Comparison of Anterior Segment Optical Coherence Tomography Parameters Among Vietnamese, Chinese, and Whites.

PURPOSE: To evaluate the difference in anterior segment optical coherence tomography (AS-OCT) parameters among ethnic Vietnamese, Chinese, and whites. DESIGN: Cross-sectional study. METHODS: We prospectively recruited phakic nonglaucomatous participants (based on intraocular pressure < 21 mm Hg, normal optic nerve head and retinal nerve fiber, and cup-to-disc ratio < 0.6) of 3 ethnic groups-Vietnamese, Chinese, and white. AS-OCT parameters including angle (angle opening distance, trabecular-iris space area, and angle recess area), iris (iris thickness, iris area, and iris curvature [Icurv]), and anterior chamber parameters (anterior chamber depth, width [ACW], area, and lens vault [LV]) were obtained. A linear regression model with adjustment for age, sex, axial length, and pupillary diameter was used for analysis. Subgroup analysis was performed in closed- and open-angle subgroups, which were defined in eyes with gonioscopically visible posterior trabecular meshwork for less than 2 quadrants (closed) and equal to or more than 2 quadrants (open). RESULTS: Data were obtained from 126 Vietnamese, 124 Chinese, and 112 white participants. Mean (standard deviation) age of participants was 60.2 (11.8) years. Among overall subjects, Vietnamese had significantly smaller values in all angle and anterior chamber parameters than Chinese and white subjects (all P ≤ .001) except LV. Vietnamese had significantly higher Icurv than Chinese (B -0.065, P < .001) and white subjects (B -0.073, P < .001). In the open-angle subgroup, Vietnamese had significantly smaller angle and anterior chamber dimensions (P ≤ .001) and thicker iris (P < .001) than whites, whereas there was no significant difference in all parameters compared to Chinese except for smaller ACW in Vietnamese (P < .001). In the closed-angle subgroup, Vietnamese possessed smaller angle, smaller anterior chamber dimensions, and higher Icurv than Chinese and white subjects (P < .001). Vietnamese had significantly greater LV compared to Chinese (B -204.8, P = .001). CONCLUSION: Vietnamese subjects had more of the anatomic characteristics associated with risk for angle closure, including small anterior chamber dimension, high iris curvature, and high lens vault, compared to the other 2 ethnicities.

Normative Database and Color-code Agreement of Peripapillary Retinal Nerve Fiber Layer and Macular Ganglion Cell-inner Plexiform Layer Thickness in a Vietnamese Population.

PURPOSE: Evaluate the distribution and the color probability codes of the peripapillary retinal nerve fiber layer (RNFL) and macular ganglion cell-inner plexiform layer (GCIPL) thickness in a healthy Vietnamese population and compare them with the original color-codes provided by the Cirrus spectral domain-optical coherence tomography. METHODS: This is a cross-sectional study. We recruited nonglaucomatous Vietnamese subjects and constructed a normative database for peripapillary RNFL and macular GCIPL thickness. The probability color-codes for each decade of age were calculated. We evaluated the agreement with κ coefficient between optical coherence tomography color probability codes with Cirrus built-in original normative database and the Vietnamese normative database. RESULTS: In total, 149 eyes of 149 subjects were included. The mean age of enrollees was 60.77 (±11.09) years, with a mean spherical equivalent of +0.65 (±1.58) D and mean axial length of 23.4 (±0.87) mm. Average RNFL thickness was 97.86 (±9.19) µm and average macular GCIPL was 82.49 (±6.09) µm. Agreement between original and adjusted normative database for RNFL was fair for average and inferior quadrant (κ=0.25 and 0.2, respectively); and good for other quadrants (range, κ=0.63 to 0.73). For macular GCIPL κ agreement ranged between 0.39 and 0.69. After adjusting with the normative Vietnamese database, the percent of yellow and red color-codes increased significantly for peripapillary RNFL thickness. CONCLUSIONS: Vietnamese population has a thicker RNFL in comparison with Cirrus normative database. This leads to a poor color-code agreement in average and inferior quadrant between the original and adjusted database. These findings should encourage to create a peripapillary RNFL normative database for each ethnicity.