Macular and choroidal thicknesses in a healthy Hispanic population evaluated by high-definition spectral-domain optical coherence tomography (SD-OCT)

Purpose To report normal values of macular and choroidal thickness obtained from a healthy Hispanic population using Optovue (Optovue Inc, Freemont CA, USA) spectral domain optical coherence tomography (SD-OCT). Design Observational, cross-sectional, correlation study. Methods A total of 290 eyes (145 healthy subjects) were included; 69% of subjects were female. The median age was 39 ± 29 years (IQR), with a range between 18 and 89 years. The study sample was stratified into three age groups: Group 1, 18–40 years (50.3%), Group 2, 41–60 years (30.7%), and Group 3, older than 61 years (19%). Central macular, perifoveal (inner quadrants), and parafoveal (outer quadrants) thicknesses were estimated. In addition, central and peripheral choroidal thicknesses were estimated. Data analysis was performed to calculate the standardized mean difference according to the variance (Student’s t-test) and its differences with Epidat 4.1. Results Median macular central thickness was 250 ±30 µm (IQR) with Optovue. Median central choroidal thickness was 263 ± 48 µm (IQR). Median central choroidal thickness was greater than mean peripheral thickness. Macular evaluation showed a statistically significant difference in central, perifoveal, and parafoveal thicknesses, with lower values being recorded for the study sample compared with the manufacturer’s data. Conclusions SD-OCT has become a useful tool to obtain high-resolution images of the macula and choroid. This method allows precise assessment of the retinal and choroidal layers to diagnose and follow up posterior segment diseases. We are reporting normal cut-off values of macular and choroidal thicknesses in healthy Hispanic subjects evaluated with Optovue SD-OCT as new diagnostic normal parameters for research and clinical activities.


Introduction
Since the introduction of Optical Coherence Tomography (OCT) in the 20th century, it has been possible to assess retinal and choroidal diseases more precisely.
OCT is a non-invasive transpupillary method aimed with a laser system to obtain accurate measurements in vivo of the retina and choroid layers [1]. These devices use two different technologies for imaging, known as time domain (TD) and spectral domain (SD). Spectral Domain OCT, also known as Fourier-domain OCT, acquires images 100 times faster than TD technology, giving higher image resolution [2,3].

Open Access
International Journal of Retina and Vitreous *Correspondence: fjrodriguez@fon.org.co 1 Fundación Oftalmológica Nacional, Calle 50 # 13-50, Bogotá, Bogota, Colombia Full list of author information is available at the end of the article Page 2 of 10 Cortés et al. Int J Retin Vitr (2020) 6:66 One of the most significant contributions of OCT is the quantitative assessment of retinal and choroidal thickness [1]. Some reports suggest that these measurements vary according to age and ethnicity. Therefore, all these thickness measurements and their respective variations in data obtained through OCT must be standardized to demographic data from healthy subjects belonging to various age and ethnic groups [2]. Macular thickness variations are commonly seen in eyes with retinal diseases such as macular edema, age-related macular degeneration, diabetic retinopathy, vascular occlusions, uveitis, and macular atrophy [4][5][6][7].
Published evidence confirms that thinning or thickening of the macula is well correlated with visual function [8]. Achieving normal macular and choroidal thickness values provides a parameter to evaluate patients with posterior segment diseases and becomes a benchmark for clinical and research activities. However, there is limited information from different ethnic groups, especially the Hispanic population, from which to obtain a reliable parameter to compare findings across Latin America with confidence [9][10][11][12][13]. Current clinical practice demands knowledge of normal values of macular and choroidal thickness in the Hispanic population for comparison with normal cut-off values included by manufacturers as a reference in technology manuals. Current normal SD-OCT cut-off values in Optovue have been developed by including data from different ethnic groups, with a small proportion of the sample representing the Hispanic population, which could induce a classification bias for this ethnic group. Of the 480 subjects enrolled in Optovue's normal cut-off value study, 33% were Caucasian, 22% Asian, 20% African-American, 12% Hispanic, 12% Indian, and 1% comprised other ethnic groups [14]. Therefore, due to the absence of valid information regarding macular and choroidal thicknesses in the Hispanic population, our study aims to obtain this information from subjects evaluated at the Fundación Oftalmológica Nacional, in Bogotá Colombia.

Methods
A cross-sectional and correlation study was carried out including 290 eyes from 145 healthy Hispanic subjects that were evaluated at the Fundación Oftalmológica Nacional in Bogotá, Colombia. The global median (± IQR) age was 39 ± 29 and 69% were female (Table 1). All subjects underwent a complete ophthalmological exam, including refractive error determined by an autorefractor. OCT scans were performed with Optovue. The obtained values were compared with the manufacturer's normal values.
The ethnic category, Hispanics, as defined by the Office of Management and Budget (OMB) in 1978, refers to persons or descendants of people from Latin American countries or other Spanish cultures. Under this definition Hispanics are culturally and genetically a heterogeneous group. In Latin America, each country has its own demographic and genetic structure, with its own distinct migration history between regions. All Hispanics are basically trihybrid, their ancestral populations being European, African, and Native American [15]. For this study we included Hispanics who had at least two generations of Hispanic ancestors.
Healthy subjects were included who met the following criteria: age greater than 18 years, written informed consent to participate in the study, and visual acuity of 20/20 in all eyes included in the study. The exclusion criteria were myopia greater than − 5.00D, hyperopia greater than + 5.00D, diagnosis of glaucoma, history of eye diseases (retinal detachment, age-related macular degeneration, history of venous or arterial occlusions, retinal dystrophies, central serous chorioretinopathy, uveitis, intraocular tumors), systemic diseases (diabetes mellitus, high blood pressure), history of eye surgery (vitreoretinal surgery, intravitreal injections, complicated cataract surgery), presence of degenerative neurological diseases, and poor image quality taken by the Optovue.
Images were obtained using six radial macular probes centered on the fovea with equal angular distance and 20 tracking lines spaced by 200 µm, to achieve an axial and transverse resolution of 7-10 µm, respectively. The crosssectional images were analyzed with a software program that automatically performs segmentation of the two edges on each OCT scan, one at the vitreoretinal interface and the other one in the retinal pigment epithelium (RPE), Bruch's membrane complex.

Study sample
Simple randomized sampling was conducted for gender and age. The study sample was stratified into three groups by age: Group 1 from 18 to 40 years old, Group 2 from 41 to 60, and Group 3, which included subjects older than 61 years. Sample size was calculated assuming a 50% proportion of normal patients attending the comprehensive eye clinic at Fundación Oftalmológica Nacional with a 95% confidence level and an absolute precision of 5%, reaching an estimated sample of 369 eyes. The total number of eyes included in the study was 290.

Data analysis
A univariate analysis was performed for quantitative variables such as age and retinal and choroid thicknesses. A global analysis was performed and stratified by gender and age using a Student's t-test for a standardized mean difference according to variance. The Epidat 4.1 statistical package was used for analysis.

Ethical considerations
The study was conducted in accordance with the tenets of the Declaration of Helsinki and National regulations. All patients signed a statement of informed consent before enrollment, and all procedures were reviewed and approved by the appropriate institutional review boards and ethics committees.
Thinner parafoveal areas (external macular) were observed in the study sample. To assess whether or not there was a difference between the normal reported values for the equipment and the studied population, a standardized difference of means was used, for a confidence level of 95% using the Student's t-test.
Statistically significant differences were found in central macular thickness, internal macular thickness (perifoveal), and external macular thickness (parafoveal), with the exception of the external nasal macular thickness, between the normative database reported by the Optovue manufacturer and the population studied. The central retina was significantly thinner in the study sample, with a standard mean deviation (SMD) of 5.59 µm (95%CI 2.01-9.13, p < 0.002). Internal superior macular thickness showed a statistically significant difference of 7.37 µm (SMD, 95%CI 4.60-10.13 p < 0.000), being thinner in study sample. Inner inferior macular thickness showed a statistically significant difference of 7.67 µm (SMD, CI: 95% 5.18-10.15 p < 0.001), being thinner in the study sample. Inner nasal macular thickness showed a statistically significant difference of 7.78 µm (SMD, 95%CI (4.95-10.60, p < 0.001), being thinner in the study sample. Inner temporal macular thickness showed a statistically significant difference of 7.65 µm (SMD, 95%CI 4.81-10.48 p < 0.000), being thinner in the study sample.

Results stratified by age
The study sample was stratified into three age groups. Macular thickness showed variability when all groups were compared. Group 1 (age range 18-40 years) had greater macular thickness values at the central, perifoveal, and parafoveal levels than Group 2 (age range 41 to 60) and Group 3 (age range: older than 61 years).
Group 2 had greater macular thickness values at the central, perifoveal, and parafoveal levels than Group 3 (age range: older than 60 years), evidencing a decrease in macular thickness in elderly patients (Fig. 3).  We included 152 eyes (69.74% from female subjects) with a median (± interquartile range, IQR) age of 29 ± 11 years. The median (± IQR) central macular thickness was 252 ± 32 µm. Regarding inner macular areas, the results showed an inner superior with a mean (± SD) thickness of 318 µm ± 19 µm, an inner inferior with a mean (± SD) thickness of 312 ± 15.41 µm, an inner nasal area with a mean (± SD) thickness of 317 ± 19 µm, and an inner temporal area with a median (± IQR) thickness of 303 ± 23 um.
Ninety eyes from male subjects (31.03%, n = 290) were included with a median (± IQR) age of 44 ± 18 years. Retinal area thicknesses showed a median (± IQR) central thickness of 268 ± 28 um. Regarding inner macular areas, the inner superior showed a median (± IQR) thickness of 344 ±23 um, the inner inferior a mean (± SD) thickness of 340 ±15 um, the inner nasal a mean (± SD) thickness of 347 ± 17 um, and the inner temporal a mean (± SD) thickness of 333 ± 16 um. Regarding outer retinal areas, the outer superior showed a median (± IQR) thickness of 304 ± 20 um, the outer inferior a median (± IQR) thickness of 287 ± 21 um, the outer nasal a mean (SD) thickness of 314 ± 18 um, and the outer temporal a median (IQR) thickness of 290 ± 19 µm (Fig. 4.2). There was no statistically significant difference between the normal value database from Optovue technology and our Hispanic study sample.

Choroid thickness
There were no previous publications regarding normal choroidal thickness values from Optovue technology. In our healthy Hispanic study sample, the median (± IQR) central choroidal thickness found with Optovue was 263 ± 48 um.

Discussion
In recent years, spectral domain optical coherence tomography (SD-OCT) has become a useful tool that provides high-resolution images and valuable information in different diseases of the retina, choroid, and optic     [16]. Subjects included in Group 3 (> 60 years) were less than subjects included in Groups 1 and 2 and showed thinner retinal thickness values in all areas with progressive thinning in retinal layers with age. The central macula was the thinnest area and decreased progressively in the perifoveal and parafoveal areas according to a normal anatomical distribution. This finding was reported previously by Appukuttan et al. [2] and Grover et al. [3] in Indian and Caucasian populations, respectively. Due to the absence of a normal value database stratified by age from the manufacturer, we could not compare our results in age Groups 1, 2, and 3 to either confirm or dismiss differences in macular thickness. Our study, like others before it, found that gender and age are factors that influence macular thickness [16]. Adhi [19]. Both reports concluded that the choroid layer is thicker in the subfoveal area, and the nasal area is thinner than the temporal one as well. The results obtained with Spectralis showed a central choroidal thickness of 287 ± 76 µm (n = 30) and 272 ± 81 µm with Cirrus (n = 34) [18,19]. In our study, according to the findings of Margolis and Manjunath, central or subfoveal thickness in global analysis was the thickest area in this layer, and this measurement was confirmed in age and gender subgroup analysis (263 ± 48 um).
A comparison of our central choroidal thickness results with the literature revealed a thinner central choroid in the Hispanic study sample, a novel finding with high relevance as a new parameter for ocular and cerebral vascular disease evaluation. Nasal and temporal choroidal thicknesses did not show a statistically significant difference in our results.

Conclusions
Finally, our study shared the first normal value database to measure macular and choroidal thicknesses in the Hispanic population using Optovue. This novel dataset will allow a more objective and precise comparison between Hispanic patients in global analysis and adjusted for age and gender, in contrast to values reported previously by manufacturers or clinicians with these technologies based on other ethnic groups. The limitations of the study are mainly related to difficulty reaching the estimated sample size (n = 369) and an imbalance of gender participation including more women than men. Regarding age stratification, Group 1 included more subjects than Group 2 and 3, a risk for selection bias. To the best of our knowledge, this is the first report in the Hispanic population regarding normal cut-off values of retina and choroid thickness using SD-OCT Optovue. These new data give accurate parameters in Hispanics to rule