Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jun 1.
Published in final edited form as: Curr Eye Res. 2020 Oct 27;46(6):895–902. doi: 10.1080/02713683.2020.1836229

Characterization of Prelaminar Wedge-Shaped Defects in Primary Open Angle Glaucoma

Carolina A Chiou 1, Mengyu Wang 1,2, Elise V Taniguchi 1,3, Rafaella Nascimento e Silva 1, Anna Khoroshilov 1, Dian Li 1,2, Haobing Wang 1, Scott H Greenstein 1, Stacey C Brauner 1, Angela V Turalba 1,4, Louis R Pasquale 5, Lucy Q Shen 1
PMCID: PMC8767987  NIHMSID: NIHMS1769751  PMID: 33054505

Abstract

Purpose

To determine the clinical relevance of prelaminar wedge defects (PLWDs) detected by swept-source optical coherence tomography (SS-OCT) in primary open angle glaucoma (POAG).

Materials and Methods

In this retrospective case-control study, PLWDs were defined as triangular-shaped defects at the surface of the optic nerve prelaminar tissue, not adjacent to blood vessels, present on cross-sectional SS-OCT scans. Two observers masked to diagnosis independently reviewed scans to detect PLWDs and lamina cribrosa defects. History of disc hemorrhage, occurring within 2 years prior to imaging, was obtained from chart review. One eye per subject was randomly selected. Two-sided t-tests, analysis of variance with Bonferroni correction, and multivariable logistic regression analysis were performed to explore demographic and clinical features associated with PLWDs.

Results

40 POAG and 23 control eyes were included. PLWDS were found in 27.5% of POAG (n=11) and 4.3% of controls (n=1, p=0.04). Eyes with repeat SS-OCT imaging (7 POAG and 0 controls) had persistent PLWDs. More POAG eyes with PLWDs had a history of disc hemorrhage (45.5%) than POAG eyes without PLWDs (3.4%, p=0.004). On multivariable analysis, compared to POAG without PLWDs, POAG with PLWDs had increased odds of observed disc hemorrhage (OR=21.6, 95% CI, 2.2–589.0, p=0.02) after adjusting for age, gender, visual field mean deviation and maximum intraocular pressure (IOP). POAG with PLWDs had more lamina cribrosa defects (45.5%) than POAG without PLWDs (3.4%, p=0.01) but did not differ significantly from controls (8.7%, p=0.07). Compared to all patients without PLWDs, patients with PLWDs had increased odds of having lamina cribrosa defects (OR=44.8; 95% CI, 6.3–703.6, p<0.001) after adjusting for age, gender, and maximum IOP.

Conclusions

PLWDs were more frequently found in POAG than control eyes and were associated with a history of disc hemorrhage and lamina cribrosa defects. PLWDs may be a useful imaging biomarker of glaucomatous damage.

Keywords: prelaminar tissue, primary open angle glaucoma, swept-source OCT, prelaminar wedge defects, lamina cribrosa defects, disc hemorrhage

Introduction

Glaucoma is a heterogeneous group of diseases that result in progressive loss of retinal ganglion cells and axons,1 manifesting as optic nerve head (ONH) excavation on fundoscopic examination. This structural appearance is secondary to changes in the prelaminar and lamina cribrosa (LC) tissue.2, 3 The prelaminar tissue, located within Bruch’s membrane opening (BMO) of the ONH between the internal limiting membrane and anterior LC surface,2 is composed of neuronal, connective and vascular tissue, and is an important structure in glaucoma pathophysiology.4 Quantitative changes in prelaminar tissue, such as decreased thickness, have been observed in glaucoma.5

The new generation of imaging devices has allowed for high resolution and detailed characterization of ONH microarchitecture. Compared to conventional spectral-domain optical coherence tomography (SD-OCT), swept-source OCT (SS-OCT) has a longer wavelength centered at 1050 nm and higher scanning speed ranging from 100,000 to 400,000 axial scans per second,6, 7 allowing for deeper tissue penetration while minimizing motion artifacts. Furthermore, this device can simultaneously image superficial and deep tissue layers in high resolution enabling differentiation between tissue defects and vascular shadowing artifacts.8 In a previous SS-OCT study, we identified focal triangular-shaped defects on the surface of the prelaminar layer, which we termed prelaminar wedge defects (PLWDs). These defects were more prevalent in primary open angle glaucoma (POAG) than chronic angle closure glaucoma.9

Focal prelaminar changes, which are difficult to discern clinically due to the translucent nature of the neuronal tissue, have not been fully characterized in POAG. We hypothesize that PLWDs, morphologic abnormalities observed in the ONH prelaminar tissue in SS-OCT scans, carry clinical significance and may become a biomarker for glaucomatous damage in POAG.

Materials and Methods

This is a retrospective study approved by the Institutional Review Board of Massachusetts Eye and Ear (MEE) conducted in compliance with the Health Insurance Portability and Accountability Act and adhered to the tenets of the Declaration of Helsinki.

Description of the Study Population

Adult patients with POAG and control subjects were initially recruited from the Glaucoma Consultation Service and Comprehensive Ophthalmology Service of MEE between September 2013 and March 2016 for a previous prospective study9 and were consecutively included in the current retrospective study. The study subjects had best corrected visual acuity (BCVA) of 20/50 or better in both eyes and spherical equivalence between +6 diopters (D) and −6 D. Patients with significant retinal pathology, such as age-related macular degeneration or diabetic retinopathy, were excluded. All subjects provided informed written consent, and were imaged with SS-OCT (Deep Range Imaging [DRI] OCT-1, Topcon, Tokyo, Japan) and SD-OCT (Spectralis, Heidelberg Engineering GmbH, Heidelberg, Germany).

Primary Open Angle Glaucoma Definition

Inclusion criteria were: (1) open angles on gonioscopy, (2) glaucomatous ONH damage characterized by neuroretinal rim thinning, and (3) retinal nerve fiber layer (RNFL) defects with corresponding reproducible functional loss on Humphrey Visual Field (HVF, Carl Zeiss Meditec, Dublin, CA) 24-2 tests on the Swedish Interactive Thresholding Algorithm. Exclusion criteria were: (1) secondary open angle glaucoma, such as exfoliation glaucoma and pigmentary glaucoma, (2) unreliable HVF with fixation loss >33% or false positive and false negative >20%, (3) HVF mean deviation (MD) worse than −12dB, (4) prior penetrating glaucoma surgery such as trabeculectomy and glaucoma drainage device implantation, due to their reported effects on LC depth,10 (5) non-glaucomatous optic nerve disease and retinal disease(s) causing visual field loss, and (6) optic disc torsion (longest axis rotation) of ≥15° outside the vertical meridian and/ or tilt ratio between the shortest and longest disc diameter of <0.75.11,12

Control Definition

Inclusion criteria were: (1) maximum intraocular pressure (IOP) <21 mmHg, (2) no family history of glaucoma, (3) no history of glaucomatous damage such as a disc hemorrhage, (4) normal optic nerve appearance on exam with a cup-to disc ratio (CDR) ≤0.6 and no localized thinning of the neuroretinal rim, (5) CDR asymmetry ≤0.2 in both eyes, and (6) retinal nerve fiber layer thickness within normal limits on SD-OCT. The criteria for no-glaucoma status was based on optic disc evaluation and IOP only, as HVFs were not performed for control subjects. The following exclusion criteria are the same as for the POAG group: (1) non-glaucomatous optic nerve disease and retinal disease(s) causing visual field loss and (2) optic disc torsion (longest axis rotation) of ≥15° outside the vertical meridian and/or tilt ratio between the shortest and longest disc diameter of <0.75.

Image Analysis

Based on the fixed scanning patterns from the manufacturer, SS-OCT images from twelve radial scans spaced 15° apart and ten 5-line cross scans with 5 horizontal and 5 vertical lines spaced 250 μm apart (online supplementary fig 1) were used to evaluate the prelaminar and laminar tissue. All images were obtained through a dilated pupil and had 32 frames averaged to achieve best image quality. Two observers (CAC, LQS) masked to diagnosis independently reviewed the images for defects in the prelaminar and laminar tissues. Adjudication was performed by a third glaucoma specialist (EVT) when there was a disagreement. Specifically, PLWDs were defined as focal thinning within the prelaminar tissue, which includes the descending fibers from Bruch’s membrane opening (BMO) and the tissue anterior to the LC, where the curved and smooth delineation of the cup contour is lost and a triangular defect can be outlined (fig 1). The PLWD may or may not reach the anterior surface of the LC. Ambiguous PLWDs, i.e. a defect immediately adjacent to a blood vessel with no visible separation from the vessel and that may be secondary to distortion of the cup contour from the vessel, were excluded. Images of patients included in the study with subsequent SS-OCT scans were evaluated for persistence of previously identified PLWDs and new PLWDs. LC defects were defined as discontinuities of the anterior LC surface on SS-OCT not due to blood vessels or vascular shadowing (fig 1; online supplementary fig 2).

Figure 1: Examples of Eyes with Primary Open Angle Glaucoma and Prelaminar Wedge Defects.

Figure 1:

Open in a new tab

In both examples, unmarked images are shown in the top row and annotated images are shown below.

Example 1: Humphrey visual field (HVF) pattern deviation plot (a1) and corresponding retinal nerve fiber layer (RNFL) thickness scan from spectral-domain optical coherence tomography (OCT) (a2) of a right eye. Thinnest RNFL thickness of 37 μm is located in the inferotemporal sector. Disc photo of the same eye (b1), where the prelaminar wedge defect (PLWD) is located at the inferotemporal border of the optic nerve as indicated by the white circle (b2). The PLWD, a focal thinning of the prelaminar tissue in the shape of a triangular defect and not adjacent to a blood vessel, is identified on the swept-source OCT radial scan (c1) as indicated by the red arrow (c2). The orientation of the SS-OCT scan is depicted by green arrows (b2, c2, d2). The lamina cribrosa (LC) defect is contiguous with the PLWD in the color-inverted SS-OCT image (d1) and is indicated by black arrowheads with the anterior LC surface outlined by the blue dashed line (d2).

Example 2: HVF pattern deviation plot (e1) and corresponding RNFL thickness scan from spectral-domain OCT (e2) of a right eye. The thinnest RNFL thickness of 60 μm is located in the inferotemporal sector. Disc photo of the same eye (f1), where the PLWD is located at the superior border of the optic nerve as indicated by the white circle (f2). The PLWD is a triangular-shaped defect identified on the vertical SS-OCT scan (g1) and indicated by the red arrow (g2). The orientation of the SS-OCT scan is depicted by green arrows (f2, g2, h2). The LC defect is contiguous with the PLWD as shown in the color-inverted SS-OCT image (h1) and is indicated by black arrowheads with the anterior LC surface outlined by the blue dashed line (h2). The LC defect has a slightly globular shape when compared with the PLWD.

Quantifications of the ONH were performed independently by two observers (CAC, EVT) masked to diagnoses utilizing SS-OCT scans and customized ImageJ plugins (ImageJ, National Institutes of Health, Bethesda, Maryland).13 In brief, BMO-MRW was measured manually as the minimum distance between the internal limiting membrane and BMO (online supplementary fig 3a). Global BMO-MRW was the average of 24 measurements from 12 radial scans; the minimum BMO-MRW was the lowest value among these 24 measurements of BMO-MRW per eye; scaled BMO-MRW was normalized by disc circumference according to a published method.14 Horizontal lamina cribrosa depth (LCD) was measured from the BMO distance line to the anterior surface of the LC on a horizontal scan centered on the ONH and calculated as the mean of perpendicular lines spaced 100 μm apart over the entire ONH surface (online supplementary fig 3b).15 RNFL thickness measurements were not available from SS-OCT at the time of imaging. Hence, average RNFL thickness was acquired directly from SD-OCT scans, while the clock-hour sector with the lowest value was used as the thinnest RNFL thickness (fig 1 a2 and e2).

Clinical Data Collection

Clinical data including demographic and baseline characteristics were collected prior to the determination of PLWDs. Medical records and disc photos (Visucam Pro NM, Carl Zeiss, Meditec, Dublin, CA, USA) two years prior to initial SS-OCT imaging were reviewed for documentation and location (superior vs. inferior) of disc hemorrhages (DHs). Paracentral visual field loss was defined as a cluster of 3 or more contiguous points each with pattern deviation value of −5 dB or worse or two contiguous points with combined pattern deviation value of −15 dB or worse within the central 10° on the pattern deviation plot from HVF 24–2 tests.16

Statistical Analysis

One eye per subject was included. If both eyes met inclusion criteria, one eye was randomly selected. R language (version 3.4.3, R Foundation, Vienna, Austria) was utilized for statistical analyses. For continuous variables, two-sided t-test was performed between two groups and analysis of variance (ANOVA) between three groups. Fisher’s exact test was utilized for categorical variables. Bonferroni correction was applied for multiple comparisons to adjust for the alpha error. Multivariable logistic regression analyses were performed to assess the association between PLWD and a history of DH or the presence of LC defect, while adjusting for systemic and ophthalmic characteristics. All patients without PLWDs served as reference for the analysis on LC defects, whereas POAG patients without PLWDs served as reference for the analysis on DH, which was an exclusion criterion for control subjects. Statistical significance was considered at a p-value <0.05.

RESULTS

Seventy subjects were initially included in this study; of those, 7 (6 POAG, 1 control; 10.0%) were excluded due to ambiguous PLWDs. Therefore, 40 POAG and 23 control subjects were included. PLWDs were found in 27.5% (n=11) of POAG patients versus 4.3% (n=1) of control subjects (p=0.04). The inter-reader agreement for assessing the presence of PLWDs between two independent readers was 0.70 (two-way inter-reader correlation coefficient, p<0.001). PLWDs were primarily located in the periphery of the ONH at the superior (25.0%, 3 out of 12) and inferior borders (66.7%, 8 out of 12, fig 1). The one (8.3%) PLWD found near the center of the ONH belonged to a control eye (online supplementary fig 4 b1, b2). PLWDs were detected on 1–2 scans in either the radial or 5-line scans. Of the eyes with PLWDs, 7 (7 POAG, 0 controls; 58.3%) had subsequent imaging in a time frame ranging from 6.9 to 37.3 months (mean 23.0±10.4 months), all demonstrating persistent PLWDs and no new PLWDs. LC defects were found in 15.0% (n=6) of POAG and 8.7% (n=2) of control eyes (p=0.70). In POAG eyes, most LC defects (5 out of 6) were found in in eyes with PLWDs (45.5% of eyes with PLWDs) and one was found in an eye without a PLWD (3.4%, p=0.01, table 1). In POAG eyes with PLWDs, all LC defects were visualized to be directly connected to PLWDs on the same scan but some were shaped differently from PLWDs (fig 1 h1 and h2).

Table 1.

Demographic and Ophthalmic Characteristics of Study Subjects

POAG PLWD (n=11) POAG no PLWD (n=29) Control Group (n=23) POAG PLWD vs. POAG no PLWD (p-value) POAG PLWD vs. Control (p-value) POAG no PLWD vs. Control (p-value)
Gender, Male (%) 27.3 58.6 39.1 0.47 > 0.99 0.79
Age (Years) 66.7 ± 8.1 66.5 ± 9.3 66.4 ± 7.1 > 0.99 > 0.99 > 0.99
Ethnicity, Caucasian (%) * 100 85.7 90.5 > 0.99 > 0.99 >0.99
BCVA (LogMAR) 0.00 ± 0.00 0.06 ± 0.11 0.13 ± 0.13 0.02 < 0.001 0.14
Myopia (%) ** 36.4 61.5 15.8 0.48 >0.99 0.006
Visual Field MD (dB) −3.64 ± 3.4 −4.49 ± 4.1 NA 0.52 NA NA
Visual Field PSD (dB) 5.75 ± 4.3 4.34 ± 3.3 NA 0.34 NA NA
IOP, Current (mmHg) 13.6 ± 2.9 13.9 ± 2.6 14.4 ± 2.4 > 0.99 > 0.99 > 0.99
IOP, Maximum (mmHg) 18.1 ± 3.2 21.1 ± 5.1 16.7 ± 2.5 0.10 0.68 < 0.001
CCT (μm) 546 ± 52 553 ± 35 NA 0.72 NA NA
Phakic (%) 90.9 82.8 82.6 > 0.99 > 0.99 > 0.99
History of DH (%) 45.5 3.4 NA*** 0.004 NA NA
Paracentral Loss on HVF (%) 45.5 24.1 NA 0.25 NA NA
LC Defect (%) 45.5 3.4 8.7 0.01 0.07 > 0.99
Open in a new tab

Data expressed as mean ± standard deviation for continuous variables unless otherwise specified. Statistically significant differences are indicated by bold italic font.

*

Data for ethnicity available for 36 POAG patients and 21 control subjects.

**

Myopia defined as spherical equivalence of −0.75D or worse in phakic eyes or prior to cataract surgery.

***

A history of DH was considered evidence of ON damage and was an exclusion criterion for the control group.

Abbreviations: POAG, primary open angle glaucoma; PLWD, prelaminar wedge defect; BCVA, best corrected visual acuity; LogMAR, logarithm of the minimal angle of resolution; MD, mean deviation; PSD, pattern standard deviation, IOP, intraocular pressure; CCT, central corneal thickness; DH, disc hemorrhage; HVF, Humphrey Visual Field; LC, lamina cribrosa; NA, not applicable.

Baseline Systemic and Ophthalmic Characteristics

POAG patients with PLWDs, POAG patients without PLWDs and control subjects did not differ in age (66.5 years overall, p=0.99 by ANOVA, table 1). Most subjects were Caucasian (93.0% overall, p=0.73). Gender distribution (46.0% male, overall) was comparable between the groups (p=0.18). POAG eyes with PLWDs had better BCVA (0.00±0.00, logarithm of the minimal angle of resolution [LogMAR]) than control eyes (0.13±0.13 LogMAR, p<0.001) and POAG eyes without PLWDs (0.06±0.11 LogMAR, p=0.02). At the time of imaging, 85.0% of the POAG patients and 82.6% of the control subjects were phakic (p>0.99). Information for refractive error was available for all POAG eyes with PLWDs, 95.8% without PLWDs, and 82.6% of control eyes. Prevalence of myopia, defined as spherical equivalence of −0.75 D or worse17 in phakic eyes or prior to cataract surgery, was similar between POAG eyes with (36.4%) and without PLWDs (61.5%, p=0.48). Control eyes had similar rates of myopia (15.8%) compared to POAG eyes with PLWDs (36.4%, p>0.99) but lower rates compared to POAG without PLWDs (61.5%, p=0.006). POAG eyes without PLWDs had more myopic refractive error (−3.3±1.4 D) than control eyes (−2.0±1.2 D, p<0.001), but did not differ significantly from POAG eyes with PLWDs (−3.2±2.4 D); POAG eyes with PLWDs and control eyes had similar mean refractive error (p=0.74).

Mean IOP on the day of imaging was similar between POAG eyes with PLWDs (13.6±2.9 mmHg), POAG eyes without PLWDs (13.9±2.6 mmHg) and control eyes (14.4±2.4 mmHg, p=0.67 by ANOVA). All POAG eyes were on topical IOP-lowering therapy. Untreated maximum IOP of POAG eyes with PLWDs (18.1±3.2 mmHg) was similar to controls (16.7±2.5 mmHg, p=0.68), whereas POAG eyes without PLWDs had higher maximum IOP (21.1±5.1 mmHg) than controls (p<0.001). Central corneal thickness was available for 39 out of 40 POAG patients and not available for control subjects; POAG eyes with and without PLWDs showed similar values (546±52 vs. 553±35 μm, p=0.72). POAG eyes with and without PLWDs had similar HVF MDs (−3.64±3.4 dB and −4.49±4.1 dB, respectively, p=0.52). Eight POAG eyes with PLWDs (72.7%) had HVF functional loss that corresponded to the hemisphere (superior vs. inferior) of the PLWD (fig 1 a1). Paracentral loss was present in 45.5% of POAG eyes with PLWDs and 24.1% of POAG eyes without PLWDs (p=0.25, table 1).

A chart review was conducted to assess for history of DH in all POAG patients up to 2 years prior to imaging. POAG patients with and without PLWDs had similar numbers of clinic appointments during that time period (5.5±4.9 vs 6.2±3.7, respectively, p=0.64). POAG eyes with PLWDs had a more frequent history of DH (n=5, 45.5%) compared to POAG eyes without PLWDs (n=1, 3.4%, p=0.004, table 1). Mean time from documentation of DH to initial SS-OCT imaging was 3.8±6.1 months for POAG eyes with PLWDs and 19.2 months for POAG eyes without PLWDs (p-value not applicable). Locations of DHs (superior vs. inferior) and PLWDs were concordant in 80.0% (4 out of 5) of cases for POAG eyes with PLWDs. Four out of five POAG eyes with PLWDs and DHs (80.0%) also had LC defects, and the location of the DHs (superior vs. inferior), PLWDs, and LC defects were concordant in 75.0% (3 of 4) of cases.

On multivariable analysis of POAG eyes (n=40), PLWDs were associated with increased odds of a prior DH (odds ratio [OR] =21.6, 95% CI, 2.2–589.0, p=0.02) compared to those without PLWDs after adjusting for age (p=0.73), gender (p=0.26), HVF MD (p=0.95), and untreated maximum IOP (p=0.24, table 2). In addition, comparing all patients with and without PLWDs (including controls), patients with PLWDs had increased odds of having a LC defect (OR=44.8; 95% CI, 6.3–703.6, p<0.001) compared to those without PLWDs after adjusting for age (p=0.40), gender (p=0.55), and untreated maximum IOP (p=0.16, table 3).

Table 2.

Multivariable Logistic Regression to Assess for History of Disc Hemorrhage in POAG Patients

POAG PLWD (n=11) vs. POAG no PLWD (n=29)
Variables Odds Ratio [95% CI] p-value
Age (Years) 0.98 [0.87, 1.10] 0.73
Gender, Male 0.34 [0.04, 2.06] 0.26
Visual Field MD (dB) 0.99 [0.73, 1.38] 0.95
IOP, maximum (mmHg) 0.87 [0.65, 1.08] 0.24
History of DH 21.6 [2.2, 589.0] 0.02
Open in a new tab

Logistic regression model used to analyze POAG patients with PLWDs vs. POAG patients without PLWD. POAG patients without PLWDs served as reference. Statistically significant differences are indicated by bold italic font.

Abbreviations: POAG, primary open angle glaucoma; PLWD, prelaminar wedge defect; CI, confidence interval; MD, mean deviation; IOP, intraocular pressure; DH, disc hemorrhage.

Table 3.

Multivariable Logistic Regression to Assess for Lamina Cribrosa Defects in All Subjects

All Eyes with PLWD (n=12) vs. without PLWD (n=51)
Variables Odds Ratio [95% CI] p-value
Age (Years) 1.05 [0.95, 1.18] 0.40
Gender, Male 0.61 [0.10, 3.07] 0.55
IOP, maximum (mmHg) 0.86 [0.69, 1.04] 0.16
LC Defect 44.8 [6.3, 703.6] < 0.001
Open in a new tab

Logistic regression model used to analyze all subjects with PLWDs vs. subjects without PLWDs, including control subjects. Statistically significant differences are indicated by bold italic font.

Abbreviations: PLWD, prelaminar wedge defect; CI, confidence interval; MD, mean deviation; IOP, intraocular pressure; LC, lamina cribrosa.

Quantitative Structural Characteristics

The groups did not differ in horizontal LCD (405.2±64.1 μm for POAG with PLWDs, 443.9±112.3 μm for POAG without PLWDs, 405.2±112.4 μm for controls, p=0.36 by ANOVA). POAG eyes with and without PLWDs had similar mean global BMO-MRW (201.1±38.3 μm vs 214.9±60.4 μm, p>0.99), normalized global BMO-MRW (211.3±51.0 μm vs 214.9±64.9 μm, p>0.99), minimum BMO-MRW (81.1±21.1 μm vs 108.8±58.0 μm, p=0.10), average RNFL (76.1±12.0 μm vs 75.6±15.2 μm, p>0.99), and thinnest RNFL (50.0±10.4 μm vs 50.3±12.5 μm, p>0.99, table 4). Control eyes had higher global BMO-MRW (326.4±65.2 μm), normalized global BMO-MRW (317.7±69.4 μm) and minimum BMO-MRW (224.5±38.3 μm) than POAG groups (p<0.001 for all). Control eyes also had higher average RNFL (97.5±11.3 μm) and thinnest RNFL (63.0±9.5 μm) compared to POAG groups (p≤0.007 for all).

Table 4.

Quantitative Structural Characteristics of Study Subjects

POAG PLWD (n=11) POAG no PLWD (n=29) Control Group (n=23) POAG PLWD vs POAG no PLWD (p-value) POAG PLWD vs Control (p-value) POAG no PLWD vs Control (p-value)
Horizontal LCD (μm) 405.2 ± 64.1 443.9 ± 112.3 405.2 ± 112.4 0.55 > 0.99 0.67
Global BMO-MRW (μm) 201.1 ± 38.3 214.9 ± 60.4 326.4 ± 65.2 > 0.99 < 0.001 < 0.001
Normalized BMO-MRW (μm) 211.3 ± 51.0 214.9 ± 64.9 317.7 ± 69.4 > 0.99 < 0.001 < 0.001
Minimum BMO-MRW (μm) 81.1 ± 21.1 108.8 ± 58.0 224.5 ± 38.3 0.10 < 0.001 < 0.001
Average RNFL (μm) 76.1 ± 12.0 75.6 ± 15.2 97.5 ± 11.3 > 0.99 < 0.001 < 0.001
Thinnest RNFL (μm) 50.0 ± 10.4 50.3 ± 12.5 63.0 ± 9.5 > 0.99 0.007 < 0.001
Open in a new tab

Data expressed as mean ± standard deviation for continuous variables unless otherwise specified. Swept-source optical coherence tomography (OCT) scans were used to measure horizontal LCD, global BMO-MRW, normalized global BMO-MRW, and minimum BMO-MRW, while spectral-domain OCT scans were used for average and thinnest clock-hour RNFL thickness. Statistically significant differences are indicated by bold italic font.

Abbreviations: POAG, primary open angle glaucoma; PLWD, prelaminar wedge defect; LCD, lamina cribrosa depth; BMO-MRW, Bruch’s membrane opening minimum rim width; RNFL, retinal nerve fiber layer.

DISCUSSION

Glaucomatous damage can manifest as focal ONH changes, such as DH, focal notching,18 and acquired optic nerve pits.19,20 Therefore, localized features of the prelaminar tissue may be a better imaging surrogate for assessing glaucoma subtypes with focal ONH damage than global characterization such as decreased mean prelaminar tissue thickness and increased anterior prelaminar depth.5,21 Fundoscopic assessment of prelaminar pathology in glaucoma can be challenging as this tissue is generally translucent and axons are only rarely myelinated. Development of imaging technology has allowed for high-resolution visualization of prelaminar and laminar ONH structures.8,22 We utilized SS-OCT scans of POAG and control eyes to identify wedge-shaped prelaminar defects that have previously been described on histology.19 We focused our efforts on patients with mild to moderate glaucoma, to be consistent with larger clinical trials,23,24 and detected PLWDs more frequently in POAG patients (27.5%) compared to controls (4.3%, p=0.04). Furthermore, we showed significant correlations with prior DH and LC defects.

The association between PLWDs and DHs was further supported by the finding that PLWDs and DHs were located in the same half of the ONH (superior or inferior) in the majority of the eyes. DH is a well-established risk factor for glaucoma progression.25,26 One study reported that 81.3% of eyes experienced additional VF loss within five years after a DH.27 Another study showed that eyes with DHs had a 2-fold faster rate in RNFL thinning than in non-DH eyes.28 However, the transient nature of DHs makes them likely to be missed on clinical visits if they occur on an interim basis.29 In addition, the Ocular Hypertension Treatment Study demonstrated that hemorrhages are often overlooked – only 25.0% of DHs were detected by clinical examination.30 These detection rates are concerning given that the presence of a DH typically prompts aggressive treatment to preserve visual function.31 Hence, a disease biomarker, such as a PLWD, that is persistent, readily identified on imaging, and is significantly associated with DHs, can be clinically useful. Unfortunately, the retrospective nature of our study did not allow for a history of DH, nor the precise location of the DH beyond superior and inferior hemisphere of the disc, to be ascertained in a systematic fashion. On the other hand, eyes with PLWDs and subsequent imaging (an average of 23 months later) all demonstrated persistent PLWDs, suggesting a more permanent nature of this structural change compared to DHs. We believe that PLWDs may be an imaging biomarker for eyes prone to DHs and have the potential to become a surveillance tool for glaucomatous damage. Longitudinal studies are needed to assess the association between PLWDs and subsequent thinning of RNFL or other structural parameters consistent with glaucomatous progression.

In our study, LC defects were also more frequently detected in POAG eyes (15.0%) than control eyes (8.7%) but did not reach statistical significance (p=0.70); the sample size may have limited the power of the study to detect this difference. The detection rate of LC defects in POAG and control subjects in our study is similar to those reported in the literature.32,33 Interestingly, most LC defects in POAG eyes were found in eyes with PLWDs (5 out of 6) and all LC defects were contiguous with PLWDs identified on the same scan. The multivariable analysis, which was performed with all eyes, also showed a significant association between PLWDs and LC defects. These LC defects are unlikely to be shadowing artifacts of the PLWDs as some of they have been observed to have different shapes than PLWDs (fig 1, h1 and h2). Eyes with focal LC defects have been shown to experience more rapid glaucomatous VF progression.34 Their connection with PLWDs suggests that these structural abnormalities have much in common. Higher prevalence of PLWDs makes them a more sensitive marker of glaucomatous damage than LC defects. It is unlikely that our findings are confounded by the presence of myopia, which has been associated with certain types of LC defects.35 We excluded patients with high myopia, or spherical equivalence worse than −6 D, and did not find worse myopia in the POAG eyes with PLWDs compared to POAG eyes without PLWDs.

The correlation of PLWDs with DHs and LC defects may suggest that PLWDs are clinically relevant as a disease biomarker. Previous studies have shown correlations between DHs, LC defects, and more advanced glaucoma status.36,37 It has also been found that eyes with spatial concordance of DHs and LC defects are more likely to develop VF progression compared to eyes with an isolated finding of a DH or LC alteration.38 In our study, PLWDs were associated with DHs and they co-localized in 80% of the eyes. We also found that PLWDs were associated and contiguous with LC defects. The spatial relationship between PLWDs, DHs and LC defects additionally supports PLWDs as a potential biomarker for glaucomatous damage. A recent study by our group suggested an arterial origin for DHs, implicating vascular dysregulation in POAG.39 It is possible that dysregulation of arterioles supplying the ONH causes DHs and the associated compression from the blood or ischemic injury directly leads to focal damage of the retinal ganglion cell axons. These changes may manifest as PLWDs which may involve the LC and present as LC defects as well, although the causal relationship cannot be proven in this retrospective study. In addition, DHs are more commonly observed in normal-tension glaucoma patients.40 This is partly supported by our study, where the maximum IOP in POAG eyes with PLWDs was similar to controls while the maximum IOP in POAG eyes without PLWDs was higher, although POAG patients might be on treatment when the maximum IOP was recorded in this retrospective study. Furthermore, PLWDs are located at the superior or inferior disc rim in POAG patients; these regions are considered vulnerable in POAG.41 Taken together, we believe that PLWD characterizes the vulnerability of the glaucomatous optic nerve, in which IOP-independent factors, such as vascular dysfunction, plays an important role in neuronal damage. Also, these characteristics may describe a POAG subgroup with a distinct pattern of glaucomatous damage and have prognostic implications.

This study has several limitations. First, the sensitivity for detecting PLWDs is limited by the number of cross-sectional images in the fixed scanning patterns of the SS-OCT device. Defects were present in 1–2 images of the radial and/or 5-line cross scan patterns. Detection may be improved if the SS-OCT imaging protocol allowed for denser scanning patterns. Second, we cannot rule out differential ascertainment of DH in eyes with PLWDs versus those without these defects, although DHs were documented in clinical records prior to imaging and POAG patients without PLWDs had more clinical visits on average than patients with PLWDs during the same time period before imaging. Third, a PLWD was found in one control subject, suggesting that it may not be a specific finding to glaucoma. However, the PLWD in the control eye was located in the center of the ONH, while the PLWDs in POAG eyes were all located in the peripheral ONH. This finding suggests that the etiology of PLWDs may be different for controls. Fourth, ambiguous PLWDs (10.0%) and moderate inter-reader agreement between two independent observers masked to diagnosis allude to the subjective nature of image interpretation. Fifth, the study has a small number of patients and a larger study may identify additional quantitative optic nerve parameters, such as minimum BMO-MRW, between the POAG with and without PLWD groups. Lastly, the cross-sectional nature of this study does not allow for evaluation of cause and effect relationships or the assessment of glaucoma progression in eyes with PLWDs.

In conclusion, SS-OCT delineated focal ONH structural damage in the form of wedge-shaped defects in the prelaminar tissue of POAG patients. Our results showed that PLWDs are associated with prior DHs and LC defects. Given that DHs and LC defects are established signs of glaucoma progression, PLWDs may become a clinically useful imaging biomarker in identifying glaucomatous damage, particularly because of the persistent nature and relatively high prevalence of these defects.

Supplementary Material

1

Acknowledgements:

The authors thank the Massachusetts Eye and Ear Fluorescein Lab Photographers for assistance with the imaging, and the Massachusetts Eye and Ear study coordinators for patient recruitment.

Funding:

This work was supported by the Glaucoma Center of Excellence and Miller Research Funds at the Massachusetts Eye and Ear, Eleanor and Miles Shore Fellowship, Harvard Medical School, Boston MA, and NIH K99 EY028631 (M.W.). The sponsors and funding organizations had no role in the design or conduct of this research.

Disclosure Statement:

Dr. Lucy Shen receives research support from Topcon. In work that is not relevant to this research, Dr. Louis Pasquale is a consultant to Bausch + Lomb, Inc., Verily, Emerald Biosciences and on the Advisory Board of Eyenovia, Inc. Dr. Angela Turalba is a consultant for Beaver-Visitec International.

Footnotes

Meeting Presentation: American Glaucoma Society Annual Meeting, San Francisco, March, 2018.

References:

  • 1.Quigley HA, Dunkelberger GR, Green WR. Retinal ganglion cell atrophy correlated with automated perimetry in human eyes with glaucoma. Am J Ophthalmol. 1989;107:453–464. [DOI] [PubMed] [Google Scholar]
  • 2.Yang H, Downs JC, Bellezza A, Thompson H, Burgoyne CF. 3-D histomorphometry of the normal and early glaucomatous monkey optic nerve head: Prelaminar neural tissues and cupping. Invest Ophthalmol Vis Sci. 2007;48:5068–5084 [DOI] [PubMed] [Google Scholar]
  • 3.Lee SH, Kim TW, Lee EJ, Girard MJA, Mari JM. Diagnostic power of lamina cribrosa depth and curvature in glaucoma. Invest Ophthalmol Vis Sci. 2017;58:755–762. [DOI] [PubMed] [Google Scholar]
  • 4.Strouthidis NG, Grimm J, Williams GA, Cull GA, Wilson DJ, Burgoyne CF. A comparison of optic nerve head morphology viewed by spectral domain optical coherence tomography and by serial histology. Invest Ophthalmol Vis Sci. 2010;51:1464–1474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Jung YH, Park HYL, Jung KI, Park CK. Comparison of prelaminar thickness between primary open angle glaucoma and normal tension glaucoma patients. PLoS One. 2015;10:1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Potsaid B, Baumann B, Huang D, Barry S, Cable AE, Schuman JS, Duker JS, Fujimoto JG. Ultrahigh speed 1050nm swept source/Fourier domain OCT retinal and anterior segment imaging at 100,000 to 400,000 axial scans per second. Opt Express. 2010;18:20029–20048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Mwanza JC, Budenz DL. New developments in optical coherence tomography imaging for glaucoma. Curr Opin Ophthalmol. 2018;29:121–129. [DOI] [PubMed] [Google Scholar]
  • 8.Takayama K, Hangai M, Kimura Y, Morooka S, Nukada M, Akagi T, Ikeda HO, Matsumoto A, Yoshimura N. Three-dimensional imaging of lamina cribrosa defects in glaucoma using swept-source optical coherence tomography. Invest Ophthalmol Vis Sci. 2013;54:4798–4807. [DOI] [PubMed] [Google Scholar]
  • 9.Li D, Li T, Paschalis EI, Wang H, Taniguchi EV, Choo ZN, Shoji MK, Greenstein SH, Brauner SC, Turalba AV, et al. Optic Nerve Head Characteristics in Chronic Angle Closure Glaucoma Detected by Swept-Source OCT. Curr Eye Res. 2017;42:1450–1457. [DOI] [PubMed] [Google Scholar]
  • 10.Shin JW, Sung KR, Uhm KB, Jo J, Moon Y, Song MK, Song JY. Peripapillary microvascular improvement and lamina cribrosa depth reduction after trabeculectomy in primary open-angle glaucoma. Invest Ophthalmol Vis Sci. 2017;58:5993–5999. [DOI] [PubMed] [Google Scholar]
  • 11.How ACS, Tan GSW, Chan YH, Wong TTL, Seah SK, Foster PJ, Aung T. Population prevalence of tilted and torted optic discs among an adult Chinese population in Singapore: The Tanjong Pagar study. Arch of Ophthalmol. 2009;127:894–899. [DOI] [PubMed] [Google Scholar]
  • 12.Jonas JB, Kling F, Grundler AE. Optic disc shape, corneal astigmatism, and amblyopia. Ophthalmology. 1997;104:1934–1937. [DOI] [PubMed] [Google Scholar]
  • 13.Li D, Taniguchi EV, Cai S, Paschalis EI, Wang H, Miller JB, Turalba AV, Greenstein SH, Brauner S, Pasquale LR, et al. Comparison of swept-source and enhanced depth imaging spectral-domain optical coherence tomography in quantitative characterisation of the optic nerve head. Br J Ophthalmol. 2016;101:299–304. [DOI] [PubMed] [Google Scholar]
  • 14.Patel NB, Sullivan-Mee M, Harwerth RS. The relationship between retinal nerve fiber layer thickness and optic nerve head neuroretinal rim tissue in glaucoma. Invest Ophthalmol Vis Sci. 2014;55:6802–6816. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Taniguchi EV, Paschalis EI, Li D, Nouri-Mahdavi K, Brauner SC, Greenstein SH, Turalba AV, Wiggs JL, Pasquale LR, Shen LQ. Thin minimal rim width at Bruch’s membrane opening is associated with glaucomatous paracentral visual field loss. Clin Ophthalmol. 2017;11:2157–2167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Loomis SJ, Kang JH, Weinreb RN, Yaspan BL, Bailey JNC, Gaasterland D, Gaasterland T, Lee RK, Lichter PR, Budenz DL, et al. Association of CAV1/CAV2 genomic variants with primary open-angle glaucoma overall and by gender and pattern of visual field loss. Ophthalmology. 2014;121:508–516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Kleinstein RN, Jones LA, Hullett S, Kwon S, Lee RJ, Friedman NE, Manny RE, Mutti DO, Yu JA, Zadnik K. Refractive error and ethnicity in children. Arch Ophthalmol. 2003;121:1141–1147. [DOI] [PubMed] [Google Scholar]
  • 18.Garway-Heath DF, Hitchings RA. Quantitative evaluation of the optic nerve head in early glaucoma. Statistics. 1998:352–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Radius RL, Maumenee AE, Green WR. Pit-like changes of the optic nerve head in open-angle glaucoma. Br J Ophthalmol. 1978;62:389–393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ugurlu S, Weitzman M, Nguaguba C, Caprioli J. Acquired pit of the optic nerve: A risk factor for progression of glaucoma. Am J Ophthalmol. 1998;125:457–464. [DOI] [PubMed] [Google Scholar]
  • 21.Kim DW, Jeoung JW, Kim YW, Girard MJA, Mari JM, Kim YK, Park KH, Kim DM. Prelamina and lamina cribrosa in glaucoma patients with unilateral visual field loss. Invest Opthalmol Vis Sci. 2016;57:1662–1670. [DOI] [PubMed] [Google Scholar]
  • 22.Omodaka K, Horii T, Takahashi S, Kikawa T, Matsumoto A, Shiga Y, Maruyama K, Yuasa T, Akiba M, Nakazawa T. 3D evaluation of the lamina cribrosa with swept-source optical coherence tomography in normal tension glaucoma. PLoS ONE. 2015;10:1–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Gordon MO, Kass MA. The ocular hypertension treatment study: design and baseline description of the participants. Arch Ophthalmol. 1999;117:573–583. [DOI] [PubMed] [Google Scholar]
  • 24.Leske MC, Hejil A, Hyman L, Bengtsson B. Early manifest glaucoma trial: design and baseline data. Ophthalmology. 1999;106:2144–2153. [DOI] [PubMed] [Google Scholar]
  • 25.De Moraes CG, Demirel S, Gardiner SK, Liebmann JM, Cioffi GA, Ritch R, Gordon MO, Kass MA. Rate of visual field progression in eyes with optic disc hemorrhages in the ocular hypertension treatment study. Arch Ophthal. 2012;130:1541–1546. [DOI] [PubMed] [Google Scholar]
  • 26.Drance S, Anderson DR, Schulzer M. Risk factors for progression of visual field abnormalities in normal-tension glaucoma. Am J Ophthalmol. 2001;131:699–708. [DOI] [PubMed] [Google Scholar]
  • 27.Ishida K, Yamamoto T, Sugiyama K, Kitazawa Y. Disk hemorrhage is a significantly negative prognostic factor in normal-tension glaucoma. Am J Ophthalmol. 2000;129:707–714. [DOI] [PubMed] [Google Scholar]
  • 28.Hou H, Moghimi S, Zangwill LM, Proudfoot JA, Akagi T, Shoji T, Girkin CA, Liebmann JM, Weinreb RN. Association between rates of retinal nerve fiber layer thinning after intraocular pressure–lowering procedures and disc hemorrhage. Ophthalmol Glaucoma. 2020;3:7–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Bengtsson B, Holmin C, Krakau C. Disc haemorrhage and glaucoma. Acta Ophthalmol. 1981;59:1–14. [DOI] [PubMed] [Google Scholar]
  • 30.Budenz DL, Anderson DR, Feuer WJ, Beiser JA, Schiffman J, Parrish RK, Piltz-Seymour JR, Gordon MO, Kass MA. Detection and prognostic significance of optic disc hemorrhages during the ocular hypertension treatment study. Ophthalmology. 2006;113:2137–2143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Medeiros FA, Alencar LM, Sample PA, Zangwill LM, Susanna R, Weinreb RN. The relationship between intraocular pressure reduction and rates of progressive visual field loss in eyes with optic disc hemorrhage. Ophthalmology. 2010;117:2061–2066. [DOI] [PubMed] [Google Scholar]
  • 32.You JY, Park SC, Su D, Teng CC, Liebmann JM, Ritch R. Focal lamina cribrosa defects associated with glaucomatous rim thinning and acquired pits. JAMA Ophthalmol. 2013;131:314–320. [DOI] [PubMed] [Google Scholar]
  • 33.Tatham AJ, Miki A, Weinreb RN, Zangwill LM, Medeiros FA. Defects of the lamina cribrosa in eyes with localized retinal nerve fiber layer loss. Ophthalmology. 2014;121:110–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Faridi OS, Park SC, Kabadi R, Su D, De Moraes CG, Liebmann JM, Ritch R. Effect of focal lamina cribrosa defect on glaucomatous visual field progression. Ophthalmology. 2014;121:1524–1530. [DOI] [PubMed] [Google Scholar]
  • 35.Sawada Y, Araie M, Kasuga H, Ishikawa M, Iwata T, Murata K, Yoshitomi T. Focal lamina cribrosa defect in myopic eyes with nonprogressive glaucomatous visual field defect. Am J Ophthalmol. 2018;190:34–49. [DOI] [PubMed] [Google Scholar]
  • 36.Sharpe GP, Danthurebandara VM, Vianna JR, Alotaibi N, Hutchison DM, Belliveau AC, Shuba LM, Nicolela MT, Chauhan BC. Optic disc hemorrhages and laminar disinsertions in glaucoma. Ophthalmology. 2016;123:1949–1956. [DOI] [PubMed] [Google Scholar]
  • 37.Park SC, Hsu AT, Su D, Simonson JL, Al-Jumayli M, Liu Y, Liebmann JM, Ritch R. Factors associated with focal lamina cribrosa defects in glaucoma. Invest Ophthalmol Vis Sci. 2013;54:8401–8407. [DOI] [PubMed] [Google Scholar]
  • 38.Park HYL, Lee J, Jung Y, Park CK. Optic disc hemorrhage and lamina cribrosa defects in glaucoma progression. Sci Rep 2017;7:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Chou JC, Cousins CC, Miller JB, Song BJ, Shen LQ, Kass MA, Wiggs JL, Pasquale LR. Fundus densitometry findings suggest optic disc hemorrhages in primary open-angle glaucoma have an arterial origin. Am Journal of Ophthal. 2018;187:108–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Suh MH, Park KH. Period prevalence and incidence of optic disc haemorrhage in normal tension glaucoma and primary open-angle glaucoma. Clin and Exp Ophthalmol. 2011;39:513–519. doi: 10.1111/j.1442-9071.2010.02482.x [DOI] [PubMed] [Google Scholar]
  • 41.Quigley HA, Addicks EM, Green R. Optic Nerve Damage In Human Glaucoma. Ophthalmology. 1982;100:135–146. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1769751-supplement-1.pdf (2.9MB, pdf)

RESOURCES