Abstract
Purpose:
The relative contribution of mechanical and vascular factors to the pathogenesis of myopic macular degeneration (MMD) is unclear. To address this gap, we examined the association of choroidal thickness (CT) and scleral thickness (ST) with MMD.
Methods:
Prospective, clinic-based case series of 62 eyes of 41 patients with high myopia (≤−6 diopters or axial length ≥26.5 mm). Swept-source optical coherence tomography (SSOCT) was performed to measure subfoveal CT and ST. Myopic macular degeneration was graded from fundus photographs according to the International Meta-Analysis for Pathologic Myopia (META-PM) classification. Presence of MMD was defined as META-PM category ≥ 2 and severe MMD was defined as category ≥ 3.
Results:
The distribution of MMD severity was 15 (24.2%) in category 1, 28 (45.2%) in category 2, 10 (16.1%) in category 3, and 9 (14.5%) in category 4. Correlation of MMD severity was strong for subfoveal CT (r = −0.70, P < 0.001) but weak for subfoveal ST (r = −0.31, P = 0.01). Subfoveal CT, but not ST, was independently associated with presence of MMD (age and gender adjusted odds ratio [OR] per 10 μm decrease in CT 1.41, P = 0.002), and subfoveal CT, but not subfoveal ST, was significantly thinner in eyes with severe MMD (≥ category 3) than in eyes with mild MMD (CT: 31.5 ± 40.5 μm versus 82.0 ± 57.1 μm, P < 0.001; ST: 261.6 ± 78.5 μm versus 297.0 ± 73.8 μm, P = 0.09).
Conclusions:
We demonstrated significant thinning of the choroid with increasing MMD severity. In contrast, ST was weakly correlated with MMD. These data suggest progressive loss of choroid may be important in the pathogenesis of MMD.
Myopic macular degeneration (MMD) is one of the leading causes of blindness or low vision in developed nations, particularly in East Asia where the prevalence of myopia is high.
1–5 It is characterized by excessive axial elongation of the eye, scleral thinning, posterior staphyloma, chorioretinal atrophy, and choroidal neovascularization.
6 The prevalence of MMD in East Asia is estimated to be 1.7% to 3.1%,
7,8 but because MMD also affects young and middle-aged working adults,
9,10 it carries a significant socioeconomic burden.
11 Furthermore, MMD has been shown to be a progressive disease over time and with aging, leading to significant visual loss.
7,12 Thus, it is important to understand the risk factors and pathogenesis of MMD to allow the development of possible preventative and therapeutic strategies.
5
However, the exact mechanisms that lead to the development of the macular and retinal changes in MMD are currently unknown.
13 A widely accepted view is the “mechanical” theory of MMD in which axial elongation of the globe causes mechanical stretching of the retina that leads to decreased photoreceptor density and the ultimate development of pathologic features such as lacquer cracks and chorioretinal atrophy seen in MMD.
14,15 The sclera plays a pivotal role in myopia development. Studies show complex biochemical changes within the sclera in eyes with high myopia, including reduced collagen synthesis, increased collagen degradation, decreased glycosaminoglycan production, and decreased fibroblast to myofibroblast differentiation.
16 Indeed, high myopes have been shown to have thinner sclera in the posterior pole than emmetropic eyes, independent of the presence of posterior staphyloma.
17 In addition to the “mechanical” theory, it has been suggested that choroidal thinning leading to decreased choroidal perfusion and ischemia and subsequent upregulation of angiogenic factors in some eyes may also be a mechanism for the development of myopic choroidal neovascularization and other features of MMD.
18
However, the relative contribution of “mechanical” versus “ischemic/vascular” factors in the pathogenesis of MMD is unclear. This may be clarified by clinically studying structural properties of the choroid and sclera simultaneously. Previously, the study of choroidal and scleral structure in vivo has been hampered by limitations in imaging technology, which do not allow for the visualization of the posterior scleral border. This limitation has been circumvented with swept-source optical coherence tomography (SSOCT), which provides a previously unattainable depth of visualization and is particularly useful in high myopes with deep posterior staphyloma. To examine the relative importance and contribution of mechanical versus ischemic/vascular factors in the pathogenesis of myopic MMD, we compared scleral and choroidal thickness (CT) using SSOCT in high myopes with different severity of MMD.
This was a prospectively planned clinical study in which patients with high myopia aged 40 to 80 years, with manifest spherical equivalent (SE) of ≤−6.0 dioptres (D) or axial length (AL) of ≥26.5 mm, were recruited from the outpatient clinics of the Singapore National Eye Centre from April 2015 to October 2015. Both eyes of the same patient may be included in the study. Exclusion criteria included any coexisting or previous ocular disease in either eye that may confound measurements of the choroid and sclera, such as corneal opacities, dense cataract, diabetic retinopathy or diabetic macular edema, central serous choroioretinopathy, retinal detachment, retinal dystrophies, non-myopia–related macular scarring and previous retinal photocoagulation or photodynamic therapy, previous anti-vascular endothelial growth factor treatment, active myopic choroidal neovascularization, and eyes with dome-shaped macula. This study was approved by the Centralized Institutional Review Board of SingHealth, Singapore (protocol number 2014/405/A) and conducted in accordance with the Declaration of Helsinki. All participants provided signed informed consent for their participation.
All patients underwent comprehensive ophthalmic evaluation including best corrected visual acuity (BCVA), measured with the logarithm of the minimal angle of resolution (logMAR) chart (Lighthouse International, New York, NY, USA) at 4 m, slit lamp biomicroscopy examination, and manifest refraction performed by certified study optometrists. Spherical equivalent was calculated as the sum of the spherical power and half of the cylindrical power. The IOL Master (Carl Zeiss Meditec AG, Jena, Germany) was used for ocular biometry including AL and keratometry measurements. Fundus photography was performed with a 45-degree digital retinal camera after pupillary dilation with tropicamide 1% and phenylephrine 2.5%, using Canon CR-DGi with Canon EOS 10D SLR backing (Canon, Inc., Tokyo, Japan). Two fields of each eye were photographed, with one centered at the optic disc and another centered at the fovea, following the Early Treatment for Diabetic Retinopathy Study standard photograph numbers 1 and 2.
High resolution digital photographs were stored and retrieved digitally on the Singapore Eye Research Institute server. Fundus photographs were viewed using Photoshop CS5 (Adobe Systems, Inc., San Jose, CA, USA). Two ophthalmologists, masked to participant characteristics, performed grading of the fundus photographs using the International Meta-Analysis for Pathologic Myopia (META-PM) classification system.
6 Discrepancies were adjudicated by a senior retinal specialist. Intra- and interobserver agreement and the kappa statistics (κ) were calculated. Mean inter- and intraobserver agreement and κ of 62 fundus images graded by two independent observers are shown in
Table 1. Intraobserver agreement was ≥87% and κ was ≥0.6 for MMD fundus features and plus lesions. Interobserver agreement was ≥84.1% and κ was ≥0.6, suggesting excellent agreement between observers for MMD grading using the META-PM classification system. Observers agreed most often (100%) on the presence of tessellation and least often on the presence of lacquer cracks (84.1%).
The presence of MMD was defined as META-PM catetory ≥2, mild MMD was defined as META-PM category 2, and severe MMD was defined as META-PM categories 3 and 4.
Swept-source optical coherence tomography was performed using the Topcon Atlantis DRI OCT-1 system (Topcon Medical Systems, Paramus, NJ, USA). Scan lengths of 12 mm were used. Two trained graders measured the choroidal and scleral thickness (ST) independently at the subfoveal region and at 3000 μm nasal, temporal, superior, and inferior to the fovea center. Choroidal thickness was defined as the distance between the hyperreflective line corresponding to the Bruch membrane beneath the retinal pigment epithelium and the choroid-sclera interface. Scleral thickness was defined as the distance from the inner sclera border to the outer sclera border (
Fig. 1). Measurements from both graders were averaged. Both graders were masked to the participants' information.
The outcomes of interest were the correlation of CT and ST with MMD severity and in particular, difference in CT and ST between eyes with and without severe MMD. All data were expressed as mean ± standard deviation or proportions as appropriate. Categorical variables were analyzed with the χ2 test and continuous variables were analyzed with the independent t-test. Spearman correlation coefficients were calculated to analyze the correlation between (1) AL, SE, IOP, and BCVA with MMD severity and (2) CT/ST with MMD severity. Increasing MMD severity categories were considered as equally spaced ordinal variables. Multivariable analysis was performed to evaluate the independent associations of ST and CT with MMD, adjusting for age, gender, and AL. Generalized estimating equations were used to account for correlation between left and right eyes of the same patient. A P value of <0.5 was considered to be statistically significant. All statistical analyses were performed using SPSS version 21 (SPSS, Chicago, IL, USA).
Ninety-two eyes of 52 patients met the inclusion criteria. Of these, 12 eyes with dome shaped macula, 14 with nonvisible posterior scleral border, and 4 eyes with active choroidal neovascularization (CNV)/previous anti-VEGF treatment were excluded. A total of 62 eyes of 41 patients were enrolled in the study. Twenty-two patients had both eyes enrolled in the study.
Table 2 shows the baseline characteristics of these eyes. The mean age was 60.5 ± 8.2 years, mean AL was 29.58 ± 2.53 mm, and the mean SE was −12.1 ± 5.0 D.
The prevalence of MMD categories is shown in
Figure 2. Fifteen (24.2%) eyes had tessellated fundus only (category 1), 28 (45.2%) eyes had diffuse chorioretinal atrophy (category 2), 10 (16.1%) eyes had patchy chorioretinal atrophy (category 3), and 9 (14.5%) eyes had macular atrophy (category 4). Severe MMD (categories 3 and 4) was present in 19 (30.6%) eyes. Of the 22 patients with bilateral MMD, 15 (68.2%) had MMD of the same severity in both eyes, whereas 7 (31.8%) had bilateral MMD of different severity. Plus lesion was present in 17 eyes (27.5%). Myopic macular degeneration severity was significantly correlated with AL (correlation coefficient [
r] = 0.60,
P < 0.001), BCVA (
r = 0.44,
P = 0.001), and SE (
r = −0.47,
P < 0.001). There was no correlation of MMD severity with mean intraocular pressure (IOP;
r = 0.26,
P = 0.07). Patients with severe MMD were older, more likely to be female, Chinese, pseudophakic, and had longer AL, more myopic refractive error, and worse BCVA compared with eyes without severe MMD (
Table 1).
Table 3 shows the CT and ST in eyes with and without severe MMD. Subfoveal CT was significantly thinner in eyes with severe MMD (31.5 ± 40.5 μm versus 82.0 ± 57.1 μm,
P < 0.001). The superior, inferior, nasal, and temporal CT were thinner in eyes with severe MMD, but the difference was not statistically significant. Scleral thickness was thinner in all positions, but none of these reached statistical significance.
We further evaluated the correlation between MMD severity and ST, and between MMD severity and CT (
Table 4). Choroidal thickness, in particular subfoveal CT, was strongly correlated with MMD severity based on META-PM classification (
r = −0.70,
P < 0.001). Subfoveal ST was also correlated with MMD severity (
r = −0.31,
P = 0.01), but the correlation was weaker than that between CT and MMD severity. In addition, both subfoveal CT and subfoveal ST had moderate correlation with AL and weak correlation with BCVA. There was weak correlation between subfoveal CT and subfoveal ST (
r = 0.30,
P = 0.02).
Multivariable analysis adjusted for age, gender, and correlation between left and right eyes of the same patient showed an independent association between subfoveal CT, but not subfoveal ST, with the presence of MMD (odds ratio [OR] per 10 μm decrease in CT 1.41, 95% confidence interval [CI] 1.13–1.76, P = 0.002, and OR per 10 μm decrease in ST 0.99, 95% CI 0.89–1.11, P = 0.90). In addition, we repeated the multivariable analysis after excluding eyes in which measurements of the subfoveal ST were confounded by the presence of posterior ciliary arteries and/or episcleral fibers. In this analysis, subfoveal CT remained significantly associated with the presence of MMD (OR per 10 μm decrease in CT 1.88, 95% CI 1.18–3.01, P = 0.008, and OR per 10 μm decrease in ST 1.08, 95% CI 0.89–1.32, P = 0.42).
In this study, we demonstrated strong correlation of CT with MMD severity and significant thinning of the choroid in eyes with severe MMD compared to eyes without. In contrast, ST was only weakly correlated with MMD, and was not significantly thinner in eyes with severe MMD than in highly myopic eyes with milder severity of MMD. Our data support the importance of ischemic/vascular factors (as reflected by CT), rather than mechanical changes (as reflected by ST), in the pathogenesis of MMD.
In this study, we used the international photographic classification and grading system for MMD developed by META-PM study group to provide a framework for future clinical and epidemiologic studies. The study group found a high intraobserver (≥85%) and satisfactory interobserver agreement (κ ≥ 0.4) using this classification system.
6 This is the first study since the publication of the META-PM classification to validate the reliability of the classification system using photographs taken in a clinic-based setting. Using the new META-PM classification system, we found excellent inter- and intraobserver agreement for MMD severity grading. Agreement for plus lesions were slightly lower, particularly for lacquer cracks, which may be difficult to diagnose from fundus photographs alone. In addition, our findings showed that MMD severity based on the META-PM classification was moderately well correlated with anatomical and functional parameters (AL, SE, and BCVA). The prevalence of severe MMD (categories 3 and 4) in our study was 30%, and the prevalence of any plus lesions was 30%. Bilateral symmetrical MMD was seen in 70% of patients.
How or why some eyes with high myopia develop the typical atrophic and degenerative changes seen in MMD, while others do not, are currently unclear, and likely involves multiple mechanisms.
13 Mechanical stretching of the retina by axial elongation and choroidal ischemia are the most likely mechanisms for MMD, but the relative importance and contribution of each process has not been previously studied. In this study, we assessed these mechanisms, using ST as a biomarker for mechanical stretching and CT for choroidal vascularity. Our results showed that subfoveal CT, but not ST, was significantly decreased in eyes with severe MMD. Although both subfoveal CT and ST were significantly correlated with MMD severity, the strength of correlation was stronger for CT than for ST. In addition, subfoveal CT, but not ST, was independently associated with the presence of MMD. Both CT and ST were significantly correlated with BCVA.
In previous studies, both ST and CT have been reported to be thinner in highly myopic eyes compared to nonhighly myopic eyes.
19–22 Except for Shen et al.,
20 who studied CT in postmortem eyes and therefore unlikely to reflect choroidal vascularity, none of these studies examined ST and CT concurrently. Subfoveal ST varied depending on the method of measurement: mean subfoveal ST ranged from 281 to 335 μm on OCT
23,24 and 602 to 670 μm in histological studies.
20,22 Scleral thickness may be underestimated when measured with OCT because the posterior scleral border may not be visible in eyes with thicker sclera. Park et al.
25 recently compared enhanced depth imaging spectral-domain optical coherence tomography (EDI-OCT) with SSOCT and found a better detection rate for the posterior scleral border with SSOCT (31% vs. 53%,
P = 0.008).
25 Previously reported rates of detection of the posterior scleral border ranged from 53% to 68.4% on SSOCT.
25,26 In our study, of the 92 highly myopic eyes that met the inclusion criteria, 84.7% had visible posterior scleral border and subfoveal ST ranged from 130 to 478 μm in these eyes. As expected, the posterior scleral border was better seen as the AL increased and ST decreased: detection rate increased to 98.5% in eyes with AL > 27.5 mm in our study. Aside from difficulties in visualizing the posterior scleral border, the measurement of ST may be confounded by two other factors: First, the long and short posterior ciliary arteries have been observed within the sclera as linear hyporeflective structures on SSOCT
27 and second, a separation of episcleral fibers from scleral fibers have been observed in highly myopic eyes (
Fig. 3). The presence of either of these features may result in a spurious overestimation of ST. After excluding eyes with at least one of these features, the conclusions of our study remain unchanged.
Hayashi et al.
17 measured subfoveal ST with EDI-OCT in 75 eyes with high myopia (≤−8 D or AL ≥26.5 mm). Scleral thickness was measurable in all eyes. They found a mean subfoveal ST of 284.0 ± 70.4 μm and a subfoveal CT of 41.7 ± 20.6 μm. There were no significant differences in subfoveal ST between eyes with and without myopic retinochoroidal pathologies (including tractional, degenerative, and neovascular myopic maculopathy lesions). In contrast to our study, Hayashi et al.
17 did not apply the META-PM classification and included the whole spectrum of both tractional and degenerative changes in their analysis. Their study did not specifically analyze ST in eyes with MMD nor compare the relative association of CT and ST with the presence of myopic retinochoroidal pathologies. Ohno-Matsui et al.
28 measured ST in highly myopic eyes and observed that the scleral was thinnest inferiorly. The findings from our study are consistent with this observation, and suggest that the inferior sclera expands the most in myopic eyes (
Fig. 4); they also observed that MMD was most frequently seen in eyes with irregular scleral curvatures. Incidentally, these were also the eyes with the thinnest choroid.
28
Although ST may decrease as a result of mechanical stretching, scleral volume may remain unchanged. To study the correlation of AL with ST and scleral volume, Jonas et al.
19 measured ST and scleral volume in 214 histologic samples of Caucasian eyes and found significant correlation of ST with AL at the posterior pole, equator, midpoint between equator and posterior pole and optic nerve head border.
19 In contrast to thickness measurements, scleral volume measurements did not vary significantly with AL. These results suggest that axial elongation results in a remodeling of existing scleral tissue rather than the formation of new tissue.
Based on previous studies and ours, a few key points about the sclera in highly myopic eyes could be summarized as follows. First, because scleral volume does not increase with axial elongation, the sclera must thin with increasing AL. Second, although axial elongation leads to a decrease in both ST and CT, CT appears to decrease to a much greater extent than ST in eyes with MMD than highly myopic eyes without MMD. Third, irregularity of scleral contour and asymmetry of scleral thinning may play an important role in choroidal thinning and pathogenesis of MMD. Based on these observations, we hypothesize that choroidal ischemia may play a greater role than mechanical stretching alone in the pathogenesis of MMD. Reduction in choroidal blood flow and attenuation of choroidal blood vessels have been observed in eyes with pathologic/degenerative myopia.
29,30 In addition, asymmetrical scleral thinning and irregular scleral contour (
Fig. 3) may result in kinking and occlusion of the choroidal vessels with consequent choroidal ischemia and eventual atrophy of the elements supplied by the choroid (i.e., the Bruch membrane, RPE, and outer retina).
Our study has several limitations. The relatively small number of eyes with severe MMD may be a reason for the lack of statistical significance for association of CT with severe MMD. We did not account for systemic risk factors, other than age, that may confound the relationship between CT and MMD, although previous studies have not found significant systemic risk factors for MMD. We excluded eyes if the posterior scleral border was not clearly visible. This may result in an underestimation of the mean ST, particularly in eyes without MMD. However, the effect of this underestimation is small as only 15% of eyes were excluded for this reason. Choroidal thickness itself is a measure of the structure of choroid, which includes choriocapillaris, vasculature within Sattler's and Haller's layer, and also a stromal component. Therefore, it may not truly reflect the functional aspect of choroidal blood flow. Novel imaging techniques such as OCT angiography and other biomarkers of choroidal vascularity, such as the choroidal vascularity index (CVI),
31,32 may further clarify the role of choroidal blood flow in MMD pathogenesis. Lastly, due to the cross-sectional design, causality cannot be inferred.
In conclusion, we demonstrated that CT but not ST was independently associated with MMD. These data suggest the need for further studies to evaluate whether reduction in choroidal vasculature leads to changes within the RPE and outer retina, which are responsible for the eventual development of MMD. Future studies with optical coherence tomography angiography (OCTA) may also add to the understanding of choriocapillaris status and blood flow velocity in these eyes.
Supported by the Singapore National Eye Centre Health Research Endowment Fund (grant number: R1122/24/2014).
Disclosure: C.W. Wong, None; V. Phua, None; S.Y. Lee, None; T.Y. Wong, None; C.M.G. Cheung, None