Choroidal thickness and effect of sex and age in Africans.

The choroid provides nutrients and oxygen to the retinal pigment epithelium (RPE), the avascular fovea, and the prelaminar portion of the optic nerve,[ 1 ] therefore, diseases of the choroid can have considerable effect on these structures. For this reason and more, imaging of the choroid has been given a priority over the years and resulted in the development of enhanced depth (ED) optical coherence tomography (OCT) imaging of the choroid.[ 2 ] Before now, the choroid could not be properly imaged with ease. With the availability of ED-OCT, imaging and study of the choroid in health and in disease has become a reality. Imaging of the choroid provides useful understanding into the pathogenesis of several diseases, including polypoidal choroidal vasculopathy, central serous chorioretinopathy, pachychoroid pigment epitheliopathy, and age-related macular degeneration to mention a few.[ 3 , 4 , 5 , 6 ] ED OCT gives an opportunity for in vivo , easy, reproducible, noninvasive choroidal imaging and measurement of the choroidal thickness (CT).[ 2 ] This has resulted in several publications on the CT in various ethnic populations including Chinese,[ 7 ] Japanese,[ 8 ] Hispanics,[ 9 ] and Indians[ 10 ] as well as the CT in several ocular disease and nondisease states.[ 11 , 12 , 13 , 14 ] Factors that could affect CT include axial length, diurnal changes, refractive error, intraocular pressure, blood pressure, and other ocular and systemic conditions. It has been suggested that CT may be affected also by ethnicity; therefore, Karapetyan et al . compared CT in three different ethnic groups and included 30 Africans among the study population of 88 volunteers who were aged between 20 and 40 years.[ 15 ] Limitations of this study include that a limited number of African eyes, and a narrow age bracket was considered. Therefore, it may not provide information that can be generalized to a broader age group of Africans. It is possible that increased pigmentation in the negroid choroid may result in considerable variation in CT from other ethnic groups.[ 10 ] This study set out to determine the CT in a wider age bracket and larger population of black African volunteers, and to determine the effect of gender and age on CT. It also compared findings with CT in other nonblack ethnic population.

One hundred and seventy-two eyes of 88 volunteers were used for the study analysis. The mean age for all study participants was 53.7 years (SD ± 14.5 years), with ages ranging from 30 to 80 years. The volunteers were categorized based on age groupings as shown in Table 1 and there was no significant difference in the distribution of volunteers between the age groups ( P = 0.548). There were slightly more males (46, 52.3%) than females; but the gender difference was not significant ( P = 0.819). Summary of the age and sex distribution of participants a 1 eye each of 2 participants excluded, b,c1 eye of 1 participant excluded The mean CT in the CSF area and 4 quadrants for all study participants is shown in Table 2 . The mean CT was thickest in the CSF area, mean of 265.4 µm ± 76.2 µm and thinnest in the nasal quadrant with a mean of 199.5 µm ± 69.9 µm. In order of decreasing CT, the CSF was thickest, followed by the inferior, then superior, temporal following and the nasal quadrant was the thinnest. Details of the CT by quadrant are presented in Table 2 . Average quadratic choroidal thickness for (n=172) eyes SD=Standard deviation There was a reduction in mean CT with increasing age, with a statistically significant association between age groups of volunteers and mean CSF CT ( F = 8.039, P < 0.001). The reduction in mean CT with increasing age is shown in Table 3 and Figure 2 . There was a decrease in CT with increasing age in the CSF and all the other four quadrants. The thickest choroid as demonstrated in Table 3 was the CSF choroid in the youngest age group (30–39 years) while the thinnest choroid was the nasal choroid in the oldest age (>70 years age group). Figure 2 shows a diagrammatic representation of the effect of age on CT using bar charts and shows the progressive decrease in CT across all the measured quadrants with increasing age. Mean quadratic choroidal thickness for each age group *The thickest and thinnest choroid. CSF=Central subfoveal A bar chart of age grouping versus mean choroidal thickness. The central sub foveal (CSF) and all 4 quadrants are represented for each age group. There is an appreciable decrease in mean choroidal thickness with increasing age in the CSF and 4 quadrants In the temporal quadrant, the mean CT was 235.5 µm ± 70.3 µm. There was a negative correlation with increasing age as shown in Figure 3 a ( r = −0.43). With increasing age, there was a statistically significant decrease in thickness by 2.13 µm/year (B = −2.13, 95% confidence interval [CI] = −2.8, −1.5, P < 0.001). (a) A scatter plot of the average temporal choroidal thickness with age for all volunteers. (b) A scatter plot of the average nasal choroidal thickness with age for all volunteers. (c) A scatter plot of the average superior choroidal thickness with age for all volunteers. (d) A scatter plot of the average inferior choroidal thickness with the age for all volunteers. (e) A scatter plot of the average central choroidal thickness with age for all volunteers. CT = Choroidal thickness Likewise in the nasal quadrant, the mean CT was 199.5 µm ± 69.9 µm. There was a negative correlation as shown in Figure 3 b ( r = −0.49). With increasing age, there is a decrease in thickness by 2.39 µm/year. This was statistically significant (B = −2.39, 95% CI = −3.0, −1.7, P < 0.001). As in the temporal and nasal quadrants, the superior quadrant with a mean CT of 249.9 µm ± 76.2 µm showed a negative correlation ( r = −0.503). Mean CT decreased by 2.68 µm/year (B = −2.68, 95% CI −3.38, −1.98, P < 0.001). This is shown in Figure 3 c. In the inferior quadrant, the mean CT was 256.2 µm ± 70.6 µm. There was a negative correlation ( r = −0.528), and a statistically significant decrease in thickness by 2.61 µm/year of age (B = −2.61, 95% CI − 3.24, −1.97, P < 0.001). Figure 3 d depicts this. Finally, in the CSF area, in which the mean CT which was 265.4 µm ± 76.2 µm, there was a negative correlation ( r = −0.562), with increasing age. There was a statistically significant decrease in thickness by 3.00 µm per increase in year of age (B = −3.00 95% CI − 3.66, −2.33, P < 0.001). Figure 3 e depicts this. Therefore, the CSF showed the highest decrease in CT per year. Figure 3 a-e shows the temporal, nasal, superior, inferior, and CSF scatter plot images for all the 172 eyes. There were 88 male eyes and 84 female eyes. The mean CSF CT for all male eyes was 266.5 µm ± 77.4 µm while for females, it was 264.4 µm ± 75.4 µm. The difference in mean central CT between males and females was 2.085 µm ( P = 0.858) and not statistically significant. We further investigated the difference in mean CSF CT between premenopausal (45 years), it was noted that there were 28 and 56, premenopausal and postmenopausal eyes, respectively (ratio 1:2). The mean CSF-CT of premenopausal eyes measured 320.8 µm ± 53.7 µm while postmenopausal eyes measured 236.2 µm ± 68.9 µm. The mean difference between these two groups was statistically significant ( P < 0.001). As shown in Figure 4 , the mean CSF CT in eyes of males 45 years. Similarly, a statistically significant difference was also observed between these two groups of male eyes [ Figure 4 ]. A comparison of the CT between males and females >45 years showed that though males had a thicker choroid, there was no statistically significant difference in mean CSF CT between male and female study eyes in participants >45 years of age ( t = 0.660, df = 118, mean difference = 8.33 µm, P = 0.51). A bar chart showing comparison of mean choroidal thickness between eyes of volunteers 45 years, for male and female genders. CT = Choroidal thickness As shown in Table 4 , the mean CSF CT of 265 µm seen in this study of black African eyes is very similar to the 262 µm observed in the Chinese study; but 22 µ thinner than the study reported from New York. The CT of 354 µm reported from the Japanese study is by far the thickest. A comparison of mean sub-foveal choroidal thickness from this study with previous studies NA=Information was not available from the study, CT=Choroidal thickness, CSF=Central subfoveal An age-wise comparison of CSF CT in Africans with CSF CT in India and Chinese is shown in Table 5 . This reveals that the mean CSF CT is within 50 µm of the data from the Chinese and Indian eyes, across all the represented age groups. Age wise comparison of central subfoveal choroidal thickness amongst the Chinese, African and Indian population CSF CT=Central subfoveal choroidal thickness

CT measurements using SD-OCT among Africans appear to agree with findings from other ethnic nationalities,[ 15 ] and measurements are closest to the findings from the Chinese study. The CSF CT is thickest, and the nasal quadrant is always thinnest. This study agrees with other studies on the effect of increasing age by the finding of a significant decrease in CT with age, which was consistent in all the CSF and 4 quadrants measured. There was no gender difference observed in CT. Although CT in postmenopausal women was significantly thinner than in premenopausal women, it is likely that the thin choroid in females >45 years is related to increasing age, and less dependent on reduced estrogen effect. The normative data we have provided on CT in black African eyes can be used for clinical and experimental purposes and certainly can be improved on by increasing the sample size, controlling for axial length, and diurnal variation in future studies. Nil. There are no conflicts of interest.

This study contributes to existing studies on CT in healthy eyes, using the SD-OCT to determine CT. The study of the healthy choroid gives us normative data to compare and identify pathologic variants of choroidal anatomy and structure. The CT in an exclusive group of normal Africans has not been previously reported, though Karapetyan et al .[ 15 ] reported on a group of volunteers in which Africans constituted a subset. We report our findings using healthy volunteers, while eliminating the effect of refractive error, which has been found to significantly affect CT,[ 7 , 9 ] by excluding eyes with significant refractive errors >6 diopters from our study. Furthermore, we tried to obtain a balance across age groups and between the two genders. We also selected wider age groupings compared to Karapetyan's study which considered a narrower age group, to make the sampled population more representative of an adult African population. CT has been found by previous researchers to be thickest in the central area and rapidly thinning out nasal to the fovea.[ 13 ] Choroidal volume has also been found to be lowest in the nasal quadrant.[ 16 ] This study agrees with these previous studies in finding that the CSF CT was the thickest and the nasal quadrant had the thinnest CT. This study also confirms reports that CT varies significantly around the globe and is lowest in the nasal zone as shown in Table 3 . In our observation, the choroid is thickest in the central area, this is an adaptive feature necessary to meet the needs of the metabolically active foveo-macular area, which is rich in cone photoreceptors and retinal pigment epithelial cells. The choroid is, therefore, able to meet the requirements of the outer retina and optic nerve head.[ 17 , 18 ] For this reason, it has the highest blood flow per unit weight of any tissue in the body.[ 18 ] CT has been shown to be affected by age. In our study, we found that the CT significantly reduced with increasing age in not only the CSF but in all 4 quadrants of the eyes studied. The CSF area had the most CT reduction per year of age. Although we recorded a 3.0 µm reduction in CSF-CT per year, other researchers have recorded a much slower rate of reduction of 1.56 and 2.31 µm.[ 13 ] The reason for this difference is not clear. This could suggest that Africans may have a higher rate of choroidal thinning in the CSF area. This observation needs to be further investigated, using a larger sample size of volunteers. While some studies have found a significant effect of gender on CT, others have not and believe that CT is not affected by gender.[ 9 , 19 ] We did not find any significant variation in CT due to gender. We demonstrated a clinically significant difference in CT between postmenopausal (>45 years) and premenopausal (<45 years) females. There have been studies on CT in physiological and pathological states in females; such studies include CT in oral contraceptive pill (OCP) use, in endometriosis and in eclampsia. While some researchers found no effect of OCP on CT, others have reported a reduction in CT after 1 year of OCP use in females of reproductive age.[ 20 ] Estrogen receptors have been reported to occur in the normal choroid. The resultant effect of the stimulation of these receptors is unclear since both conditions of estrogen lack, as seen in postmenopausal women, as well as estrogen stimulation by estrogen containing OCPs, have both been reported to result in thinning of the CSF CT by different researchers.[ 20 ] Considerable controversy therefore appears to exist on the effect of estrogen on the CT. We wanted to investigate the hypothesis that the reduction in CT observed in our study population of >45-year-old women was as a result of the increasing age. Therefore, we performed similar analysis in the male counterpart (>45-year-old males) and found the result to be similar to that in females. Ageing was likely the reason for the thinner CT, rather than estrogen lack from menopause since CT in males (>45 years) was 244.5 ± 69.1 and for females it was 236.2 ± 68.9. Although postmenopausal females had a thinner CT than their male counter parts, this CT difference was not significant and could not be ascribed to be due to gender difference or postmenopausal state ( P = 0.51). Since CT changes with distance from the fovea; some previous studies have considered a wider range of distances from the CSF area.[ 9 , 13 ] We measured the CT at 2.5 mm distance from the CSF area. Other studies measured 1.0 mm, 3.0 mm, and others every 500 µm from the CSF area. This variation amongst studies provides a wide range of distances to compare different studies. On the other hand, it could also make comparison challenging. However, all studies measure the CT in the CSF and this serves as a useful index for comparison as shown in Tables 4 and 5. Another challenge with a comparison of normal CT study values is the use of different OCT machines and software technology in the different studies (industry-related differences in OCT technology, algorithms, and normative database), which can introduce some variation to the comparison. Despite this, it was our goal to compare CT in our study with normal from other nonblack populations. Since this comparison has been scarcely investigated, it is therefore not certain if CT findings from other studies can be directly extrapolated to Africans. The study on ethnicity and CT,[ 15 ] found that cognitive stimulation therapy CT did not significantly correlate with ethnicity. Our study provides data, which can be used in diagnosis, monitoring of disease activity and response to therapy in Africans; but it also compared CT data from other regions of the world as shown in Tables 4 and 5. As shown in Table 4 , the overall mean CSF CT of 262 µ seen in Africans is very similar to the 265 µ in the Chinese. This is much thinner than the CT from the Japanese study at 354 µ. This variation can be accounted for by the much younger age group of participants in the Japanese study compared to the Chinese and African studies. Table 5 gives age-specific comparison, which shows that our African data are within 50 µ of the data from the Chinese and Indian study across all age groups. In the 40–49 years of age group, our data is only 1 µ thicker than the Chinese and 4 µ thinner than the Indian. This similarity with other ethnic nationalities means that previous and future work relating to CT done by these other nationalities could be potentially extrapolated to the black African eye. It has been suggested that the CT also varies with the time of the day and that a person with a normal choroid may manifest differences in thickness at intervals of a few hours or days.[ 21 ] In the light of this information, CT in an individual ought to be measured at a specific time for comparison. We measured CT of volunteers in our study between 10 am and 12 noon daily. An ideal situation will be to take the scans at the same time for all the volunteers. It was not stated in the other CT studies cited by us and used for our comparison if diurnal variation was corrected for. Perhaps the time of CT measurement ought to be documented as is done during intraocular pressure measurement documentation, to enhance future comparison.

We conducted a prospective, single-center cross-sectional nonrandomized study assessing CT in a population of Africans. The study was conducted according to the tenets of the Helsinki declaration. Ethical approval was obtained from the Institutional Research Ethics Committee, with approval number EFHREC/019/011. Informed consent was obtained from all study volunteers before recruiting into the study. All volunteers were informed in detail of the requirements of the study and the process each participant would undergo. All volunteers had a Snellen visual acuity, refraction, intraocular pressure measurement, and anterior segment examination using a slit lamp. Inclusion criteria include healthy adults, 30 years or more in age, no previous ocular surgery, no history of prior ocular trauma, no history of ocular inflammation, no diagnosis of glaucoma, and no retinal or choroidal pathology, refractive error was <6 diopters. A medical history was taken to exclude any systemic disease that could affect the CT. Specifically, blood pressure levels were within normal limits for all study participants. Exclusion criteria include <30 years of age, previous history of ocular trauma or surgery, ocular examination finding of glaucoma or findings suggestive of intraocular inflammation, presence of retinal or choroidal disease, and refractive error > 6 diopters. Corrected Snellen visual acuity was 6/6 (20/20) in all study eyes. Intraocular pressure was <20 mmHg, and anterior segment examination was within normal limits in all eyes. Dilated fundus examination showed no abnormal retinal findings. There was specific attention to exclude presence of glaucomatous optic neuropathy, retinovascular disease as diabetic and hypertensive retinopathy, macular degeneration, choroidal pathology including focal or generalized chorioretinal disease and any other abnormality of the retina or choroid. A normal retina was confirmed using the spectral-domain (SD) OCT enface retinal image appearance which was performed before the choroidal scan. Two trained technicians performed the choroidal scans using the Optovue Avanti RTVue XR OCT machine, version 2018.0.0.18 (Optovue Inc, Fremont, CA, USA) for choroidal imaging in all eyes. Repeated A-scans at a rate of 70,000 A-scans per second were used to obtain composite B scan images. The machine uses an 840 nm wavelength beam and achieves an axial resolution of 5 μm in tissue (3 μm digital) and transverse resolution of 15 μm. Eye tracking was activated. The RTVue system captures two consecutive B-scans (M-B frames) each containing 304 A-scans (304 B-scan locations, each separated by 9.9 μm). For each eye, a 6 mm × 6 mm OCT scan centered on the fovea and a cross sectional image of the macula was taken as horizontal and vertical cross-line scans. The quality of all the images was assessed; and a scan quality index <50 was excluded from the study (manufacturers signal quality range 0–100) [ Figure 1 ]. Cross line optical coherence tomography scan and fundus pictures of the right eye of a volunteer, showing the cross sectional vertical and horizontal scans as well as the image viewer caliper markings highlighting the choroidal thickness and dimensions in green Two of the authors reviewed all the choroidal scans together and agreed on the points to place the calipers for CT measurement to be taken. For the CT measurement of each eye, the inbuilt horizontal calipers on the software image viewer were placed from the center of the fovea depression to 2.5 mm outward in the superior, inferior, temporal, and nasal quadrants respectively [ Figure 1 ]. The vertical calipers were placed at this 2.5 mm mark from the inferior margin of the hyperreflective RPE layer and onto the choroidoscleral junction. The vertical distance between the hyperreflective RPE layer and the choroidoscleral junction was measured using same software caliper on the OCT image viewer. Both authors had to agree on the exact line marking the choroidoscleral interface for each scan. The measures were documented using an excel spread sheet format. CT therefore was measured at the central subfoveal (CSF) area and 2.5 mm nasal, temporal, superior, and inferior to the fovea respectively [ Figure 1 ]. Eighty-eight volunteers had both eyes scanned. Of the 176 eyes scanned, images from 4 eyes had a scan quality <50 or were decentered (entire extent of choroid could not be easily assessed) and were removed from the study. The remaining 172 eye scan measurements were analyzed. The analysis was done using IBM SPSS Statistics version 22 (IBM Corp. Armonk, NY, USA). Categorical variables were expressed as frequency and percentages. CT measurements were expressed as means and standard deviation. Simple linear regression analysis was used to determine the effect of age on CT measured in the CSF, nasal, temporal, superior, and inferior quadrants. Analysis of variance was used to compare mean CT and age groups of participants. Independent samples t -test was done to compare between mean CT and gender. Statistically significant level was set at P < 0.05.