Characteristics of the Ciliary Body in Healthy Chinese Subjects Evaluated by Radial and Transverse Imaging of Ultrasound Biometric Microscopy

Background: The imaging and analysis of the ciliary body (CB) are valuable in many potential clinical applications. This study aims to demonstrate the anatomy characteristics of CB using radial and transverse imaging of ultrasound biometric microscopy (UBM) in healthy Chinese subjects, and to explore the determining factors. Methods: Fifty-four eyes of 30 healthy Chinese subjects were evaluated. Clinical data, including age, body mass index (BMI), intraocular pressure (IOP), axial length (AL), and lens thickness (LT), were collected. Radial and transverse UBM measurements of the ciliary body were performed. Anterior chamber depth (ACD), ciliary sulcus diameter (CSD), ciliary process length (CPL), ciliary process density (CPD), ciliary process area (CPA), ciliary muscle area (CMA), ciliary body area (CBA), ciliary body thickness (CBT0, CBT1, and CBTmax), anterior placement of ciliary body (APCB), and trabecular-ciliary angle (TCA) of four (superior, nasal, inferior, and temporal) quadrants were measured. Results: The average CPL was 0.513 ± 0.074 mm, and the average CPA was 0.890 ± 0.141 mm2. CPL and CPA tended to be longer and larger in the superior quadrant (p < 0.001) than in the other three quadrants. Average CPL was significantly correlated with AL (r = 0.535, p < 0.001), ACD (r = 0.511, p < 0.001), and LT (r = −0.512, p < 0.001). Intraclass correlation coefficient (ICC) scores were high for CPL (0.979), CPD (0.992), CPA (0.966), CMA (0.963), and CBA (0.951). Conclusions: In healthy Chinese subjects, CPL was greatest in the superior quadrant, followed by the inferior, temporal, and nasal quadrants, and CPA was largest in the superior quadrant, followed by the tempdoral, inferior, and nasal quadrants. Transverse UBM images can be used to measure the anatomy of the ciliary process with relatively good repeatability and reliability.


Introduction
The ciliary body (CB) is the middle part of the anterior uvea. Anatomically, the CB spans the portion of the eye between the scleral spur and the ora serrata. In the sagittal section, the CB is triangular and divided into two parts: the pars plicata, characterized by a longitudinal radial process called the ciliary process (CP), and the pars plana, which is flat and approximately 4 mm behind the CP [1]. The CB has an important relationship with neighboring structures and has a variety of functions, including aqueous humor production, regulation of aqueous humor output through the uveal sclera pathway, and regulation via the ciliary muscle and suspensory ligament [2]. However, the CB and CP cannot be directly visualized due to the posterior position.
Ultrasound biometric microscopy (UBM) has been widely used to analyze anterior chamber structures, especially the structure behind the iris, such as the CB [3,4]. The UBM imaging and analysis of the CB, based on the radial scan, is valuable in many potential clinical applications, such as analyzing the pathogenesis of different types of angle-closure glaucoma [5][6][7][8]; describing the effects of pharmacologic agents on CB [9][10][11]; revealing the relationship between CBT and refractive error [12]; and assessing the development of uveitis, tumors, and cysts [13][14][15].
In radial UBM images, only one CB can be seen at one time, and only the pars plana and plicata of the CB can be visualized. In addition, it cannot clearly characterize the morphological features of the CPs due to their irregular arrangement, similar to protuberance. However, a transverse scan of CB is available by UBM. In transverse images, the transducer probe can be aligned with an entire row of CPs to illustrate the complex anatomy of the CB. Such images provide the ability to analyze a group of CPs rather than a single CP as in a radial scan. However, as far as we know, an objective and repeatable protocol of transverse scanning of UBM has never been reported. In this study, we aim to demonstrate the anatomy characteristics of the ciliary body using both radial and transverse scans of UBM in healthy Chinese subjects, and to explore the determining factors for further study of the pathogenesis, prevention, and follow-up of CB-related disease in vivo.

Participants
This was a cross-sectional study consisting of healthy Chinese subjects. The research conformed to the Helsinki Declaration's guidelines; was approved by the Ethics Board of the Zhongshan Ophthalmic Center (ZOC), Sun Yat-Sen University; and participants signed informed consents. All subjects underwent detailed ocular examinations, including slit-lamp examination, fundus examination with a 90-diopter lens, and intraocular pressure (IOP) measurement by Goldmann applanation tonometry. Axial length (AL), lens thickness (LT), white-to-white (WTW) corneal diameter, and central corneal thickness (CCT) were measured by the same trained observer (X.G.) via IOL Master 700 (Carl Zeiss Meditec AG, Jena, Germany, version 1.7). High-definition images of the anterior segment and the CB structures were provided by UBM.

Image Acquisition
UBM imaging was performed using an sw-3200L UBM and 50 MHz linear transducer (Tianjin Suowei Electronic Technology Co., Ltd., Tianjin, China). The highest axial and lateral resolutions of the UBM were no less than 40 µm. All images included in this study were obtained by the same examiner (L.C.). Radial scans were performed in the positions of 9, 12, 3, and 6 o'clock centered over the corneal limbus, and perpendicular sulcus-to-sulcus scans were obtained over the pupil center.
Specifically, transverse scans of UBM images were obtained as follows. The probe was perpendicular to the corneal limbus, and the first and most precise CPs image at the time of disappearance of the ciliary sulcus was obtained from four different quadrants (superior, nasal, inferior, and temporal) of each eye. Measurements of the superior quadrant were performed repeatedly over for 1 h by the same physician while masking the initial results. The images were analyzed by the same observer. Figure 1A-D show the probe directions. Figure 1E shows a transverse (superior) CPs image sample. initial results. The images were analyzed by the same observer. Figure 1A-D show the probe directions. Figure 1E shows a transverse (superior) CPs image sample.

Image Measurement
As Figure 2A shows: (1) the ciliary process length (CPL) was determined by calculating the average length of each individual CP within a 3-mm line in a row; (2) the ciliary process density (CPD) was defined as the overall number of ciliary processes in a 3-mm segment of the transverse CB; (3) the ciliary process area (CPA), ciliary muscle area (CMA), and ciliary body area (CBA) were measured by calculating the area of ciliary processes, ciliary muscle, and CB, respectively, calculating the area within the boundaries of CPs within a 3-mm linear distance using ImageJ software 1.51 (ImageJ Software Inc., Bethesda, MD, USA). The method and the parameters, such as CPL, CPA, CPD, CMA, and CBA of transverse UBM scans, were first defined in this study, so we analyzed the intraobserver reproducibility.
As Figure 2B shows: anterior chamber depth (ACD) was defined as the axial distance between the corneal endothelium and the anterior lens surface. Ciliary sulcus diameter (CSD) was the perpendicular sulcus-to-sulcus distance from 12 to 6 o'clock.
As Figure 2C shows: (1) CBT0 was the CBT at the point of the scleral spur, and CBT1 was the CBT at a distance of 1 mm from the scleral spur; (2) maximum CBT (CBTmax) was defined as the distance between the innermost point of the CB and the inner surface of the

Image Measurement
As Figure 2A shows: (1) the ciliary process length (CPL) was determined by calculating the average length of each individual CP within a 3-mm line in a row; (2) the ciliary process density (CPD) was defined as the overall number of ciliary processes in a 3-mm segment of the transverse CB; (3) the ciliary process area (CPA), ciliary muscle area (CMA), and ciliary body area (CBA) were measured by calculating the area of ciliary processes, ciliary muscle, and CB, respectively, calculating the area within the boundaries of CPs within a 3-mm linear distance using ImageJ software 1.51 (ImageJ Software Inc., Bethesda, MD, USA). The method and the parameters, such as CPL, CPA, CPD, CMA, and CBA of transverse UBM scans, were first defined in this study, so we analyzed the intra-observer reproducibility.
As Figure 2B shows: anterior chamber depth (ACD) was defined as the axial distance between the corneal endothelium and the anterior lens surface. Ciliary sulcus diameter (CSD) was the perpendicular sulcus-to-sulcus distance from 12 to 6 o'clock.
As Figure 2C shows: (1) CBT 0 was the CBT at the point of the scleral spur, and CBT 1 was the CBT at a distance of 1 mm from the scleral spur; (2) maximum CBT (CBT max ) was defined as the distance between the innermost point of the CB and the inner surface of the sclera; (3) anterior placement of the ciliary body (APCB) was the distance from the most anterior point of the CB to the vertical line drawn from the inner surface of the scleral spur; (4) the trabecular-ciliary angle (TCA) refers to the angle between the posterior corneal surface and the anterior surface of the CB. sclera; (3) anterior placement of the ciliary body (APCB) was the distance from the most anterior point of the CB to the vertical line drawn from the inner surface of the scleral spur; (4) the trabecular-ciliary angle (TCA) refers to the angle between the posterior corneal surface and the anterior surface of the CB.

Statistical Analysis
All data were imported and sorted by two authors. The statistical analysis and description were performed in SPSS 23 (IBM Corporation, Chicago, IL, USA). One-way analysis of variance (ANOVA) was performed to compare the CPL, CPD, CPA, CMA, CBA, CBT, and APCB in each of the four quadrants. The CPL, CPD, CPA, CMA, CBA, CBT, and APCB values obtained in the four quadrants were averaged, and the Pearson's correlation coefficient test was used to assess the relationships of average CPL, CPD, CPA, CMA, CBA, CBT, and APCB with the following parameters: age, BMI, AL, ACD, LT, and CSD. The intraclass correlation coefficient (ICC) and the Bland-Altman plots were used to analyze the consistency of each parameter for two repeat examinations of the superior CP. p < 0.05 was considered to be statistically significant.

Statistical Analysis
All data were imported and sorted by two authors. The statistical analysis and description were performed in SPSS 23 (IBM Corporation, Chicago, IL, USA). One-way analysis of variance (ANOVA) was performed to compare the CPL, CPD, CPA, CMA, CBA, CBT, and APCB in each of the four quadrants. The CPL, CPD, CPA, CMA, CBA, CBT, and APCB values obtained in the four quadrants were averaged, and the Pearson's correlation coefficient test was used to assess the relationships of average CPL, CPD, CPA, CMA, CBA, CBT, and APCB with the following parameters: age, BMI, AL, ACD, LT, and CSD. The intraclass correlation coefficient (ICC) and the Bland-Altman plots were used to analyze the consistency of each parameter for two repeat examinations of the superior CP. p < 0.05 was considered to be statistically significant.

Discussion
The study of CB has gained significant attention because of its involvement in glaucoma and myopia. Several studies have been conducted on humans to provide histological information about the CB, but all have encountered postmortem shrinkage problems [16,17]. UBM can be used to measure any quadrant on living subjects, as it is unaffected by shrinkage. Since its inception, UBM has been used in many clinical and preclinical studies associated with CB-related disease, and it is irreplaceable for the study of the posterior chamber and the innermost structures of the iridociliary region compared to AS-OCT [4,[18][19][20]. To the best of our knowledge, this is the first report that demonstrated the morphology parameters of the CB from both radial and transverse scans in four different quadrants in vivo of Chinese people and the correlations between them with systemic and ocular parameters.
In our study, four quadrantal analyses of the ciliary process revealed that the CPL was longest in the superior quadrant, followed by the inferior, temporal, and nasal quadrants. CPA tended to be larger in the superior quadrant than in the other three quadrants. In contrast to CPL, CMA was largest in the nasal quadrant, followed by the temporal, inferior, and superior quadrants. This may be due to the embryonic development of the eye. Numerous reports about ocular morphology have been published. Retinal nerve fiber layer (RNFL) thickness was found to be thicker in the superior and inferior quadrants, followed by the temporal and nasal quadrants in normal Chinese students aged 6 to 17 years using optical coherence tomography (OCT) [21]. Corneal thickness was lower in the inferotemporal quadrant and higher in the superonasal quadrant using anterior segmental-optical coherence tomography (AS-OCT) [22]. A previous study compared the ciliary body morphology between Caucasians and Chinese individuals aged 40 to 80 years, and found that Chinese individuals had a thinner CBT using UBM [23]. It was reported that CBT1 was significantly thicker in the superior quadrant than in the nasal, temporal, and inferior quadrants using UBM in Asian subjects aged 11-86 years [24]. However, in our study, no significant difference in CBT0, CBT1, and CBTmax among the four quadrants was found. The inconsistency may arise from the inclusion of subjects of different races. APCB and TCA were inversely related to anterior rotation of the ciliary body. According to a previous study in Japan, the mean TCA was 79.2 degrees using UBM [3], which is similar to our findings. The longer the AL, the deeper the ACD, and the larger CSD were, the longer the CPL was. The smaller the BMI and the thinner the LT were, the longer the CPL was. The younger the subject was, the longer the CPL was. The solid line represents the best-fit line.

Discussion
The study of CB has gained significant attention because of its involvement in glaucoma and myopia. Several studies have been conducted on humans to provide histological information about the CB, but all have encountered postmortem shrinkage problems [16,17]. UBM can be used to measure any quadrant on living subjects, as it is unaffected by shrinkage. Since its inception, UBM has been used in many clinical and preclinical studies associated with CB-related disease, and it is irreplaceable for the study of the posterior chamber and the innermost structures of the iridociliary region compared to AS-OCT [4,[18][19][20]. To the best of our knowledge, this is the first report that demonstrated the morphology parameters of the CB from both radial and transverse scans in four different quadrants in vivo of Chinese people and the correlations between them with systemic and ocular parameters.
In our study, four quadrantal analyses of the ciliary process revealed that the CPL was longest in the superior quadrant, followed by the inferior, temporal, and nasal quadrants. CPA tended to be larger in the superior quadrant than in the other three quadrants. In contrast to CPL, CMA was largest in the nasal quadrant, followed by the temporal, inferior, and superior quadrants. This may be due to the embryonic development of the eye. Numerous reports about ocular morphology have been published. Retinal nerve fiber layer (RNFL) thickness was found to be thicker in the superior and inferior quadrants, followed by the temporal and nasal quadrants in normal Chinese students aged 6 to 17 years using optical coherence tomography (OCT) [21]. Corneal thickness was lower in the inferotemporal quadrant and higher in the superonasal quadrant using anterior segmental-optical coherence tomography (AS-OCT) [22]. A previous study compared the ciliary body morphology between Caucasians and Chinese individuals aged 40 to 80 years, and found that Chinese individuals had a thinner CBT using UBM [23]. It was reported that CBT 1 was significantly thicker in the superior quadrant than in the nasal, temporal, and inferior quadrants using UBM in Asian subjects aged 11-86 years [24]. However, in our study, no significant difference in CBT 0 , CBT 1 , and CBT max among the four quadrants was found. The inconsistency may arise from the inclusion of subjects of different races. APCB and TCA were inversely related to anterior rotation of the ciliary body. According to a previous study in Japan, the mean TCA was 79.2 degrees using UBM [3], which is similar to our findings.
Our study provided a new protocol to scan transverse (quadrant) CP images using UBM; in particular, a method to locate the anatomical point to achieve a repeatable fashion. For the first time, we clearly defined CPL as the distance parallel to the long axis of the CP from the farthest end to the midpoint of the line of the bilateral ciliary sulcus according to the morphology. CPD, CPA, CMA, and CBA were all measured within the middle 3-mm linear distance of the transverse CB image. Due to the spherical shape of the eyeball, the surrounding CPs appear deformed in the image at a greater distance. Additionally, if it is at a shorter distance, the number of CPs will be too small, resulting in statistical errors. Therefore, the CPs in the area we chose were the most precise and closest to the actual shape in the scan. To date, there has been no standardized method for reliable measurement of the ciliary process. Only one study published in 2021 obtained CP images of normal human eyes from the temporal pars plicata using UBM, and the ICC values were >0.8 for CP parameters (CPL, CPD, CPA) [25]. However, there was an advantage that all five parameters (CPL, CPD, CPA, CMA, CBA) in our study had ICC values > 0.9, suggesting relatively better repeatability and reliability than the study mentioned above. The reason for this might be that we obtained CP images in our study by focusing the ultrasonic probe perpendicular to the corneal limbus at the time of disappearance of the ciliary sulcus, and this might prove a more reliable method to measure CPs from transverse UBM images.
Additionally, the correlations between parameters of the CP and systemic and ocular parameters in healthy Chinese subjects were revealed, which has not been reported before, as far as we know. In addition, average CPL had a negative correlation with age and BMI, indicating that people who were younger or with lower BMI tended to have a longer CP. In our study, average CPL and average TCA showed a positive correlation with AL and CSD, and showed a negative correlation with LT; average CPD, average CMA, and average APCB showed the opposite correlation. In other words, healthy Chinese subjects, who had smaller eyes with shorter AL, shorter CSD, and thicker lenses, had shorter but denser ciliary processes, thicker ciliary muscle, and more anteriorly located CB, which might provide new inspiration for the occurrence and mechanism of malignant glaucoma in patients with primary angle-closure (PAC) and primary angle-closure glaucoma (PACG). With transverse direction UBM scans, CP can be seen more clearly, which may provide a more accurate way to explore the pathogenesis, and evaluate the surgical prognosis of glaucoma.
In previous studies, CBT 1 was positively correlated with AL in normal subjects [24], and CBT 2 and CBT 3 (2 mm and 3 mm posterior to the scleral spur) were positively correlated with AL in myopic eyes [12]. It was reported that ciliary muscle thickness (CMT 2 and CMT 3 , 2 mm and 3 mm posterior to the scleral spur) was increased in myopic eyes compared with nonmyopic eyes, suggesting that a large CB may be associated with a greater contraction of the ciliary muscles to resist equatorial sclera thinning [26,27]. However, a correlation between CBT 0 , CBT 1 , CBT max , and AL was not found in our study, probably because we used different measured parameters, and the difference was not significant.
As the site of aqueous humor production, the CP is also the site of surgical interventions, such as trans-scleral cyclo-photocoagulation (TSCPC), endoscopic cyclophotocoagulation (ECP), and ultrasound cycloplasty (UCP), which are aimed at destroying the CB to reduce IOP in glaucoma patients [28][29][30]. Therefore, images of the ciliary process using UBM could provide a more refined observation method to evaluate the morphological changes of the ciliary process in TSCPC-, ECP-, and UCP-treated glaucoma and the relationship between the morphological changes and IOP reduction. The ability to study drug or surgery action on the morphology of the CP itself may offer valuable insights regarding the mechanisms of action.
The main limitation of our study is that the resolution limitations of the UBM technique may have contributed to morphometric errors. The spatial resolution of the UBM is superior in the radial direction compared to the transverse direction. The examination was performed to apply the probe as vertically to the corneal limbus as possible to obtain a precise image. Another significant limitation is the small sample size. Therefore, we need to compensate for the lack of accuracy caused by these limitations by increasing the number of subjects. The third limitation is that we only included Chinese subjects in this study, which limited the generalizability of the results. In order to verify the generalizability of the results, further multiracial studies are warranted to confirm our findings. In addition, even though it does not affect the clinical conclusions, the analysis could be more accurate if the inter-eye correlation was considered.

Conclusions
In conclusion, we found a topographical distribution of the ciliary body in healthy Chinese subjects using the application of radial and transverse UBM imaging, and provided more information on the ciliary anatomy. The average CPL and CPA were greatest in the superior quadrant. Given the relatively good repeatability and reliability of ciliary process imaging, further studies should be performed to explore the pathogenesis of CB-related diseases, such as malignant glaucoma, and to evaluate the role of glaucoma interventions aimed at the CB, such as TSCPC, ECP, and UCP.