Central Visual Field Defects in Patients with Distinct Glaucomatous Optic Disc Phenotypes

Abstract

PURPOSE:

To investigate central visual field (VF) defects among four phenotypes of glaucomatous optic discs.

DESIGN:

Cross-sectional study.

METHODS:

Optic disc phenotypes were determined in eyes with definite or suspected glaucoma that had a 24–2 VF with mean deviation (MD) better than −12 dB and a 10–2 VF. 10–2 VFs were classified as abnormal based on a cluster criterion. Additionally, the average of the total deviation values at each 10–2 test point was compared by optic disc phenotype.

RESULTS:

Four glaucomatous optic disc phenotypes were identified in 448 eyes of 309 patients: focal ischemic (FI)(n=121), generalized cup enlargement (GE)(n=109), myopic glaucoma (MY)(n=66), and senile sclerotic (SS)(n=152). Although VF 24–2 MD values were similar among optic disc phenotypes, GE eyes had higher VF 10–2 MD(p=0.004), as well as lower VF 24–2 pattern standard deviation(PSD)(p<0.001) and VF 10–2 PSD(p<0.001) compared to other phenotypes. The prevalence of an abnormal VF 10–2 was highest in FI eyes (78.5%) and lowest in GE eyes (50.5%)(p<0.001). In glaucoma suspects, the prevalence of an abnormal 10–2 VF was highest in the MY (31.2%) eyes and FI (23.5%) eyes and lowest in GE eyes (8.6%). In mild glaucoma, the prevalence of abnormal 10–2 VF tests was highest in FI eyes (79.2%) and lowest in GE eyes (44.4%)(p=0.013).

CONCLUSION:

The severity and prevalence of central visual field loss vary among different glaucomatous optic disc phenotypes. Glaucomatous eyes with FI and MY optic disc phenotypes are more likely to have 10–2 VF loss, particularly in early disease, and especially may benefit from testing with both 10–2 and 24–2 VF tests.

INTRODUCTION

Early detection and intervention is vital to reduce the progression and minimize loss of vision of glaucoma.¹ It has been suggested that different patterns of glaucomatous optic disc injury are related to distinctive pathophysiologic mechanisms. Nicolela and Drance^2,3 categorized glaucomatous eyes into four optic disc phenotypes based on their appearance, including focal ischemic (FI), generalized cup enlargement (GE), myopic glaucomatous (MY), and senile sclerotic (SS). Each phenotype is associated with several distinctive clinical features. For instance, while optic discs with generalized enlargement were associated with elevated levels of intraocular pressure (IOP), optic discs with focal notches were associated with localized visual field (VF) loss.^3–5

The detection of central VF loss in patients with glaucoma is important as it may affect more tasks such as reading, watching, recognizing faces, walking, and driving than peripheral VFs.⁶ When evaluating the central VF, the 10–2 test pattern is often used because it has denser and more test points within the central 10° of the VF than the 24–2 VF test pattern. Moreover, glaucomatous VF defects can be detected close to fixation even with early disease. Several studies suggest that these central VF changes detected by 10–2 VFs may be missed by 24–2 or 30–2 VFs.^7–10 Other studies suggest that using PSD or cluster criteria a similar number of eyes are identified with central VF damage by 24–2 VF and 10–2 VF testing; both tests can miss some central VF damage that the other test identifies.^11,12

Although the patterns of glaucomatous VF defects have been reported to be different among the four optic disc phenotypes,^2,5 these results were derived from analyses of solely 24–2 VF results and the central VFs were not fully evaluated. We hypothesized that each glaucomatous optic disc phenotype represents a unique degree and pattern of central VF defect. The purpose of the present study was to investigate central VF damage detected on 10–2 VF tests in four optic disc phenotypes of glaucoma. Moreover, we compared regional VF defect patterns and severity of central VF damage shown on 10–2 tests in definite and suspected glaucoma among different glaucomatous optic disc phenotypes.

METHODS

Glaucoma patients and healthy subjects were recruited from the longitudinal University of California, San Diego (UCSD) based Diagnostic Innovations in Glaucoma Study (DIGS) and the multicenter African Descent and Glaucoma Evaluation Study (ADAGES). The ADAGES is a collaboration of the UCSD Hamilton Glaucoma Center, Columbia University Irving Medical Center Edward S. Harkness Eye Institute (New York, NY), and the University of Alabama at Birmingham Department of Ophthalmology (Birmingham, AL). This report is a cross-sectional analysis. The institutional review boards at all sites approved the study methodology, which adheres to the tenets of the Declaration of Helsinki for research involving human subjects and to the Health Insurance Portability and Accountability Act. Informed consent was obtained from all participants.

Details of the DIGS and ADAGES protocols and eligibility have been described previously.¹³ In brief, all the patients underwent an annually comprehensive ophthalmologic examination, including review of medical history, best-corrected visual acuity, slit-lamp biomicroscopy, IOP measurement with Goldmann applanation tonometry, gonioscopy, dilated fundus examination, stereoscopic optic disc photography (Nidek 3Dx Stereo Camera [Nidek Inc, Palo Alto, CA] after maximal pupil dilation), and ultrasound pachymetry for central corneal thickness (CCT) measurements in both eyes. The semiannual examination included IOP, SD-OCT imaging, and both 24–2 and 10–2 VF testing.

Glaucoma patients in this study were defined as individuals who had the presence of glaucomatous optic neuropathy (GON) (defined as excavation, the presence of focal thinning, notching of neuroretinal rim, or localized or diffuse atrophy of the RNFL on the basis of masked grading of optic disc photographs by 2 graders or clinical examination by a glaucoma specialist and with (definite glaucoma) or without (glaucoma suspect) reliable (fixation losses ≤33%, false negatives ≤ 33% and false positives≤ 15%) and repeatable abnormal 24–2 VF tests. Inclusion criteria for this study were older than 18 years of age, open angles with gonioscopy, best-corrected visual acuity of 20/40 or better, and refraction less than ±5.0 diopters of sphere and 3.0 diopters of cylinder. Participants with a history of intraocular surgery (except for uncomplicated cataract surgery or uncomplicated glaucoma surgery), retinal pathologies including diabetic retinopathy and hypertensive retinopathy, non-glaucomatous optic neuropathy, uveitis, ocular trauma, Parkinson disease, Alzheimer disease, or stroke affecting VF were excluded. Those with unreliable VFs were also excluded from this report.

Participants who underwent automated VF testing using the 24–2 and 10–2 VF pattern on the Humphrey Field Analyzer (Carl Zeiss Meditec, Dublin, CA) within 6 months of imaging were enrolled. Although the 10–2 tests were not used at baseline to define the diagnostic groups, they had to meet the same reliability criteria as the 24–2 tests.

The quality of VF tests were reviewed by the Visual Field Assessment Center staff to identify and exclude VFs with evidence of inattention, inappropriate fixation, artifacts such as eyelid and lens rim artifacts, fatigue effects, macula pathology, and abnormal results caused by diseases other than glaucoma. Only those patients who had at least 2 reliable 10–2 VF were included. Glaucoma severity was classified based on the 24–2 only into glaucoma suspect, mild (MD ≥ −6 dB) and moderate (−12 ≤ MD < −6 dB) glaucomatous defects. Severity of glaucoma was restricted from early to moderate with MD ≥ −12 dB to avoid central field defects due to severe glaucoma.

Classifying 10–2 Visual Field Tests

For 10–2 VFs, the same reliability criteria were required. 10–2 VF was considered abnormal if the hemifield on the total deviation (TD) or pattern deviation (PD) plot was abnormal on 2 consecutive visits. Hemifields were classified as abnormal if there was a cluster of 3 contiguous points (5%, 5%, and 1% or 5%, 2%, and 2%) within a hemifield on either TD or PD plots.¹⁴ The specificity of this cluster criteria has been reported to be approximately 95%.¹⁴ The VF hemifields that were abnormal by this cluster test were classified into 3 categories based on the pattern/shape of abnormal points: (1) arcuate-like, (2) diffuse, and (3) other (Figure 1). The arcuate-like category included arcuate (a continuous, dense defect that involves both quadrants); partial arcuate (a continuous defect that involves both quadrants, but is less dense than an arcuate); and nasal (a defect restricted largely to the nasal quadrant).^14,15 The diffuse category was defined as a loss in all 4 quadrants on TD and PD plots that does not appear arcuatelike. Abnormal hemifields that didn’t fall into either of these categories were classified as other, which were predominately scattered across the whole field or predominately located in the temporal quadrants.¹⁴ Moreover, the average of the total deviation values at each 10–2 test point were compared by optic disc phenotype.

Figure 1.

Examples of 10–2 Visual Field (VF) Categories. All VFs are presented in the right-eye view. The graphs show VFs with loss consistent with an arcuate-like pattern (A), VF with diffuse loss (B), and VFs with temporal loss classified as other (C).

Optic disc Phenotype Assessment

Disc types of glaucomatous and suspect eyes were classified into four groups as described by Nicolela and Drance or as mixed or unclassified phenotypes^2,5 (Figure 2). With focal ischemic (FI), the optic disc had focal loss of neuroretinal rim with the other areas being normal. With generalized cup enlargement (GE), there was a large and deep concentric circular cup without a localized defect of the neuroretinal rim. With myopic glaucomatous (MY), there was temporal parapapillary atrophy with temporal cupping and a slightly tilted and ellipsoid optic disc. With senile sclerotic (SS), there was an atrophic halo (chorioretinal atrophy) around the optic disc, saucerized and shallow cupping, and a ‘moth-eaten’ and pale neuroretinal rim. For combination optic disc phenotypes , the optic disc was classified as mixed/unclassified. Two glaucoma specialists (E.E., S.M.), who were masked to the participant’s identity, diagnostic status, race, and other results, classified the disc types independently. If initial classifications did not agree, the consensus was obtained after the photographs were reviewed again. Any optic disc in which consensus could not be reached was categorized as unclassified. Mixed /unclassified eyes (n = 80) were not included in the statistical analysis.

Figure 2.

Glaucomatous optic disc types classified into four groups as described by Nicolela and Drance.

Statistical Analysis

Descriptive statistics were calculated as the mean (95% confidence interval) for continuous variables and count (percentage) for categorical variables. The statistical significance of continuous and categorical patient-level variables was compared among different phenotypes with ANOVA and chi-squared tests, respectively. Eye-level continuous characteristics were compared across groups with generalized estimating equation (GEE) models, assuming an exchangeable working correlation matrix to account for within-patient clustering. Similarly, logistic GEE models were used to assess the rates of abnormal 10–2 VF defect among the optic disc phenotypes, with and without additional effects for age and 24–2 VF MD in the model. We consider p-values less than 0.05 to indicate statistical significance. All analyses were performed using the statistical software R (version 3.6.1).

RESULTS

448 eyes of 309 glaucoma patients were included in the analysis. 94 eyes were glaucoma suspects and 354 eyes had definite glaucoma. The inter-examiner agreement on classification into FI, GE, MY, SE, or mixed/unclassified disc appearance types was good with a Kappa value of 0.83 (95% CI: 0.81 – 0.84).

Demographic and ocular characteristics of the study population are presented in Table 1. There was no significant difference in age, axial length (AL), spherical equivalent, and Bruch’s membrane opening (BMO) area between glaucoma suspect, and the definite glaucoma groups. The definite glaucoma group had a higher proportion of males and African American, lower VF 10–2 MD, lower VF 24–2 MD, greater VF 10–2 PSD, and greater VF 24–2 PSD compared to the glaucoma suspects.

Table 1.

Patient and eye characteristics in the study groups.

		Classification
	Glaucoma suspect (n= 41, eye=94)	Glaucoma (n= 268, eye=354)	p-value
Patient Characteristics
Age	69.8 (66.8, 72.9)	73.0 (71.6, 74.4)	0.133
Sex
Female	30 (73.2%)	131 (48.9%)	0.021
Male	11 (26.8%)	137 (51.1%)
Race
African Descent	12 (29.3%)	108 (40.3%)	0.016
European Descent/Other	29 (70.7%)	160 (59.7%)
Eye Characteristics
IOP	15.8 (14.6, 17.0)	14.1 (13.5, 14.7)	0.008
AL	24.1 (23.9, 24.2)	24.1 (24.0, 24.2)	0.743
SE	−0.31 (−0.69, 0.06)	−0.44 (−0.63, −0.24)	0.836
BMO Area	2.09 (1.96, 2.22)	2.16 (2.09, 2.23)	0.512
VF 24–2 MD	−1.38 (−1.87, −0.89)	−9.65 (−10.59, −8.72)	<0.001
VF 24–2 PSD	1.97 (1.74, 2.20)	7.39 (6.97, 7.81)	<0.001
VF 10–2 MD	−0.84 (−1.28, −0.40)	−8.82 (−9.73, −7.91)	<0.001
VF 10–2 PSD	1.55 (1.30, 1.79)	7.23 (6.69, 7.78)	<0.001
Phenotype
FI	17 (18.1%)	104 (29.4%)
GE	35 (37.2%)	74 (20.9%)
MY	16 (17.0%)	50 (14.1%)
SS	26 (27.7%)	126 (35.6%)

Abbreviations: FI,focal ischemic; GE,generalized cup enlargement; MY,myopic glaucomatous; SS,senile sclerotic disc; VF, visual field; MD, mean deviation; PSD, pattern standard deviation; IOP,intraocular pressure; AL,axial length; SE,spherical equivalent; BMO,Bruch’s membrane opening

Continuous data is presented as mean (95% confidence interval) and categorical data is presented as count (percentage). Significance is determined by ANOVA and chi-squared tests for patient level continuous and categorical data, and by generalized estimating equation (GEE) models for eye level data.

Among 448 glaucomatous (definite or suspect) eyes, the optic discs of 121 (27%) eyes were classified as FI, 109 (24%) eyes as GE, 66 eyes as MY (15%), and 152 (34%) eyes as SS (Table 2). SS eyes are significantly older (p<0.001) compared to other optic disc phenotypes. Race distribution differed among the phenotypes and the GE group had more patients of African descent than the others (p=0.002). AL was the highest (p=0.002) and BMO area was the lowest (p=0.041) in MY eyes. There was no significant difference in sex, IOP, or spherical equivalent, among different optic disc phenotypes. Although VF 24–2 MD were comparable among glaucomatous optic disc phenotypes (p=0.260), GE eyes had the greatest VF 10–2 MD (p=0.004), the lowest VF 24–2 PSD (p<0.001), and lowest VF 10–2 PSD (p<0.001).

Table 2.

Patient and eye characteristics across different glaucoma phenotypes.

	Phenotype				p-value
	FI (n=89, Eye=121)	GE (n=72, Eye=109)	MY (n=41, Eye=66)	SS (n=107, Eye=152)
Patient Characteristics
Age	70.2 (67.8, 72.6)	69.6 (66.8, 72.5)	71.9 (68.9, 74.9)	76.7 (74.8, 78.7)	<0.001
Sex
Female	48 (53.9%)	38 (52.8%)	21 (51.2%)	54 (50.5%)	0.967
Male	41 (46.1%)	34 (47.2%)	20 (48.8%)	53 (49.5%)
Race
African Descent	39 (43.8%)	38 (52.8%)	16 (39.0%)	27 (25.2%)	0.002
European Descent / Other	50 (56.2%)	34 (47.2%)	25 (61.0%)	80 (74.8%)
Eye Characteristics IOP	14.6 (13.7, 15.5)	15.0 (13.9, 16.1)	14.2 (12.4, 16.0)	14.0 (13.1, 15.0)	0.557
AL	23.9 (23.8, 24.1)	24.0 (23.9, 24.2)	24.3 (24.0, 24.5)	24.2 (24.0, 24.4)	0.002
SE	−0.38 (−0.69, −0.08)	−0.25 (−0.52, 0.01)	−0.82 (−1.31, −0.32)	−0.38 (−0.70, −0.07)	0.212
BMO Area	2.11 (2.02, 2.21)	2.18 (2.08, 2.28)	2.01 (1.88, 2.13)	2.21 (2.10, 2.33)	0.041
VF 24–2 MD	−7.65 (−8.84, −6.46)	−6.90 (−8.51, −5.28)	−8.82 (−10.91, −6.73)	−8.80 (−10.26, −7.34)	0.260
VF 24–2 PSD	7.20 (6.46, 7.94)	4.90 (4.23, 5.58)	6.45 (5.38, 7.52)	6.48 (5.78, 7.18)	<0.001
VF 10–2 MD	−7.60 (−8.83, −6.37)	−5.07 (−6.47, −3.67)	−8.05 (−10.31, −5.78)	−8.18 (−9.64, −6.73)	0.004
VF 10–2 PSD	7.76 (6.78, 8.74)	4.47 (3.68, 5.27)	5.79 (4.60, 6.97)	6.02 (5.24, 6.81)	<0.001

Abbreviations: FI,focal ischemic; GE,generalized cup enlargement; MY,myopic glaucomatous; SS,senile sclerotic disc; VF, visual field; MD, mean deviation; PSD, pattern standard deviation; IOP,intraocular pressure; AL,axial length; SE,spherical equivalent; BMO,Bruch’s membrane opening

Continuous data is presented as mean (95% confidence interval) and categorical data is presented as count (percentage). Significance is determined by ANOVA and chi-squared tests for patient level continuous and categorical data, and by generalized estimating equation (GEE) models for eye level data.

No significant difference in VF 10–2 and VF 24–2 global indices was found among optic disc phenotypes in the glaucoma suspects. In glaucoma group, VF 24–2 MD was comparable among the phenotypes. However, VF 24–2 PSD (p=0.004) and VF 10–2 PSD (p<0.001) were the lowest in the GE eyes (6.11 (5.31, 6.91) dB and 5.59 (4.55, 6.62) dB, respectively) and highest in the FI eyes (7.94(7.23, 8.64) dB and 8.55(7.58, 9.52) dB, respectively). In mild glaucoma, while there was no significant group-difference in VF 24–2 MD among optic disc phenotypes, VF 10–2 MD (p<0.001) was the lowest and VF 10–2 PSD (p<0.001) was the highest in the FI eyes. In moderate glaucomatous eyes, no significant group-difference in VF 24–2 MD and VF 10–2 MD was found. However, VF 24–2 PSD (p=0.002) and VF 10–2 PSD (p<0.001) were the lowest in GE eyes and highest in FI eyes in these eyes. (Table 3)

Table 3.

Visual field characteristics of different optic disc glaucoma phenotype across disease severity

	FI	GE	MY	SS	P-value
Glaucoma suspect eyes	n=17	n=35	n=16	n=26
VF 24–2 MD	−0.59 (−1.14,−0.05)	−0.18 (−0.60, 0.24)	−1.31 (−2.20,−0.42)	−0.64 (−1.28,−0.00)	0.097
VF 24–2 PSD	1.91 (1.64, 2.18)	1.85 (1.51, 2.19)	1.91 (1.53, 2.30)	2.00 (1.64, 2.35)	0.950
VF 10–2 MD	0.13 (−0.35, 0.60)	−0.04 (−0.60, 0.53)	−1.00 (−1.82,−0.19)	−0.53 (−1.29, 0.23)	0.068
VF 10–2 PSD	1.27 (1.15, 1.39)	1.39 (1.31, 1.47)	1.42 (1.18, 1.67)	1.43 (0.97, 1.88)	0.337
VF 10–2 Foveal Sensitivity	36.4 (35.3, 37.4)	35.8 (35, 36.6)	35.3 (33.7, 36.8)	34.5 (33.6, 35.4)	0.319
Glaucoma eyes	n=104	n=74	n=50	n=126
VF 24–2 MD	−8.44 (−9.75,−7.14)	−9.36 (−11.47,−7.25)	−11.02 (−13.38,−8.66)	−10.51 (−12.07,−8.96)	0.109
VF 24–2 PSD	7.94 (7.23, 8.64)	6.11 (5.31, 6.91)	7.83 (6.79, 8.88)	7.51 (6.80, 8.22)	0.004
VF 10–2 MD	−8.42 (−9.72,−7.12)	−6.98 (−8.83,−5.14)	−10.23 (−12.87,−7.58)	−9.78 (−11.34,−8.22)	0.079
VF 10–2 PSD	8.55 (7.58, 9.52)	5.59 (4.55, 6.62)	7.12 (5.84, 8.40)	7.11 (6.28, 7.94)	<0.001
VF 10–2 Foveal Sensitivity	35 (34.3, 35.7)	34.4 (33.5, 35.4)	34.3 (33.2, 35.3)	33 (32.2, 33.8)	0.2928
Mild glaucoma	n=48	n=36	n=21	n=46
VF 24–2 MD	−2.97 (−3.47,−2.48)	−2.81 (−3.25,−2.38)	−3.12 (−3.94,−2.30)	−3.18 (−3.66,−2.69)	0.711
VF 24–2 PSD	5.20 (4.37, 6.02)	3.84 (3.14, 4.54)	4.98 (3.94, 6.03)	3.86 (3.36, 4.35)	0.013
VF 10–2 MD	−3.97 (−5.02,−2.92)	−1.95 (−2.46,−1.44)	−2.63 (−4.00,−1.25)	−3.41 (−4.38,−2.45)	<0.001
VF 10–2 PSD	5.81 (4.45, 7.16)	2.58 (1.87, 3.28)	3.96 (2.43, 5.49)	3.96 (2.96, 4.97)	<0.001
VF 10–2 Foveal Sensitivity	35.7 (35.1, 36.3)	34.7 (33.5, 35.9)	35 (33.5, 36.4)	33.1 (32.2, 34.1)	0.065
Moderate glaucoma	n=32	n=13	n=10	n=30
VF 24–2 MD	−8.35 (−8.89,−7.80)	−8.17 (−9.09,−7.24)	−8.42 (−9.21,−7.64)	−8.45 (−8.99,−7.91)	0.961
VF 24–2 PSD	9.41 (8.39, 10.43)	7.08 (5.86, 8.29)	7.42 (6.62, 8.23)	8.89 (7.66, 10.11)	0.002
VF 10–2 MD	−8.75 (−12.30,−5.19)	−5.44 (−9.61,−1.28)	−5.20 (−14.46, 4.06)	−8.61 (−11.87,−5.35)	0.567
VF 10–2 PSD	9.90 (8.36, 11.43)	5.65 (4.07, 7.22)	6.83 (4.92, 8.73)	7.60 (6.03, 9.17)	<0.001
VF 10–2 Foveal Sensitivity	34 (32.6, 35.3)	33.6 (31.8, 35.3)	32.9 (31.7, 34.1)	32.9 (31.7, 34.1)	0.475

Abbreviations: FI,focal ischemic; GE,generalized cup enlargement; MY,myopic glaucomatous;; SS,senile sclerotic disc; VF, visual field; MD, mean deviation; PSD, pattern standard deviation

The prevalence of abnormal VF 10–2 clusters in the all glaucoma (defenite and suspected) eyes was the highest in FI (78.5%) eyes (p<0.001) and the lowest in GE eyes (50.5%). In glaucoma suspects, the MY (31.2%) and FI group (23.5%) had the highest and GE eyes (8.6%) had the lowest 10–2 abnormality. The prevalence of abnormal VF 10–2 tests in mild glaucoma was the lowest in GE eyes (44.4%) and the highest in FI (79.2%) eyes (p=0.013). (Table 4)

Table 4.

Prevalence of abnormal VF 10–2 by phenotype, stratified by severity group.

	Phenotype
	FI	GE	MY	SS	p-value
Glaucoma suspect	n = 17	n = 35	n = 16	n = 26	- (−)
Abnormal VF 10–2, n (%)	4 (23.5%)	3 (8.6%)	5 (31.2%)	3 (11.5%)
Mild glaucoma	n = 48	n = 36	n = 21	n = 46	0.013 (0.014)
Abnormal VF 10–2, n (%)	38 (79.2%)	16 (44.4%)	13 (61.9%)	31 (67.4%)
Moderate glaucoma	n = 32	n = 13	n = 10	n = 30
Abnormal VF 10–2, n (%)	29 (90.6%)	11 (84.6%)	10 (100.0%)	30 (100.0%)	- (−)
Total (Definite and suspected glaucoma)	n = 121	n = 109	n = 66	n = 152	<0.001 (0.005)
Abnormal VF 10–2, n (%)	95 (78.5%)	55 (50.5%)	47 (71.2%)	113 (74.3%)

Abbreviations: GVFD,glaucomatous visual field defect; GON,glaucomatous optic neuropathy; MY,myopic glaucomatous; SS,senile sclerotic disc; VF,visual field; MD,mean deviation; FI,focal ischemic; GE,generalized cup enlargement; MY, myopic glaucomatous; SS,senile sclerotic disc; VF,visual field; MD,mean deviation.

Significance is determined by a binomial GEE model (assuming an exchangeable working correlation matrix). The p-values in parenthesis are adjusted for VF 24–2 MD and age. Due to sparsity, no p-value is reported for GON only eyes or moderate glaucoma eyes.

Classification of 10–2 Visual Fields

The results of the 10–2 hemifields classification are shown in Table 5. In glaucomatous eyes (definite and suspected), the superior hemifield of VF 10–2 was affected more often than the inferior hemifield in all optic disc phenotypes and the arcuatelike pattern was the most common pattern of 10–2 VF defects. In mild glaucoma, similarly, the superior hemifield was affected more prominently in all optic disc phenotypes and the arcuate-like pattern was the most common pattern of 10–2 VF defect in superior and inferior hemifield in all glaucomatous optic disc phenotypes. Of note, the superior hemifield was more preferentially affected in the MY and FI group.

Table 5.

10–2 visual field defect location and pattern in eyes with glaucoma.

	Optic Disc Phenotype
	FI	GE	MY	SS
Mild glaucoma	(n=48)	(n=36)	(n=21)	(n=46)
Defect Location
None	10 (20.8%)	20 (55.6%)	8 (38.1%)	15 (32.6%)
Inferior Only	10 (20.8%)	5 (13.9%)	2 (9.5%)	7 (15.2%)
Superior Only	21 (43.8%)	7 (19.4%)	9 (42.9%)	10 (21.7%)
Both	7 (14.6%)	4 (11.1%)	2 (9.5%)	14 (30.4%)
Superior Pattern
None	20 (41.7%)	25 (69.4%)	10 (47.6%)	22 (47.8%)
Arcuate-Like	25 (52.1%)	10 (27.8%)	9 (42.9%)	20 (43.5%)
Diffuse	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
Other	3 (6.2%)	1 (2.8%)	2 (9.5%)	4 (8.7%)
Inferior Pattern
None	31 (64.6%)	27 (75.0%)	17 (81.0%)	25 (54.3%)
Arcuate-Like	14 (29.2%)	9 (25.0%)	3 (14.3%)	16 (34.8%)
Diffuse	0 (0.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
Other	3 (6.2%)	0 (0.0%)	1 (4.8%)	5 (10.9%)
Total (Definite and suspected glaucoma)	(n=121)	(n=109)	(n=66)	(n=152)
Defect Location
None	26 (21.5%)	54 (49.5%)	19 (28.8%)	39 (25.7%)
Inferior Only	21 (17.4%)	8 (7.3%)	6 (9.1%)	16 (10.5%)
Superior Only	34 (28.1%)	14 (12.8%)	15 (22.7%)	19 (12.5%)
Both	40 (33.1%)	33 (30.3%)	26 (39.4%)	78 (51.3%)
Superior Pattern
None	47 (38.8%)	62 (56.9%)	25 (37.9%)	55 (36.2%)
Arcuate-Like	70 (57.9%)	44 (40.4%)	33 (50.0%)	80 (52.6%)
Widespread	0 (0.0%)	0 (0.0%)	1 (1.5%)	6 (3.9%)
Other	4 (3.3%)	3 (2.8%)	7 (10.6%)	11 (7.2%)
Inferior Pattern
None	60 (49.6%)	68 (62.4%)	34 (51.5%)	58 (38.2%)
Arcuate-Like	51 (42.1%)	37 (33.9%)	27 (40.9%)	74 (48.7%)
Widespread	0 (0.0%)	0 (0.0%)	1 (1.5%)	6 (3.9%)
Other	10 (8.3%)	4 (3.7%)	4 (6.1%)	14 (9.2%)

Abbreviations: GVFD,glaucomatous visual field defect; GON,glaucomatous optic neuropathy; MY,myopic glaucomatous; SS,senile sclerotic disc; VF, visual field; MD, mean deviation; FI,focal ischemic; GE,generalized cup enlargement; MY,myopic glaucomatous; SS,senile sclerotic disc; VF,visual field; MD,mean deviation

Pattern of Glaucomatous Visual Field Damage

In mild glaucoma, the proportion of eyes with at least 1 abnormal point depressed <1% in the most central 16 points and most central 4 points of the 10–2 VF PD plot was lower in GE eyes [7 (19.4%), and 12 (25.0%), respectively] compared to FI eyes [30 (62.5%), and 12 (25.0%), respectively], MY eyes [12 (57.1%) , and 7 (33.3%), respectively], and SS eyes [27 (58.7%) , and 14 (30.4%), respectively](p=0.002, and p=0.183, respectively). The pattern of glaucomatous VF damage in the macula in different optic disc phenotypes is presented in Figure 3 using pseudocolors. In mild glaucoma, FI eyes and MY eyes were more severely affected. Furthermore, in these optic disc phenotypes, the deepest defects (yellow and red) were close to fixation in the superior VF and the superior papillomacular region (blue rectangles) and inferior hemifield were less affected. In all glaucoma cases (Figure 3, the second row), a similar pattern was found with FI and MY phenotypes demonstrating predomninant involvement of the superior VF. Although the 10–2 VF was affected more diffusely in SS and GE phenotypes, still superior VF defects close to fixation and inferior defects are at the nasal edge of the inferior field. Likewise, superior papillomacular region and inferior hemifield were less affected. Figure 4 depicts the average total deviation values for the 24–2 visual field. Similarly, the most central 12 and most central 4 most inner points were more affected in eyes with FI and MY phenotypes.

Figure 3.

Pseudocolor map for average total deviation values at each 10–2 location in mild glaucoma (Top) and all glaucoma (Bottom). The blue rectangle corresponds to the papillomacular region, which is preserved in the superior hemifield. Both maps are in the right-eye view.

Figure 4.

Pseudocolor map for average total deviation values at each 24–2 location in mild glaucoma (Top) and all glaucoma (Bottom). The red cross contains central 12 points and corresponds to the central 10 degree of visual field. Both maps are in the right-eye view.

DISCUSSION

Our results show that central VF damage was less common in eyes with the GE phenotype. In eyes classified as glaucoma suspect or mild glaucoma (MD better than −6 dB in 24–2 tests), abnormal 10–2 VFs were more common in FI and MY phenotypes despite no significant differences in the 24–2. Abnormal central visual fields were found in up to one quarter and three- quarters of eyes with FI phenotypes in glaucoma suspect and mild glaucoma, respectively. This information may help the clinicians to better understand the role of 10–2 VF tests in standard care glaucoma and provides clinical clues to predict the presence of parafoveal scotoma.

Different optic disc phenotypes have been found to be associated with distinct clinical characteristics that may be involved in glaucoma pathogenesis.^2,3 FI eyes in which there is a markedly focal change of the optic disc have highly characteristic VF changes.⁴ Discs of this type are characterized by localized neuroretinal rim tissue loss primarily at the inferior and/or superior pole of the optic nerve head (just temporal to the midline) and less frequent involvement of the temporal or nasal disc.³ The age of subjects with the FI phenotype is slightly younger than the mean age of all primary open-angle glaucoma patients, and women are more likely to be affected than men.² We found that the GE group had proportionately more patients of African descent than the other groups despite similar axial length and SE in the African and European descent groups . A previous study has shown that there are structural differences within the optic nerve complex between these groups and that African descent eyes were less likely to have focal damage compared to European descent eyes.¹⁶ In the present study, the FI phenotype in definite glaucoma were more frequent than in the glaucoma suspect. This is in line with previous inestigations that showed glaucomatous eyes with the FI phenotype had more rapid neuroretinal rim loss and VF progression compared to the other phenotypes.⁴

Different pathophysiologic mechanisms may be involved in producing the four types of optic disc phenotypes.^2,3 For example, mechanical distortion of the lamina cribrosa with impingement on the retinal ganglion cell axons might result in diffuse damage.¹⁷ Similarly, generalized compromise of optic nerve head blood flow also may result in diffuse damage.¹⁸ Localized damage may be related to a focal area of weakness in the lamina, or localized vascular event, such as one heralded by a disc hemorrhage or focal loss of choriocapillaris.^16,19,20 Localized optic disc change associated with glaucomatous paracentral scotomas lie closer to the papillomacular bundle than that due to peripheral VF loss.²¹ As another example, Sawada et al.²² reported that the severity of LC defects in myopic glaucoma usually correlated with the extent of disc tilting, and the location of LC defects usually were observed at the temporal region of the optic disc. With these observations, they suggest reasons why the central scotomas appear in the early stage in myopic glaucoma.

Early glaucomatous damage often involves the macula. Even in glaucoma suspect eyes, we observed abnormal central VF defects in as many as one-third of FI and MY eyes. This finding is in line with the study by de Moraes et al. which demonstrated that abnormal central VF were prevalent in 35% and 39% of ocular hypertensive and glaucoma suspect eyes, respectively, while 21.5%, and 24.8% classified as normal based on 10–2 were classified as abnormal by the 24–2.²³ In another study, Traynis et al.¹⁰ have shown that as many as 16% of eyes with a normal 24–2 VF result had significant abnormalities on a 10–2 VF in eyes with early glaucoma. Park et al²⁴ found that 74% of eyes had a parafoveal scotoma detected on the 10–2 VF test in a population with mild glaucoma. The present study demonstrated that 65% of eyes with mild glaucoma had abnormal 10–2 VF test results and central VF was more affected in the FI group compared to other glaucomatous optic disc phenotypes. Although the foveal sensitivities were similar among the phenotypes, FI, MY and SS groups had proportionately more eyes with abnormal points in most central 4 and 16 points of 10–2 VF compared to GE group. Similarly, Nicolela and Drance evaluated 24–2 VF tests in distinct glaucomatous optic disc phenotypes and demonstrated that eyes in FI and MY groups typically had dense localized scotomas; fixation was frequently threatened by these scotomas, particularly in the FI group (81%).² The marked predominance of superior scotomas in the FI group corresponded to the great frequency of focal loss in the inferior pole of the disc.² These studies were done using 24–2 and may underestimate central involvement. Early glaucomatous damage near fixation can be relatively subtle, as it involves a loss of only a few dB of sensitivity, very local and/or diffuse.^25–27 Patients with glaucoma with VF defects within 5 degrees of fixation are at greater risk of losing visual acuity and global VF,^28,29 and VF defects within 3 degrees of fixation in more than 2 adjacent quadrants may cause reading difficulty.³⁰ The macula is particularly important for daily functioning and loss in that region is strongly associated with self-reported quality of life.^31,32 A recent study has shown that the 10–2 VF was abnormal in nearly as many hemifields as was the 24–2 VF, including some with normal 24–2 VF, suggesting that the 24–2 test is not optimal for detecting early damage of the macula. They recommended 10–2 exams or their equivalents are necessary to detect central abnormalities and early glaucomatous damage.³³ Others have suggested that close examination of the 24–2 based on PSD or cluster criteria can detect a similar number of eyes with central VF damage.^11,12

Our findings have important clinical implications. Patients with central VF damage have functional loss which can affect their quality of life. They also have advanced glaucomatous damage according to many classification systems which consider the presence of central damage a criteria for severe disease^31,32 Notably, patients in the FI group of the current study were more likely to have central VF abnormalities especially in early stages of the disease, which underscores the importance of clinically evaluating the optic disc phenotypes.

In contrast to the results found in the FI and MY optic disc phenotypes, only 10 % and 40% of eyes with glaucoma suspect and mild glaucoma had central VF damage, respectively, with the GE optic disc phenotype; this was significantly lower than observed with other optic disc phenotypes. Remarkably, the severity of central functional damage also was lower in GE phenotypes in mild and moderate glaucoma as indicated by significantly better 10–2 MD. Similar findings have been shown by previous investigators.² Specifically, in a cross-sectional study diffuse VF loss in 24–2 VF was the only finding in 40% of eyes with GE optic disc phenotype.² However, once glaucomatous eyes become moderately damaged, most of the eyes showed central VF defect with 10–2 VF tests. In the current study, the superior hemifield of VF 10–2 was preferentially affected in the MY and GE groups. In both definite and glaucoma suspect eyes, an arcuate-like pattern was the most common pattern of 10–2 VF defect in each of the glaucomatous optic disc phenotypes. This is in agreement with Traynis et al. who found that more than two-thirds of abnormal central visual defects were arcuate-like in eyes with early glaucoma.¹⁰ Our results are also supported by the model proposed by Hood et al.³ This model assumes that the retinal ganglion cells (RGCs) in the inferior retinal region of the macula largely project to the most vulnerable (inferior) region of the disc and can be damaged early in the glacoma. The arcuate defects seen in the upper macular VF are associated with arcuate RNFL defects that are centered primarily in the macular vulnerable zone (MVZ) of the disc. The severity of the damage was also greater in superior central hemifield as observed by heatmap plots, and superior arcuate VF defect was the most common pattern seen in these groups.

There are some limitations to the current study. First, optic disc phenotypes were based on subjective observations, although the two experienced graders in this study had good inter-observer agreement in determining the final optic disc phenotype (κ = 0.83). Second, the differences in some ocular and demographic characteristics among the optic disc phenotype groups may suggest that all the factors that might affect central VF damage were not adequately controlled despite adjusting for confounders in multivariable analysis. Finally, the study was a cross-sectional study that did not evaluate subjects over time, limiting our understanding of the relationship between changes in central VF damage and optic disc phenotypes that a longitudinal study would provide.

In conclusion, macular damage in early glaucoma, as detected with 10–2 VFs, appears more frequently in FI and MY phenotypes than in the GE phenotype despite similar 24–2 results. Central VF damage was less frequent and less severe in eyes with GE glaucomatous optic disc phenotype compared to the other groups. The pattern of central visual field damage shown in 10–2 tests was mostly arcuate-like in each of the optic disc phenotypes. Assessing the 10–2 VF, particularly in those patients with FI and MY optic disc phenotypes, will enhance glaucoma diagnosis and management.

Highlights:

• The severity and prevalence of central visual field loss vary among different glaucomatous optic disc phenotypes.

• Glaucomatous eyes with focal ischemic and myopic optic disc phenotypes are more likely to have 10–2 visual field loss, particularly in early disease.

• Glaucomatous eyes with focal ischemic and myopic optic disc phenotypes optic disc phenotypes may benefit from testing with both 10–2 and 24–2 visual field tests, particularly in early disease.

Synopsis:

This cross-sectional study demonstrates that the severity and prevalence of central visual field loss varies among different glaucomatous optic disc phenotypes and Glaucomatous eyes with focal ischemic and myopic glaucoma phenotypes especially may benefit from testing with both 10–2 and 24–2 visual field tests.