Grid-based analysis of three-dimensional choroidal Sattler’s and Haller’s layer vascularity volume index and its relation to age and axial length in adults by ultra-widefield swept-source optical coherence tomography angiography Authors: Yuyao Qu, Jixian Ma, Rong Di, Yue Bian, Yuchen Li, Hanze Zhang, Yazhou Qin, Qiuping Liu, Jianqin Lei, Jingming Li Published: 2024 ABSTRACT BackgroundThis study was to evaluate the features of choroidal Sattler’s and Haller’s layer (CSHL) vessels in adults and explore its relationship with factors such as age, axial length (AL) and body mass index (BMI) by using ultra-widefield swept-source optical coherence tomography angiography (ultra-widefield SS-OCTA).Methods24mm × 20mm fovea centered rectangular SS-OCTA scans were acquired. They were divided into 9 identical grids, with each equal size of 8 mm × 6.7 mm. OCTA artifacts were evaluated. Three-dimensional choroidal vascularity volume index (CVI) and choroidal vascular volume per unit area (CVV/a or CVV) of CSHL were calculated. The correlations of CVI and CVV with age, AL, and BMI were analyzed.ResultsArtifacts account for 19.30%, 18.60%, 30.23% of three inferior grids, respectively. CVI and CVV of left eyes were 30.82 ± 0.28% and 76.76 ± 2.19μm, respectively. Similary, CVI and CVV of right eyes were 31.16 ± 0.34% and 80.49 ± 3.10μm. Strong relationships of CVI (0.877, P = 0.000) and CVV (0.866, P = 0.000) were found between the two eyes in the same individual. The covariance of CVI is lower than that of CVV. Higher CVI was associated with younger age and shorter AL. Neither CVI nor CVV has association with BMI.ConclusionsCVI can be a more stable parameter for describing choroidal medium and large-sized vessels compared to CVV. Aging and longer AL were associated with attenuated CSHL’s vascularity. Taken together, these findings may provide helpful and referential CSHL vascular parameters for retinal specialists to reveal risk factors of choroidal diseases. Introduction The choroid is a highly vascularized structure in the eye that lies between the retinal pigment epithelium (RPE)-Bruch’s membrane complex and the sclera [1]. It participates in the regulation of blood vessel formation and regulate the imaging process of the eye [2]. Therefore, alterations in choroidal blood flow may be implicated in a variety of ocular diseases, including age-related macular degeneration (AMD), diabetic retinopathy (DR) [3], central serous choroidal retinopathy (CSC) [4], polypoid choroidal vasculopathy (PCV) [5], etc. Thus, knowing more about the choroidal blood flow status becomes imminent. Choroidal vascular structure has three main layers. The innermost is the choriocapillaris layer. The outer layer consists of medium and large-sized vessels, known as choroidal Sattler's layer and Haller's layer (CSHL) [6]. Currently, fundus blood flow could be evaluated by fundus fluorescein angiography (FA), indocyanine green angiography (ICGA), and optical coherence tomography angiography (OCTA) [7]. Compared with FA and ICGA, OCTA has the advantages of non-invasiveness, high resolution, less time consuming, and good repeatability. Importantly, it also allows for assessing the volume of the choroidal drainage system [8] Recently, it has been widely used in the cross-sectional examination of fundus structure and quantification of the retinal and inner choroidal blood flow [9, 10]. Due to the light scattering and light absorption of RPE and inner choroid, detecting and imaging deeper structure of the choroidal vasculature once remains a challenge till swept-source (SS)-OCTA has achieved better accuracy and deeper structural imaging [11]. However, earlier SS-OCT can reach only 2.9 mm of the depth of field (DOF) and 12 mm of width of field (WOF), it was difficult to observe deeper and wider structures including the CSHL, especially that of peripheral region. Recently, ultra-widefield SS-OCTA has proven useful to further explore the choroidal vascular structures, including the vortex veins, by enabling DOF of 6 mm and achieving WOF of 24 mm [12, 13]. Several parameters, such as Choroidal thickness (CT), choroidal volume (CV), choroidal vascularity volume index (CVI), and choroidal vascular volume per unit area (CVV/a or CVV), are used for assessing choroidal vascular structure [14]. In the present study, we utilized two parameters three-dimensional CVI and CVV to evaluated CSHL vascularity at both central and peripheral regions in adults. Three-dimensional CVI is defined as the ratio of vascular volume to the total choroidal volume, which is presented as a percentage (%). Moreover, three-dimensional CVV is defined as the ratio of CSHL vascular volume to the same area, which is presented as μm. The structure changes of the choroid have been reported to be associated with aging, myopia, and obesity [15–17]. Previous studies preferentially focused on choroidal capillaries within the central choroid zone due to the limitations of OCTA scan technique. In this regards, we herein not only used the latest high-speed ultra-widefield SS-OCTA device to obtain normal adults’ choroidal images with VOF of 120° at a size of 24 mm × 20 mm to analyze the blood flow of CSHL, but also tried to find the characteristics of CSHL in different subfields among population, and its relation to age, axial length (AL), and body mass index (BMI). Materials and methods Study subjects This cross-sectional, observational study was conducted at First Affiliated Hospital of Xi’an Jiaotong University. Informed consent was obtained from all individual participants included in the study. All procedures performed in the study were in accordance with the Helsinki declaration. The research protocol was reviewed and approved by the Institutional Review Board (IRB) of Biomedical Ethics Review Committee of Xi’an Jiaotong University, with the approval number 2022–138. We recruited participants by publicizing and explaining the purpose of the research, and initially screened out participants by filling out questionnaires online. Participants eligible for the research design then underwent blood pressure measurement and comprehensive ophthalmic examinations including best-corrected visual acuity (BCVA, decimal), intraocular pressure (IOP, mmHg), axial length (AL, mm) and ultra-widefield SS-OCTA. The inclusion criteria were as follows: 1. Age ≥ 18 years; 2. Able to fill in the online questionnaire; 3. BCVA ≥ 0.8 and IOP within 10–21 mmHg; 4. Without pathologic myopia; 5. Without systemic diseases according to medical records, such as diabetes, hypertension, hyperlipidemia, etc., clinically diagnosed. The exclusion criteria were as follows: 1. Systolic blood pressure > 140 mmHg; 2. Suffering from eye diseases, such as cataract, glaucoma, etc., clinically diagnosed; 3. Undergone eye surgery, such as refractive surgery, retinal laser, etc.; 4. Eyes with trauma; 5. Eyes with severe artifacts in ultra-widefield SS-OCTA images; 6. High myopia with ocular pathological changes. A total of 156 subjects were recruited from August 2022 to December 2022. 131 subjects including 128 right eyes and 124 left eyes were eligible for this study, according to the criteria given above. Examination procedures Online questionnaires were collected, followed by blood pressure measurement to assess the eligibility of the participants. IOP measurements were conducted using full auto tonometer (TX-20, Canon, Tokyo, Japan), AL measurements were achieved by using IOL Mater (Zeiss 700; Carl Zeiss Meditech AG, Jena, Germany). Body weight, height, and BMI were collected through the questionnaires above. Choroidal imaging was performed with ultra-widefield SS-OCTA, as detailed below. The examination time was from 10:00am to 12:00 pm and 14:00 pm to 17:00 pm. A quiet rest for more than 10 min before measurement was required. Ultra-widefield SS-OCTA imaging and analysis The 24 mm × 20 mm sized fundus images were acquired by ultra-widefield SS-OCTA (TowardPi BMizar; TowardPi Medical Technology, Beijing, China) with a 400 kHz speed. It utilizes a swept-source vertical-cavity surface-emitting laser (VCSEL) with a wavelength of 1060 nm, providing a transverse resolution of 10 µm and in-depth resolution of 3.8 µm in tissue [9, 18]. The fovea centered rectangular SS-OCTA scan with an image size of 24 mm × 20 mm consists of 1536 A-scans and 1280 B-scans was selected, corresponding to a 120° field of view. Typical OCTA artifacts as demonstrated in Fig. 1a, b and c including shadow, movement, and tilt were evaluated [19, 20]. The exclusion criteria of images include: 1. Severe shadow or tilt artifacts, more than 50% of the whole image; 2. Severe movement, including more than 6 moving lines in total and with break of vessel continuity; 3. Images with severe defocus (as detailed below). To control quality, the images were collected by two well-trained technicians and manually reviewed to exclude significant artifacts. The built-in software was utilized to achieve automatic stratification and measurements of the fundus image based on structural OCT volume scans. The software identified the choriocapillaris layer, which was defined as a slab from the basal border of the RPE-Bruch’s membrane complex to 20 μm beneath it [21]. The rest of the choroidal vasculature was CSHL as shown in Fig. 1d. Based on automatic stratification images, we analyzed three-dimensional CVI and CVV of CSHL. The analyzing procedure was primarily conducted by dividing the 24 mm × 20 mm size image into 9 identical grids, with each grid at the equal size of 8 mm × 6.7 mm. Each grid was evaluated by two well-trained technicians to identify whether it has artifact or not. Those girds met any criteria of the following were defined as grids with artifacts: 1. Shadow or tilt, more than 10% of the grid; 2. Movement, more than 2 lines; 3. more than one type of artifact. The overall and within each grid statistics of CVV and CVI as shown in Fig. 1e and 1f were calculated respectively by the built-in software.Fig. 1Ultra-widefield swept-source optical coherence tomography angiography (SS-OCTA) within 24 mm × 20 mm enface images of retina and choroid. (a-c) enface images of retina, showing types of artifacts (three volunteer, right eyes) (a) tilt in inferior and temporal field (b) eyelash in inferior-temporal field (c) movement (d-f) representative choroid images of a 22-year-old healthy male (right eye) (d) enface image of the choroid (e) choroidal vascularity volume index (CVI) statistics in the 9 grids of choroidal Sattler’s and Haller’s layer (CSHL) (f) choroidal vascular volume per unit area (CVV) statistics in the 9 grids of CSHL Ultra-widefield swept-source optical coherence tomography angiography (SS-OCTA) within 24 mm × 20 mm enface images of retina and choroid. (a-c) enface images of retina, showing types of artifacts (three volunteer, right eyes) (a) tilt in inferior and temporal field (b) eyelash in inferior-temporal field (c) movement (d-f) representative choroid images of a 22-year-old healthy male (right eye) (d) enface image of the choroid (e) choroidal vascularity volume index (CVI) statistics in the 9 grids of choroidal Sattler’s and Haller’s layer (CSHL) (f) choroidal vascular volume per unit area (CVV) statistics in the 9 grids of CSHL Statistical analysis The distribution of continuous numerical data was assessed by the Shapiro–Wilk normality test. The means and standard deviations (SDs), as well as medians and ranges, of all continuous variables were calculated. The coefficients of variation (COV) for choroidal metrics were also calculated. The proportion of different type of artifacts were calculated. Generalized linear mixed model (GLMM) was built to calculate the relationship among age, AL, or BMI and CVI or CVV. Results Demographics and ocular biometrics 131 subjects in total were eligible for the study. Demographics and ocular biometrics of all participants are shown in Table 1. The mean age was 38.69 ± 1.01 years. The female subjects were 108 (82.4%). The mean BMI was 22.49 ± 0.21 kg/m2. Of the 128 right eyes, the average value of AL, and IOP were 24.51 ± 0.13 mm, and 14.96 ± 0.20 mmHg, respectively. Of the 124 left eyes, the average value of AL, and IOP were 24.41 ± 0.13 mm, and 15.23 ± 0.20 mmHg, respectively. The S-W normality test showed that age and AL obey the normal distribution (P < 0.05). Table 1Demographics and ocular biometrics of the participantsTotal (n = 131)Mean ± SDMedianRangeGender (female)n = 108 (P = 82.4%)Age, y38.69 ± 1.0133.0020 ~ 77Weight, kg60.51 ± 0.8860.0041 ~ 98Height, cm163.94 ± 0.61163.00152 ~ 192BMI, kg/m222.49 ± 0.2122.1316.41 ~ 27.55AL, mmright eye24.51 ± 0.1324.6321.68 ~ 28.44left eye24.41 ± 0.1324.4321.80 ~ 28.49IOP, mmHgright eye14.96 ± 0.2014.98.9 ~ 20.4left eye15.23 ± 0.2015.09.9 ~ 20.5Abbreviations: SD Standard deviations, BMI Body mass index, AL Axial length, IOP Intraocular pressure Table 1Demographics and ocular biometrics of the participantsTotal (n = 131)Mean ± SDMedianRangeGender (female)n = 108 (P = 82.4%)Age, y38.69 ± 1.0133.0020 ~ 77Weight, kg60.51 ± 0.8860.0041 ~ 98Height, cm163.94 ± 0.61163.00152 ~ 192BMI, kg/m222.49 ± 0.2122.1316.41 ~ 27.55AL, mmright eye24.51 ± 0.1324.6321.68 ~ 28.44left eye24.41 ± 0.1324.4321.80 ~ 28.49IOP, mmHgright eye14.96 ± 0.2014.98.9 ~ 20.4left eye15.23 ± 0.2015.09.9 ~ 20.5Abbreviations: SD Standard deviations, BMI Body mass index, AL Axial length, IOP Intraocular pressure Demographics and ocular biometrics of the participants Abbreviations: SD Standard deviations, BMI Body mass index, AL Axial length, IOP Intraocular pressure Choroidal Characteristics Using the 3 × 3 grid, the 24 mm × 20 mm CSHL enface image was subdivided into 9 identical grids with each equal size of 8 mm × 6.7 mm. Firstly, the type and grid counts of artifacts were calculated and shown in Table 2. Tilt and shadow appeared most frequently (ngrid = 181), accounting for 7.98%, then was artifacts caused by eyelash blocking (ngrid = 144, 6.35%). Movement can also be seen (ngrid = 37, 1.63%). In addition, there were also some grids with more than one type of artifacts (ngrid = 68, 3.00%). Then, the distribution of artifacts among 9 grids were further analyzed. As shown in Table 3, artifacts mostly appeared in the 3 inferior grids, the proportion of variables among all grids with artifacts was also calculated. Table 2The proportion of different type of artifactType of artifactGrid Counts (n)Proportion (%)Tilt and shadow1817.98Eyelash blocking1446.35Movement371.63 ≥ 2 types683.00 Table 2The proportion of different type of artifactType of artifactGrid Counts (n)Proportion (%)Tilt and shadow1817.98Eyelash blocking1446.35Movement371.63 ≥ 2 types683.00 The proportion of different type of artifact Secondly, the mean of CVI and CVV of the CSHL were calculated. Of right eye, the mean of CVI and CVV was 30.82 ± 0.28% and 76.76 ± 2.19 μm, respectively. Of left eye, the mean was 31.16 ± 0.34% and 80.49 ± 3.10 μm. The mean coefficient of variation (COV) of CVI was 0.15 ± 0.01 (right eye) and 0.15 ± 0.01 (left eye). The mean COV of CVV was 0.30 ± 0.01 (right eye) and 0.31 ± 0.01 (left eye). Thus, the COV of CVI were lower than that of CVV. The spearman’s correlation coefficient of CVI and CVV between the right eye and left eye were 0.877 and 0.866, respectively (both P < 0.000), which suggested strong consistency of both eyes in the same individual. Thus, in subsequent analysis, only the right eyes were included unless poor quality of ultra-widefield SS-OCTA images were obtained due to fixation issues or an OCT signal strength < 6, in which case, the left eyes were chosen for analysis instead (n = 13). Table 3Distribution of artifacts and covariance of CVI and CVV variables in different gridsName of variableArtifact Grid Counts and Proportion (n, %)COV of CVI (median centered)*COV of CVV(median centered)*superior-nasal31 (7.21)0.1490.402superior-middle17(3.95)0.1040.349superior-temporal29 (6.74)0.1320.361middle-nasal20 (4.65)0.1520.470middle11 (2.56)0.1090.385middle-temporal29 (6.74)0.1440.412inferior-nasal83 (19.30)0.2330.778inferior-middle80 (18.60)0.1840.566inferior-temporal130 (30.23)0.2000.535Abbreviations: COV Covariance, CVI Choroidal vascularity volume index, CVV Choroidal vascular volume per unit area*Median centered COV was calculated because all variables unsatisfied the normal distribution Table 3Distribution of artifacts and covariance of CVI and CVV variables in different gridsName of variableArtifact Grid Counts and Proportion (n, %)COV of CVI (median centered)*COV of CVV(median centered)*superior-nasal31 (7.21)0.1490.402superior-middle17(3.95)0.1040.349superior-temporal29 (6.74)0.1320.361middle-nasal20 (4.65)0.1520.470middle11 (2.56)0.1090.385middle-temporal29 (6.74)0.1440.412inferior-nasal83 (19.30)0.2330.778inferior-middle80 (18.60)0.1840.566inferior-temporal130 (30.23)0.2000.535Abbreviations: COV Covariance, CVI Choroidal vascularity volume index, CVV Choroidal vascular volume per unit area*Median centered COV was calculated because all variables unsatisfied the normal distribution Distribution of artifacts and covariance of CVI and CVV variables in different grids Abbreviations: COV Covariance, CVI Choroidal vascularity volume index, CVV Choroidal vascular volume per unit area *Median centered COV was calculated because all variables unsatisfied the normal distribution Finally, the COV and mean of CVI and CVV among 9 grids were further demonstrated. Neither CVI nor CVV in the 9 grids followed normal distribution (P < 0.05). As shown in Table 3, the distribution of COV for CVI and CVV among 9 grids were consistent. The highest was in inferior-nasal grid, while the lowest was in superior-middle grid. The COV of CVV in all the 9 grids were higher than those of CVI. As shown in Fig. 2a, among 9 grids, the fovea was in the middle subfield; while the optic disc was mostly located in the nasal subfield. The middle subfield had the highest mean CVI of 34.54% and the lowest mean CVI of 24.25% was in the inferior-nasal subfield (Fig. 2b). Similarly, the highest mean CVV of 99.62 μm was in the middle grid; while the lowest was in the inferior-nasal grid with the mean of 48.34 μm (Fig. 2c).Fig. 2The division of the choroidal Sattler’s and Haller’s layer (CSHL) and its characteristics (a) the 9 grids of the CSHL enface image (b) the mean of choroidal vascularity volume index (CVI) in 9 grids (c) the mean of choroidal vascular volume per unit area (CVV) in 9 grids The division of the choroidal Sattler’s and Haller’s layer (CSHL) and its characteristics (a) the 9 grids of the CSHL enface image (b) the mean of choroidal vascularity volume index (CVI) in 9 grids (c) the mean of choroidal vascular volume per unit area (CVV) in 9 grids Correlations By building GLMM, some correlations between variables showed statistical significance in Table 4. Age and AL showed a significant correlation with CVI (coefficient = −0.084, P < 0.000 and coefficient = −0.700, P = 0.001, respectively). There was also statistically significant correlation between AL and CVV (coefficient = −3.448, P = 0.049); the correlation was only marginally different but not statistically significant between age and CVV (coefficient = −0.287, P = 0.069). BMI showed no statistically significant correlation to either CVI or CVV. Table 4Coefficients between age, AL or BMI and CVI or CVVCVICVVAge−0.084 (P = 0.000)−0.287 (P = 0.069)AL−0.700 (P = 0.001)−3.448 (P = 0.049)BMI0.029 (P = 0.733)0.341 (P = 0.632)Abbreviations: CVI Choroidal vascularity volume index, CVV Choroidal vascular volume per unit area, AL Axial length, BMI Body mass indexThe bold values denote statistical significance with p < 0.05 Table 4Coefficients between age, AL or BMI and CVI or CVVCVICVVAge−0.084 (P = 0.000)−0.287 (P = 0.069)AL−0.700 (P = 0.001)−3.448 (P = 0.049)BMI0.029 (P = 0.733)0.341 (P = 0.632)Abbreviations: CVI Choroidal vascularity volume index, CVV Choroidal vascular volume per unit area, AL Axial length, BMI Body mass indexThe bold values denote statistical significance with p < 0.05 Coefficients between age, AL or BMI and CVI or CVV Abbreviations: CVI Choroidal vascularity volume index, CVV Choroidal vascular volume per unit area, AL Axial length, BMI Body mass index The bold values denote statistical significance with p < 0.05 Discussion In this study, we used the latest ultra-widefield SS-OCTA to elucidate a detailed comprehensive choroidal medium and large-sized vessels of 120° VOF, further mapping the CSHL vascularity in 9 subfields and investigating correlations between specific choroidal vascular parameters and independent factors including age, AL and BMI. We found that CVI had lower covariance than that of CVV when describing the vascularity of CSHL. Higher CVI was associated with younger age and shorter AL. Moreover, longer AL was associated with attenuated CVV. The choroid supplies nutrients and oxygen to the outer retina and removes waste from the visual cycle [22]. Thus, evaluating the vascularity of the choroid has vital significance. The choroid is categorized into five layers from innermost to outermost: Bruch’s membrane (BM), choriocapillaris (CC), Sattler’s layer, Haller’s layer and the suprachoroidal space. Venous drainage from the choroid is primarily through the four vortex veins. Before ultra-widefield OCTA is put into clinical practice, the vortex vein had never been captured [13]. Based on the structural image of OCT B-scans, OCTA provides clear segmentation slabs of choroid-retinal vascularization as well as CC [10]. Due to the choroidal vessel variation and subtle segmentation inaccuracy, the quantitative evaluation of CSHL in vivo is still a challenge [23]. Retinal specialists have tried hard to visualize and quantify the mysterious CSHL with various algorithms or segmentation methods including global or local thresholding, deep learning, clustering, active contour models, edge detection or machine learning [24, 25]. Ultra-widefield OCTA was firstly mentioned in 2016 [26]. In 2019, the International Widefield Imaging Study Group identified the term ultra-widefield as showing retinal anatomic features anterior to the vortex vein ampullae in all 4 quadrants [13], giving a standard across all retinal imaging methods, including OCTA. Upon using ultra-widefield SS-OCTA, we herein clearly demonstrated the enface images of CSHL with 120° VOF and 24 mm × 20 mm scan size with reflectance-based three-dimensional threshold segmentation algorithm. OCTA is based on the recognition of erythrocyte motion to non-invasively visualize retinal and inner choroidal vessels through repetition of multiple OCT scans and differential analysis of the sequential scans over time. Thus, artifacts in the enface images are inevitable. They originate in relation to image acquisition, intrinsic ocular characteristics, eye motion, image processing and displaying strategies [27]. It is crucial to identify and label different types of artifacts while analyzing OCTA enface images [28]. In this study, we evaluated and marked the artifacts in 9 grids respectively. It has been found that the major type of artifact was tilt and shadow, artifacts mostly appear in the three inferior grids. Inevitably, image artifacts on OCTA are commonly encountered and appear to be more frequent in eyes with pathology and poor visual acuity [29]. Analyzing the characteristics of image artifacts in normal adults can be helpful to evaluate the image quality in pathological situations. Choroidal parameters have been widely used in clinical practice and research to recognize, diagnose, and follow-up ocular conditions. The quantitative parameter of two-dimensional CVI was introduced recently by Agrawal et al. in 2016 and calculated by a binarization algorithm to assess the ratio of the luminal choroidal area to the total choroidal area [30]. Three-dimensional CVI is merely reported. Literature suggests that CVI has less variability compared to other parameters such as choroidal thickness [31]. Moreover, previous studies usually used ETDRS subfields with relatively small area to explore the spatial distribution of choroidal vascularization. We herein demonstrated that three-dimensional CVI of CSHL within 24 mm × 20 mm scanning range is around 30% among the normal adults with a low covariance of 0.15, confirming CVI is relatively stable for evaluating the choroidal vasculature. Though previous study reported the two-dimensional CVI range within the sub-foveal choroidal area [30], the variation of three-dimensional CVI in different subfields of 120° VOF was not fully reported. We found the covariance of CVI showed good stability among overall and 9 subfields. Of note, CVI was negatively correlated with age in the population, suggesting luminal choroidal volume decreased with age. Ruiz-Medrano et al. also reported that two-dimensional CVI was statistically higher in the < 18-year-old group versus the > 18-year-old group within the central 9 mm choroidal region [32]. Furthermore, significantly correlation suggested longer AL could be a risk factor for attenuated choroidal blood supply [33]. Thus, the contribution of age and AL to CVI variation should not be underestimated. As suggested by investigators, CVI should not be viewed as an isolated marker [34]. Therefore, we used another parameter CVV to further explore CSHL vascular status. In the present study, we found that CVV varied tomographically among 9 individual subfields with the lowest value at the inferior 3 grids. Moreover, CVV was found negatively associated with longer AL. However, unlike CVI, we didn’t get the same statistical significance between CVV and age. CVV or CVV/a, defined as the ratio of CSHL structure vascular volume to the same area, is novel biomarker of choroidal medium and large-sized vessel caliber thickness with minor impact by choroidal matrix. Thus,we speculated this may explain the discrepancy with previous study indicating that the whole CT were decreased with aging [35]. We acknowledge several limitations of the present study. Although we aimed to establish normal reference values for CSHL metrics, the population, as well as the application of different algorithms from different laboratories and manufacturers, made it hard to provide representative meanings. Meanwhile, the statistics of height and weight were obtained from the online questionnaire a few days before participants got their ultra-widefield SS-OCTA images. We gathered our participants mainly via online notices, so gender proportion was probably thus unbalanced. Moreover, the research was conducted cross-sectional, images were obtained once at specific time of a day, diurnal variation of choroidal morphology would be conducted basing on further research. We acknowledge the artifacts of the images reduced the reliability of the statistics in certain regions. Since the present study included normal adults with a mean BMI of 22.49 ± 0.21, the CVI and CVV in patients with metabolic disorders such as diabetes, dyslipidemia, or obesity required further investigation. In addition, the sex was reported as a factor influencing choroidal morphology, which should also be evaluated further [36]. Finally, choroidal vasculature was analyzed only with en-face scans, whereas a B-scan analysis may also provide additional and meaningful information for choroidal vasculature. Taken together, we quantified the vascular tomography of CSHL using ultra-widefield SS-OCTA. In a population of normal Chinese adults, we found CVI more stable than CVV to describe the CSHL vascularity. Longer AL was associated with attenuated choroidal blood supply. The correlation between age and CSHL vascularity was also significant but not as obvious as that of AL. Further investigations will facilitate us to have a better understanding of the physiological and pathological mechanisms responsible for choroidal homeostasis and abnormal conditions.