Purpose To investigate the factors associated with variability in the longitudinal visual field (VF) test in patients with normal‐tension glaucoma (NTG). Methods This retrospective study enrolled patients with NTG who underwent ≥ 12 reliable VF tests over a follow‐up period of ≥ 6 years. For each eye, 52 total deviation values (TDVs) were linearly regressed against time (years), while the root‐mean‐squared error (RMSE) of the residuals was calculated to quantify the long‐term VF variability. Hierarchical cluster analysis using squared Spearman correlation coefficients was conducted to assess collinearity and select variables. Six linear mixed‐effects models were constructed, each including clinical parameters such as intraocular pressure (IOP), retinal nerve fiber layer thickness (RNFLT), and VF parameters. Mixed‐effects regression was applied to evaluate the factors associated with long‐term VF variability, adjusting for intereye correlation. Results The study analyzed 114 eyes of 77 patients with NTG. In models including baseline RNFLT, lower baseline IOP, lower mean IOP, greater IOP fluctuation, thinner baseline RNFLT, and a steeper mean deviation (MD) slope were significantly associated with increased long‐term VF variability (all p ≤ 0.031). In models that included baseline MD, worse baseline MD and a steeper MD slope were significantly associated with increased long‐term VF variability (all p ≤ 0.004). Conclusions Lower baseline and mean IOP, greater IOP fluctuation, thinner baseline RNFLT, worse baseline MD, and steeper MD slope are all associated with a greater long‐term VF variability in patients with NTG. NTG eyes with thin RNFL or rapid VF deterioration may require more frequent VF testing to avoid misinterpreting fluctuation as true progression.
Building similarity graph...
Analyzing shared references across papers
Loading...
Seunghee Ha
Sangwoo Moon
EunAh Kim
Journal of Ophthalmology
Sungkyunkwan University
Pusan National University
Changwon National University
Building similarity graph...
Analyzing shared references across papers
Loading...
Ha et al. (Thu,) studied this question.
www.synapsesocial.com/papers/69c37adcb34aaaeb1a67cc71 — DOI: https://doi.org/10.1155/joph/7211382