AIM: To evaluate the repeatability of ocular biometrical measurements obtained from the optical low-coherence reflectometry(Lenstar LS900^® version 1.10), as well as its agreement with the partial coherence interferometry(IOL Master^® 500 version 7.1)in Chinese school-aged children.

METHODS: A prospective comparison of ocular biometrical measurements made by the Lenstar and IOL Master was performed on right eye of each participant. The intraobserver repeatability of Lenstar was assessed by coefficient of variation(CV). Agreement was analyzed using Bland-Altman plots.

RESULTS: The mean age of the 110 subjects(range 6-15y)was 10.9±2.0y, and 54.5% were female. The Lenstar yielded repeatable measurements, with the smallest CV was obtained for axial length(AL)(CV<0.1%). The Lenstar produced a greater mean value for AL(23.90±1.28 vs 23.88±1.27mm, P<0.001), anterior chamber depth(ACD)(3.62±0.26 vs 3.58±0.25mm, P<0.001), and the steepest corneal radius of curvature(CR2)(7.58±0.27 vs 7.56±0.27mm, P<0.001)than IOL Master. The 95% limits of agreement(LoA)between the two instruments were -0.025 to 0.053mm for AL, -0.047 to 0.057mm for the flattest corneal radius of curvature(CR1), -0.057 to 0.102mm for CR2, and -0.083 to 0.152mm for ACD.

CONCLUSIONS: The Lenstar yielded excellent repeatability results of AL, ACD, and corneal curvature measurements, which were interchangeable with the IOL Master measurements in school-aged children.