1999; 40: 3147–3151.
Gardiner SK, Johnson CA, Demirel S. The effect of test variability on the structure-function relationship in early glaucoma. Graefes Arch Clin Exp Ophthalmol. 2012 ; 250: 1851–1861.
Racette L, Medeiros FA, Bowd C, Zangwill LM, Weinreb RN, Sample PA. The impact of the perimetric measurement scale, sample composition, and statistical method on the structure-function relationship in glaucoma. J Glaucoma. 2007 ; 16: 676–684.
Lamparter J, Russell RA, Schulze A, Schuff A-C, Pfeiffer N, Hoffmann EM. Structure-function relationship between FDF, FDT, SAP, and scanning laser ophthalmoscopy in glaucoma patients. Invest Ophthalmol Vis Sci. 2012 ; 53: 7553–7559.
Strouthidis NG, Gardiner SK, Sinapis C, Burgoyne CF, Garway-Heath DF. The spatial pattern of neuroretinal rim loss in ocular hypertension. Invest Ophthalmol Vis Sci. 2009 ; 50: 3737–3742.
DeLeón Ortega JE, Sakata LM, Kakati B, et al. Effect of glaucomatous damage on repeatability of confocal scanning laser ophthalmoscope, scanning laser polarimetry, and optical coherence tomography. Invest Ophthalmol Vis Sci. 2007 ; 48: 1156–1163.
Lee JM, Cirineo N, Ramanathan M, et al. Performance of the visual field index in glaucoma patients with moderately advanced visual field loss. Am J Ophthalmol. 2014 ; 157: 39–43.
Lee JM, Caprioli J, Nouri-Mahdavi K, et al. Baseline prognostic factors predict rapid visual field deterioration in glaucoma. Invest Ophthalmol Vis Sci. 2014 ; 55: 2228–2236.
Ernest PJ, Schouten JS, Beckers HJ, Hendrikse F, Prins MH, Webers CA. An evidence-based review of prognostic factors for glaucomatous visual field progression. Ophthalmology. 2013 ; 120: 512–519.
Alencar LM, Zangwill LM, Weinreb RN, et al. A comparison of rates of change in neuroretinal rim area and retinal nerve fiber layer thickness in progressive glaucoma. Invest Ophthalmol Vis Sci. 2010 ; 51: 3531–3539.
Zangwill LM, Jain S, Dirkes K, et al. The rate of structural change: the confocal scanning laser ophthalmoscopy ancillary study to the ocular hypertension treatment study. Am J Ophthalmol. 2013 ; 155: 971–982.
Kerrigan-Baumrind LA, Quigley HA, Pease ME, Kerrigan DF, Mitchell RS. 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.
Harwerth RS, Crawford MLJ, Frishman LJ, Viswanathan S, Smith EL, Carter-Dawson L. Visual field defects and neural losses from experimental glaucoma. Prog Retin Eye Res. 2002 ; 21: 91–125.
Caprioli J. Discrimination between normal and glaucomatous eyes. Invest Ophthalmol Vis Sci. 1992 ; 33: 153–159.
Medeiros FA, Leite MT, Zangwill LM, Weinreb RN. Combining structural and functional measurements to improve detection of glaucoma progression using Bayesian hierarchical models. Invest Ophthalmol Vis Sci. 2011 ; 52: 5794–5803.
Owen VMF, Strouthidis NG, Garway-Heath DF, Crabb DP. Measurement variability in Heidelberg Retina Tomograph imaging of neuroretinal rim area. Invest Ophthalmol Vis Sci. 2006 ; 47: 5322–5330.
Strouthidis NG, White ET, Owen VMF, Ho TA, Hammond CJ, Garway-Heath DF. Factors affecting the test-retest variability of Heidelberg retina tomograph and Heidelberg retina tomograph II measurements. Br J Ophthalmol. 2005 ; 89: 1427–1432.
Figure 1
 
Distributions of rates of change for global rim area (A), MD (B), VFI (C), SC (D), and FC (E) visual field components in entire study sample.
Figure 1
 
Distributions of rates of change for global rim area (A), MD (B), VFI (C), SC (D), and FC (E) visual field components in entire study sample.
Figure 2
 
Scatterplot showing the influence of baseline global rim area on the rate of global rim area change in entire study samples. The rate of global rim area change was negatively correlated with baseline global rim area (r = −0.30, P < 0.001).
Figure 2
 
Scatterplot showing the influence of baseline global rim area on the rate of global rim area change in entire study samples. The rate of global rim area change was negatively correlated with baseline global rim area (r = −0.30, P < 0.001).
Table 1
 
Demographic and Clinical Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 1
 
Demographic and Clinical Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 2
 
Baseline HRT Stereometric Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 2
 
Baseline HRT Stereometric Characteristics of the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 3
 
Changes in IOP, Number of Glaucoma Medications, and Global Visual Field Parameters Among the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 3
 
Changes in IOP, Number of Glaucoma Medications, and Global Visual Field Parameters Among the Study Groups Based on the Diagnosis and Glaucoma Severity of 465 Eyes of 338 Subjects
Table 4
 
The Rates of Structural and Functional Changes Among 465 Eyes of 338 Subjects With Different Severity of Glaucoma
Table 4
 
The Rates of Structural and Functional Changes Among 465 Eyes of 338 Subjects With Different Severity of Glaucoma
Table 5
 
Results from Mixed Effects Linear Regression Model for Association Between Structural Change as a Predictor and Functional Change as Each Individual Outcome in Separate Regression Models After Adjusting for Baseline Age, Baseline Global Rim Area, and Follow-Up Time
Table 5
 
Results from Mixed Effects Linear Regression Model for Association Between Structural Change as a Predictor and Functional Change as Each Individual Outcome in Separate Regression Models After Adjusting for Baseline Age, Baseline Global Rim Area, and Follow-Up Time
×
×

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.

×