Identification of eyes at risk of acute primary angle-closure in elderly Japanese patients.

To identify from the anterior segment the structural variables of the eyes that can be used to distinguish acute primary angle-closure (APAC) eyes or primary angle-closure suspect (PACS) eyes from normal eyes. We used a Pentacam scanner to measure participants' anterior eye segments. We assessed each anterior segment structure variable on the basis of receiver operating characteristic curves using the area under the curve (AUC). AUCs for eyes in men with APAC were 1.000 for central anterior chamber depth (ACD), 0.982 for peripheral ACD, 0.916 for anterior chamber angle (ACA), and 0.992 for anterior chamber volume (ACV). AUCs for eyes in women with APAC were 0.997 for central ACD, 0.942 for peripheral ACD, 0.922 for ACA, and 0.946 for ACV. AUCs for eyes in men with PACS were 0.933 for central ACD, 0.930 for peripheral ACD, 0.887 for ACA, and 0.937 for ACV. AUCs for eyes in women with PACS were 0.960 for central ACD, 0.957 for peripheral ACD, 0.937 for ACA, and 0.937 for ACV. The negative predictive values (%) in men with APAC were 100 for all the four variables (central ACD, peripheral ACD, ACA, and ACV). The negative predictive values (%) in women with APAC were 100 for central ACD, 98.7 for peripheral ACD, 97.1 for ACA, and 97.9 for ACV. The negative predictive values (%) in men with PACS were 98.6 for central ACD, 100 for peripheral ACD, 98.5 for ACA, and 99.4 for ACV. The negative predictive values (%) in women with PACS were 100 for central ACD, 98.7 for peripheral ACD, 97.1 for ACA, and 97.9 for ACV. The central ACD, peripheral ACD, ACA, and ACV measurements seem to be excellent markers to identify eyes without APAC or PACS.