Background/aims To determine the topographic variation of macular choroidal and retinal thicknesses (RTs) in normal eyes and their relationship with refractive error.
Methods Spectral domain optical coherence tomography with enhanced depth imaging was performed on 124 healthy participants using a standardised imaging protocol. Manual segmentation of choroidal boundaries was performed by trained graders, and mean choroidal thickness (CT) was compared with mean RT in corresponding sectors of the Early Treatment Diabetic Retinopathy Study (ETDRS) grid.
Results Mean central subfield CT was 322.2 µm. The choroid was thickest at the temporal and superior sectors (323.1–338.1 µm), followed by inferior sectors (314.0–321.8 µm), and thinnest at the nasal sectors (232.8–287.8 µm). In contrast, the retina was thicker nasally (343.4 µm) and thinner temporally (287.1 µm). CT was thickest among emmetropes in all ETDRS subfields and became thinner progressively among low, moderate and high myopes (p<0.001). The variation of both choroidal and RTs among refractive error groups resulted in different topographic patterns at the macula.
Conclusion There is significant topographic variation of choroidal and RTs at different regions of the macula, with progressive change of choroidal thickness in all sectors based on the refractive status of the eye.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The choroid plays an important role in the physiology of the eye. Besides providing metabolic support to the retinal pigment epithelium (RPE) and outer retina, the choroid has been shown to regulate ocular growth1 and may play a role in the development of refractive error.2 Advances in spectral domain optical coherence tomography (SD-OCT) have enabled clinicians to visualise the structures within the retina and choroid in unprecedented detail.
Recent reports suggest that the choroid may be involved in the pathogenesis of a variety of ocular diseases, with thinning of the choroid in myopia,3 ,4 age-related macular degeneration5 and diabetic retinopathy.6 In contrast, the choroid has been reported to be significantly thicker in central serous chorioretinopathy7 and Vogt–Koyanagi–Harada (VKH) disease.8 It is possible that variations in choroidal thickness may indicate the presence of a disease, signal its progression or provide insights into its prognosis. For example, studies have reported that in eyes with central serous chorioretinopathy, choroidal thickness is significantly increased compared with normal contralateral eyes,7 and the thickness reduces following treatment.
In order to fully appreciate the significance of variations in choroidal thicknesses in ocular disease, it is essential to obtain normative values for choroidal thicknesses among different populations and demographics. Investigators have reported a range of choroidal thicknesses, with mean subfoveal choroidal thickness ranging from 283.7 µm to 354 µm.9–12 In many studies, however, choroidal thickness was measured only at one location at the foveal centre or at specific distances from the fovea along a single horizontal or vertical B-scan.9–14 However, the choroid is a three-dimensional vascular structure consisting of a highly anastomosed network of blood vessels. Measurement of point thicknesses, therefore, provides limited information on the actual topographical variation of the choroid at the macula.
Another key gap in current knowledge is the relationship between choroidal and retinal thicknesses (RTs) at different regions of the macula and whether the topographic variation of the choroid differs from that of the retina. While a few studies have reported considerable spatial variation of sectoral choroidal thickness throughout the posterior pole,15–19 they did not report the corresponding RTs. Differences in the patterns of topographic variation of choroidal and RTs, if they exist, may provide insights into the interaction between the retina and the choroid in the normal functioning of the eye and how the choroid influences the development of myopia and other ocular conditions. For example, the retina is reported to be thinner in the temporal sectors, whereas separate studies on choroidal thickness have reported that the choroid is thinnest nasally and relatively thick temporally.15–,19 If choroidal and RTs can be compared in the same population, it will help us understand the complex topographical relationship between these two structures and ascertain the influence of refractive error. Studying this relationship in healthy, young individuals will also serve as a baseline for comparisons with the retinal and choroidal thicknesses in individuals with pathologic myopia.
Our objectives were to describe the topographic variation of choroidal and RTs at the macula in normal healthy eyes, to evaluate differences in the topographic variation between the choroid and retina, and the relationship of these parameters with refractive status.
This was a prospective cross-sectional study of 124 healthy volunteers of Chinese ethnicity with no ocular disease performed by the Ophthalmic Imaging Research Group at the National Healthcare Group Eye Institute, Tan Tock Seng Hospital, Singapore. The study was approved by the Institutional Review Board of the National Healthcare Group and conformed to the tenets of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Ocular imaging protocol
A standardised imaging protocol was used to obtain the OCT scans. SD-OCT with enhanced depth imaging (EDI) was performed on both eyes using the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). Using the EDI technique,12 the objective lens is pushed closer to the eye in order to obtain an inverted image. As a result, the deeper structures, including the choroid–scleral junction, are visualised more distinctly by placing them closer to the zero delay line.
All OCT scans were performed by the same experienced operator under standardised mesopic lighting conditions. A 31 horizontal line raster scan (30o×25o, 9.2 mm×7.6 mm) centred on the fovea was performed, with 25 frames averaged in each OCT B-scan to improve the image quality. All OCT scans were reviewed by a fellowship-trained retinal specialist (CSHT) to ensure that the scans were of sufficient clarity to adequately visualise the choroid–scleral boundary on every B-scan. If the scans were of insufficient quality, they were immediately repeated.
In order to account for the effects of diurnal variation of choroidal thickness,20–22 all scans were performed within a 2-hour period from 12 PM to 2 PM.
Refractive error and keratometry were measured using the Canon RK-F1 full autorefractor-keratometer (Canon Inc, Tokyo, Japan).
Manual segmentation of the choroid
Manual segmentation of the choroidal boundaries for all 31 OCT sections was independently performed by two trained graders using the Heidelberg Eye Explorer software (V.18.104.22.168). The lower segmentation line (originally drawn automatically at the lower border of the RPE) was moved to the choroid–scleral junction (figure 1). In regions where the boundary was indistinct or uncertain, reference was made to the adjacent areas of the choroid–scleral interface as well as to OCT B-scans superior and inferior to that line to give an indication of the variation of the choroidal topography both horizontally and vertically in that region. Subsequently, the upper segmentation line (corresponding to the internal limiting membrane originally) was moved down to the lower border of the RPE.
After ensuring that the scan was centred on the fovea, the mean choroidal thicknesses in all sectors of the Early Treatment Diabetic Retinopathy Study (ETDRS) grid were calculated by the software and displayed. Corresponding RTs were obtained from the original (unadjusted) OCT scans.
Statistical analysis was performed using SPSS V.16.0 (SPSS Inc, Chicago, USA). The differences between the refractive error groups were analysed using analysis of variance (ANOVA) with Bonferroni correction. The choroidal and RTs in the same ETDRS subfield were compared using paired t tests. Intraclass correlation (ICC) was used to assess the agreement between the two graders for choroidal thickness measurements.
The mean age of the 124 participants was 23.0 years (range 21–33 years, SD±1.9), with 68 males and 56 females. All participants were of Chinese ethnicity. The mean spherical equivalent was −4.02 D (range −10.0 D to +0.5 D, SD±2.7). Among the 124 participants, 38 (30.6%) had high myopia (≤−6.0D), 38 (30.6%) had moderate myopia (−3.00D to >−6.0D), 37 (29.8%) had low myopia (−0.5D to >−3.0D) and 11 (8.9%) were emmetropic (+0.5D to >−0.5D). Comparing the four refractive error groups, the participants were comparable in terms of mean age (ANOVA p=0.686) and gender ratio (p=0.601). Due to the strong correlation of both spherical equivalent (R = 0.931, p<0.001) and central subfield choroidal thickness (R = 0.880, p<0.001), only data from a single eye of each participant are reported.
The choroidal thickness measurements obtained by the two graders were very consistent and repeatable, with intraclass correlation coefficient of 0.989.
The mean central subfield choroidal thickness (central 1 mm of the ETDRS grid) was 322.2 µm (range 123–566 µm, SD±98.2). A distinct topographic variation of the choroid was observed (figure 2): for both the inner and outer rings of the ETDRS grid, the choroid was thickest in the temporal and superior sectors of the ETDRS grid (ranging from 323.1 µm to 338.1 µm), followed by the inferior sectors (321.8 µm and 314.0 µm, respectively), and was thinnest at the nasal sectors (287.8 µm and 232.8 µm, respectively).
In contrast, the retina demonstrated a distinctly different pattern of topographic variation (figure 2). In the inner sectors of the ETDRS grid, the retina was thinnest at the temporal sector (332.7 µm) and was of similar thickness in the superior, inferior and nasal sectors (340.2–343.8 µm). In the outer sectors, the retina was thickest nasally (316.4 µm), followed by the superior sector (303.0 µm), and was of similar thickness in the temporal and inferior sectors (287.1 µm and 287.4 µm, respectively).
Comparing choroidal and RT in patients with different refractive errors
The choroidal thickness was thickest among emmetropes in all ETDRS subfields and reduced progressively among the low myopes, moderate myopes and high myopes (figure 3). The respective differences in choroidal thicknesses between all groups were significant (p<0.01).
Analysing for RT, while there was a variation among the refractive error groups, with the retina generally being thinnest in the high myopes and thickest in the emmetropes, this variation was smaller, with the differences between the thinnest and thickest groups ranging from 12.5 µm to 22.9 µm.
The patterns of topographic variation between choroidal and RTs varied between refractive error groups, resulting in different patterns (figure 4). In the central subfield, the mean choroidal thickness was greater than RT for emmetropes, low and moderate myopes (all p<0.05), while the choroid and retina were of similar thickness in high myopes. Among emmetropes, choroidal thickness was greater than RT in all sectors (p<0.001), except the outer nasal sector, where they were of similar thickness (figure 4A). Participants with low myopia (figure 4B) generally had similar choroidal and RTs in the inner subfields (all p>0.05), whereas the choroid was thicker in the outer superior, outer temporal and outer inferior subfields (all p<0.001). Among those with moderate myopia (figure 4C), the choroid was still thicker in the outer superior, outer temporal and outer inferior sectors. However, the retina was significantly thicker in the outer nasal sector and all inner sectors (all p<0.01), except for the inner temporal sector. In contrast, for high myopes (figure 4D), choroidal thickness and RT were similar in the central, outer superior, outer temporal and outer inferior sectors (all p>0.05), while RT was greater than choroidal thickness in all inner sectors and the outer nasal sector (all p<0.001). When both eyes of all participants (248 eyes) were analysed together, the same pattern of results were obtained.
In this study, we have described interesting patterns in the topographic variation of choroidal and RTs at different regions of the macula. Choroidal thickness varies significantly with refractive error (being thickest in emmetropes and thinnest in high myopes), while the variation in RT is much smaller. As a result, choroidal and RTs have different relationships depending on the refractive error of the individual (figure 4).
To our knowledge, the relationships between the thicknesses of the choroid and retina at different regions of the macula and the effect of refractive status on these topographic variations have not been described in detail. An earlier paper23 reported a weak correlation between the point thicknesses of the retina and choroid at the fovea and concluded that RT may not be directly related to choroidal thickness in normal eyes. That paper, however, did not report retina to choroid thickness correlations in other regions of the macula.
Most of the earlier papers describing choroidal thickness in normal individuals were based on point measurements.7 ,9–14 ,24 However, the choroid is a complex three-dimensional vascular structure with potentially large variations in its topography, which will not be adequately assessed by point thickness measurements. Only a few studies have reported topographic variation of choroidal thicknesses in different sectors of the ETDRS grid,15–19 as summarised in table 1.
Our analysis of the patterns of variation of both choroidal and RTs yields interesting observations. Similar to these earlier studies, we noted that the choroid is especially thin nasally.15–19 However, we have further ascertained that the RT in the nasal region was of similar thickness to, or thicker than, other corresponding retinal sectors of the ETDRS grid. A possible explanation is that the blood supply to the outer retina in that region may be supplemented by the posterior ciliary circulation around the optic disc. In contrast, in the temporal sectors, the choroid was generally thicker, whereas the retina is thinner temporally compared with other regions of the macula.
Earlier studies of choroidal thicknesses using ETDRS grids were performed on groups of Japanese,15 ,17 ,19 Korean16 and mixed ethnicities,18 while we are not aware of a similar study among participants of Chinese ethnicity. An earlier study by Tanabe et al15 performed raster OCT scans using the Spectralis OCT. However, that study was performed on the fellow eyes of patients with retinal disease, and the authors acknowledged the possibility of subclinical disease in the eyes studied,15 which may potentially affect the range of choroidal thicknesses reported. In contrast, our participants were normal in both eyes and had no history of ocular diseases. The few other studies reporting sectoral choroidal thicknesses used other OCT devices and scanning protocols,16–19 including radial scans,16 which would require interpolation of choroidal thicknesses between these scans.
In this study, we found differences in the pattern of topographic variation of choroidal and RTs among emmetropes, low myopes, moderate myopes and high myopes. These different patterns appear to arise mainly from differences in choroidal thickness among eyes with varying severity of refractive error (figures 3 and 4). Among emmetropes and low myopes, the choroid is generally thicker than, or of equal thickness to, the corresponding retinal sectors. In contrast, among high myopes, the choroid is thinner than, or of equal thickness to, the corresponding RTs in the respective sectors (figure 4). These observations may provide insights into the normal development of the eye and, possibly, the development of refractive error, and the roles played by the interaction of the choroid and retina in these processes. If these variations are indeed related to myopia, it will be interesting to determine whether these variations are the result of myopia development or the cause of it.
A recent study3 reported thicker choroidal measurements in emmetropes and low myopes compared with high myopes. However, only the differences between choroidal point thicknesses were reported in that study and these were not correlated with RTs.
The strengths of this study include the large number of participants with a wide range of refractive errors, thus allowing us to determine the effects of refractive error on topographic variation of both the choroid and the retina. All OCT scans were performed during a standardised time period to minimise the effects of diurnal variation on choroidal thickness measurements. Diurnal variation of choroidal thickness has been demonstrated in normal eyes,20–22 with amplitudes of up to 67 µm reported.20 If not properly accounted for, this may affect the accuracy of assessment of choroidal thickness.
This study is not without limitations. We excluded older participants because choroidal thickness is known to vary with age, and we wished to avoid additional factors that may confound the results. Our objective was to describe the ‘baseline’ choroidal thickness in young adulthood and its variation with refractive error. Subsequent changes of choroidal thickness with age and disease will be further examined in future studies using the current findings as a reference. In addition, imaging of older patients could be affected by possible media opacity and subclinical retinal disease, and they would be required to undergo a more thorough ophthalmic review in order to exclude these conditions. Finally, we did not include hyperopes in this study. The prevalence of hyperopia among the Chinese aged below 30 years is extremely low (as low as 0.7%),25 and it will be difficult to recruit a sufficient number of subjects for useful analysis. However, this can be explored in future studies in populations or age groups where the prevalence of hyperopia is higher.
In summary, we have demonstrated different patterns of topographic variation of the choroidal and RTs among different regions of the macula. These topographic variations differ according to the refractive status of the eye. These findings may add to our understanding of the development and functioning of the eye in health and disease.
The authors would like to thank the participants who volunteered for the OCT scans.
Contributors CSHT, KXC, LWL and KZL are responsible for formulation of the manuscript.
Funding CSHT receives research funding from the National Healthcare Group Clinician Scientist Career Scheme Grant (Code: CSCS/12005). CSHT also receives travel support from Bayer (South East Asia) Pte Ltd. (Code: R). KXC, LWL and KZL receive no funding.
Competing interests None.
Patient consent Obtained.
Ethics approval The study was approved by the Institutional Review Board of the National Healthcare Group, Singapore, and conformed to the tenets of the Declaration of Helsinki.
Provenance and peer review Not commissioned; externally peer reviewed.