MACULAR CHOROIDAL VOLUME VARIATIONS IN HIGHLY MYOPIC EYES WITH MYOPIC TRACTION MACULOPATHY AND CHOROIDAL NEOVASCULARIZATION GIULIO BARTESELLI, MD,*† SU NA LEE, MD,*‡ SHARIF EL-EMAM, MD,* HUIYUAN HOU, MD,* FEIYAN MA, MD,* JAY CHHABLANI, MD,* LAURA CONNER, MS,* LINGYUN CHENG, MD,* DIRK-UWE BARTSCH, PHD,* WILLIAM R. FREEMAN, MD* Purpose: To compare the choroidal volume (CV) between emmetropic and highly myopic eyes, and to assess if the presence of myopic fundus abnormalities, myopic traction maculopathy, or choroidal neovascularization affects the CV. Methods: We retrospectively reviewed imaging studies of 98 eyes of 98 patients who underwent CV measurement on optical coherence tomography. We included 31 emmetropic eyes (Group 1), 36 highly myopic eyes without vitreoretinal or choroidal pathologies (Group 2), 21 highly myopic eyes with traction maculopathy (Group 3), and 10 highly myopic eyes with history of choroidal neovascularization (Group 3). Eyes with chorioretinal atrophy were excluded. Regression analysis was performed to evaluate the correlation between CV and multiple variables. Results: Choroidal volume was lower in Group 2 than in Group 1 (P , 0.001), and in Groups 3 and 4 than in Group 2 (P , 0.001 and P = 0.002, respectively). Age (P = 0.002), axial length (P , 0.001), sex (P = 0.047), staphyloma (P , 0.001), and myopic group (P = 0.05) were independent predictors for the final CV (R2 = 0.645). In highly myopic eyes, CV decreased by 0.32 mm3 for every 10 years and by 0.49 mm3 per millimeter of axial length. Conclusion: Choroidal thinning is present in highly myopic eyes compared with emmetropic eyes, and is related to age, axial length, sex, and staphyloma. However, myopic eyes with coexisting myopic traction maculopathy or history of choroidal neovascularization have more severe thinning, likely leading to insufficient metabolic supplementation for the macula. RETINA 34:880–889, 2014
P
imaged and assessed by spectral domain optical coherence tomography (SD-OCT), and can be grouped into what has been called “myopic traction maculopathy.”10 In addition to clinical findings, histologic studies have demonstrated choroidal thinning in eyes with high myopia, with a lack of large vessels in some areas and pronounced thinning of the choriocapillaris in others.7,11 Indocyanine green angiography has been widely used in ophthalmology to provide information about the choroidal perfusion, and has been useful for observing various pathologic changes in highly myopic eyes.12 However, it does not allow cross-sectional imaging of the choroidal structures.
athologic myopia is one of the leading causes of vision impairment worldwide. It is characterized by congenital scleral weakness leading to the progressive elongation of the globe resulting in a variety of fundus changes, such as posterior staphyloma, progressive atrophy of the choriocapillaris and choroid, lacquer cracks, and eventually choroidal neovascularization (CNV).1 Apart from these abnormalities, the posterior retina can also be damaged by the presence of pathologic tractions induced by epiretinal membrane and residual vitreomacular adhesion, eventually leading to retinoschisis, posterior retinal detachment, and macular holes.2–9 These structural alterations induced by tractional forces may be 880
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
Recently, enhanced depth imaging (EDI) on SD-OCT has provided three-dimensional analysis of the anatomy of deeper structures in the eye including the choroid.13,14 Using EDI-OCT, the chorio-scleral junction can be easily visualized, and its distal margin can be used to measure choroidal thickness.15,16 Previous investigators have used manual choroidal measurements in the fovea and a few perifoveal locations, and they reported that the choroid in eyes with high myopia is markedly thinner when compared with normal eyes.17–20 However, measuring thickness only at selected points of the choroid may not describe comprehensively the thinning of the choroid, since the irregularity of the inner chorio-scleral border may influence the measurement at few sampling points.21 To overcome this problem and to obtain a better picture of the state of the posterior choroid, our group devised a manual method to measure the choroidal volume (CV) of the entire macula22,23 that has been extensively described in nonpathologic eyes using a commercially available SD-OCT device.24 On the basis of these previous reports, we aimed to compare macular CV measurements between emmetropic eyes and highly myopic eyes without evidence of chorioretinal atrophy, and also to assess if the presence of common myopic fundus findings (such as staphyloma, peripapillary atrophy, tilted disk, and lacquer cracks), or myopic traction maculopathy, or CNV affects the CV. Methods Starting in July 2011, all patients undergoing SD-OCT scans at the Jacobs Retina Center at Shiley Eye Center, University of California San Diego (La Jolla, CA) underwent choroidal imaging as part of their evaluation. A raster scan of the macula using Heidelberg Spectralis (Heidelberg Engineering, Carlsbad, CA) in EDI mode was incorporated into the standard From the *Department of Ophthalmology, Jacobs Retina Center at Shiley Eye Center, University of California San Diego, La Jolla, California; †Department of Clinical Sciences and Community Health, Ophthalmological Unit, Ca’ Granda Foundation-Ospedale Maggiore Policlinico, University of Milan, Milan, Italy; and ‡Department of Ophthalmology, Eulji University Hospital, Daejeon, Republic of Korea. J. Chhablani is now at L. V. Prasad Eye Institute, Hyderabad, India. Supported by NIH grants R01EY007366 and R01EY018589 (W.R.F.), R01EY020617 (L.C.), and in part by an unrestricted fund from Research to Prevent Blindness to the Department of Ophthalmology, University of California San Diego. None of the authors have any conflicting interests to disclose. G. Barteselli and S. N. Lee contributed equally as first authors. Reprint requests: William R. Freeman, MD, University of California at San Diego, Shiley Eye Center, 0946, 9415 Campus Point Drive, La Jolla, CA 92037; e-mail: [email protected]
881
scanning protocol and used to study the choroid. After approval by the institutional review board of the University of California at San Diego for chart review and data analysis, we retrospectively reviewed charts and OCT images of 98 consecutive patients who underwent EDI-OCT scans from July 2011 to June 2012. We included 31 normal emmetropic eyes (axial length $23.5 and ,24.5 mm),25 and 67 highly myopic eyes with an axial length .26.5 mm.25 Highly myopic eyes were divided into three groups: 1) eyes without vitreoretinal or choroidal pathology; 2) eyes with traction maculopathy; and 3) eyes with a history of CNV. The presence of other myopic fundus findings, such as staphyloma, peripapillary atrophy, tilted disk, and lacquer cracks, was also assessed for each eye. Eyes with evidence of patchy or diffuse chorioretinal atrophy in the macula, or eyes that received previous intravitreal antivascular endothelial growth factor injections within 6 months of the imaging study were not included in this study. Patients who had any systemic or ocular disease that could affect the choroid, including uncontrolled hypertension, diabetes, coagulopathies, CNV secondary to age-related macular degeneration, central serous chorioretinopathy, or other conditions that have been shown to affect the choroid were also excluded. The study was conducted in adherence with the tenets of the Declaration of Helsinki. Patients underwent ophthalmic evaluation, including best-corrected visual acuity, slit-lamp biomicroscopy, axial length measurement using ocular biometry (IOLMaster; Carl Zeiss Meditec, Jena, Germany), and a raster scan of the macula using the EDI-OCT. The raster scan consisted of 31 high resolution B-scans associated with a simultaneous near-infrared image; an internal fixation light was used to center the scanning area on the fovea. Each scan was 9.0 mm in length and spaced 240 mm apart. All 31 OCT B-scans were acquired in a continuous automated sequence and covered a 30° · 25° area centered on the fovea. Image averaging of 25 frames was used to obtain a good quality choroidal image by the built-in Active Eye Tracking software (TruTrack; Heidelberg Engineering). Manual choroidal segmentation of each B-scan was performed by 2 retina specialists according to a previously described well-reproducible method (Figure 1).22 Choroidal volume map centered on the fovea was automatically generated by the built-in software of the device by applying the Early Treatment Diabetic Retinopathy Study (ETDRS) grid. After checking the position of the ETDRS grid to ensure a correct foveal placement, measurements for the CV were recorded for each subfield of the grid. The ETDRS grid divides the macula into 3 concentric rings (center, inner, and outer), with the inner ring measuring 1 mm to 3 mm and the outer ring
882
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
Fig. 1. Enhanced depth imaging optical coherence tomography raster scan protocol (A), automated retinal segmentation (B), manual choroidal segmentation (C), and choroidal measurements on the standardized ETDRS grid (D) including three concentric rings with a total of nine subfields centered on the fovea. To demarcate choroidal boundaries, the internal limiting membrane line (arrows, A) was manually moved to the base of retinal pigment epithelium (arrows, B), whereas the basement membrane line (asterisks, A) was manually moved to chorio-scleral interface (asterisks, B). BM, basement membrane; ILM, internal limiting membrane.
measuring 3 mm to 6 mm from the center (referring to a ring with a diameter of 1 mm centered on the fovea). The grid further divides inner and outer rings into four quadrants (superior, inferior, temporal, and nasal). Statistical analysis was performed using the Statistical Package for the Social Sciences version 20 (SPSS Inc, Chicago, IL). Continuous variables are expressed as mean ± standard deviation. One eye per subject was included; in the group of normal emmetropic subjects, the eye was chosen randomly; in the group of highly myopic subjects, the eye was chosen based on the presence of a vitreoretinal or choroidal pathology, or the longest axial length if the pathology was bilateral. We examined the association of myopic fundus findings with age, axial length, and sex. The relationship was assessed using the chi-square test (or Fisher exact test if .20% of cells have expected count of ,5) or analysis of variance. Multiple linear regression analysis was used to evaluate the correlation between CV in highly myopic eyes and independent variables including age, axial length, sex, myopic fundus findings, and myopic groups (including the group without vitreoretinal or choroidal pathology, the group with traction maculopathy, and the group with CNV). Tukey–Kramer test was used to compare the mean CV between groups. Agreement between interobserver measurements in
a subset of emmetropic and myopic eyes was assessed using the concordance correlation coefficient. A P , 0.05 was considered to be statistically significant.
Results A total of 98 eyes of 98 patients (46 men and 52 women) with a mean age of 53 years (range, 20–91 years) were included in the study. We studied 31 normal emmetropic eyes of 31 patients with a mean age of 53 years (range, 20–85 years), and 67 highly myopic eyes of 67 patients with a mean age of 57 years (range, 21–91 years). Mean logMAR best-corrected visual acuity was 0.026 ± 0.088 in emmetropic eyes and 0.146 ± 0.237 in highly myopic eyes. Highly myopic eyes included 36 eyes without vitreoretinal or choroidal pathology, 21 eyes with traction maculopathy, and 10 eyes with a history of CNV. In the group with myopic traction maculopathy, 10 eyes had epiretinal membrane, 5 had lamellar macular hole, 3 had foveoschisis, 2 had vitreomacular traction syndrome, and 1 had fullthickness macular hole. Of the 10 eyes with a history of CNV, 6 were treated with photodynamic therapy and 4 were treated with intravitreal antivascular endothelial growth factor medications; none of the eyes was treated
Table 1. Distribution of Study Eyes Highly Myopic (n = 67)
Age (years, range) Sex (M/F) Axial length (mm, range) Staphyloma (n, %) Peripapillary atrophy (n, %) Tilted disk (n, %) Lacquer cracks (n, %)
Emmetropic (n = 31)
MTM (n = 21)
CNV (n = 10)
No Retinal or Choroidal Pathology (n = 36)
53 (20–85) 14/17 24.0 ± 0.3 0 (0%) 0 (0%) 0 (0%) 0 (0%)
64 (41–91) 9/12 28.4 ± 1.5 16 (76%) 20 (95%) 15 (71%) 7 (33%)
67 (41–90) 4/6 28.6 ± 1.5 6 (60%) 10 (100%) 4 (40%) 6 (60%)
50 (21–85) 19/17 27.6 ± 1.1 9 (25%) 22 (61%) 7 (19%) 0 (0%)
F, female; M, male; MTM, myopic traction maculopathy.
883 0 (0) 5 (21.7) 8 (24.2) 0.200 11 23 33 3 (27.3) 8 (34.8) 15 (45.5) 0.500 11 23 33 7 (63.6) 16 (69.6) 29 (87.9) 0.129 11 23 33 2 (18.2) 11 (47.8) 18 (54.5) 0.110 11 23 33 0 (0) 3 (13.0) 7 (21.2) 0.221 MTM, myopic traction maculopathy.
0 (0) 6 (26.1) 15 (45.5) 0.015 11 23 33
11 23 33
3 (9.4) 10 (28.6) 0.047 32 35 10 (31.2) 16 (45.7) 0.225 32 35 23 (71.9) 39 (82.9) 0.281 32 35 14 (43.8) 17 (48.6) 0.693 32 35 4 (12.5) 6 (17.1) 0.594 10 (31.2) 11 (31.4) 0.987 32 35
32 35
1 (3.6) 4 (23.5) 8 (36.4) 0.013 28 17 22 3 (10.7) 9 (52.9) 14 (63.6) ,0.001 28 17 22 17 (60.7) 15 (88.2) 20 (90.9) 0.019 28 17 22 5 (17.9) 12 (70.6) 14 (63.6) ,0.001 28 17 22 2 (7.1) 4 (23.5) 4 (18.2) 0.285 6 (21.4) 6 (35.3) 9 (40.9) 0.311 28 17 22
28 17 22
13 (19.4) 67 26 (38.8) 67 52 (77.6) 67 31 (46.3) 67 10 (14.9) 67 21 (31.3) 67
All Axial length (mm) 26.5–27.49 27.5–28.49 28.5+ Ptrend Sex Male Female P Age (years) 20–39 40–59 60+ Ptrend
n (%) N n (%) N
n (%)
N
n (%)
N
n (%)
N
n (%)
N
Tilted Disk Peripapillary Atrophy Staphyloma CNV MTM
Table 2. Summary of Fundus Changes With Axial Length, Sex, and Age, in Highly Myopic Eyes
within 6 months of the imaging study. Table 1 shows the distribution of subjects divided by groups according to age, sex, axial length, and myopic fundus findings (including staphyloma, peripapillary atrophy, tilted disk, and lacquer cracks). When the axial length was divided into categories (26.5–27.49 mm, 27.5–28.49 mm, and $28.5 mm), staphyloma (P = 0.001), peripapillary atrophy (P = 0.019), tilted disk (P , 0.001), and lacquer cracks (P = 0.013) increased in prevalence with increasing axial length in highly myopic eyes. We noticed that women had a greater frequency of lacquer cracks than men (P = 0.047). Sex did not alter the prevalence of the other pathologic myopia findings, whereas traction maculopathy increased in prevalence with increasing age (P = 0.015, Table 2). The interobserver concordance correlation coefficient for the total CV in a subset of 15 emmetropic and 15 myopic eyes that were randomly selected from the sample was 0.93. The mean subfoveal and total CVs were 0.24 ± 0.07 mm3 and 7.74 ± 2.19 mm3 in emmetropic eyes; 0.16 ± 0.06 mm3 and 5.56 ± 1.81 mm3 in highly myopic eyes without retinal or choroidal pathology; 0.09 ± 0.05 mm3 and 3.25 ± 1.38 mm3 in cases of traction maculopathy; 0.08 ± 0.04 mm3 and 3.10 ± 1.28 mm3 in cases of myopic CNV (Figure 2). Multiple comparisons analysis showed that the subfoveal and total CVs were significantly lower in highly myopic eyes without retinal or choroidal pathology than in emmetropic eyes (both P , 0.001, Table 3). Compared with highly myopic eyes without retinal or choroidal pathology, the total CV was lower in highly myopic eyes with traction maculopathy or CNV (P , 0.001 and P = 0.002, respectively, Table 3). No difference in CV was noticed between the eyes with traction maculopathy and eyes with myopic CNV. Considering all eyes, CV was greater in the superior and temporal quadrants than in the inferior (P , 0.05) and nasal quadrants (P , 0.001); the nasal quadrant showed the lowest CV among the 4 quadrants (P , 0.001). When comparing the CV between emmetropic and highly myopic eyes (Table 3), we found that volume in each quadrant was significantly lower in highly myopic eyes (P , 0.001). Considering only highly myopic eyes, CV in the eyes without retinal or choroidal pathology was significantly greater in each quadrant than in the eyes with traction maculopathy or CNV (P , 0.01). Table 4 shows the results of the multiple linear regression analysis for mean total macular CV in highly myopic eyes. A stepwise method was used to determine the most unexpected factors. The model determined by age, axial length, sex, staphyloma, and myopic group had the best regression. This model showed a high coefficient of determination, which was 0.803 (R2 = 0.645).
Lacquer Cracks
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
884
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
Fig. 2. Enhanced depth imaging optical coherence tomography scans with arrowheads indicating the chorio-scleral junction (First column); choroidal thickness and volume measurements in the ETDRS grid centered on the fovea (Second column); and color-coded CV maps (Third column). In an emmetropic eye (axial length: 23.8 mm), the choroid-scleral junction was well visible, and the macular CV was 6.93 mm3 (A). The choroid of a nonpathologic highly myopic eye (axial length: 26.6 mm) was thin, showing a macular CV of 4.34 mm3 (B). In a case of highly myopic eye (axial length: 27.6 mm), epiretinal membrane and retinoschisis in the outer plexiform layer were observed above thin choroid with a CV of 2.37 mm3 (C). In a case of highly myopic eye (axial length: 28.1 mm), hyperreflective juxtafoveal CNV was observed over thin choroid with a CV of 3.55 mm3 (D).
In highly myopic eyes, univariate regression analysis showed a decrease in CV of approximately 0.32 mm3 for every 10 years and a decrease of 0.49 mm3 per millimeter of axial length. Figure 3, A and B shows that the total CV in the ETDRS map was negatively correlated with age (P = 0.002, R2 = 0.178), and axial length (P , 0.001, R2 = 0.272). Men had a greater CV than women (P = 0.047, Figure 3C). The presence of traction maculopathy or CNV lead to lower CV than in cases of highly myopic eyes without vitreoretinal or choroidal pathologies (P = 0.05, Figure 3D). The presence of a staphyloma significantly reduced the CV (P , 0.001, Figure 3E). Discussion This study demonstrated differences in macular CV between emmetropic eyes and highly myopic eyes,
and investigated variations of CV between highly myopic eyes without retinal or choroidal pathologies and highly myopic eyes with traction maculopathy or CNV. Even in the absence of visible chorioretinal atrophy in the macula, highly myopic eyes have lower CV than emmetropic eyes, as expected from a previous study investigating choroidal thinning in the highly myopic eyes.26 Beside age, axial length, or presence of a posterior staphyloma—that are known to affect the choroid in highly myopic eyes14,17,19—we demonstrated that sex and presence of traction maculopathy or CNV are independent predictors for the CV as well. So far, the choroid in highly myopic eyes has been investigated bi-dimensionally using conventional SD-OCT devices.14,17,18 Previous studies manually measured the choroidal thickness only in selected points of the macula.14,17–19,26 However, thickness measurements taken at multiple single points could
ETDRS Subfields Eyes Emmetropic Highly myopic without retinal or choroidal pathology Highly myopic without retinal or choroidal pathology Myopic traction maculopathy Highly myopic without retinal or choroidal pathology Myopic choroidal neovascularization Myopic traction maculopathy Myopic choroidal neovascularization
Subfoveal (Mean ± SD)
P
Superior (Mean ± SD)
P
Nasal (Mean ± SD)
P
Inferior (Mean ± SD)
P
Temporal (Mean ± SD)
0.24 ± 0.07 ,0.001 2.07 ± 0.57 ,0.001 1.62 ± 0.57 ,0.001 1.93 ± 0.57 ,0.001 1.89 ± 0.55 0.16 ± 0.06 1.54 ± 0.51 1.07 ± 0.42 1.33 ± 0.48 1.46 ± 0.43
P
Total (Mean ± SD)
P
0.001 7.74 ± 2.19 ,0.001 5.56 ± 1.81
0.16 ± 0.06 ,0.001 1.54 ± 0.51 ,0.001 1.07 ± 0.42 ,0.001 1.33 ± 0.48 ,0.001 1.46 ± 0.43 ,0.001 5.56 ± 1.81 ,0.001
0.09 ± 0.05
0.91 ± 0.45
0.65 ± 0.28
0.75 ± 0.30
0.86 ± 0.36
3.25 ± 1.38
0.16 ± 0.06 ,0.001 1.54 ± 0.51
0.002 1.07 ± 0.42
0.002 1.33 ± 0.48
0.005 1.46 ± 0.43
0.004 5.56 ± 1.81
0.08 ± 0.04
0.89 ± 0.35
0.49 ± 0.22
0.75 ± 0.35
0.89 ± 0.39
3.10 ± 1.28
0.09 ± 0.05
0.872 0.91 ± 0.45
1.000 0.65 ± 0.28
0.788 0.75 ± 0.30
1.000 0.86 ± 0.43
0.997 3.25 ± 1.38
0.08 ± 0.04
0.89 ± 0.35
0.49 ± 0.22
0.75 ± 0.35
0.89 ± 0.39
3.10 ± 1.28
ETDRS, Early Treatment Diabetic Retinopathy Study; SD, standard deviation.
0.002
0.996
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
Table 3. Multiple Comparison Between Choroidal Volume Measurements is the ETDRS Grid Subfields
885
886
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
Table 4. Multiple Regression Analysis for Total Choroidal Volume by Age, Axial Length, Sex, Staphyloma, and Group, in Highly Myopic Eyes Factor
Coefficients
Standard Error
Intercept Age Sex Axial length Staphyloma Group*
23.057 −0.032 −0.628 −0.495 −1.364 −0.488
3.598 0.010 0.310 0.125 0.354 0.248
R2
P
0.645 ,0.001 0.645 0.002 0.645 0.047 0.645 ,0.001 0.645 ,0.001 0.645 0.050
*Including highly myopic eyes without retinal or choroidal pathology, with traction maculopathy, and with CNV.
be misleading in the overall assessment of choroidal involvement by the pathologic myopia,22 especially because the chorio-scleral border is irregular. Since the choroid is a three-dimensional structure, we believe that a volumetric analysis of the choroid would be helpful to assess the disease course. As previously shown by our group in normal eyes,24 this study confirmed that the eyes with high myopia present a significantly lower CV than emmetropic eyes.
Since the choroidal blood flow is the highest of any tissue in the body to satisfy the normal metabolic demands of the outer retina,27 a very thin choroid as found in the highly myopic eyes may deliver decreased amounts of oxygen and nutrients to the outer retina. This decreased delivery may affect signal generation by the photoreceptors17 and eventually may lead to decreased visual function, especially in highly myopic elderly patients. The decreased CV in highly myopic eyes with traction maculopathy supports the evidence that progressive eyeball elongation accompanying choroidal and scleral thinning contributes to the development of these vitreoretinal tractional abnormalities. Myopic traction maculopathy is caused by a relative tautness or noncompliance involving the inner retina as compared with the outer retina, typically within the concavity of a posterior staphyloma.28 With eyeball elongation and chorio-scleral thinning, the relatively strong attachment of the internal limiting membrane to the posterior vitreous hyaloid generates inward vector forces that detach or split the neural retina, facilitated by
Fig. 3. Graphs showing the association between CV and age, axial length, sex, myopic group, and staphyloma in highly myopic eyes. A. Scatter plot of age and CV shows a significant negative correlation (P = 0.002; R2 = 0.178). B. Scatter plot of axial length and macular CV shows a significant negative correlation (P , 0.001; R2 = 0.272). C. Error bar of sex and macular CV shows a significantly lower volume in men than in women (P = 0.047). D. Error bar of myopic groups and CV shows that eyes without vitreoretinal or choroidal pathology have greater CV than (1) eyes with myopic traction maculopathy (2) eyes with CNV (3) (P = 0.05). E. Error bar of staphyloma and CV shows that eyes with the presence of posterior staphyloma have significantly lower CV than those with the absence of it (P , 0.001).
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
the presence of vitreous cortex shrinkage, epiretinal membranes, or rigid internal limiting membrane.4,6,10 The decreased CV in cases of myopic CNV confirms and strengthens the previous finding of decreased subfoveal choroidal thickness in highly myopic eyes with CNV in comparison to the absence of CNV.14 Although the pathogenesis of CNV in high myopia has not been fully clarified, previous studies have suggested a pathogenic relationship between lacquer cracks and the development of CNV.29 Lacquer cracks occur with the rupture of Bruch membrane, which is caused by excessive elongation of the eyeball30; eventually, wound-healing mechanisms after an acute rupture of the Bruch membrane may promote the development of CNV. As recently demonstrated by Wang et al,31 the choroid is thinner in eyes with stellate lacquer cracks than those without, suggesting that choroidal thickness may be a good predictor for the formation of lacquer cracks. Choroidal neovascularization may also result from the attenuation of the choroid resulting from increasing age14 and increased retinal pigment epithelium/choroid curvature32 in highly myopic eyes, leading to insufficient blood support of the outer retina, the retinal pigment epithelium, and even the choroid itself. A similar finding was recently reported by Ellabban et al21 using manual segmentation of the choroid on a prototype of swept-source OCT; they found that highly myopic eyes with inferior posterior staphyloma had marked choroidal thinning along the superior border of the staphyloma, and that the reduction of the choroidal thickness progressed with age and seemed to be involved in the development of CNV associated with the staphyloma. However, since we included eyes that were previously treated for myopic CNV, we cannot exclude that those treatments were involved in the choroidal thinning. Indeed, previous reports showed that choroidal thickness was decreased after photodynamic therapy for polypoidal choroidal vasculopathy,33 and also after intravitreal ranibizumab injections for wet age-related macular degeneration.34 However, the subfoveal choroidal thickness after ranibizumab treatment decreased by only 18 mm.34 In addition, Ikuno et al32 demonstrated that eyes with newly diagnosed myopic CNV have thinner choroid than the unaffected fellow eyes. They also demonstrated that choroidal thinning after intravitreal bevacizumab injections is transient in eyes with myopic CNV; after 3 months, the subfoveal thickness returned to the baseline value.35 Based on this evidence, previous treatment does not sufficiently explain the decrease in macular choroidal thickness in eyes with neovascular complications. Further longitudinal prospective studies are needed to clarify if choroidal thinning is a cause or consequence of the development of myopic CNV.
887
Interestingly, we also found sexual differences in CV; regression analysis showed that men had a thicker choroid than women. This finding is consistent with the results of previous studies of choroidal thickness36 and volume.24 The explanation of this finding remains uncertain; however, one biologic explanation may be a different hormonal exposure between men and women. Indeed, it has been indicated that sex and hormonal status may influence the choroidal blood flow,37,38 and therefore also the CV. The distribution pattern of CVs in highly myopic eyes and emmetropic eyes was similar. In all the eyes, the lowest macular CV was noted in the nasal quadrants of the ETDRS grid, and the greatest in the superior quadrants. Ikuno and Tano20 suggested that 1 possible reason for this observation might be the presence of a watershed zone between the macula and optic nerve, observed in 60% of the eyes.39 The posterior choroid is supplied by 2 major posterior ciliary arteries (lateral and medial) in approximately 90% of the eyes.39 The watershed zone indicates the isolation of the choroidal capillary bed supplied by an independent posterior ciliary artery that does not anastomose with another end artery, making it prone to ischemia. Otherwise, the presence of a superior posterior ciliary artery (reported in 10% of the eyes)20 may explain why the superior quadrants show the greatest CV. Our study has some limitations. First, it is retrospective in nature; however, we carefully excluded eyes/patients with conditions that have been suggested to affect the choroid and eye with evidence of chorioretinal atrophy. Second, we performed a manual segmentation of the choroid because of the lack of an automatic built-in software; previous studies have demonstrated good reliability of this manual technique.22,24 Third, the few highly myopic eyes with CNV may not be sufficient to assess differences in CV between the eyes with myopic traction maculopathy and eyes with myopic CNV. Fourth, another potential problem with the accuracy of CV measurements in SD-OCT devices is the different transverse magnification of OCT scans in eyes with different axial lengths.40 However, we used Spectralis SD-OCT that uses a telecentric optical system and uses software correction for ocular transverse magnification errors based upon the corneal curvature and defocus setting of the instrument. Measures with Spectralis SD-OCT are therefore likely to be less affected by ocular magnification effects than other SD-OCT devices that do not compensate for magnification effects, however small errors in transverse measurement location could still be possible. Based upon the measures of a model eye with Spectralis SD-OCT, we estimated that the maximum error in transverse magnification based
888
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
upon the axial length differences in this study is likely to be ,10%, which would result in ,3% effect on the measured CV. It is therefore unlikely that these effects substantially confounded the comparisons made in this study. In conclusion, our study described variations in CV in eyes with myopic pathologies using the EDI modality on a commercially available SD-OCT device. The choroid has a lower volume in highly myopic eyes than in emmetropic eyes; the reduction of CV depends on age, axial length, sex, and presence/ absence of posterior staphyloma. However, as a new finding we demonstrated that eyes with coexisting myopic traction maculopathy or CNV have more severe choroidal thinning, most likely leading to insufficient metabolic supplementation for the macula in elderly subjects. Therefore, such patients should receive a closer follow-up than emmetropic eyes and highly myopic eyes without vitreoretinal pathologies or CNV, since choroidal thinning may play a role in the pathogenesis of visual dysfunctions. Key words: choroidal volume, choroidal thickness, high myopia, myopic traction maculopathy, myopic choroidal neovascularization, enhanced depth imaging. References 1. Wang NK, Lai CC, Chu HY, et al. Classification of early drytype myopic maculopathy with macular choroidal thickness. Am J Ophthalmol 2012;153:669–677, 677 e1–2. 2. Phillips CI, Dobbie JG. Posterior staphyloma and retnal detachment. Am J Ophthalmol 1963;55:332–335. 3. Siam A. Macular hole with central retinal detachment in high myopia with posterior staphyloma. Br J Ophthalmol 1969;53: 62–63. 4. Takano M, Kishi S. Foveal retinoschisis and retinal detachment in severely myopic eyes with posterior staphyloma. Am J Ophthalmol 1999;128:472–476. 5. Baba T, Ohno-Matsui K, Futagami S, et al. Prevalence and characteristics of foveal retinal detachment without macular hole in high myopia. Am J Ophthalmol 2003;135:338–342. 6. Sayanagi K, Ikuno Y, Tano Y. Tractional internal limiting membrane detachment in highly myopic eyes. Am J Ophthalmol 2006;142:850–852. 7. Grossniklaus HE, Green WR. Pathologic findings in pathologic myopia. Retina 1992;12:127–133. 8. Akiba J, Konno S, Yoshida A. Retinal detachment associated with a macular hole in severely myopic eyes. Am J Ophthalmol 1999;128:654–655. 9. Morita H, Ideta H, Ito K, et al. Causative factors of retinal detachment in macular holes. Retina 1991;11:281–284. 10. Panozzo G, Mercanti A. Optical coherence tomography findings in myopic traction maculopathy. Arch Ophthalmol 2004; 122:1455–1460. 11. Okabe S, Matsuo N, Okamoto S, Kataoka H. Electron microscopic studies on retinochoroidal atrophy in the human eye. Acta Med Okayama 1982;36:11–21. 12. Ohno-Matsui K, Akiba M, Moriyama M, et al. Imaging retrobulbar subarachnoid space around optic nerve by swept-source
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30. 31.
32.
optical coherence tomography in eyes with pathologic myopia. Invest Ophthalmol Vis Sci 2011;52:9644–9650. Spaide RF, Koizumi H, Pozzoni MC. Enhanced depth imaging spectral-domain optical coherence tomography. Am J Ophthalmol 2008;146:496–500. Fujiwara T, Imamura Y, Margolis R, et al. Enhanced depth imaging optical coherence tomography of the choroid in highly myopic eyes. Am J Ophthalmol 2009;148:445–450. Ho J, Branchini L, Regatieri C, et al. Analysis of normal peripapillary choroidal thickness via spectral domain optical coherence tomography. Ophthalmology 2011;118:2001–2007. Manjunath V, Taha M, Fujimoto JG, Duker JS. Choroidal thickness in normal eyes measured using Cirrus HD optical coherence tomography. Am J Ophthalmol 2010;150:325–329 e1. Nishida Y, Fujiwara T, Imamura Y, et al. Choroidal thickness and visual acuity in highly myopic eyes. Retina 2012;32: 1229–1236. Imamura Y, Iida T, Maruko I, et al. Enhanced depth imaging optical coherence tomography of the sclera in dome-shaped macula. Am J Ophthalmol 2011;151:297–302. Maruko I, Iida T, Sugano Y, et al. Morphologic analysis in pathologic myopia using high-penetration optical coherence tomography. Invest Ophthalmol Vis Sci 2012;53:3834–3838. Ikuno Y, Tano Y. Retinal and choroidal biometry in highly myopic eyes with spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci 2009;50:3876–3880. Ellabban AA, Tsujikawa A, Matsumoto A, et al. Macular choroidal thickness measured by swept source optical coherence tomography in eyes with inferior posterior staphyloma. Invest Ophthalmol Vis Sci 2012;53:7735–7745. Chhablani J, Barteselli G, Wang H, et al. Repeatability and reproducibility of manual choroidal volume measurements using enhanced depth imaging optical coherence tomography. Invest Ophthalmol Vis Sci 2012;53:2274–2280. Chhablani J, Barteselli G, Bartsch DU, et al. Influence of scanning density on macular choroidal volume measurement using spectral-domain optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 2013;251:1303–1309. Barteselli G, Chhablani J, El-Emam S, et al. Choroidal volume variations with age, axial length, and sex in healthy subjects: a three-dimensional analysis. Ophthalmology 2012; 119:2572–2578. Llorente L, Barbero S, Cano D, et al. Myopic versus hyperopic eyes: axial length, corneal shape and optical aberrations. J Vis 2004;4:288–298. Flores-Moreno I, Lugo F, Duker JS, Ruiz-Moreno JM. The relationship between axial length and choroidal thickness in eyes with high myopia. Am J Ophthalmol 2013;155:314–319 e1. Linsenmeier RA, Padnick-Silver L. Metabolic dependence of photoreceptors on the choroid in the normal and detached retina. Invest Ophthalmol Vis Sci 2000;41:3117–3123. VanderBeek BL, Johnson MW. The diversity of traction mechanisms in myopic traction maculopathy. Am J Ophthalmol 2012;153:93–102. Noble KG, Carr RE. Pathologic myopia. Ophthalmology 1982; 89:1099–1100. Klein RM, Green S. The development of lacquer cracks in pathologic myopia. Am J Ophthalmol 1988;106:282–285. Wang NK, Lai CC, Chou CL, et al. Choroidal thickness and biometric markers for the screening of lacquer cracks in patients with high myopia. PloS One 2013;8:e53660. Ikuno Y, Jo Y, Hamasaki T, Tano Y. Ocular risk factors for choroidal neovascularization in pathologic myopia. Invest Ophthalmol Vis Sci 2010;51:3721–3725.
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL 33. Maruko I, Iida T, Sugano Y, et al. Subfoveal retinal and choroidal thickness after verteporfin photodynamic therapy for polypoidal choroidal vasculopathy. Am J Ophthalmol 2011; 151:594–603 e1. 34. Yamazaki T, Koizumi H, Yamagishi T, Kinoshita S. Subfoveal choroidal thickness after ranibizumab therapy for neovascular age-related macular degeneration: 12-month results. Ophthalmology 2012;119:1621–1627. 35. Sayanagi K, Jo Y, Ikuno Y. Transient choroidal thinning after intravitreal bevacizumab injection for myopic choroidal neovascularization. J Clinic Experiment Ophthalmol 2011;2:6. 36. Li XQ, Larsen M, Munch IC. Subfoveal choroidal thickness in relation to sex and axial length in 93 Danish
37.
38.
39. 40.
889
university students. Invest Ophthalmol Vis Sci 2011;52: 8438–8441. Centofanti M, Bonini S, Manni G, et al. Do sex and hormonal status influence choroidal circulation? Br J Ophthalmol 2000; 84:786–787. Kavroulaki D, Gugleta K, Kochkorov A, et al. Influence of gender and menopausal status on peripheral and choroidal circulation. Acta Ophthalmol 2010;88:850–853. Hayreh SS. The ophthalmic artery: III. Branches. Br J Ophthalmol 1962;46:212–247. Odell D, Dubis AM, Lever JF, et al. Assessing errors inherent in OCT-derived macular thickness maps. J Ophthalmol 2011; 2011:692574.
P
imaged and assessed by spectral domain optical coherence tomography (SD-OCT), and can be grouped into what has been called “myopic traction maculopathy.”10 In addition to clinical findings, histologic studies have demonstrated choroidal thinning in eyes with high myopia, with a lack of large vessels in some areas and pronounced thinning of the choriocapillaris in others.7,11 Indocyanine green angiography has been widely used in ophthalmology to provide information about the choroidal perfusion, and has been useful for observing various pathologic changes in highly myopic eyes.12 However, it does not allow cross-sectional imaging of the choroidal structures.
athologic myopia is one of the leading causes of vision impairment worldwide. It is characterized by congenital scleral weakness leading to the progressive elongation of the globe resulting in a variety of fundus changes, such as posterior staphyloma, progressive atrophy of the choriocapillaris and choroid, lacquer cracks, and eventually choroidal neovascularization (CNV).1 Apart from these abnormalities, the posterior retina can also be damaged by the presence of pathologic tractions induced by epiretinal membrane and residual vitreomacular adhesion, eventually leading to retinoschisis, posterior retinal detachment, and macular holes.2–9 These structural alterations induced by tractional forces may be 880
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
Recently, enhanced depth imaging (EDI) on SD-OCT has provided three-dimensional analysis of the anatomy of deeper structures in the eye including the choroid.13,14 Using EDI-OCT, the chorio-scleral junction can be easily visualized, and its distal margin can be used to measure choroidal thickness.15,16 Previous investigators have used manual choroidal measurements in the fovea and a few perifoveal locations, and they reported that the choroid in eyes with high myopia is markedly thinner when compared with normal eyes.17–20 However, measuring thickness only at selected points of the choroid may not describe comprehensively the thinning of the choroid, since the irregularity of the inner chorio-scleral border may influence the measurement at few sampling points.21 To overcome this problem and to obtain a better picture of the state of the posterior choroid, our group devised a manual method to measure the choroidal volume (CV) of the entire macula22,23 that has been extensively described in nonpathologic eyes using a commercially available SD-OCT device.24 On the basis of these previous reports, we aimed to compare macular CV measurements between emmetropic eyes and highly myopic eyes without evidence of chorioretinal atrophy, and also to assess if the presence of common myopic fundus findings (such as staphyloma, peripapillary atrophy, tilted disk, and lacquer cracks), or myopic traction maculopathy, or CNV affects the CV. Methods Starting in July 2011, all patients undergoing SD-OCT scans at the Jacobs Retina Center at Shiley Eye Center, University of California San Diego (La Jolla, CA) underwent choroidal imaging as part of their evaluation. A raster scan of the macula using Heidelberg Spectralis (Heidelberg Engineering, Carlsbad, CA) in EDI mode was incorporated into the standard From the *Department of Ophthalmology, Jacobs Retina Center at Shiley Eye Center, University of California San Diego, La Jolla, California; †Department of Clinical Sciences and Community Health, Ophthalmological Unit, Ca’ Granda Foundation-Ospedale Maggiore Policlinico, University of Milan, Milan, Italy; and ‡Department of Ophthalmology, Eulji University Hospital, Daejeon, Republic of Korea. J. Chhablani is now at L. V. Prasad Eye Institute, Hyderabad, India. Supported by NIH grants R01EY007366 and R01EY018589 (W.R.F.), R01EY020617 (L.C.), and in part by an unrestricted fund from Research to Prevent Blindness to the Department of Ophthalmology, University of California San Diego. None of the authors have any conflicting interests to disclose. G. Barteselli and S. N. Lee contributed equally as first authors. Reprint requests: William R. Freeman, MD, University of California at San Diego, Shiley Eye Center, 0946, 9415 Campus Point Drive, La Jolla, CA 92037; e-mail: [email protected]
881
scanning protocol and used to study the choroid. After approval by the institutional review board of the University of California at San Diego for chart review and data analysis, we retrospectively reviewed charts and OCT images of 98 consecutive patients who underwent EDI-OCT scans from July 2011 to June 2012. We included 31 normal emmetropic eyes (axial length $23.5 and ,24.5 mm),25 and 67 highly myopic eyes with an axial length .26.5 mm.25 Highly myopic eyes were divided into three groups: 1) eyes without vitreoretinal or choroidal pathology; 2) eyes with traction maculopathy; and 3) eyes with a history of CNV. The presence of other myopic fundus findings, such as staphyloma, peripapillary atrophy, tilted disk, and lacquer cracks, was also assessed for each eye. Eyes with evidence of patchy or diffuse chorioretinal atrophy in the macula, or eyes that received previous intravitreal antivascular endothelial growth factor injections within 6 months of the imaging study were not included in this study. Patients who had any systemic or ocular disease that could affect the choroid, including uncontrolled hypertension, diabetes, coagulopathies, CNV secondary to age-related macular degeneration, central serous chorioretinopathy, or other conditions that have been shown to affect the choroid were also excluded. The study was conducted in adherence with the tenets of the Declaration of Helsinki. Patients underwent ophthalmic evaluation, including best-corrected visual acuity, slit-lamp biomicroscopy, axial length measurement using ocular biometry (IOLMaster; Carl Zeiss Meditec, Jena, Germany), and a raster scan of the macula using the EDI-OCT. The raster scan consisted of 31 high resolution B-scans associated with a simultaneous near-infrared image; an internal fixation light was used to center the scanning area on the fovea. Each scan was 9.0 mm in length and spaced 240 mm apart. All 31 OCT B-scans were acquired in a continuous automated sequence and covered a 30° · 25° area centered on the fovea. Image averaging of 25 frames was used to obtain a good quality choroidal image by the built-in Active Eye Tracking software (TruTrack; Heidelberg Engineering). Manual choroidal segmentation of each B-scan was performed by 2 retina specialists according to a previously described well-reproducible method (Figure 1).22 Choroidal volume map centered on the fovea was automatically generated by the built-in software of the device by applying the Early Treatment Diabetic Retinopathy Study (ETDRS) grid. After checking the position of the ETDRS grid to ensure a correct foveal placement, measurements for the CV were recorded for each subfield of the grid. The ETDRS grid divides the macula into 3 concentric rings (center, inner, and outer), with the inner ring measuring 1 mm to 3 mm and the outer ring
882
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
Fig. 1. Enhanced depth imaging optical coherence tomography raster scan protocol (A), automated retinal segmentation (B), manual choroidal segmentation (C), and choroidal measurements on the standardized ETDRS grid (D) including three concentric rings with a total of nine subfields centered on the fovea. To demarcate choroidal boundaries, the internal limiting membrane line (arrows, A) was manually moved to the base of retinal pigment epithelium (arrows, B), whereas the basement membrane line (asterisks, A) was manually moved to chorio-scleral interface (asterisks, B). BM, basement membrane; ILM, internal limiting membrane.
measuring 3 mm to 6 mm from the center (referring to a ring with a diameter of 1 mm centered on the fovea). The grid further divides inner and outer rings into four quadrants (superior, inferior, temporal, and nasal). Statistical analysis was performed using the Statistical Package for the Social Sciences version 20 (SPSS Inc, Chicago, IL). Continuous variables are expressed as mean ± standard deviation. One eye per subject was included; in the group of normal emmetropic subjects, the eye was chosen randomly; in the group of highly myopic subjects, the eye was chosen based on the presence of a vitreoretinal or choroidal pathology, or the longest axial length if the pathology was bilateral. We examined the association of myopic fundus findings with age, axial length, and sex. The relationship was assessed using the chi-square test (or Fisher exact test if .20% of cells have expected count of ,5) or analysis of variance. Multiple linear regression analysis was used to evaluate the correlation between CV in highly myopic eyes and independent variables including age, axial length, sex, myopic fundus findings, and myopic groups (including the group without vitreoretinal or choroidal pathology, the group with traction maculopathy, and the group with CNV). Tukey–Kramer test was used to compare the mean CV between groups. Agreement between interobserver measurements in
a subset of emmetropic and myopic eyes was assessed using the concordance correlation coefficient. A P , 0.05 was considered to be statistically significant.
Results A total of 98 eyes of 98 patients (46 men and 52 women) with a mean age of 53 years (range, 20–91 years) were included in the study. We studied 31 normal emmetropic eyes of 31 patients with a mean age of 53 years (range, 20–85 years), and 67 highly myopic eyes of 67 patients with a mean age of 57 years (range, 21–91 years). Mean logMAR best-corrected visual acuity was 0.026 ± 0.088 in emmetropic eyes and 0.146 ± 0.237 in highly myopic eyes. Highly myopic eyes included 36 eyes without vitreoretinal or choroidal pathology, 21 eyes with traction maculopathy, and 10 eyes with a history of CNV. In the group with myopic traction maculopathy, 10 eyes had epiretinal membrane, 5 had lamellar macular hole, 3 had foveoschisis, 2 had vitreomacular traction syndrome, and 1 had fullthickness macular hole. Of the 10 eyes with a history of CNV, 6 were treated with photodynamic therapy and 4 were treated with intravitreal antivascular endothelial growth factor medications; none of the eyes was treated
Table 1. Distribution of Study Eyes Highly Myopic (n = 67)
Age (years, range) Sex (M/F) Axial length (mm, range) Staphyloma (n, %) Peripapillary atrophy (n, %) Tilted disk (n, %) Lacquer cracks (n, %)
Emmetropic (n = 31)
MTM (n = 21)
CNV (n = 10)
No Retinal or Choroidal Pathology (n = 36)
53 (20–85) 14/17 24.0 ± 0.3 0 (0%) 0 (0%) 0 (0%) 0 (0%)
64 (41–91) 9/12 28.4 ± 1.5 16 (76%) 20 (95%) 15 (71%) 7 (33%)
67 (41–90) 4/6 28.6 ± 1.5 6 (60%) 10 (100%) 4 (40%) 6 (60%)
50 (21–85) 19/17 27.6 ± 1.1 9 (25%) 22 (61%) 7 (19%) 0 (0%)
F, female; M, male; MTM, myopic traction maculopathy.
883 0 (0) 5 (21.7) 8 (24.2) 0.200 11 23 33 3 (27.3) 8 (34.8) 15 (45.5) 0.500 11 23 33 7 (63.6) 16 (69.6) 29 (87.9) 0.129 11 23 33 2 (18.2) 11 (47.8) 18 (54.5) 0.110 11 23 33 0 (0) 3 (13.0) 7 (21.2) 0.221 MTM, myopic traction maculopathy.
0 (0) 6 (26.1) 15 (45.5) 0.015 11 23 33
11 23 33
3 (9.4) 10 (28.6) 0.047 32 35 10 (31.2) 16 (45.7) 0.225 32 35 23 (71.9) 39 (82.9) 0.281 32 35 14 (43.8) 17 (48.6) 0.693 32 35 4 (12.5) 6 (17.1) 0.594 10 (31.2) 11 (31.4) 0.987 32 35
32 35
1 (3.6) 4 (23.5) 8 (36.4) 0.013 28 17 22 3 (10.7) 9 (52.9) 14 (63.6) ,0.001 28 17 22 17 (60.7) 15 (88.2) 20 (90.9) 0.019 28 17 22 5 (17.9) 12 (70.6) 14 (63.6) ,0.001 28 17 22 2 (7.1) 4 (23.5) 4 (18.2) 0.285 6 (21.4) 6 (35.3) 9 (40.9) 0.311 28 17 22
28 17 22
13 (19.4) 67 26 (38.8) 67 52 (77.6) 67 31 (46.3) 67 10 (14.9) 67 21 (31.3) 67
All Axial length (mm) 26.5–27.49 27.5–28.49 28.5+ Ptrend Sex Male Female P Age (years) 20–39 40–59 60+ Ptrend
n (%) N n (%) N
n (%)
N
n (%)
N
n (%)
N
n (%)
N
Tilted Disk Peripapillary Atrophy Staphyloma CNV MTM
Table 2. Summary of Fundus Changes With Axial Length, Sex, and Age, in Highly Myopic Eyes
within 6 months of the imaging study. Table 1 shows the distribution of subjects divided by groups according to age, sex, axial length, and myopic fundus findings (including staphyloma, peripapillary atrophy, tilted disk, and lacquer cracks). When the axial length was divided into categories (26.5–27.49 mm, 27.5–28.49 mm, and $28.5 mm), staphyloma (P = 0.001), peripapillary atrophy (P = 0.019), tilted disk (P , 0.001), and lacquer cracks (P = 0.013) increased in prevalence with increasing axial length in highly myopic eyes. We noticed that women had a greater frequency of lacquer cracks than men (P = 0.047). Sex did not alter the prevalence of the other pathologic myopia findings, whereas traction maculopathy increased in prevalence with increasing age (P = 0.015, Table 2). The interobserver concordance correlation coefficient for the total CV in a subset of 15 emmetropic and 15 myopic eyes that were randomly selected from the sample was 0.93. The mean subfoveal and total CVs were 0.24 ± 0.07 mm3 and 7.74 ± 2.19 mm3 in emmetropic eyes; 0.16 ± 0.06 mm3 and 5.56 ± 1.81 mm3 in highly myopic eyes without retinal or choroidal pathology; 0.09 ± 0.05 mm3 and 3.25 ± 1.38 mm3 in cases of traction maculopathy; 0.08 ± 0.04 mm3 and 3.10 ± 1.28 mm3 in cases of myopic CNV (Figure 2). Multiple comparisons analysis showed that the subfoveal and total CVs were significantly lower in highly myopic eyes without retinal or choroidal pathology than in emmetropic eyes (both P , 0.001, Table 3). Compared with highly myopic eyes without retinal or choroidal pathology, the total CV was lower in highly myopic eyes with traction maculopathy or CNV (P , 0.001 and P = 0.002, respectively, Table 3). No difference in CV was noticed between the eyes with traction maculopathy and eyes with myopic CNV. Considering all eyes, CV was greater in the superior and temporal quadrants than in the inferior (P , 0.05) and nasal quadrants (P , 0.001); the nasal quadrant showed the lowest CV among the 4 quadrants (P , 0.001). When comparing the CV between emmetropic and highly myopic eyes (Table 3), we found that volume in each quadrant was significantly lower in highly myopic eyes (P , 0.001). Considering only highly myopic eyes, CV in the eyes without retinal or choroidal pathology was significantly greater in each quadrant than in the eyes with traction maculopathy or CNV (P , 0.01). Table 4 shows the results of the multiple linear regression analysis for mean total macular CV in highly myopic eyes. A stepwise method was used to determine the most unexpected factors. The model determined by age, axial length, sex, staphyloma, and myopic group had the best regression. This model showed a high coefficient of determination, which was 0.803 (R2 = 0.645).
Lacquer Cracks
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
884
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
Fig. 2. Enhanced depth imaging optical coherence tomography scans with arrowheads indicating the chorio-scleral junction (First column); choroidal thickness and volume measurements in the ETDRS grid centered on the fovea (Second column); and color-coded CV maps (Third column). In an emmetropic eye (axial length: 23.8 mm), the choroid-scleral junction was well visible, and the macular CV was 6.93 mm3 (A). The choroid of a nonpathologic highly myopic eye (axial length: 26.6 mm) was thin, showing a macular CV of 4.34 mm3 (B). In a case of highly myopic eye (axial length: 27.6 mm), epiretinal membrane and retinoschisis in the outer plexiform layer were observed above thin choroid with a CV of 2.37 mm3 (C). In a case of highly myopic eye (axial length: 28.1 mm), hyperreflective juxtafoveal CNV was observed over thin choroid with a CV of 3.55 mm3 (D).
In highly myopic eyes, univariate regression analysis showed a decrease in CV of approximately 0.32 mm3 for every 10 years and a decrease of 0.49 mm3 per millimeter of axial length. Figure 3, A and B shows that the total CV in the ETDRS map was negatively correlated with age (P = 0.002, R2 = 0.178), and axial length (P , 0.001, R2 = 0.272). Men had a greater CV than women (P = 0.047, Figure 3C). The presence of traction maculopathy or CNV lead to lower CV than in cases of highly myopic eyes without vitreoretinal or choroidal pathologies (P = 0.05, Figure 3D). The presence of a staphyloma significantly reduced the CV (P , 0.001, Figure 3E). Discussion This study demonstrated differences in macular CV between emmetropic eyes and highly myopic eyes,
and investigated variations of CV between highly myopic eyes without retinal or choroidal pathologies and highly myopic eyes with traction maculopathy or CNV. Even in the absence of visible chorioretinal atrophy in the macula, highly myopic eyes have lower CV than emmetropic eyes, as expected from a previous study investigating choroidal thinning in the highly myopic eyes.26 Beside age, axial length, or presence of a posterior staphyloma—that are known to affect the choroid in highly myopic eyes14,17,19—we demonstrated that sex and presence of traction maculopathy or CNV are independent predictors for the CV as well. So far, the choroid in highly myopic eyes has been investigated bi-dimensionally using conventional SD-OCT devices.14,17,18 Previous studies manually measured the choroidal thickness only in selected points of the macula.14,17–19,26 However, thickness measurements taken at multiple single points could
ETDRS Subfields Eyes Emmetropic Highly myopic without retinal or choroidal pathology Highly myopic without retinal or choroidal pathology Myopic traction maculopathy Highly myopic without retinal or choroidal pathology Myopic choroidal neovascularization Myopic traction maculopathy Myopic choroidal neovascularization
Subfoveal (Mean ± SD)
P
Superior (Mean ± SD)
P
Nasal (Mean ± SD)
P
Inferior (Mean ± SD)
P
Temporal (Mean ± SD)
0.24 ± 0.07 ,0.001 2.07 ± 0.57 ,0.001 1.62 ± 0.57 ,0.001 1.93 ± 0.57 ,0.001 1.89 ± 0.55 0.16 ± 0.06 1.54 ± 0.51 1.07 ± 0.42 1.33 ± 0.48 1.46 ± 0.43
P
Total (Mean ± SD)
P
0.001 7.74 ± 2.19 ,0.001 5.56 ± 1.81
0.16 ± 0.06 ,0.001 1.54 ± 0.51 ,0.001 1.07 ± 0.42 ,0.001 1.33 ± 0.48 ,0.001 1.46 ± 0.43 ,0.001 5.56 ± 1.81 ,0.001
0.09 ± 0.05
0.91 ± 0.45
0.65 ± 0.28
0.75 ± 0.30
0.86 ± 0.36
3.25 ± 1.38
0.16 ± 0.06 ,0.001 1.54 ± 0.51
0.002 1.07 ± 0.42
0.002 1.33 ± 0.48
0.005 1.46 ± 0.43
0.004 5.56 ± 1.81
0.08 ± 0.04
0.89 ± 0.35
0.49 ± 0.22
0.75 ± 0.35
0.89 ± 0.39
3.10 ± 1.28
0.09 ± 0.05
0.872 0.91 ± 0.45
1.000 0.65 ± 0.28
0.788 0.75 ± 0.30
1.000 0.86 ± 0.43
0.997 3.25 ± 1.38
0.08 ± 0.04
0.89 ± 0.35
0.49 ± 0.22
0.75 ± 0.35
0.89 ± 0.39
3.10 ± 1.28
ETDRS, Early Treatment Diabetic Retinopathy Study; SD, standard deviation.
0.002
0.996
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
Table 3. Multiple Comparison Between Choroidal Volume Measurements is the ETDRS Grid Subfields
885
886
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
Table 4. Multiple Regression Analysis for Total Choroidal Volume by Age, Axial Length, Sex, Staphyloma, and Group, in Highly Myopic Eyes Factor
Coefficients
Standard Error
Intercept Age Sex Axial length Staphyloma Group*
23.057 −0.032 −0.628 −0.495 −1.364 −0.488
3.598 0.010 0.310 0.125 0.354 0.248
R2
P
0.645 ,0.001 0.645 0.002 0.645 0.047 0.645 ,0.001 0.645 ,0.001 0.645 0.050
*Including highly myopic eyes without retinal or choroidal pathology, with traction maculopathy, and with CNV.
be misleading in the overall assessment of choroidal involvement by the pathologic myopia,22 especially because the chorio-scleral border is irregular. Since the choroid is a three-dimensional structure, we believe that a volumetric analysis of the choroid would be helpful to assess the disease course. As previously shown by our group in normal eyes,24 this study confirmed that the eyes with high myopia present a significantly lower CV than emmetropic eyes.
Since the choroidal blood flow is the highest of any tissue in the body to satisfy the normal metabolic demands of the outer retina,27 a very thin choroid as found in the highly myopic eyes may deliver decreased amounts of oxygen and nutrients to the outer retina. This decreased delivery may affect signal generation by the photoreceptors17 and eventually may lead to decreased visual function, especially in highly myopic elderly patients. The decreased CV in highly myopic eyes with traction maculopathy supports the evidence that progressive eyeball elongation accompanying choroidal and scleral thinning contributes to the development of these vitreoretinal tractional abnormalities. Myopic traction maculopathy is caused by a relative tautness or noncompliance involving the inner retina as compared with the outer retina, typically within the concavity of a posterior staphyloma.28 With eyeball elongation and chorio-scleral thinning, the relatively strong attachment of the internal limiting membrane to the posterior vitreous hyaloid generates inward vector forces that detach or split the neural retina, facilitated by
Fig. 3. Graphs showing the association between CV and age, axial length, sex, myopic group, and staphyloma in highly myopic eyes. A. Scatter plot of age and CV shows a significant negative correlation (P = 0.002; R2 = 0.178). B. Scatter plot of axial length and macular CV shows a significant negative correlation (P , 0.001; R2 = 0.272). C. Error bar of sex and macular CV shows a significantly lower volume in men than in women (P = 0.047). D. Error bar of myopic groups and CV shows that eyes without vitreoretinal or choroidal pathology have greater CV than (1) eyes with myopic traction maculopathy (2) eyes with CNV (3) (P = 0.05). E. Error bar of staphyloma and CV shows that eyes with the presence of posterior staphyloma have significantly lower CV than those with the absence of it (P , 0.001).
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL
the presence of vitreous cortex shrinkage, epiretinal membranes, or rigid internal limiting membrane.4,6,10 The decreased CV in cases of myopic CNV confirms and strengthens the previous finding of decreased subfoveal choroidal thickness in highly myopic eyes with CNV in comparison to the absence of CNV.14 Although the pathogenesis of CNV in high myopia has not been fully clarified, previous studies have suggested a pathogenic relationship between lacquer cracks and the development of CNV.29 Lacquer cracks occur with the rupture of Bruch membrane, which is caused by excessive elongation of the eyeball30; eventually, wound-healing mechanisms after an acute rupture of the Bruch membrane may promote the development of CNV. As recently demonstrated by Wang et al,31 the choroid is thinner in eyes with stellate lacquer cracks than those without, suggesting that choroidal thickness may be a good predictor for the formation of lacquer cracks. Choroidal neovascularization may also result from the attenuation of the choroid resulting from increasing age14 and increased retinal pigment epithelium/choroid curvature32 in highly myopic eyes, leading to insufficient blood support of the outer retina, the retinal pigment epithelium, and even the choroid itself. A similar finding was recently reported by Ellabban et al21 using manual segmentation of the choroid on a prototype of swept-source OCT; they found that highly myopic eyes with inferior posterior staphyloma had marked choroidal thinning along the superior border of the staphyloma, and that the reduction of the choroidal thickness progressed with age and seemed to be involved in the development of CNV associated with the staphyloma. However, since we included eyes that were previously treated for myopic CNV, we cannot exclude that those treatments were involved in the choroidal thinning. Indeed, previous reports showed that choroidal thickness was decreased after photodynamic therapy for polypoidal choroidal vasculopathy,33 and also after intravitreal ranibizumab injections for wet age-related macular degeneration.34 However, the subfoveal choroidal thickness after ranibizumab treatment decreased by only 18 mm.34 In addition, Ikuno et al32 demonstrated that eyes with newly diagnosed myopic CNV have thinner choroid than the unaffected fellow eyes. They also demonstrated that choroidal thinning after intravitreal bevacizumab injections is transient in eyes with myopic CNV; after 3 months, the subfoveal thickness returned to the baseline value.35 Based on this evidence, previous treatment does not sufficiently explain the decrease in macular choroidal thickness in eyes with neovascular complications. Further longitudinal prospective studies are needed to clarify if choroidal thinning is a cause or consequence of the development of myopic CNV.
887
Interestingly, we also found sexual differences in CV; regression analysis showed that men had a thicker choroid than women. This finding is consistent with the results of previous studies of choroidal thickness36 and volume.24 The explanation of this finding remains uncertain; however, one biologic explanation may be a different hormonal exposure between men and women. Indeed, it has been indicated that sex and hormonal status may influence the choroidal blood flow,37,38 and therefore also the CV. The distribution pattern of CVs in highly myopic eyes and emmetropic eyes was similar. In all the eyes, the lowest macular CV was noted in the nasal quadrants of the ETDRS grid, and the greatest in the superior quadrants. Ikuno and Tano20 suggested that 1 possible reason for this observation might be the presence of a watershed zone between the macula and optic nerve, observed in 60% of the eyes.39 The posterior choroid is supplied by 2 major posterior ciliary arteries (lateral and medial) in approximately 90% of the eyes.39 The watershed zone indicates the isolation of the choroidal capillary bed supplied by an independent posterior ciliary artery that does not anastomose with another end artery, making it prone to ischemia. Otherwise, the presence of a superior posterior ciliary artery (reported in 10% of the eyes)20 may explain why the superior quadrants show the greatest CV. Our study has some limitations. First, it is retrospective in nature; however, we carefully excluded eyes/patients with conditions that have been suggested to affect the choroid and eye with evidence of chorioretinal atrophy. Second, we performed a manual segmentation of the choroid because of the lack of an automatic built-in software; previous studies have demonstrated good reliability of this manual technique.22,24 Third, the few highly myopic eyes with CNV may not be sufficient to assess differences in CV between the eyes with myopic traction maculopathy and eyes with myopic CNV. Fourth, another potential problem with the accuracy of CV measurements in SD-OCT devices is the different transverse magnification of OCT scans in eyes with different axial lengths.40 However, we used Spectralis SD-OCT that uses a telecentric optical system and uses software correction for ocular transverse magnification errors based upon the corneal curvature and defocus setting of the instrument. Measures with Spectralis SD-OCT are therefore likely to be less affected by ocular magnification effects than other SD-OCT devices that do not compensate for magnification effects, however small errors in transverse measurement location could still be possible. Based upon the measures of a model eye with Spectralis SD-OCT, we estimated that the maximum error in transverse magnification based
888
RETINA, THE JOURNAL OF RETINAL AND VITREOUS DISEASES 2014 VOLUME 34 NUMBER 5
upon the axial length differences in this study is likely to be ,10%, which would result in ,3% effect on the measured CV. It is therefore unlikely that these effects substantially confounded the comparisons made in this study. In conclusion, our study described variations in CV in eyes with myopic pathologies using the EDI modality on a commercially available SD-OCT device. The choroid has a lower volume in highly myopic eyes than in emmetropic eyes; the reduction of CV depends on age, axial length, sex, and presence/ absence of posterior staphyloma. However, as a new finding we demonstrated that eyes with coexisting myopic traction maculopathy or CNV have more severe choroidal thinning, most likely leading to insufficient metabolic supplementation for the macula in elderly subjects. Therefore, such patients should receive a closer follow-up than emmetropic eyes and highly myopic eyes without vitreoretinal pathologies or CNV, since choroidal thinning may play a role in the pathogenesis of visual dysfunctions. Key words: choroidal volume, choroidal thickness, high myopia, myopic traction maculopathy, myopic choroidal neovascularization, enhanced depth imaging. References 1. Wang NK, Lai CC, Chu HY, et al. Classification of early drytype myopic maculopathy with macular choroidal thickness. Am J Ophthalmol 2012;153:669–677, 677 e1–2. 2. Phillips CI, Dobbie JG. Posterior staphyloma and retnal detachment. Am J Ophthalmol 1963;55:332–335. 3. Siam A. Macular hole with central retinal detachment in high myopia with posterior staphyloma. Br J Ophthalmol 1969;53: 62–63. 4. Takano M, Kishi S. Foveal retinoschisis and retinal detachment in severely myopic eyes with posterior staphyloma. Am J Ophthalmol 1999;128:472–476. 5. Baba T, Ohno-Matsui K, Futagami S, et al. Prevalence and characteristics of foveal retinal detachment without macular hole in high myopia. Am J Ophthalmol 2003;135:338–342. 6. Sayanagi K, Ikuno Y, Tano Y. Tractional internal limiting membrane detachment in highly myopic eyes. Am J Ophthalmol 2006;142:850–852. 7. Grossniklaus HE, Green WR. Pathologic findings in pathologic myopia. Retina 1992;12:127–133. 8. Akiba J, Konno S, Yoshida A. Retinal detachment associated with a macular hole in severely myopic eyes. Am J Ophthalmol 1999;128:654–655. 9. Morita H, Ideta H, Ito K, et al. Causative factors of retinal detachment in macular holes. Retina 1991;11:281–284. 10. Panozzo G, Mercanti A. Optical coherence tomography findings in myopic traction maculopathy. Arch Ophthalmol 2004; 122:1455–1460. 11. Okabe S, Matsuo N, Okamoto S, Kataoka H. Electron microscopic studies on retinochoroidal atrophy in the human eye. Acta Med Okayama 1982;36:11–21. 12. Ohno-Matsui K, Akiba M, Moriyama M, et al. Imaging retrobulbar subarachnoid space around optic nerve by swept-source
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30. 31.
32.
optical coherence tomography in eyes with pathologic myopia. Invest Ophthalmol Vis Sci 2011;52:9644–9650. Spaide RF, Koizumi H, Pozzoni MC. Enhanced depth imaging spectral-domain optical coherence tomography. Am J Ophthalmol 2008;146:496–500. Fujiwara T, Imamura Y, Margolis R, et al. Enhanced depth imaging optical coherence tomography of the choroid in highly myopic eyes. Am J Ophthalmol 2009;148:445–450. Ho J, Branchini L, Regatieri C, et al. Analysis of normal peripapillary choroidal thickness via spectral domain optical coherence tomography. Ophthalmology 2011;118:2001–2007. Manjunath V, Taha M, Fujimoto JG, Duker JS. Choroidal thickness in normal eyes measured using Cirrus HD optical coherence tomography. Am J Ophthalmol 2010;150:325–329 e1. Nishida Y, Fujiwara T, Imamura Y, et al. Choroidal thickness and visual acuity in highly myopic eyes. Retina 2012;32: 1229–1236. Imamura Y, Iida T, Maruko I, et al. Enhanced depth imaging optical coherence tomography of the sclera in dome-shaped macula. Am J Ophthalmol 2011;151:297–302. Maruko I, Iida T, Sugano Y, et al. Morphologic analysis in pathologic myopia using high-penetration optical coherence tomography. Invest Ophthalmol Vis Sci 2012;53:3834–3838. Ikuno Y, Tano Y. Retinal and choroidal biometry in highly myopic eyes with spectral-domain optical coherence tomography. Invest Ophthalmol Vis Sci 2009;50:3876–3880. Ellabban AA, Tsujikawa A, Matsumoto A, et al. Macular choroidal thickness measured by swept source optical coherence tomography in eyes with inferior posterior staphyloma. Invest Ophthalmol Vis Sci 2012;53:7735–7745. Chhablani J, Barteselli G, Wang H, et al. Repeatability and reproducibility of manual choroidal volume measurements using enhanced depth imaging optical coherence tomography. Invest Ophthalmol Vis Sci 2012;53:2274–2280. Chhablani J, Barteselli G, Bartsch DU, et al. Influence of scanning density on macular choroidal volume measurement using spectral-domain optical coherence tomography. Graefes Arch Clin Exp Ophthalmol 2013;251:1303–1309. Barteselli G, Chhablani J, El-Emam S, et al. Choroidal volume variations with age, axial length, and sex in healthy subjects: a three-dimensional analysis. Ophthalmology 2012; 119:2572–2578. Llorente L, Barbero S, Cano D, et al. Myopic versus hyperopic eyes: axial length, corneal shape and optical aberrations. J Vis 2004;4:288–298. Flores-Moreno I, Lugo F, Duker JS, Ruiz-Moreno JM. The relationship between axial length and choroidal thickness in eyes with high myopia. Am J Ophthalmol 2013;155:314–319 e1. Linsenmeier RA, Padnick-Silver L. Metabolic dependence of photoreceptors on the choroid in the normal and detached retina. Invest Ophthalmol Vis Sci 2000;41:3117–3123. VanderBeek BL, Johnson MW. The diversity of traction mechanisms in myopic traction maculopathy. Am J Ophthalmol 2012;153:93–102. Noble KG, Carr RE. Pathologic myopia. Ophthalmology 1982; 89:1099–1100. Klein RM, Green S. The development of lacquer cracks in pathologic myopia. Am J Ophthalmol 1988;106:282–285. Wang NK, Lai CC, Chou CL, et al. Choroidal thickness and biometric markers for the screening of lacquer cracks in patients with high myopia. PloS One 2013;8:e53660. Ikuno Y, Jo Y, Hamasaki T, Tano Y. Ocular risk factors for choroidal neovascularization in pathologic myopia. Invest Ophthalmol Vis Sci 2010;51:3721–3725.
CHOROIDAL VOLUME IN HIGHLY MYOPIC EYES BARTESELLI ET AL 33. Maruko I, Iida T, Sugano Y, et al. Subfoveal retinal and choroidal thickness after verteporfin photodynamic therapy for polypoidal choroidal vasculopathy. Am J Ophthalmol 2011; 151:594–603 e1. 34. Yamazaki T, Koizumi H, Yamagishi T, Kinoshita S. Subfoveal choroidal thickness after ranibizumab therapy for neovascular age-related macular degeneration: 12-month results. Ophthalmology 2012;119:1621–1627. 35. Sayanagi K, Jo Y, Ikuno Y. Transient choroidal thinning after intravitreal bevacizumab injection for myopic choroidal neovascularization. J Clinic Experiment Ophthalmol 2011;2:6. 36. Li XQ, Larsen M, Munch IC. Subfoveal choroidal thickness in relation to sex and axial length in 93 Danish
37.
38.
39. 40.
889
university students. Invest Ophthalmol Vis Sci 2011;52: 8438–8441. Centofanti M, Bonini S, Manni G, et al. Do sex and hormonal status influence choroidal circulation? Br J Ophthalmol 2000; 84:786–787. Kavroulaki D, Gugleta K, Kochkorov A, et al. Influence of gender and menopausal status on peripheral and choroidal circulation. Acta Ophthalmol 2010;88:850–853. Hayreh SS. The ophthalmic artery: III. Branches. Br J Ophthalmol 1962;46:212–247. Odell D, Dubis AM, Lever JF, et al. Assessing errors inherent in OCT-derived macular thickness maps. J Ophthalmol 2011; 2011:692574.
