Purpose: The aim of the study was to compare and explore the agreement between the nomogram tool and ultrasound biomicroscopy (UBM) images method to calculate the ultrasound cycloplasty (UCP) probe model in Chinese glaucoma patients. Methods: Retrospective analysis of Chinese glaucoma patients who visited Zhongshan Ophthalmic Center in Guangzhou from January to December 2019 and were eligible for UCP surgery. Visual acuity, intraocular pressure (IOP), ocular axial length (AL), and horizontal corneal diameter (white to white [WTW]) were measured. UBM images with clear ciliary body imaging and AL and WTW data were sent to trained personnel for probe model measurements. The data calculated by both methods were analyzed using unweighted and weighted κ statistics. The level of agreement refers to Landis and Koch’s guideline for the strength of agreement indicated with weighted κ values. Results: 1,061 eyes of 642 patients were involved, with a mean age of 61.66 ± 11.66 years. Their best-corrected visual acuity converted to logarithm of minimal-angle-of-resolution (logMAR) scores of −0.18–3.00 with a mean value of 0.69 ± 0.77. IOP was 22.0–60.0 mm Hg with a mean of 27.97 ± 5.66 mm Hg. The mean AL and WTW were 22.88 ± 1.33 (19.15–32.14) mm and 11.52 ± 0.49 (10.00–12.90) mm, respectively. The agreement between the two methods was fair (weighted κ = 0.299), matching in 62.86% of eyes (weighted κ = 0.299, κ = 0.264). The agreement in primary open angle glaucoma, acute primary angle-closure glaucoma, chronic primary angle-closure glaucoma, and secondary glaucoma patients was 60.85% (weighted κ = 0.336, κ = 0.301), 65.06% (weighted κ = 0.146, κ = 0.127), 62.26% (weighted κ = 0.204, κ = 0.184), and 57.97% (weighted κ = 0.332, κ = 0.280) of eyes, respectively. Conclusion: The agreement between UBM images and the nomogram tool to calculate the UCP probe model of Chinese patients is at a fair level. The nomogram tool prefers to use larger probes. Improvements to the nomogram tool, such as including data from more ethnic groups and being able to calculate separately for different types of glaucoma, are needed to improve accuracy. The inclusion of parameters or images from more directions of the eye may help measure probe models more accurately for both the nomogram tool and the UBM image measurement.
Read full abstract