|Year : 2021 | Volume
| Issue : 4 | Page : 344-349
To compare and correlate visual field changes detected by perimetry with retinal nerve fiber layer and ganglion cell layer thickness observed using spectral domain optical coherence tomography in primary open angle glaucoma
Priya Gupta, Anita Minj, Suprava Das, Pradeep Kumar Panigrahi
Department of Ophthalmology, Institute of Medical Sciences and SUM Hospital, SOA (Deemed to be) University, Bhubaneswar, Odisha, India
|Date of Submission||19-Apr-2021|
|Date of Decision||10-Jul-2021|
|Date of Acceptance||10-Jul-2021|
|Date of Web Publication||21-Dec-2021|
Dr. Pradeep Kumar Panigrahi
Department of Ophthalmology, Institute of Medical Sciences and SUM Hospital, SOA (Deemed to be) University, 8-Kalinga Nagar, Bhubaneswar - 751 003, Odisha
Source of Support: None, Conflict of Interest: None
Aim: To evaluate the damage pattern of retinal ganglion cells at different stages of glaucoma in patients with primary open-angle glaucoma (POAG) by observing peripapillary retinal nerve fiber layer (RNFL) and macular ganglion cell layer-inner plexiform layer (GCL-IPL) thickness measured by spectral domain optical coherence tomography (SD-OCT) and their correlation with visual field (VF) changes. Materials and Methods: Descriptive cross-sectional study done on patients with POAG presenting to a tertiary eye center in Eastern India. Peripapillary RNFL thickness and macular GCL-IPL thickness were measured using SD-OCT. VF assessment was done using the Humphrey Field analyzer. Patients were divided into early, moderate, and advanced (severe) groups following Hodapp, Parrish, and Anderson's criteria. Results: Seventy-seven eyes of 40 patients with POAG were included in the study. The average age of the patients was 59.23 ± 9.56 years. Early, moderate, and advanced glaucoma changes were seen in 34 (44.15%), 23 (29.87%), and 20 (25.98%) eyes, respectively. Superior peripapillary RNFL thinning was noted in early glaucoma. As the disease progressed, inferior peripapillary RNFL showed significant thinning in the advanced glaucoma group as compared to early and moderate groups. Inferior macular GCL-IPL thickness showed more thinning as the severity of the disease progressed. In all stages of glaucoma, peripapillary RNFL and GCL-IPL thickness showed positive correlation with the VF mean deviation. Conclusion: Both peripapillary RNFL and macular GCL-IPL thicknesses should be assessed along with VF defects to monitor the severity of the disease in POAG. Macular GCL-IPL thickness changes are well correlated with changes in visual function and RNFL structure in glaucoma.
Keywords: Macular ganglion cell, peripapillary retinal nerve fiber layer, primary open angle glaucoma, visual field
|How to cite this article:|
Gupta P, Minj A, Das S, Panigrahi PK. To compare and correlate visual field changes detected by perimetry with retinal nerve fiber layer and ganglion cell layer thickness observed using spectral domain optical coherence tomography in primary open angle glaucoma. TNOA J Ophthalmic Sci Res 2021;59:344-9
|How to cite this URL:|
Gupta P, Minj A, Das S, Panigrahi PK. To compare and correlate visual field changes detected by perimetry with retinal nerve fiber layer and ganglion cell layer thickness observed using spectral domain optical coherence tomography in primary open angle glaucoma. TNOA J Ophthalmic Sci Res [serial online] 2021 [cited 2022 Jan 24];59:344-9. Available from: https://www.tnoajosr.com/text.asp?2021/59/4/344/333171
| Introduction|| |
Primary open-angle glaucoma (POAG) is an adult onset, bilateral and almost symmetrical disease. It is characterized by an open anterior chamber angle, glaucomatous optic nerve head (ONH) changes, visual field (VF) defect, and an intraocular pressure (IOP) more than 21 mm Hg on more than one occasion. Standard automated perimetry (SAP), white on white has remained the gold standard method in diagnosis and monitoring the progression of glaucoma. Depending upon the severity of disease either 30-2, 24-2, or 10-2 testing algorithm is being used. But an early glaucomatous threshold VF defect manifests only when approximately 40% of retinal ganglion cells (RGCs) are lost. The earliest observable defect in glaucoma is atrophy of the nerve fiber layer. Clinically, recognizable loss of the neuro-retinal rim in the ONH can precede VF defects. This is observed in 60% of eyes approximately 6 years before any detectable VF defects in glaucoma.
It has been reported that spectral-domain optical coherence tomography (SD-OCT), a noninvasive, reproducible and objective versatile tool can measure peripapillary retinal nerve fiber layer (RNFL) thickness and it can be useful in detecting glaucoma in the preperimetric stage. The advent of SD-OCT has renewed interest in the potential uses of macular imaging in glaucoma due to its ability to segment and measure individual retinal layers better. A recent study showed that inner retinal layer thinning at the macula could be demonstrated by custom-designed automatic segmentation software. This was apparent before VF changes. The macular region contains over 50% of all RGC and 10% or fewer axons may remain in advanced glaucoma. Therefore, it is an ideal region to detect early cell loss and changes over a period of time because of the high density of cells.
VF testing is a subjective method and prone to inter-test variability whereas OCT imaging of RFNL and ganglion cell layer (GCL) thickness measurements are objective methods of evaluation. Detecting structural changes that precede visual function loss may be the key to vision preservation in glaucomatous patients. Therefore, it becomes important to detect glaucoma in its preperimetric stage and monitor disease progression. In this study, we sought to evaluate the pattern of RGC damage at different stages of glaucoma by observing peripapillary RNFL and macular GCL-inner plexiform layer (GCL-IPL) thickness measured by SD-OCT and their correlation with VF changes, which clinically help us to monitor the disease and its management.
| Materials and Methods|| |
The present study was a descriptive cross-sectional study on cases of POAG presenting to the Department of Ophthalmology of a tertiary care hospital in Eastern India between December 2018 and March 2020. The study adhered to the basic tenets of the Declaration of Helsinki. Institutional ethical committee clearance was obtained before the start of the study. Written informed consent was obtained from each patient before entering into the study.
The study population was selected from the patients attending the outpatient department of the Ophthalmology department between December 2018 and March 2020. All patients of either sex diagnosed with POAG were included in the study. Exclusion criteria included significant ocular media opacities, anterior segment dysgenesis, best-corrected visual acuity (BCVA) <20/40, diabetic retinopathy or any other disease or treatment that could independently affect the retinal thickness or VFs, refractive errors (> ±6.00 D), history of past intraocular surgery and all other types of glaucoma except POAG. Raosoft™ software was used for sample size calculation. Using a 95% confidence limit, the sample size of 40 was deemed to be adequate for the study.
After obtaining informed consent, detailed history regarding patient's name, age, sex, occupation, address, presenting symptoms, associated systemic conditions, and history of past intraocular surgery, if any were recorded. A brief systemic examination was done in each case. Distant and near visual acuity was measured using Snellen's and Jaeger's chart, respectively. Anterior segment examination was done using slit- lamp biomicroscope. The angle of the anterior chamber was assessed using 4 mirror gonio lens. IOP was measured using Goldmann' applanation tonometer. Pupils were dilated using a combination of tropicamide (0.5%) and phenylephrine (10%) eye drops. Dilated stereoscopic examination of the optic disc and posterior segment was done using +78 diopters slit-lamp biomicroscopy and indirect ophthalmoscope.
SD-OCT was done for the measurement of peripapillary RNFL and macular GCL-IPL thickness. RNFL and macular scan were done using three-dimensional (3D) OCT-1 Maestro version 8.3 (Topcon Inc, Tokyo, Japan). RNFL scan was done using 3D disc protocol with scan length of 6 mm × 6 mm and resolution of 512 × 128. The thickness calculated in each cell is compared with a normative database which is already fed in the machine. Values within the normal range are represented as green, borderline as yellow, and outside normal limits are shown in red. Macular scan was done using 3D macula V protocol with scan length of 7 mm × 7 mm and resolution of 512 × 128. Humphrey field analyzer 745i (Carl Zeiss Meditec, Germany) was used for conducting standard automated white on white perimetry employing the Swedish Interactive Threshold Algorithm standard (SITA 24-2/10-2).
A VF test was considered reliable when fixation losses were <20% and false-positive and false-negative errors were <20%. Only reliable VF tests were included in the analysis. A VF defect was defined when there were 3 significant (P < 5%) nonedge contiguous points with at least one point P < 1% or two points P < 5% at the level on the same side of the horizontal meridian in the pattern deviation plot, pattern standard deviation (PSD) P < 5% and Glaucoma hemifield test abnormal.
Glaucomatous damage can also be classified additionally according to Hodapp Parrish Anderson Scale based on VF mean deviation (MD).
- Glaucoma stage 1 (Early): VF MD > −6 dB
- Glaucoma stage 2 (moderate): VF MD between − 6 dB and − 12 dB
- Glaucoma stage 3 (severe): VF MD >−12 dB.
Depending on the severity of glaucoma, patients were divided into early, moderate, and advanced (severe) groups.
Appropriate statistical analysis was done at the end of the study period. Data obtained were entered into a Microsoft Excel sheet. Descriptive statistics has been presented as actual number, percentage, mean and standard deviation. Analysis of variance test was used to obtain information about the relationship between dependent and independent variables. Pearson's correlation coefficient was used to determine the correlation between quantitative variables. The null hypothesis was rejected for P < 0.05.
| Results|| |
A total of 77 eyes of 40 patients with POAG were included in the study, of which 39 (50.65%) were right eyes and 38 (49.35%) were left eyes. The average age of the patients was 59.23 ± 9.56 years (range: 43–81 years). Out of 40, males were 26 (65%) and females were 14 (35%). Depending on the severity of glaucoma, patients were classified into early, moderate, and advanced (severe) stages of glaucoma (based on Hodapp, Parrish, and Anderson classification).
According to this classification 34 (44.15%) eyes had early, 23 (29.87%) had moderate, and 20 (25.98%) had advanced glaucoma, respectively. Average IOP in early, moderate, and advanced groups were 19.94 ± 3.81, 22.69 ± 5.2, and 24.55 ± 9.1 mm Hg, respectively. Average BCVA in early, moderate, and advanced groups were 0.112 ± 0.09, 0.222 ± 0.22, and 0.51 ± 0.31 logarithm of the minimum angle of resolution, respectively [Table 1].
Mean peripapillary superior RNFL thickness in the three groups was 88.09 ± 16.67, 71.52 ± 25.02, and 64.95 ± 29.02 μ, respectively. Mean peripapillary inferior RNFL thickness in the three groups was 91.59 ± 28.75, 70.26 ± 28.01, and 54.25 ± 26.08 μ, respectively [Table 2]. In early glaucoma, superior peripapillary RNFL showed thinning. But as the disease progressed, inferior peripapillary RNFL showed significant thinning in advanced glaucoma as compared to early and moderate groups [Figure 1]. Mean macular GCL-IPL thickness in the superior quadrant in the three groups was 61.03 ± 7.79, 59.04 ± 6.04, and 57.7 ± 11.38 μ, respectively. Mean macular GCL-IPL thickness in the inferior quadrant in the three groups was 60.53 ± 6.64, 58.91 ± 7.11, and 56.35 ± 10.39 μ, respectively [Table 3]. Inferior macular GCL-IPL thickness showed more thinning as the severity of the disease progressed. Overall as the disease progresses from early to moderate to advance, progressive thinning of macular GCL-IPL is seen [Figure 2].
|Table 2: Quadrant wise mean peripapillary retinal nerve fiber layer thickness in the three groups|
Click here to view
|Table 3: Quadrant-wise mean macular ganglion cell layer - inner plexiform layer thickness in the three groups|
Click here to view
|Figure 1: Average peripapillary retinal nerve fiber layer thickness in early, moderate and advanced glaucoma|
Click here to view
|Figure 2: Average macular ganglion cell layer-inner plexiform layer thickness in early, moderate and advanced glaucoma|
Click here to view
The MD in the three groups was −3.55 ± 1.94, −8.47 ± 1.85, and −19.55 ± 6.02 decibels (db.), respectively. The PSD in the three groups was 2.86 ± 1.68, 5.17 ± 3.38, and 9.66 ± 3.09 db, respectively. In all stages of glaucoma, peripapillary RNFL and GCL-IPL thickness showed positive correlation with the MD. In early glaucoma, inferior peripapillary RNFL thickness (r = 0.60688) showed highly significant statistical correlation with the MD (P = 0.0001) followed by inferior macular GCL-IPL (r = 0.56404) and superior peripapillary RNFL (r = 0.47778) thicknesses. Average peripapillary RNFL and macular GCL-IPL thicknesses were also statistically significant with MD (P = 0.0041 and 0.0037), respectively [Table 4].
|Table 4: Correlation coefficient for mean deviation on Humphrey visual field analysis with peripapillary retinal nerve fiber layer and macular ganglion cell layer - inner plexiform layer thicknesses parameters on optical coherence tomography in different stages of glaucomatous eyes|
Click here to view
In moderate glaucoma, superior peripapillary RNFL (r = 0.52785) thickness was statistically very significant with the MD followed by superior macular GCL-IPL (r = 0.50271) thickness. In advanced glaucoma, superior peripapillary RNFL (r = 0.4526) and macular GCL-IPL (r = 0.47802) thicknesses were statistically significant with the MD. The correlation coefficient of the inferior peripapillary RNFL thickness with MD was significantly stronger in early glaucoma (P = 0.0001) as compared to the advanced stage of glaucoma. In all stages of glaucoma, superior peripapillary RNFL thickness showed statistical significance with the MD. However, in moderate glaucoma, this statistical correlation was very significant (P = 0.0080).
| Discussion|| |
Glaucoma is a chronic, progressive optic neuropathy causing a group of ocular conditions, which lead to damage of the optic nerve with loss of visual function. Structural and/or functional measurements are necessary in determining the progression of glaucoma. When the progression of glaucoma is confirmed in patients, modification of the treatments is necessary to prevent further irreversible loss of visual function. Our study was designed with the aim of comparing and correlating the RNFL and GCL thicknesses with VF defects in POAG patients, which helps in the accuracy of diagnosis and monitoring the progression of glaucoma. The patients were divided into early, moderate, and advanced stages of glaucoma according to the “Hodapp, Parrish, and Anderson's classification.” Both the structural and functional measurements were used to assess the parameters associated with disease severity. In the present study, the functional loss was measured by VF loss considering VF MD (dB) and structural loss was assessed using SD-OCT. In our study among early, moderate, and advanced stages of POAG, BCVA, and VF MD showed deterioration with the advancement of the disease. Significant worsening of visual acuity in patients with POAG was noted in the study by Chan et al., who studied the influence of visual acuity deterioration on patients with glaucoma.
Glaucomatous ONH damage is mainly caused by the loss of RGCs and their axons. At the initial stage, these damages remain localized involving the inferior and superior pole of ONH. In our study, thickness of peripapillary RNFL at superior and inferior quadrants was measured and it was observed that the mean superior peripapillary RNFL thickness (88.09 ± 16.67 μm) was thinner compared to mean inferior peripapillary RNFL thickness (91.59 ± 28.75 μm) in the early stage of glaucoma. As the disease progressed to moderate and advanced forms, inferior RNFL thinning was noted. Similar findings of a thinner RNFL in the superior quadrant in the early stages of glaucoma were seen in the study by El-Naby et al. Another study conducted by Soliman et al. reported significant thinning of average superior and inferior quadrant RNFL thickness in POAG patients using SD-OCT. Elbendary and Mohamed Helal too had assessed the role of SD-OCT in different stages of glaucoma and found results similar to our study. Further in their report, a significant correlation was detected between the mean VF index (indicating mean VF deviation from normal) and the mean value of average RNFL thickness loss. The loss of RNFL is responsible for structural alternation of ONH appearance and functional changes in the form of reduction of VF sensitivity. Our study found a significant association between VF MD and average RNFL loss, more so in the early stage of glaucoma (P = 0.0041). Significant VF loss and RNFL thickness correlation were also noted in the study by El-Naby et al. Thus, quantification of the average loss of RNFL is a simple and objective method. It seems possible to use the average loss of RNFL measured by SD-OCT, as an adjunct to perimetry. The RNFL thickness measurement had better reproducibility using SD-OCT, as proved by Kim et al. in their study.
Quantification of peripapillary RNFL thickness is the most common use of SD-OCT in glaucoma. It only analyses the axonal portion of the RGCs without considering the cell bodies and dendrites, which are also affected in glaucoma and reside in the GCL and IPL, respectively. Macular thickness can be affected in glaucomatous eyes, suggesting that it may represent an additional indicator of RGCs damage, given the prominent distribution of these cells within the macular region. In the present study, macular GCL-IPL thickness in the superior and inferior quadrants showed thinning in all stages of glaucoma as measured by SD-OCT. This was in accordance with the study conducted by Kim et al. using SD-OCT, where they noted reduction in GCL thickness values in glaucomatous eyes compared with normal eyes, which was statistically significant in both superior and inferior quadrants.
The present study reported that in advanced glaucoma superior quadrant macular GCL-IPL (57.7 ± 11.38 μm) was thicker in comparison to inferior quadrant macular GCL-IPL (56.35 ± 10.39 μm). The superior and inferior quadrants as well as average macular GCL-IPL thicknesses were statistically more significant with VF MD in early glaucoma (P = 0.0037). In moderate and advanced stages of glaucoma, superior macular GCL-IPL thickness was statistically correlated with VF MD. This finding correlates with the study done by Choi et al., where superior GCL-IPL was significantly thicker on average than the inferior GCL-IPL. They found statistical significance of macular GCL-IPL thickness with MD in early and moderate glaucoma. The relative sparing of the RGCs in the superior macula at the advanced stage remains obscure. This may be due to the inferior temporal lamina cribrosa having larger pores and the least connective tissue support, rendering this region more susceptible for glaucomatous damage. The findings suggest that relatively more ganglion cells exist in the superior macula compared to the inferior macula. Their report further showed that the average and inferior peripapillary RNFL thicknesses had significant correlations with disease severity assessed by MD in early to moderate glaucoma. However, our study had significant correlations of inferior and average peripapillary RNFL thicknesses with MD in early glaucoma.
In a study conducted by Lin et al., it was observed that the superior and inferior quadrants macular GCL thicknesses were significantly correlated with VF MD (R2 = 0.071, P = 0.004; R2 = 0.08, P = 0.002) in early glaucoma among POAG patients. Our results are comparable with this study. The structural damage in glaucoma occurs primarily in the RGC and these cells are predominantly located in the macula, so the structural change in the macular GCL-IPL thickness significantly correlates with functional loss in VF of the early glaucomatous eyes.
Cho et al. demonstrated a significant correlation between SAP and GCL measurements on glaucomatous eyes, indicating GCL measurements are closely associated with functional loss. In our study, significant correlation was demonstrated between SAP and GCL-IPL measurements on glaucomatous eyes, more in early glaucoma followed by moderate and advanced glaucoma. Shin et al. in their study have reported relationship between VF sensitivity and the average GCL-IPL values and peripapillary RNFL values in glaucomatous eyes. In comparative analysis, the association between MD and average GCL-IPL thickness was significantly stronger than that of peripapillary RNFL thickness using the decibel scale (P < 0.001). The association between regional VF sensitivities and inferior GCL-IPL thickness were significantly stronger than peripapillary RNFL thickness (P = 0.001 and 0.007), though in our study inferior peripapillary RNFL thickness was significantly stronger than inferior GCL-IPL thickness (P = 0.0001 and 0.0004). They have recommended macular GCL-IPL over peripapillary RNFL measurements in assessing the structural and functional relationship in the macular region.
The limitations of the present study include the small sample size as a result of which the results cannot be generalized to the entire population. Our study design is cross- sectional. However, these patients need to be followed- up at regular intervals and repeated investigations are needed to confirm our clinical findings. Potential misclassification of the study groups and inaccurate assessment of disease severity based on the VF MD are possible due to the high test-retest variability of the VF test.
| Conclusion|| |
In the present study, both peripapillary RNFL and macular GCL-IPL thickness showed statistically significant correlation with VF MD. In the early stage of glaucoma, inferior peripapillary RNFL thickness was significantly correlating with VF MD. In the moderate stage, superior peripapillary RNFL thickness followed by superior GCL-IPL thickness was significantly correlating with the MD and in the advanced stage superior macular GCL-IPL was thicker than inferior macular GCL-IPL. Thus, both peripapillary RNFL and macular GCL-IPL thicknesses should be assessed along with VF defects to monitor the severity of the disease. We recommend both structural and functional analysis in monitoring the severity of glaucoma as macular thickness changes were well correlated with changes in visual function and RNFL structure in glaucoma.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Thomas R, George R. Interpreting automated perimetry. Indian J Ophthalmol 2001;49:125-40.
] [Full text]
Quigley HA, Dunkelberger GR, Green WR. Retinal ganglion cell atrophy correlated with automated perimetry in human eyes with glaucoma. Am J Ophthalmol 1989;107:453-64.
Hoyt WF, Newman NM. The earliest observable defect in glaucoma? Lancet 1972;1:692-3.
Funk J. Early detection of glaucoma by longitudinal monitoring of the optic disc structure. Graefes Arch Clin Exp Ophthalmol 1991;229:57-61.
Subbiah S, Sankarnanarayanan S, Thomas PA, Jesudasam CA. Comparative evaluation of glaucoma in ocular hypertensives, glaucomatous and normal eyes. Indian J Ophthalmol 2007;55:283-7.
] [Full text]
Lee JR, Jeoung JW, Choi J, Choi JY, Park KH, Kim YD. Structure-function relationships in normal and glaucomatous eyes determined by time- and spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci 2010;51:6424-30.
Tan O, Li G, Lu AT, Varma R, Huang D; Advanced Imaging for Glaucoma Study Group. Mapping of macular substructures with optical coherence tomography for glaucoma diagnosis. Ophthalmology 2008;115:949-56.
Turalba AV, Grosskreutz C. A review of current technology used in evaluating visual function in glaucoma. Semin Ophthalmol 2010;25:309-16.
Anderson DR, Patella VM. Automated Perimetry. 2nd
ed. St. Louis: Mosby and Co; 1999.
Hodapp E, Parrish RK 2nd
, Anderson DR. Clinical Decisions in Glaucoma. St Louis: The CV Mosby Co; 1993. p. 52-61.
Chan EW, Chiang PP, Liao J, Rees G, Wong TY, Lam JS, et al.
Glaucoma and associated visual acuity and field loss significantly affect glaucoma-specific psychosocial functioning. Ophthalmology 2015;122:494-501.
Quigley HA, Sanchez RM, Dunkelberger GR, L'Hernault NL, Baginski TA. Chronic glaucoma selectively damages large optic nerve fibers. Invest Ophthalmol Vis Sci 1987;28:913-20.
El-Naby AE, Abouelkheir HY, Al-Sharkawy HT, Mokbel TH. Correlation of retinal nerve fiber layer thickness and perimetric changes in primary open-angle glaucoma. J Egypt Ophthalmol Soc 2018;111:7-14. [Full text]
Soliman TT, Alazeem Gad EA, Selim SM. Ganglion cell analysis versus retinal nerve fiber layer thickness in glaucoma diagnosis. J Egypt Ophthalmol Soc 2019;112:122-9. [Full text]
Elbendary AM, Mohamed Helal R. Discriminating ability of spectral domain optical coherence tomography in different stages of glaucoma. Saudi J Ophthalmol 2013;27:19-24.
Kim JS, Ishikawa H, Sung KR, Xu J, Wollstein G, Bilonick RA, et al
. Retinal nerve fibre layer thickness measurement reproducibility improved with spectral domain optical coherence tomography. Br J Ophthalmol 2009;93:1057-63.
Choi JA, Shin HY, Park HL, Park CK. The pattern of retinal nerve fiber layer and macular ganglion cell-inner plexiform layer thickness changes in glaucoma. J Ophthalmol 2017;2017:6078365.
Quigley HA, Green WR. The histology of human glaucoma cupping and optic nerve damage: Clinicopathologic correlation in 21 eyes. Ophthalmology 1979;86:1803-30.
Lin PW, Chang HW, Lin JP, Lai IC. Analysis of peripapillary retinal nerve fiber layer and inner macular layers by spectral-domain optical coherence tomography for detection of early glaucoma. Int J Ophthalmol 2018;11:1163-72.
Cho JW, Sung KR, Lee S, Yun SC, Kang SY, Choi J, et al
. Relationship between visual field sensitivity and macular ganglion cell complex thickness as measured by spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci 2010;51:6401-7.
Shin HY, Park HY, Jung KI, Park CK. Comparative study of macular ganglion cell-inner plexiform layer and peripapillary retinal nerve fiber layer measurement: Structure-function analysis. Invest Ophthalmol Vis Sci 2013;54:7344-53.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]