Purpose: To investigate factors associated with variability in the longitudinal visual field (VF) test in patients with primary open-angle glaucoma (POAG). Methods: This retrospective study enrolled POAG patients with VF ≥ 12 who were followed up for ≥ 6 years. First, 52 total deviation values of VF series for each eye were linearly regressed against time (years), and the root mean square error (RMSE) of the residuals was used to measure long-term pointwise VF variability. The mean deviation (MD) of the VF series for each eye was linearly regressed against time (years), and the RMSE of the residuals was used to measure long-term global VF variability. Using hierarchical clustering, variable clustering was performed to select variables and Spearman’s correlation was used as a similarity index. We fitted a mixed effect linear regression model and evaluated factors associated with the long-term VF variability in each regression model. Results: The study included 246 eyes of 157 patients with POAG. Worse baseline peripapillary retinal nerve fiber layer (RNFL) thickness, lower baseline intraocular pressure (IOP), lower mean IOP, greater IOP fluctuation, and a faster VF decay rate were associated with increased long-term pointwise VF variability in the RNFL thickness model (all p ≤ 0.037). Worse baseline MD, lower baseline IOP, lower mean IOP, and faster VF decay rate were associated with increased long-term global VF variability in the MD model (all p ≤ 0.035). Conclusions Lower baseline and mean IOP, greater IOP fluctuation, worse glaucoma severity, and a faster VF decay rate were associated with greater long-term VF variability in patients with POAG. These factors should be considered when evaluating VF progression.

Purpose: To investigate factors associated with variability in the longitudinal visual field (VF) test in patients with primary open-angle glaucoma (POAG). Methods: This retrospective study enrolled POAG patients with VF ≥ 12 who were followed up for ≥ 6 years. First, 52 total deviation values of VF series for each eye were linearly regressed against time (years), and the root mean square error (RMSE) of the residuals was used to measure long-term pointwise VF variability. The mean deviation (MD) of the VF series for each eye was linearly regressed against time (years), and the RMSE of the residuals was used to measure long-term global VF variability. Using hierarchical clustering, variable clustering was performed to select variables and Spearman’s correlation was used as a similarity index. We fitted a mixed effect linear regression model and evaluated factors associated with the long-term VF variability in each regression model. Results: The study included 246 eyes of 157 patients with POAG. Worse baseline peripapillary retinal nerve fiber layer (RNFL) thickness, lower baseline intraocular pressure (IOP), lower mean IOP, greater IOP fluctuation, and a faster VF decay rate were associated with increased long-term pointwise VF variability in the RNFL thickness model (all p ≤ 0.037). Worse baseline MD, lower baseline IOP, lower mean IOP, and faster VF decay rate were associated with increased long-term global VF variability in the MD model (all p ≤ 0.035). Conclusions Lower baseline and mean IOP, greater IOP fluctuation, worse glaucoma severity, and a faster VF decay rate were associated with greater long-term VF variability in patients with POAG. These factors should be considered when evaluating VF progression.

Purpose: To investigate factors associated with variability in the longitudinal visual field (VF) test in patients with primary open-angle glaucoma (POAG). Methods: This retrospective study enrolled POAG patients with VF ≥ 12 who were followed up for ≥ 6 years. First, 52 total deviation values of VF series for each eye were linearly regressed against time (years), and the root mean square error (RMSE) of the residuals was used to measure long-term pointwise VF variability. The mean deviation (MD) of the VF series for each eye was linearly regressed against time (years), and the RMSE of the residuals was used to measure long-term global VF variability. Using hierarchical clustering, variable clustering was performed to select variables and Spearman’s correlation was used as a similarity index. We fitted a mixed effect linear regression model and evaluated factors associated with the long-term VF variability in each regression model. Results: The study included 246 eyes of 157 patients with POAG. Worse baseline peripapillary retinal nerve fiber layer (RNFL) thickness, lower baseline intraocular pressure (IOP), lower mean IOP, greater IOP fluctuation, and a faster VF decay rate were associated with increased long-term pointwise VF variability in the RNFL thickness model (all p ≤ 0.037). Worse baseline MD, lower baseline IOP, lower mean IOP, and faster VF decay rate were associated with increased long-term global VF variability in the MD model (all p ≤ 0.035). Conclusions Lower baseline and mean IOP, greater IOP fluctuation, worse glaucoma severity, and a faster VF decay rate were associated with greater long-term VF variability in patients with POAG. These factors should be considered when evaluating VF progression.

Purpose: To investigate factors associated with variability in the longitudinal visual field (VF) test in patients with primary open-angle glaucoma (POAG). Methods: This retrospective study enrolled POAG patients with VF ≥ 12 who were followed up for ≥ 6 years. First, 52 total deviation values of VF series for each eye were linearly regressed against time (years), and the root mean square error (RMSE) of the residuals was used to measure long-term pointwise VF variability. The mean deviation (MD) of the VF series for each eye was linearly regressed against time (years), and the RMSE of the residuals was used to measure long-term global VF variability. Using hierarchical clustering, variable clustering was performed to select variables and Spearman’s correlation was used as a similarity index. We fitted a mixed effect linear regression model and evaluated factors associated with the long-term VF variability in each regression model. Results: The study included 246 eyes of 157 patients with POAG. Worse baseline peripapillary retinal nerve fiber layer (RNFL) thickness, lower baseline intraocular pressure (IOP), lower mean IOP, greater IOP fluctuation, and a faster VF decay rate were associated with increased long-term pointwise VF variability in the RNFL thickness model (all p ≤ 0.037). Worse baseline MD, lower baseline IOP, lower mean IOP, and faster VF decay rate were associated with increased long-term global VF variability in the MD model (all p ≤ 0.035). Conclusions Lower baseline and mean IOP, greater IOP fluctuation, worse glaucoma severity, and a faster VF decay rate were associated with greater long-term VF variability in patients with POAG. These factors should be considered when evaluating VF progression.