Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com Clinical science The height of the posterior staphyloma and corneal hysteresis is associated with the scleral thickness at the staphyloma region in highly myopic normal-tension glaucoma eyes Jong Hyuk Park,1 Kyu-Ryong Choi,2 Chan Yun Kim,1 Sung Soo Kim1

1Department of Ophthalmology, ABSTRACT optic nerve head (ONH) in eyes with high – Institute of Vision Research, Aims To evaluate the characteristics of the posterior myopia.8 10 Yonsei University College of Medicine, Seoul, Republic of segments of eyes with high myopia and normal-tension Recent studies have demonstrated the importance Korea glaucoma (NTG) and identify which ocular factors are of the posterior ocular structures such as the lamina 2Department of Ophthalmology most associated with scleral thickness and posterior cribrosa and sclera in the pathogenesis of glau- – and Institute of Ophthalmology staphyloma height. coma.8 10 The mechanical influence of the peripa- & Optometry, Ewha Womans Methods The study included 45 patients with highly pillary sclera on the lamina cribrosa is considered University School of Medicine, Seoul, Republic of Korea myopic NTG and 38 controls with highly myopic eyes to be important in that scleral deformations affect (≤−6D or axial length ≥26.0 mm). The subfoveal the stiffness and thickness of the sclera and affect Correspondence to retinal, choroidal, scleral thickness and the posterior the ONH biomechanics by exacerbating strain and Professor Sung Soo Kim, staphyloma heights were examined from enhanced depth stress.11 The increased risk of glaucoma in myopic Department of Ophthalmology, College of Medicine, Yonsei imaging spectral-domain optical coherence tomography eyes may be related in part to the mechanical prop- University, 134 Sinchon-dong, and compared between two groups. erties of the sclera. There have been several Seodaemun-gu, Results Highly myopic NTG eyes had thinner subfoveal hypotheses regarding scleral remodelling mechan- Seoul 120-752, Korea; scleral thickness (473.03±43.75 vs 579.46±75.87 mm, isms including narrowing and dissociation of [email protected] p<0.001) and higher posterior staphyloma (97.80 collagen fibres or downregulation of certain extra- m Received 9 June 2015 ±70.10 vs 62.83±32.01 m, p=0.027) than highly cellular matrix components constituting the 12 13 Revised 29 October 2015 myopic, non-glaucomatous eyes. Subfoveal scleral sclera. However, few studies have investigated Accepted 1 November 2015 thickness was significantly correlated with age, axial the association between scleral mechanical changes Published Online First length, corneal hysteresis and the posterior staphyloma and the increased risk of glaucoma in highly 11 December 2015 height of the superior quadrant, the nasal quadrant and myopic eyes. the arithmetic mean of four quadrants in highly myopic Posterior staphyloma is a protrusion of the pos- NTG eyes. Corneal hysteresis (β=2.694, p=0.015), terior shell of the globe, which is considered to be corneal resistance factor (β=−2.916, p=0.010) and the a hallmark of high myopia. The scleral shell of posterior staphyloma height of the nasal quadrant highly myopic eyes has increased elasticity and a (β=−0.463, p=0.017) were significantly associated with tendency to expand. Recent reports showed that the subfoveal scleral thickness in highly myopic NTG eyes. there is a significant difference in subfoveal scleral Conclusion Subfoveal scleral thinning and non-uniform thickness depending on the presence of posterior posterior staphyloma were closely related in highly myopic staphyloma, and the sclera at the posterior pole in NTG eyes. Corneal hysteresis, corneal resistance factor and highly myopic eyes might stretch inhomogen- the nasal posterior staphyloma height were associated eously.14 Mechanical stretching nasally and super- with the scleral thickness. iorly around the fovea most significantly influences subfoveal scleral thinning, and staphyloma in this area may be associated with axial elongation.15 Thus, posterior staphyloma formation is recognised INTRODUCTION as an important risk factor for high myopia-related High myopia is one of the leading causes of visual diseases. disturbance in the world. Asian countries have Therefore, the aim of this study is to evaluate the higher prevalence of severe myopia (≤−6D), which characteristics of the posterior segments of eyes is reported to be 5.5% in Japan,1 3.6% in with high myopia and normal-tension glaucoma Singapore,2 2.8% in China3 and 2.3% in Korea,4 (NTG) and identify which ocular factors are most but it is estimated that 1% of the global population associated with scleral thickness and posterior sta- exhibits high myopia.5 High myopia is associated phyloma height. with an axial elongation of the globe, leading to the stretching and thinning of the posterior ocular PATIENTS AND METHODS tissues and further pathological changes.6 This retrospective study was approved by the Pathological study show that highly myopic eyes Institutional Review Board (IRB) of Ewha Womans have a significantly thinner retina, choroid and University Mokdong Hospital, Seoul, Korea, and 7 To cite: Park JH, Choi K-R, sclera than age-matched controls without myopia. was conducted in accordance with the tenets of the Kim CY, et al. Br J Structural changes in posterior ocular tissues are Declaration of Helsinki. This study investigated 45 Ophthalmol considered to be the foundation of the visual adults with NTG and high myopia and 38 controls – 2016;100:1251 1256. impairments and the glaucomatous changes in the with high myopia. Participants were recruited from

Park JH, et al. Br J Ophthalmol 2016;100:1251–1256. doi:10.1136/bjophthalmol-2015-307292 1251 Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com Clinical science

Ewha Womans University Mokdong Hospital between October Measurement of posterior ocular tissues thickness 2013 and March 2014. The inclusion criteria were (1) phakic Retinal, choroidal and scleral thicknesses were measured at highly myopic eyes, defined as a spherical equivalent refractive the fovea. Two independent observers ( JHP, So-Yeon Kim) error less than −6.0 D or an axial length longer than 26.0 mm; used the calliper function of the built-in optical coherence (2) no posterior abnormalities such as choroidal neovascularisa- tomography (OCT) software to make the measurement in a tion, macular schisis, macular hole, retinal detachment and (3) blind fashion. Retinal thickness was defined as the vertical best-corrected visual acuity (BCVA) better than 20/40. The distance from the retinal pigment epithelium (RPE) (the exclusion criteria were (1) eyes with media opacities such as outermost hyper-reflective line at the retina–choroidal inter- dense cataracts that prevent detailed fundus observation; (2) face) to the retinal surface. Choroidal thickness was defined aphakic or pseudophakic eyes; (3) eyes with axial lengths longer as the vertical distance from the RPE line to the hyper- than 30.0 mm and (4) eyes undergoing refractive surgery. After reflective line behind the large vessel layers of the choroid, ophthalmic evaluation, the right eye was selected for inclusion which was presumed to be the choroid–sclera interface. Both in cases in which both eyes of the patient were eligible for the retinal and choroidal thicknesses were measured in the hori- study. zontal and vertical sections running through the centre of the NTG was defined as the presence of glaucomatous optic disc fovea, and the arithmetic mean of both two measurements damage, abnormal visual field (VF) test results and intraocular was used in this study. Scleral thickness was defined as the dis- pressure (IOP) ≤21 mm Hg without topical medical treatment. tance from the choroid–sclera interface to the outer scleral Signs of glaucomatous optic disc damage were as follows: border and was measured in eyes whose outer scleral border diffuse or localised neuroretinal rim loss, excavation and/or could be clearly distinguished from fat tissues in the retinal nerve fibre layer (RNFL) defects. An abnormal VF was retro-ocular structures. Subfoveal scleral thickness was defined as a pattern SD outside of the 95% normal confidence defined as the average of three measurements—from the sub- limits or a Glaucoma Hemifield Test result outside normal foveal point and 1000 mm temporal and 1000 mm nasal from limits. At least two consecutive abnormal VF examinations were the subfoveal point. All measurements were performed per- required, with the most recent test performed within 12 months pendicular to the RPE line. of enrolment. The control group was defined as those having IOP ≤21 mm Hg, with no history of increased IOP, the absence of glaucomatous disc appearance and normal VF results. Measurement of the posterior staphyloma height The presence and location of posterior staphyloma was deter- mined by stereoscopic fundus examination using binocular indir- ect ophthalmoscopy. Selection of posterior staphyloma was Examinations performed in eyes with type II staphyloma (classification by The clinical examinations included measurement of BCVA, Curtin16), which involves the macular region, and in eyes in slit-lamp biomicroscopy with or without contact lens, which the curvature of the inner scleral surface was symmetric- Goldmann applanation tonometry, refractive error (spherical ally centred on the fovea in enhanced depth imaging OCT equivalent) by an autorefractometer (Topcon, Tokyo, Japan), (EDI-OCT) images. We excluded eyes with OCT images in axial length by partial coherence interferometry (IOL Master; which the fovea was not situated at the bottom of the curvature Carl Zeiss Meditec, Dublin, California, USA), dilated fundo- and/or the contour of the inner scleral surface was irregular or scopic examination using a 90-D lens, automated perimetry curved posteriorly with the curvature asymmetrical around the using the 30-2 Swedish Interactive Threshold Algorithm stand- central fovea. The posterior staphyloma height or depth was ard programme (Humphrey Visual Field Analyzer; Carl Zeiss defined as the distance from the RPE line beneath the fovea to Meditec, Dublin, California, USA) and Spectralis SD-OCT the nasal and temporal edge of the horizontal scan and superior (Heidelberg Instruments, Heidelberg, Germany). Central and inferior vertical scan including the fovea. Specifically, each corneal thickness (CCT) and corneal biomechanical properties posterior staphyloma height was measured as the vertical dis- such as corneal hysteresis (CH) or corneal resistance factor tance between the subfoveal RPE and the peripheral RPE 2 mm (CRF) were measured using ultrasound pachymetry (Tomey, from the fovea in the four quadrants. The arithmetic mean of Nagoya, Japan) and the Ocular Response Analyzer (Reichert these four measurements was used as the posterior staphyloma Ophthalmic Instruments, Depew, New York, USA). height in this study (figure 1).

Figure 1 The posterior staphyloma height or depth was defined as the distance from the RPE line beneath the fovea to the nasal and temporal edge of the horizontal scan and superior and inferior vertical scan including the fovea. Specifically, each posterior staphyloma height was measured as the horizontal distance between the subfoveal RPE and the peripheral RPE 2 mm from the fovea in each of the four quadrants. The arithmetical mean of these four measurements was used as the posterior staphyloma height. HI, height inferior; HN, height nasal; HS, height superior; HT, height temporal.

1252 Park JH, et al. Br J Ophthalmol 2016;100:1251–1256. doi:10.1136/bjophthalmol-2015-307292 Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com Clinical science

Measurement reproducibility Table 1 Baseline characteristics of patients with normal-tension In this study, the points of the chorioscleral border and the glaucoma (NTG) and controls with myopia outer scleral border were decided manually by two blinded observers. The intraobserver reproducibility was evaluated with Control NTG the intraclass correlation coefficient (ICC) from a two-way (n=38) (n=45) p Value mixed-effect model. The interobserver reproducibility was eval- Age (year) 44.75±10.71 49.44±11.15 0.065 uated with the averaged ICC values calculated by the Gender, female (%) 19 (50.0%) 15 (33.3%) 0.179 Spearman-Brown prophecy formula. logMAR visual acuity 0.01±0.02 0.02±0.05 0.150 Spherical equivalent (D) −7.98±1.44 −7.36±2.28 0.137 Statistical analysis Axial length (mm) 26.59±0.91 27.11±1.11 0.055 Differences for continuous variables including age, refraction Mean K (D) 43.94±0.59 43.22±1.11 0.005 error, axial length, CH and scleral thickness were compared Central corneal thickness (mm) 571.72±36.61 549.01±36.33 0.018 between the controls and the eyes with NTG using a Student’st CH (mm Hg) 10.97±1.76 10.43±1.70 0.039 test. Differences for categorical data including gender and the CRF (mm Hg) 11.43±1.89 10.37±1.81 0.033 2 prevalence of a posterior staphyloma were compared using a χ Intraocular pressure (mm Hg) 13.95±2.55 13.11±2.95 0.061 test. The correlations between the subfoveal scleral and chor- Average RNFL thickness (mm) 84.11±8.83 69.84±13.78 <0.001 oidal thicknesses and the various ocular factors were analysed Mean deviation of perimetry (dB) −1.14±1.11 −5.64±4.48 <0.001 using Pearson tests for rank correlation coefficients. A multivari- Posterior staphyloma, no. of eyes (%) 24 (63.2%) 32 (71.1%) 0.487 ate linear regression model was created to investigate the associ- Values are presented as mean±SD. ation between the subfoveal scleral thickness and the patient’s CH, corneal hysteresis; CRF, corneal resistance factor; dB, decibel; RNFL, retinal nerve age and measured continuous variables. Statistical significance fibre layer. was defined as p value less than 0.05. Statistical analyses were performed using the SPSS software (V.21.0; SPSS, Chicago, 89.31±121.82 mm in the NTG group (p=0.037). The posterior Illinois, USA). staphyloma heights of the inferior and temporal areas were not significantly different between the control and NTG groups RESULTS (table 2). EDI-OCT images of 45 patients with NTG and 38 controls were obtained. Of these 83 eyes, 32 of the patients with NTG Relationships between ocular factors and subfoveal (71.1%) and 24 (63.2%) of the controls satisfied two criteria; scleral thickness the curvature of the inner scleral surface was symmetrically Before the regression analysis, univariate analysis using the centred on the fovea and the outer scleral border was clear and Pearson correlation test was performed to determine the correl- identifiable. These qualified eyes were considered for the meas- ation between subfoveal scleral thickness and various ocular urement and analysis of posterior ocular factors. Intraobserver factors. In the control group, there were no significant relation- and interobserver reproducibility showed excellent measurement ships between subfoveal scleral thickness and ocular factors such reproducibility, with an ICC of 0.970 and 0.996, respectively, as age, BCVA, axial length, CCT, CH, CRF, mean deviation and a mean ICC of 0.959 (0.926–0.977) for the subfoveal (MD) of perimetry and posterior staphyloma heights. scleral thickness. The interobserver reproducibility of retinal In the NTG group, the subfoveal scleral thickness showed a and choroidal thickness measurements also showed excellent significant negative correlation with age (r=−0.452, p=0.014), reproducibility with a mean ICC of 0.995 (0.991–0.997) and axial length (r=−0.343, p=0.040) and the superior (r=−0.424, 0.989 (0.982–0.994), respectively. p=0.022), nasal (r=−0.538, p=0.003) and the arithmetic mean of posterior staphyloma heights (r=−0.519, p=0.004). In con- Baseline characteristics trast, the subfoveal scleral thickness had a significant positive Age, gender, logMAR visual acuity, refractive error and axial correlation with CH (r=0.460, p=0.010). The subfoveal scleral length were similar in the control and NTG groups. There were thickness was not significantly associated with BCVA, CCT, fi signi cant differences between the control and NTG groups with CRF, MD of perimetry or average RNFL thickness in the NTG regard to the following corneal factors: mean K value (p=0.005), group (table 3). Also, in the NTG group, the posterior CCT (p=0.018), CH (p=0.039) and CRF (p=0.033) (table 1).

Thicknesses of the posterior ocular factors Table 2 Posterior ocular factors of patients with normal-tension The mean thicknesses of the subfoveal retina were 227.21 glaucoma (NTG) and controls with myopia ±15.78 mm (mean±SD) in the control group and 217.75 Control (n=24) NTG (n=32) p Value ±20.91 mm in the NTG group (p=0.408). The subfoveal chor- oidal thicknesses were 192.83±66.98 mm in the control group Subfoveal retinal thickness (mm) 227.21±15.78 217.75±20.91 0.408 and 148.25±72.85 mm in the NTG group (p=0.104). The sub- Subfoveal choroidal thickness (mm) 192.83±66.98 148.25±72.85 0.104 foveal scleral thicknesses were 579.46±75.87 μm in the control Subfoveal scleral thickness (mm) 579.46±75.87 473.03±43.75 <0.001 group and 473.03±43.75 mm in the NTG group (p<0.001). Posterior staphyloma height* (mm) 62.83±32.01 97.80±70.10 0.027 The posterior staphyloma heights of the arithmetic mean of Superior 96.38±82.01 161.78±159.35 0.044 four directions were 62.83±32.01 mm in the control group and Inferior 27.71±59.81 29.97±97.63 0.921 97.80±70.10 mm in the NTG group (p=0.027). The posterior Nasal 41.63±55.17 89.31±121.82 0.037 staphyloma heights of the superior area were 96.38±82.01 mm Temporal 85.50±59.05 110.03±84.53 0.230 in the control group and 161.78±159.35 mm in the NTG Values are presented as mean±SD. group (p=0.044). The posterior staphyloma heights of the *The arithmetical mean of four measurements was used as posterior staphyloma height. nasal area were 41.63±55.17 mm in the control group and

Park JH, et al. Br J Ophthalmol 2016;100:1251–1256. doi:10.1136/bjophthalmol-2015-307292 1253 Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com Clinical science

Table 3 Relationship of scleral thickness with age, axial length, Table 4 Relationship of the PSH with age, spherical equivalent, central corneal thickness, corneal hysteresis, corneal resistance axial length, central corneal thickness, corneal hysteresis, corneal factor, mean deviation (MD) of visual field and posterior resistance factor and mean deviation (MD) of visual field in patients staphyloma height in patients with normal-tension glaucoma (NTG) with normal-tension glaucoma (NTG) and myopia and myopia Control NTG Control NTG r p Value r p Value r p Value r p Value PSH* Subfoveal scleral thickness Age 0.999 0.646 0.527 0.003 Age −0.237 0.157 −0.453 0.014 Spherical equivalent −0.262 0.216 −0.576 0.001 Axial length −0.101 0.591 −0.343 0.040 Axial length 0.330 0.155 0.836 <0.001 Central corneal thickness 0.192 0.381 −0.141 0.443 Central corneal thickness 0.555 0.074 −0.350 0.111 Corneal hysteresis −0.082 0.718 0.460 0.010 Corneal hysteresis 0.169 0.547 −0.523 0.010 Corneal resistance factor −0.045 0.842 0.220 0.242 Corneal resistance factor 0.099 0.725 −0.502 0.015 MD of perimetry −0.042 0.807 0.336 0.075 MD of perimetry −0.092 0.670 −0.431 0.020 RNFL thickness 0.458 0.024 0.020 0.916 RNFL thickness −0.282 0.182 −0.371 0.047 PSH* 0.235 0.268 −0.424 0.022 Subfoveal retinal thickness 0.400 0.063 −0.218 0.256 PSH, superior 0.019 0.928 −0.538 0.003 Subfoveal choroidal thickness −0.475 0.019 −0.777 <0.001 PSH, inferior 0.407 0.054 −0.093 0.571 Subfoveal scleral thickness 0.141 0.512 −0.302 0.031 − PSH, nasal 0.333 0.112 0.519 0.004 The bold values indicate associations that were statistically significant (p<0.05). PSH, temporal −0.110 0.610 0.160 0.330 *The arithmetical mean of four measurements was used as PSH. PSH, posterior staphyloma height; RNFL, retinal nerve fibre layer. The bold values indicate associations that were statistically significant (p<0.05). *The arithmetical mean of four measurements was used as PSH. PSH, posterior staphyloma height; RNFL, retinal nerve fibre layer. determining ONH biomechanics due to IOP-related stress and strain.91020Quigley et al reported that mice with a mutation in the collagen gene, the main component of the sclera, may staphyloma height showed a significant negative correlation respond differently to IOP elevation.21 Elevated IOP caused with spherical equivalent (r=−0.576, p=0.001), CH (r= scleral remodelling and alterations in the composition of the −0.523, p=0.010), CRF (r=−0.502, p=0.015), MD of perim- sclera extracellular matrix.11 In fact, however, the scleral stiff- etry (r=−0.431, p=0.020), RNFL thickness (r=−0.371, ness has been considered the most influential factor in several p=0.047) and subfoveal choroidal thickness (r=−0.777, studies.22 23 It is possible that there is a compensatory mechan- p<0.001). On the contrary, the posterior staphyloma height has ism whereby individuals with thinner sclera have tissue that is a significant positive correlation with age (r=0.527, p=0.003) stiffer, resulting in biomechanical behaviour similar to a thicker, and axial length (r=0.836, p<0.001) (table 4). weaker tissue.24 In addition, Sigal has demonstrated that levels Stepwise multiple regression analysis in the NTG group of IOP-induced strain within the lamina cribrosa can be high demonstrated that three factors were significantly associated when the sclera is stiff or low when the sclera is compliant.24 with the subfoveal scleral thickness: CH (β=2.694, p=0.015), Recently published models have pointed out that the ocular con- CRF (β=−2.916, p=0.010) and the posterior staphyloma nective tissues have non-linear viscoelasticity properties, as well height of the nasal quadrant (β=−0.463, p=0.017) (table 5). as mechanical properties, based on Laplace’slaw.25 26 The increased risk of glaucoma in myopic eyes may be related to DISCUSSION these material properties of the sclera. The goal of this study was to investigate the characteristics of the posterior segments of eyes with high myopia and NTG and Table 5 Multiple linear regression analyses with scleral thickness specifically identify which ocular factors were significantly asso- in patients with normal-tension glaucoma and myopia ciated with scleral thickness and posterior staphyloma height. This information is valuable for understanding the association Univariate model Multivariate model between high myopia and glaucomatous optic neuropathy, and Variable Coefficient p Value Coefficient p Value it may provide insight into the relationships between material − properties of the eye and the pathogenesis of glaucoma in Age 0.663 0.090 − myopic eyes. Spherical equivalent 0.454 0.142 − A major result of this study was that subfoveal scleral thick- Axial length 0.674 0.105 − ness was thinner in highly myopic NTG eyes compared with Central corneal thickness 0.517 0.337 myopic controls. According to previous studies, a thinner sclera Corneal hysteresis 2.181 0.016 2.694 0.015 − − was found in highly myopic eyes than in eyes without axial Corneal resistance factor 2.045 0.011 2.916 0.010 elongation,14 and the subfoveal scleral thinning was more prom- MD of perimetry 0.344 0.276 inent in NTG eyes than in primary open angle glaucoma eyes Average RNFL thickness 0.110 0.746 † − compared with controls with high myopia.17 Other studies have PSH 0.302 0.112 − observed scleral thinning in glaucoma by investigating the non- PSH, superior 0.066 0.732 linear properties of the sclera18 19 and have demonstrated that PSH, inferior 0.111 0.566 − − scleral deformations have significant effects on the biomechan- PSH, nasal 0.482 0.008 0.463 0.017 − ical environment of the ONH complex. Modelling studies have PSH, temporal 0.046 0.812 indicated that several properties, such as scleral geometry, thick- *The arithmetical mean of four measurements was used as PSH. ness and material properties, are all influential factors for MD, mean deviation; PSH, posterior staphyloma height; RNFL, retinal nerve fibre layer.

1254 Park JH, et al. Br J Ophthalmol 2016;100:1251–1256. doi:10.1136/bjophthalmol-2015-307292 Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com Clinical science

The outermost structures of the eye, consisting of the cornea, regression models showed that the nasal posterior staphyloma sclera and lamina cribrosa, originate from neural crest-derived height was significantly associated with subfoveal scleral thick- mesenchymal cells.27 Recent reports have shown that the ONH ness in myopic NTG eyes. This indicates that there may be a and peripapillary sclera behave as a complete system, and scleral geomorphological directivity for posterior staphyloma and the changes may result in the stretching and distortion of the optic characteristic shape of the staphyloma around the fovea may nerve fibres and may also lead to optic disc changes, such as disc influence scleral thickness at the fovea in myopic NTG eyes. tilting, disc torsion and peripapillary atrophy, leading to damage Some studies on pathological myopia have demonstrated that of the axons of the retinal ganglion cells.28 29 The cornea and the posterior staphyloma deepens and its morphological features sclera also have common histological properties, in that the change as patients age, even though the axial length does not corneal stroma consists of interwoven collagen fibrils that con- change significantly.36 Ohno-Matsui et al36 reported that as tinue to the sclera. The number, arrangement and types of colla- patients age, type IX staphyloma develops secondarily to type II gen fibrils are thought to determine the mechanical strength of staphyloma due to formation of a ridge-like protrusion temporal the outer coat of the eye.30 CH is a physical property related to to the optic disc, and increased mechanical tension caused by the ability of connective tissues to dampen pressure changes. increased posterior staphyloma depth may facilitate the develop- CH is tied to extracellular matrix constituents of the cornea that ment and progression of myopic retinal degeneration and aggra- may be related to those of the posterior ocular tissues.31 Wells vation of VF defects. Our results showed that the posterior et al32 showed that CH was associated with increased deform- staphyloma heights had a significant positive correlation with age ation of the optic nerve surface during transient IOP elevation in highly myopic NTG eyes, which suggests that there may be an and demonstrated that CH may be a surrogate measure of ageing effect on the morphological changes of the posterior sta- lamina cribrosa backward bowing/optic disc compliance in the phyloma in our study groups. However, a precise morphological pathogenesis of glaucoma. This is supported by Kotecha et al’s classification was not performed in our study, because our selec- findings that CH tends to decrease with high IOP, possibly due tion of posterior staphyloma was restricted to EDI-OCT images to remodelling in the cornea in response to IOP elevation.33 In in which the inner scleral borders were symmetrically centred on our study, CH and the subfoveal scleral thickness were signifi- the fovea, which probably represented type I or type II staphy- cantly lower in highly myopic NTG eyes than myopic controls, loma. In addition, the compound staphylomas, including type IX and the subfoveal scleral thickness had a significant positive cor- staphyloma, were at risk of being omitted. Further research is relation with CH in highly myopic NTG eyes but not in myopic necessary to investigate why the morphological features of pos- controls. CH was one of the factors significantly associated with terior staphyloma get worse with increasing age. the subfoveal scleral thickness in myopic NTG eyes using a step- We also found that the relationships between scleral thickness wise multiple regression model. These findings suggest that CH and ocular factors were significant only in highly myopic NTG can be the material property that indicates the viscoelasticity of eyes but not in myopic controls. And such was also in the rela- the cornea and the outer coat of ocular tissues including the tionships between posterior staphyloma heights and ocular cornea, sclera and ONH complex. Furthermore, CH may be a factors. From this analysis, we can assume that posterior staphy- characteristic that can be used to determine the structural sus- loma or deepening is probably due to the myopic mechanism of ceptibility of the posterior ocular tissues. With this in mind, a axial elongation, which increases tension by mechanical stretch- thinner sclera in myopic NTG eyes may be related to lower CH, ing, and/or another mechanism related to glaucomatous changes which may imply an increased stiffness of the sclera. These in the posterior fundus or a combination of glaucoma and/or hypotheses explain the increased risk of glaucoma in myopic myopia effects. Our findings suggest a biomechanical link eyes; however, more research is needed to determine the precise between high myopia and glaucoma, in which the higher degree relationship between scleral thickness and stiffness. of myopia is associated with increased axial elongation, Several studies have demonstrated that posterior staphyloma increased mechanical tension, decreased scleral thickness, formation is related to posterior retinal diseases, especially high increased scleral stiffness, deeper posterior staphyloma, lower myopia-specific diseases, including macular retinoschisis, CH and eventually the deterioration of VF defects. macular holes and choroidal neovascularisation.34 35 This con- Our study had several limitations including interpreting the nection is probably because deeper posterior staphyloma contri- ocular factors measured from OCT imaging techniques. First, butes to increased inward vector force that detaches or splits the the scleral thickness measurements were not confirmed by hist- neural retina in highly myopic eyes.29 In the present study, ology, and therefore may not be valid compared with the actual along with the scleral thinning, we detected significant increases thickness of the tissue in vivo. Second, we measured the retinal, in the posterior staphyloma heights of the nasal, superior quad- choroidal and scleral thicknesses only at the specific locations rant and the sum of four quadrants in highly myopic NTG eyes. including vertical and horizontal macular scan images. Although This indicates that posterior staphyloma expansion seems to be we included our cases presenting the posterior curvature sym- asymmetrical. The posterior staphyloma heights of these regions metrically centred on the fovea, it is reasonable to obtain mul- had a significant negative correlation with subfoveal scleral tiple measurements at different locations in that the axial thickness and a positive correlation with axial length in highly elongation of highly myopic eyes does not occur at one point. myopic NTG eyes. Maruko et al14 reported that the sclera at And the further methods that could image the posterior protru- the posterior pole in highly myopic eyes stretches non- sion of highly myopic eyes including B-scan ultrasonography or uniformly. Hayashi et al15 reported that the nasal and superior MRI will be required. Finally, staphyloma heights were measured posterior staphyloma heights were significantly associated with 2 mm from the fovea, but the staphyloma can have different subfoveal scleral thickness, which is consistent with our results. shapes because the posterior sclera expands in a non-uniform The mechanical stretching nasally and superiorly around the fashion and the contour of the inner scleral surface has different fovea by an extension of a staphyloma may have a more direct patterns, such as inclining toward the optic disc, symmetrical on effect on subfoveal scleral thinning, and the staphyloma in this the fovea centred or asymmetrical with irregular curve. In prac- area may be associated with axial elongation and the presence tice, because the type I staphyloma involving the optic disc is the of myopic glaucomatous lesions. In addition, our stepwise most prominent type and the pattern inclining toward the optic

Park JH, et al. Br J Ophthalmol 2016;100:1251–1256. doi:10.1136/bjophthalmol-2015-307292 1255 Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com Clinical science disc is more prevalent, the nasal posterior staphyloma height 10 Norman RE, Flanagan JG, Sigal IA, et al. Finite element modeling of the human may be deeper than any other directions. In the recent study by sclera: influence on optic nerve head biomechanics and connections with glaucoma. – et al 37 Exp Eye Res 2011;93:4 12. Park , the location of the posterior staphyloma was related 11 Coudrillier B, Tian J, Alexander S, et al. Biomechanics of the human posterior to the direction of disc torsion and optic disc torsion was a sclera: age- and glaucoma-related changes measured using inflation testing. Invest prominent finding in myopic NTG eyes when posterior staphy- Ophthalmol Vis Sci 2012;53:1714–28. loma was located temporal to the optic disc. Thus this study has 12 Seko Y, Shimokawa H, Tokoro T. Expression of bFGF and TGF-beta 2 in a potential for unintentional omission of compound staphylomas experimental myopia in chicks. Invest Ophthalmol Vis Sci 1995; 36:1183–7. and a risk of selection bias. Further investigations measuring the 13 Rohrer B, Stell WK. Basic fibroblast growth factor (bFGF) and transforming growth scleral inner/outer borders more clearly from more deeply pene- factor beta (TGF-beta) act as stop and go signals to modulate postnatal ocular trating OCT devices, such as swept-source OCT, and evaluating growth in the chick. Exp Eye Res 1994;58:553–61. the posterior staphyloma by adjusting different patterns, would 14 Maruko I, Iida T, Sugano Y, et al. Morphologic analysis in pathologic myopia using high-penetration optical coherence tomography. Invest Ophthalmol Vis Sci overcome these limitations. Another limitation is that the sub- 2012;53:3834–8. jects of the study were identified in a referral clinic-based prac- 15 Hayashi M, Ito Y, Takahashi A, et al. Scleral thickness in highly myopic eyes tice, rather than through population-based screening. Thus, the measured by enhanced depth imaging optical coherence tomography. Eye (Lond) subjects may represent a subgroup of patients with highly 2013;27:410–17. myopic NTG in Korea that may not represent the characteristics 16 Curtin BJ. The posterior staphyloma of pathologic myopia. Trans Am Ophthalmol Soc 1977;75:67–86. of similar patients in other populations. This was also a cross- 17 Lopilly Park HY, Lee NY, Choi JA, et al. Measurement of scleral thickness using sectional study, and a future longitudinal study will be required swept-source optical coherence tomography in patients with open-angle glaucoma to confirm the cause–effect relationships. and myopia. Am J Ophthalmol 2014;157:876–84. In conclusion, both subfoveal scleral thickness and CH showed 18 Downs JC, Roberts MD, Burgoyne CF, et al. Multiscale finite element modeling of fi the lamina cribrosa microarchitecture in the eye. Conf Proc IEEE Eng Med Biol Soc a signi cant difference in highly myopic NTG eyes compared 2009;2009:4277–80. with the myopic controls, and these variables had positive correla- 19 Roberts MD, Liang Y, Sigal IA, et al. Correlation between local stress and strain and tions only in highly myopic NTG eyes. These results may suggest lamina cribrosa connective tissue volume fraction in normal monkey eyes. Invest increased scleral stiffness and may explain the increased risk of Ophthalmol Vis Sci 2010;51:295–307. 20 Sigal IA, Flanagan JG, Tertinegg I, et al. Finite element modeling of optic nerve glaucoma in myopic eyes. The posterior staphyloma heights were – fi head biomechanics. Invest Ophthalmol Vis Sci 2004;45:4378 87. signi cantly higher in highly myopic NTG eyes than in myopic 21 Cone FE, Gelman SE, Son JL, et al. Differential susceptibility to experimental controls, and negatively correlated with subfoveal scleral thick- glaucoma among 3 mouse strains using bead and viscoelastic injection. Exp Eye Res ness. Of the various ocular factors evaluated, CH, CRF and the 2010;91:415–24. nasal posterior staphyloma height were significantly associated 22 Sigal IA, Flanagan JG, Ethier CR. Factors influencing optic nerve head biomechanics. Invest Ophthalmol Vis Sci 2005;46:4189–99. with scleral thickness. Posterior staphyloma and subfoveal scleral 23 Crawford Downs J, Roberts MD, Sigal IA. Glaucomatous cupping of the lamina thinning were closely related in highly myopic NTG eyes. cribrosa: a review of the evidence for active progressive remodeling as a mechanism. Exp Eye Res 2011;93:133–40. Acknowledgements The authors thank So-Yeon Kim (Department of 24 Sigal IA. Interactions between geometry and mechanical properties on the optic Rehabilitation Medicine, Ewha Womans University School of Medicine, Seoul, Korea) nerve head. Invest Ophthalmol Vis Sci 2009;50:2785–95. for the thickness measurements. 25 Girard MJ, Downs JC, Burgoyne CF, et al. Peripapillary and posterior scleral Contributors Conception and design of study: JHP, K-RC, CYK and SSK. Conduct mechanics—part I: development of an anisotropic hyperelastic constitutive model. of study: JHP. Collection and management of data: JHP. Analysis and interpretation J Biomech Eng 2009;131:051011. of data: JHP and SSK. Preparation of manuscript: JHP and SSK. Review and approval 26 Girard MJ, Downs JC, Bottlang M, et al. Peripapillary and posterior scleral of manuscript: JHP, K-RC, CYK, and SSK. mechanics—part II: experimental and inverse finite element characterization. J Biomech Eng 2009;131:051012. Competing interests None declared. 27 Sellheyer K, Spitznas M. Development of the human sclera: a morphological study. Patient consent Obtained. Graefe’s Arch Clinc Exp Ophthalmol 1988;226:89–100. 28 Ohno-Matsui K, Shimada N, Yasuzumi K, et al. Long-term development of Ethics approval Ethics approval was provided by the Institutional Review Board at significant visual field defects in highly myopic eyes. Am J Ophthalmol Ewha Womans University Mokdong Hospital, Seoul, Korea. All parts of the study 2011;152:256–65. were performed according to the tenets of the Declaration of Helsinki. 29 Park HY, Lee K, Park CK. Optic disc torsion direction predicts the location of Provenance and peer review Not commissioned; externally peer reviewed. glaucomatous damage in normal-tension glaucoma patients with myopia. Ophthalmology 2012;119:1844–51. 30 Hsiang HW, Ohno-Matsui K, Shimada N, et al. Clinical characteristics of REFERENCES posterior staphyloma in eyes with pathologic myopia. Am J Ophthalmol 1 Sawada A, Tomidokoro A, Araie M, et al. Refractive errors in an elderly Japanese 2008;146:102–10. population: the Tajimi study. Ophthalmology 2008;115:363–70. 31 Sigal IA, Yang H, Roberts MD, et al. IOP-induced lamina cribrosa displacement 2 Perera SA, Wong TY, Tay W-T, et al. Refractive error, axial dimensions, and primary and scleral canal expansion: an analysis of factor interactions using open-angle glaucoma: the Singapore Malay Eye Study. Arch Ophthalmol parameterized eye-specific models. Invest Ophthalmol Vis Sci 2011; 2010;128:900–5. 52:1896–907. 3 Xu L, Wang Y, Wang S, et al. High myopia and glaucoma susceptibility the Beijing 32 Wells AP, Garway-Heath DF, Poostchi A, et al. Corneal hysteresis but not corneal Eye Study. Ophthalmology 2007;114:216–20. thickness correlates with optic nerve surface compliance in glaucoma patients. 4 Chon B, Qiu M, Lin SC. Myopia and glaucoma in the South Korean population. Invest Ophthalmol Vis Sci 2008;49:3262–8. Invest Ophthalmol Vis Sci 2013;54:6570–7. 33 Kotecha A, Elsheikh A, Roberts CR, et al. Corneal thickness- and age-related 5 Attebo K, Ivers RQ, Mitchell P. Refractive errors in an older population: the Blue biomechanical properties of the cornea measured with the ocular response analyzer. Mountains Eye Study. Ophthalmology 1999;106:1066–72. Invest Ophthalmol Vis Sci 2006;47:5337–47. 6 Saw SM, Gazzard G, Shih-Yen EC, et al. Myopia and associated pathological 34 Pruett RC. Complications associated with posterior staphyloma. Curr Opin complications. Ophthalmic Physiol Opt 2005;25:381–91. Ophthalmol 1998;9:16–22. 7 Grossniklaus HE, Green WR. Pathologic findings in pathologic myopia. Retina 35 Steidl SM, Pruett RC. Macular complications associated with posterior staphyloma. 1992;12:127–33. Am J Ophthalmol 1997;123:181–7. 8 Bellezza AJ, Hart RT, Burgoyne CF. The optic nerve head as a biomechanical 36 Ohno-Matsui K, Akiba M, Modegi T, et al. Association between shape of sclera and structure: initial finite element modeling. Invest Ophthalmol Vis Sci myopic retinochoroidal lesions in patients with pathologic myopia. Invest 2000;41:2991–3000. Ophthalmol Vis Sci 2012;53:6046–61. 9 Sigal IA, Yang H, Roberts MD, et al. IOP-induced lamina cribrosa deformation and 37 Park HY, Jung Y, Park CK. Posterior staphyloma is related to optic disc morphology scleral canal expansion: independent or related? Invest Ophthalmol Vis Sci and the location of visual field defect in normal tension glaucoma patients with 2011;52:9023–32. myopia. Eye (Lond) 2015;29:333–41.

1256 Park JH, et al. Br J Ophthalmol 2016;100:1251–1256. doi:10.1136/bjophthalmol-2015-307292 Downloaded from http://bjo.bmj.com/ on August 30, 2016 - Published by group.bmj.com

The height of the posterior staphyloma and corneal hysteresis is associated with the scleral thickness at the staphyloma region in highly myopic normal-tension glaucoma eyes Jong Hyuk Park, Kyu-Ryong Choi, Chan Yun Kim and Sung Soo Kim

Br J Ophthalmol 2016 100: 1251-1256 originally published online December 11, 2015 doi: 10.1136/bjophthalmol-2015-307292

Updated information and services can be found at: http://bjo.bmj.com/content/100/9/1251

These include: References This article cites 37 articles, 14 of which you can access for free at: http://bjo.bmj.com/content/100/9/1251#BIBL Email alerting Receive free email alerts when new articles cite this article. Sign up in the service box at the top right corner of the online article.

Topic Articles on similar topics can be found in the following collections Collections Angle (997) Glaucoma (979) Intraocular pressure (993) Optic nerve (710) Optics and refraction (507)

Notes

To request permissions go to: http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to: http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to: http://group.bmj.com/subscribe/