A recent study found that visual field (VF) progression can be predicted from baseline and longitudinal OCT structure measures with clinically relevant accuracy.

The researchers tested the hypothesis in a prospective cohort study (104 eyes of 104 patients) with at least three years of follow-up (mean 4.5 years) and at least five visual field exams (mean 8.7 exams). They defined VF progression based on pointwise linear regression with 24-2 VF (three or more locations with slope ≤-1.0dB/yr) and used machine learning and statistical modeling to predict VF progression. They detected progression in 23 eyes (22%).

The best machine learning predictors were baseline superior hemimacular ganglion cell inner plexiform layer (GC-IPL) thickness and GC-IPL change rates. Models that only used GC-IPL structural variables performed better than retinal nerve fiber layer-only models, the researchers noted.

They concluded that “further refinement of proposed models would assist clinicians with timely prediction of functional glaucoma progression and clinical decision-making.”