Estimation of glaucomatous impairment in the visual field, excluding the effect of posterior capsule opacification

Posterior capsule opacification (PCO) significantly hinders the accurate assessment of glaucomatous visual field (VF) impairment. While pattern deviation (PD) is theoretically intended to correct for generalized sensitivity loss, its accuracy in eyes with PCO remains unverified. This study evaluated 52 eyes of 46 patients with open-angle glaucoma and PCO. Following laser capsulotomy, visual acuity significantly improved from 0.22 to 0.10 logarithm of the minimum angle of resolution (P < 0.01), and mean total deviation (mTD) improved from - 8.89 dB to -6.46 dB (P < 0.01). This post-capsulotomy mTD served as the reference for true impairment. A linear regression model incorporating pre-capsulotomy visual acuity and mTD was developed and validated using leave-one-out cross-validation. The absolute estimation error of this model was significantly smaller (median: 1.66 dB [interquartile range: 1.04-2.77 dB]) compared to that of PD (2.14 dB [1.10-3.99 dB], P < 0.01). Notably, PD-based errors were significantly influenced by both glaucoma severity and the degree of PCO-induced sensitivity loss, resulting in substantial underestimation in advanced cases. These findings indicate that PD is prone to inaccuracy in eyes with PCO, whereas the regression model offers a more precise estimation.