Disc Torsion and Vertical Disc Tilt Are Related to Subfoveal Scleral Thickness in Open-Angle Glaucoma Patients With Myopia

Abstract

Purpose: To find out the characteristics of optic disc morphology in glaucomatous eyes and to evaluate related ocular factors, including the scleral thickness measured from swept-source optical coherence tomography (OCT).

Methods: We compared disc torsion and tilt between 180 normal controls and 180 patients with open-angle glaucoma, matched according to age and axial length based on propensity scores. In a subset of 63 glaucoma and matched control eyes with high myopia, swept-source OCT images of the optic nerve head obtained. The disc ovality and torsion degree were measured from disc photographs, and tilt degree was measured from cross-sectional images of the optic nerve head using swept-source OCT. Scleral thickness was measured from swept-source OCT images at the subfoveal point and 1000 μm away superiorly, inferiorly, temporally, and nasally from the subfoveal point.

Results: The degree of disc tilt and torsion was significantly different between glaucoma (9.3 ± 6.3° and 28.2 ± 19.8°, respectively) and control eyes (6.2 ± 4.1° and 14.1 ± 8.0°, respectively) with similar axial length. The thickness of the inferior sclera and the superior-inferior difference were significantly different between control and glaucomatous eyes with high myopia (P < 0.001 and P = 0.031). The thickness of the inferior sclera and the superior-inferior difference were significantly related to the disc tilt (P = 0.034 for inferior sclera and P < 0.001 for superior-inferior difference) and torsion (P < 0.001 and P < 0.001, respectively) in glaucomatous eyes with high myopia.

Conclusions: Disc tilt and torsion were prominent features of glaucomatous eyes when compared to normal controls with similar axial length and were significantly related to the thickness of the inferior sclera in glaucomatous eyes with myopia.

There is growing evidence that the sclera is important in the pathogenesis of glaucoma.^1–4 The mechanical influence of the peripapillary sclera to the lamina cribrosa is thought to be especially important.⁵ When axial elongation occurs during eyeball development, which is also thought to be mediated by intraocular pressure (IOP), scleral changes occur throughout the posterior sclera. The baseline character of the sclera, including its thickness or stiffness, its components and their proportion, and the architecture or alignment of collagen, may determine scleral changes during eyeball development. From recent studies of myopic glaucoma eyes, we can assume that scleral changes influence the optic disc morphology, such as disc tilting, disc torsion, and peripapillary atrophy.^6–8 Variation in scleral properties can explain differences in glaucoma susceptibility.^9,10 There are possibilities that difference in the baseline properties of the sclera may influence change of the posterior sclera during eyeball elongation and determine the optic disc morphology and susceptibility to glaucoma. However, to date no studies have investigated the relationship between the sclera and the optic disc morphology directly in vivo. 

With advances in optical coherence tomography (OCT) imaging, swept-source OCT at a longer wavelength has made it possible to evaluate the sclera and obtain details of the optic nerve head (ONH) anatomy.^11–13 Previously, we imaged the sclera at the posterior pole using swept-source OCT and measured the subfoveal scleral thickness in glaucoma patients.¹⁴ The detection rate for the posterior border of the sclera and measurement reproducibility of subfoveal scleral thickness using swept-source OCT were found to be favorable in glaucoma patients with high myopia. 

In the present study, we compared the characteristics of optic disc morphology, such as disc torsion and disc tilt, between normal controls and patients with glaucoma who were matched based on propensity scores for age and axial length. In a subset of patients with high myopia, the thickness of the sclera was measured at several points around the posterior pole, and the relationship between measured scleral thickness and optic disc morphology was evaluated. 

Patients and Methods

This prospective study enrolled 374 consecutive patients with open-angle glaucoma seen by a glaucoma specialist (CKP) and 374 normal controls from a dry eye clinic and myopes who underwent a preoperative examination for refractive surgery (LASIK or surface ablation, including LASEK, epi-LASIK, or phakic intraocular lens insertion) at Seoul St. Mary's Hospital from July 2010 to January 2013. The study was performed with the informed consent of the participants and followed all of the guidelines for experimental investigation in human subjects required by the Institutional Review Board of Seoul St. Mary's Hospital. All investigations were performed in accordance with the Declaration of Helsinki. 

All participants underwent a comprehensive ophthalmic assessment, including measurement of best-corrected visual acuity, slit-lamp biomicroscopy, Goldmann applanation tonometry, dilated stereoscopic examination of the ONH and fundus, color disc photography, red-free retinal nerve fiber layer (RNFL) photography (VX-10; Kowa Optimed, Tokyo, Japan), and achromatic automated perimetry using the 24-2 Swedish Interactive Threshold Algorithm standard program (Humphrey Visual Field Analyzer; Carl Zeiss-Meditec, Inc., Dublin, CA, USA). 

Eyes had to have a best-corrected visual acuity of ≥20/40 to be included. Eyes with an axial length > 30.00 mm were excluded. Patients with signs of pathologic myopia or myopic retinopathy (including posterior staphyloma, myopic choroidal neovascularization, lacquer crack, angioid streak) were excluded. Patients with intraocular diseases or neurologic diseases that could cause visual field (VF) loss were excluded. Eyes with consistently unreliable VFs (defined as false negative > 25%, false positive > 25%, and fixation losses > 20%) were also excluded. In cases in which both eyes of a patient were eligible for the study, only one eye was randomly chosen for inclusion. 

Patients were defined as having glaucoma if they had a glaucomatous optic disc associated with typical reproducible VF defects on achromatic automated perimetry. The control group was defined as those with an IOP < 21 mm Hg and no history of increased IOP, an absence of glaucomatous disc, no visible defect in the RNFL, and a normal VF result. Absence of a glaucomatous disc was defined as an intact neuroretinal rim without peripapillary hemorrhage, notches, or localized pallor. A glaucomatous change in the VF was defined as the consistent presence of a cluster of three or more non-edge points on the pattern deviation plot with a probability of occurring in <5% of the normal population, with one of these points having the probability of occurring in <1% of the normal population; a pattern standard deviation with P < 5%; or a glaucoma hemifield test result outside normal limits. Visual field defects had to be repeatable on at least two subsequent tests. 

Propensity Score Matching

A propensity score analysis was performed to match eyes between the glaucoma and control groups according to age and axial length. Propensity scores were estimated using multiple logistic regression analysis¹⁵ for each patient; age and axial length were the explanatory variables. Using predicted probabilities, we sought to match an eye in the glaucoma group with the closest individual in the control group. Then, applying the Greedy 5 → 1 digit match algorithm,¹⁶ we created propensity score–matched pairs without replacement (a 1:1 match). Specifically, we matched each eye whose propensity score was identical to 5 digits. If this could not be done, the algorithm proceeded sequentially to the next highest digit match (4-, 3-, 2-, or 1-digit match) until no further matches were possible. From the initial 374 glaucomatous eyes and 374 control eyes, we were able to match 180 eyes from the glaucoma group with 180 eyes from the control group. 

Measurement of Optic Disc Tilt From Swept-Source OCT Images

Optic disc tilt was identified by the degree of tilt from OCT horizontal and vertical cross-sectional images. New confocal scanning laser ophthalmoscopy and OCT-assisted methods facilitating measurement of disc tilt have been introduced.^17,18 Horizontal disc tilt degree was measured according to a method proposed by Hosseini et al.¹⁷ Vertical disc tilt degree was additionally measured with the same method. We used a DRI-OCT system (Topcon Corp., Tokyo, Japan) with an axial scan rate of 100,000 Hz operated at the 1-μm wavelength region. This OCT uses a light source of a wavelength-sweeping laser centered at 1050 nm, with a repetition rate of 100,000 Hz, yielding 8-μm axial resolution in tissue. A single OCT image consisting of 1000 A lines can be acquired in 10 ms. When obtaining images from dry eye patients, artificial tear drops were instilled and patients were instructed to blink before the scanning. 

A 6-mm line horizontal scan passing through the center of the optic disc was obtained by aligning the scan line using the en face image. The horizontal scan was aligned with an imaginary line passing through the center of the optic disc and the fovea. A 6-mm vertical line scan 90° vertical to the horizontal scan was also obtained. Disc photographs were overlapped with transparency 50% on the en face image (Fig. 1A). The clinical disc margin points were marked using the combined disc photograph and en face OCT images (Figs. 1B, 1C, red dots). A line connecting the two points marking the clinical disc margin on the cross-sectional images was defined as the ONH plane (Figs. 1B, 1C, green line). A line connecting the inner tips of Bruch's membrane (Figs. 1B, 1C, white dots) on each side of the ONH on the cross-sectional images was drawn as the reference plane (Figs. 1B, 1C, white line). Degree of tilt was defined as the angle between the reference plane and the ONH plane. Angle measurements were performed by two observers (H-YLP and SIC) with the angle tool in the software. Degree of vertical tilt was measured from a vertical cross-sectional image (Fig. 1B), and the degree of horizontal tilt was measured from a horizontal cross-sectional image (Fig. 1C). A positive degree of tilt indicated an inferior and temporal tilt, and a negative value indicated a superior and nasal tilt. 

Figure 1

 

Disc tilt was measured from cross-sectional images of optical coherence tomography (OCT). A 6-mm line scan was centered on the optic disc, and vertical and horizontal scans were obtained (A). The disc photograph was overlapped with the en face image of the OCT, and the disc border points where it met Bruch's membrane or border tissue were marked on the scan line (B, C; red glyph). A line connecting the two points marking the clinical disc margin on the cross-sectional images was defined as the optic nerve head (ONH) plane (B, C; green line). A line connecting the inner tips of Bruch's membrane on each side of the ONH (white glyph) on the cross-sectional images was drawn as the reference plane (B, C; white line). The degree of tilt was defined as the angle between the reference plane and the ONH plane ( ). Degree of vertical tilt was measured from a vertical cross-sectional image (B), and the degree of horizontal tilt was measured from a horizontal cross-sectional image (C).

Measurement of Optic Disc Tilt and Torsion

Digital retinal photographs centered on the optic disc and macula region were obtained using standardized settings. Optic disc torsion was measured from photographs by two observers (H-YLP and SIC) using the National Institutes of Health image analysis software (ImageJ 1.40; available from http://rsb.info.nih.gov/ij/index.html, National Institutes of Health, Bethesda, MD, USA [in the public domain]). Disc tilt was identified by the disc ovality ratio, defined as the ratio between the longest and shortest diameters of the optic disc (tilt ratio = LD/SD).^19–21 Disc torsion was identified and defined as the deviation of the long axis of the optic disc from the vertical meridian.^22,23 The vertical meridian was considered a vertical line 90° from a horizontal line connecting the fovea, which is 2° to 6° below the optic disc, to the center of the optic disc. The angle between the vertical meridian and the longest diameter of the optic disc was considered the degree of torsion (Fig. 2). A positive torsion value indicated inferotemporal torsion (which is counterclockwise torsion in the right eye format), and a negative value indicated superonasal torsion (which is clockwise torsion in the right eye format). Details have been described previously.²⁴ 

Figure 2

 

Representative case of a patient with highly myopic normal-tension glaucoma and inferior disc torsion. Inferior disc torsion (7.5°) with a defect in the inferior region of the retinal nerve fiber layer is presented as a superior arcuate defect. Disc torsion was measured from photographs and was identified and defined as the deviation of the long axis of the optic disc from the vertical meridian. The vertical meridian was identified as a vertical line 90° from a horizontal line connecting the fovea to the center of the optic disc. The angle between the vertical meridian and the longest diameter of the optic disc was considered the degree of torsion. The vertical tilt of the disc was 8.3° and horizontal tilt was 3.6°. Scleral thickness was least at the inferior posterior pole (210 μm, arrowheads) compared to subfoveal (380 μm), superior (365 μm), temporal (358 μm), and nasal (372 μm) regions.

Measurement of Scleral Thickness by Swept-Source OCT

A 12-mm line scan centered on the foveal region and two vertical and horizontal scans passing the foveal center were obtained using swept-source OCT. We measured the scleral thickness from both vertical and horizontal scans with use of the caliper function in the OCT software by two observers (H-YLP and J-AC). Details are described elsewhere.¹⁴ Scleral thickness was defined as the distance between the chorioscleral interface and the outer scleral border. The outer surface of the sclera was identified carefully in the OCT scans from retro-ocular structures. Scleral tissues were identified by their lamellar morphologic features and high reflectivity values. Subfoveal scleral thickness was the average of the measurements at the subfoveal point from both vertical and horizontal scans. From the vertical scans, 1000 μm superiorly and 1000 μm inferiorly from the subfoveal point were defined as the superior and inferior thicknesses, respectively. From the horizontal scans, 1000 μm temporally and 1000 μm nasally from the subfoveal point were defined as the temporal and nasal thicknesses, respectively (Fig. 2). Superior-inferior difference was defined as the inferior scleral thickness subtracted from the superior scleral thickness. 

Identification of the VF Defect Locations

The VF examination of glaucomatous eyes was reviewed to identify the location of the VF defects that occurred in <5% of age-matched controls on the pattern deviation map. Eyes with defects within the 26 points of the superior hemifield were classified as the superior VF defect group. Eyes with defects within the 26 points of the inferior hemifield were classified as the inferior VF defect group. Eyes with defects involving both the superior and inferior hemifields were excluded from this subanalysis. 

Statistical Analysis

Statistical analyses were performed using the SPSS ver. 16.0 software (SPSS, Inc., Chicago, IL, USA). The direction of the tilt and torsion of the optic disc were considered when comparing the mean values. Absolute values were used for correlation and regression analyses. The independent t-test and χ² test for independent samples were used to assess differences between groups. Pearson's correlation analysis was used to assess the relationships between ocular parameters and optic disc morphology. To evaluate the strength of association between parameters, R values were compared using MedCalc (MedCalc Software, Inc., Mariakerke, Belgium). Univariate and multivariate linear regression analyses were conducted to identify factors related to tilt and torsion. Dependent variables were tilt and torsion, and independent variables were age, axial length, mean deviation (MD) of VF, central corneal thickness, IOP, average RNFL thickness, thickness of the inferior sclera, and superior-inferior difference of scleral thickness. The variables that retained significance at P < 0.20 in the univariate analysis were included in the multivariate model. The level of statistical significance was set at P < 0.05. 

Results

Subject Baseline Characteristics and Optic Disc Morphology

Age, sex, spherical equivalent refraction, axial length, central corneal thickness, and IOP were similar between the control and glaucomatous eyes. Significant differences were observed in average RNFL thickness among the groups (P < 0.001). Disc ovality was 1.2 ± 0.1 in control eyes and 1.3 ± 0.3 in glaucomatous eyes, which was not significantly different (P = 0.601). However, disc torsion (−0.6 ± 4.7° and 19.7 ± 15.0°) was significantly different between the groups (P < 0.001). The direction of disc torsion was mainly inferiorly in glaucomatous eyes (76.7%), which was significantly different from that of control eyes (33.3%, P < 0.001; Table 1). The relationship between axial length and optic disc morphology differed between control and glaucomatous eyes, with glaucomatous eyes having a wider range of disc torsion (Fig. 3). The distribution of horizontal tilt degree, vertical tilt degree, and torsion degree in patients with myopia is shown in Figure 4. Glaucomatous eyes have greater distribution of vertical tilt degree and torsion degree in the higher range compared to normal controls. 

Table 1

 

Subject Characteristics of the Normal Controls and the Glaucomatous Eyes Matched According to Age and Axial Length by Propensity Score Matching

 

Figure 3

 

Relationship between axial length and optic disc morphology, including disc tilt and torsion. The degree of tilt measured as disc ovality ratio (A) and degree of torsion (B) and the relationship with axial length were significantly different between normal controls and glaucomatous eyes.

Figure 4

 

Histogram showing the distribution of disc tilt and torsion degree measured by optical coherence tomography. Distribution of horizontal disc tilt degree was similar between normal controls and glaucoma patients. The degree of vertical tilt was distributed as 23.8% less than 4°, 39.7% between 4° and 8°, 23.8% between 8° and 12°, and 12.7% more than 12° in normal controls. The degree of vertical tilt was distributed as 6.3%, 23.8%, 42.9%, and 27.0%, respectively, in glaucoma patients showing higher distribution to the greatest vertical tilt. The degree of torsion was distributed as 27.0% less than 10°, 33.3% between 10° and 20°, 27.0% between 20° and 30°, and 12.7% more than 30° in normal controls. The degree of torsion was distributed as 12.7%, 27.0%, 36.5%, and 23.8%, respectively, in glaucoma patients showing higher distribution to the greater disc torsion.

Thickness of the Posterior Sclera in Highly Myopic Eyes

In the subset of patients with high myopia (axial length more than 26.00 mm), swept-source OCT images of 112 glaucomatous eyes and 102 control eyes were obtained. Among those images, the posterior border of the sclera was unclear and could not be identified in 35 (31.3%) of 112 glaucomatous eyes and 33 (32.4%) of 102 control eyes. From the remaining 77 glaucomatous eyes and 69 control eyes with high myopia, 63 pairs were matched according to axial length. 

The spherical equivalent was −6.4 ± 4.5 diopters in the highly myopic control eyes and −6.6 ± 4.0 diopters in the highly myopic glaucomatous eyes. Axial length was 27.1 ± 0.9 mm in the highly myopic control eyes and 27.1 ± 0.6 mm in the highly myopic glaucomatous eyes. 

Degree of horizontal tilt was 4.2 ± 6.2° in highly myopic control eyes and 4.7 ± 6.3° in highly myopic glaucoma eyes, which was not significantly different (P = 0.346). However, both vertical tilt (6.2 ± 4.1° in highly myopic controls and 9.3 ± 6.3° in highly myopic glaucoma) and torsion (14.1 ± 8.0° and 28.2 ± 19.8°) were significantly different between the groups (P = 0.011 and P = 0.006). The direction of disc tilt and torsion was mainly inferior in highly myopic glaucomatous eyes (67.7% and 55.6%), which was significantly different from that of highly myopic control eyes (38.1 and 33.3%, P = 0.001 and P = 0.010; Table 2). 

Table 2

 

Subject Characteristics and Comparison of Scleral Thicknesses in Highly Myopic Eyes With Measured Scleral Thickness via Swept-Source Optical Coherence Tomography

 

The mean (± standard deviation) thickness of the subfoveal sclera was 379.0 ± 61.9 μm in the highly myopic control eyes and 303.5 ± 69.2 μm in the highly myopic glaucomatous eyes, which was significantly different (P = 0.017). Superior, nasal, and temporal scleral thicknesses were not significantly different between the groups (P = 0.656, P = 0.319, and P = 0.945, respectively). However, the thickness of the inferior sclera was significantly different between the groups (P < 0.001), being 333.3 ± 61.7 μm in highly myopic control eyes and 208.4 ± 71.1 μm in highly myopic glaucomatous eyes. The difference in thickness between the superior and inferior sclera was also significant between control and glaucomatous eyes with high myopia (P = 0.031). 

Disc Tilt and Torsion According to the Location of VF Defects

Regarding the location of VF damage in the glaucoma group, 31 eyes had superior VF defects and 19 eyes had inferior VF defects. Degree of horizontal tilt was 4.4 ± 4.9° in the superior VF defect group and 4.7 ± 5.2° in the inferior VF defect group, which was not significantly different (P = 0.572). However, both vertical tilt (2.5 ± 6.5° in superior VF defect group and 9.2 ± 6.5° in inferior VF defect group) and torsion (−2.2 ± 13.3° and 16.5 ± 15.1°) were significantly different between the groups (P < 0.001 and P < 0.001). The direction of disc tilt and torsion was mainly inferior in the superior VF defect group (84.2% and 63.2%), which was significantly different from that in the inferior VF defect group (38.7% and 25.8%, P = 0.002 and P = 0.010; Table 3). 

Table 3

 

Comparison of the Optic Disc Characteristics in Highly Myopic Glaucomatous Eyes According to the Location of Visual Field (VF) Defects

 

Relationship Between the Ocular Factors and Optic Disc Morphology

The relationship between ocular factors and optic disc morphology was assessed in all glaucoma and control eyes with high myopia (Table 4). Degree of horizontal tilt (R = 0.026, P = 0.016), degree of vertical tilt (R = 0.018, P = 0.038), and degree of torsion (R = 0.862, P = 0.007) were significantly correlated with axial length. The relationship between spherical equivalent refraction and optic disc morphology were also significantly correlated. Age and VF MD were not significantly correlated with optic disc morphology. 

Table 4

 

Relationship Between Ocular Characteristics and Optic Disc Morphology in Highly Myopic Control and Glaucomatous Eyes

 

The subfoveal, superior, inferior, nasal, and temporal scleral thicknesses had a weak negative correlation with the degree of horizontal and vertical tilt. The subfoveal, inferior, nasal, and temporal scleral thicknesses had a moderate negative correlation with the degree of torsion. The thickness of the inferior sclera showed a moderate negative correlation with degree of torsion (R = −0.414, P = 0.006). The superior-inferior difference of scleral thickness was moderately correlated with the degree of vertical tilt (R = 0.329, P = 0.007) and degree of torsion (R = 0.341, P = 0.025; Table 4). 

The relationship between axial length and degree of vertical tilt was significantly different between the control eyes (R = 0.045, P = 0.011) and glaucomatous eyes (R = 0.174, P = < 0.001; P < 0.001 by comparing the correlation coefficient). The relationship between axial length and degree of torsion was significantly different between control eyes (R = −0.026, P = 0.002) and glaucomatous eyes (R = 0.199, P = 0.004; P < 0.001 by comparing the correlation coefficient). 

Univariate and multivariate regression analyses were performed to determine the relationship between the ocular parameters and the morphology of the optic disc. Axial length, thickness of the inferior sclera, and superior-inferior difference of scleral thickness were significant related to degree of vertical tilt. Only the thickness of the inferior sclera and superior-inferior difference of scleral thickness were significantly related to degree of torsion in highly myopic glaucomatous eyes (Table 5). 

Table 5

 

Univariate and Multivariate Regression Analyses of Highly Myopic Glaucomatous Eyes

 

Discussion

Glaucomatous eyes had more prominent disc torsion than control eyes of similar age and axial length, and the direction was mainly inferiorly rotated. The relationship between axial length and disc torsion was significantly different between control and glaucomatous eyes. This shows that glaucomatous eyes have more prominent optic disc morphological changes in terms of disc torsion, independent from axial length. The posterior sclera was significantly thinner in the inferior region of glaucomatous eyes compared to control eyes with high myopia. The thickness of the inferior sclera and the difference in thickness between the superior and inferior sclera were significantly correlated with the degree of vertical tilt and torsion in myopic glaucomatous eyes. Since disc tilt and torsion were significantly more frequent to the inferior direction, there are possibilities that optic disc morphological changes may be related to changes in the inferior scleral thinning. 

We have observed the role of the posterior sclera clinically in myopic eyes. Many studies, including population-based studies, have demonstrated a relatively high prevalence rate of glaucoma in patients with myopia.^25–27 Structural changes, such as an unusually large or skewed sclera canal shape, thinning of the lamina cribrosa, and marked thinning of the sclera, mainly contribute to glaucomatous changes in patients with myopia.^28–30 Tilt and torsion of the optic disc represent skewed insertions of the optic nerve into the globe and could originate during eyeball elongation due to myopia.^30,31 Our results show that glaucomatous eyes have different optic disc morphologies compared to control eyes of similar age and degree of myopia. The relationship between axial length and optic disc morphology differed between control and glaucomatous eyes, with glaucomatous eyes having more prominent disc torsion (Fig. 3). Vertical tilt and torsion and their direction were related to the location of VF defect. The thickness of the posterior sclera was significantly less in the inferior region of glaucoma eyes with high myopia, and was related to the degree of vertical tilt and torsion. As shown in a representative case in Figure 2, inferior direction of torsion and vertical tilt is associated with inferior nerve fiber layer bundle defect with superior VF defect. The inferior subfoveal scleral thickness is least compared to that in other regions around the posterior pole. These results may suggest that difference in the posterior sclera could be related to changes of the optic disc, and could contribute to glaucomatous change in the corresponding region. However, further investigation is needed to reveal the pathogenesis of the relationship found in this study. 

In vivo imaging of the sclera has become possible with the introduction of swept-source OCT. Previous studies reported that in 57% to 58% of subjects, the entire scleral layer can be visualized by swept-source OCT.^11–13 In one study, the entire scleral layer was observed in 57% of 488 highly myopic eyes with a mean axial length of 29.9 ± 2.0 mm.¹³ Another study reported that the entire scleral layer was visible in 14 (58%) of 24 myopic eyes with a mean axial length of 25.05 ± 1.26 mm.³² As the entire sclera is more visible in myopic eyes, we measured the thickness in the subfoveal region and analyzed this parameter only in highly myopic eyes. The peripapillary scleral border is more difficult to identify with the current technique. Although we indirectly measured the subfoveal scleral thickness, instead of the peripapillary scleral thickness, subfoveal scleral thickness is reported to be highly correlated with peripapillary scleral thickness in human eyes.³³ More advanced techniques that improve imaging of the sclera may further improve measurements in eyes with various axial lengths to confirm our findings. 

There are several limitations to the present study. We could not recruit all the participants to undergo swept-source OCT, and only a subset of patients with high myopia was analyzed. Imaging the sclera with swept-source OCT still has limitations. Approximately a third of obtained images were excluded in the analysis for unclear posterior border of the sclera. Measurement was possible in only a limited proportion of images, not confirmed by histology. The measured scleral thickness in vivo at present may not be valid, but comparisons can be made if the measurement is repeatable and reproducible.³⁴ Additionally, measurement of the sclera is reported to be influenced by the thickness and reflectivity of the retina and the choroid.¹⁵ Glaucomatous eyes have thinner retina due to reduced RNFL and different tissue reflectivity when compared to normal control eyes. This should be kept in mind when interpreting findings from the present study. Finally, several properties of the sclera, other than thickness, contribute to and influence the ONH. Various factors, particularly scleral stiffness, should be considered when interpreting our findings. However, measuring other factors in human eyes in vivo is not possible. Our findings should be confirmed by further studies. 

In conclusion, glaucomatous eyes had a more prominent vertical tilt and torsion of the optic disc than control eyes of similar age and axial length, and the direction was mainly inferiorly torsioned and tilted. The inferior disc tilt and torsion may be related to changes in the inferior region of the posterior sclera in glaucomatous eyes with high myopia. 

Acknowledgments

Supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (No. 2014003087). The authors alone are responsible for the content and writing of the paper. 

Disclosure: H.-Y.L. Park, None; S.I. Choi, None; J.-A Choi, None; C.K. Park, None