Anterior Segment Measurement and Reproducibility in Pediatric Eyes Using Quantitative Ultrasound Biomicroscopy.

Abstract

Purpose: The purpose of this study is to provide an evidence-based protocol for iris measurement from ultrasound biomicroscopy (UBM) images with reproducibility analysis and resulting normative iris thickness dataset of the pediatric human iris by age.

Methods: Healthy pediatric subjects (14 subjects, 25 eyes, mean age 2.0 ± 1.2 years) were recruited prospectively and underwent UBM imaging. Iris parameters were measured in 4 UBM images per eye in raw image and processed edge detection format. Agreement and variability were evaluated. Regression assessed the association between measurement differences and the variables expected to influence measurement reproducibility (image quality, orientation, and processing). Iris thickness by age was reported.

Results: Intraclass correlation was >0.6 and correlation was >0.7 for all parameters. Coefficient of variation was <30% for iris measurements not involving the ciliary body. Image quality improved reproducibility but was not statistically significant (p = 0.10). Age of subject, edge detection, and image orientation were also not significant. Iris thickness increased with increasing age (r = 0.63, p < 0.0001).

Conclusions: This study demonstrated reproducible iris measurements using a prospective protocol. We found image features, including image quality and edge detection pre-processing, were not critical to reproducibility. In the future, clinical correlations with iris morphology may be more rigorously studied using a well-defined, reproducible, and quantitative approach as presented in this UBM-based image analysis protocol.