Abstract
Purpose :
Preserved visual fields is one major advantage quoted for anti-VEGF over pan-retinal photocoagulation (PRP) for proliferative diabetic retinopathy (PDR). However, visual field changes over the natural history of untreated DR are unknown. This study analyzed visual field changes in untreated DR, and compared these with DR treated with PRP.
Methods :
Data were used from the DRS (1972-1979) and ETDRS (1980-1989), two unique historical datasets in which eyes with DR were randomized to PRP or no treatment. Visual field scores (VFS) were obtained with the Goldmann perimeter using the IV/4e stimulus (DRS), and the I/4e and I/2e stimuli (ETDRS). The current analysis included eyes with severe non-proliferative DR (SNPDR) or PDR, randomized to either the immediate or deferred argon laser PRP arms. Any eyes in the deferred arm that received argon laser were censored from the point of treatment. The outcome of annual VFS change was calculated using mixed models.
Results :
A total of 2189 DRS eyes and 1325 ETDRS eyes, followed for median of 2.3 years (IQR 1.7–4.0) and 6.3 years (5.0–7.3), respectively, were included. Mean annual VFS changes from baseline were -42.0 (95% CI -46.9, -37.1) , -11.0 (-13.8, -8.3), and -2.9 (-5.1, -0.7) in untreated eyes and -42.9 (-47.5, -38.3), -30.8 (-33.6, -28.0), and -12.8 (-15.1, -10.5) in treated eyes, using the IV/4e, I/4e, and I/2e stimuli, respectively. Mean annual VFS changes from the 4-month visit were -37.1 (-43.0, -31.2), -9.6 (-13.3, -6.0), and -4.1 (-7.1, -1.0) in untreated eyes and -29.5 (-34.2, -24.7), -18.1 (-21.3, -14.8), and -4.3 (-6.1, -2.4) in treated eyes (See Figure).
Conclusions :
The natural history of untreated SNPR and PDR involves a significant decline over time in the peripheral visual field. PRP causes a rapid initial decline in the visual field. Subsequent rates of visual field decline are relatively similar in untreated DR and DR treated with PRP. No future randomized studies will ever provide a comparison between untreated and treated PDR. These findings have implications for assessment of the relative merits of anti-VEGF versus PRP for PDR.
This is a 2020 ARVO Annual Meeting abstract.