New disease scale helps predict visual acuity change in neovascular AMD
CHICAGO — A new diagnostic scale correlated with best corrected visual acuity and could serve as an alternative to evaluate disease activity in neovascular age-related macular degeneration, according to a study.

At Retina Subspecialty Day at the American Academy of Ophthalmology meeting, Usha Chakravarthy, MBBS, PhD, said there is a known negative correlation between commonly used markers, such as foveal retinal thickness and central subfield thickness (CST), and BCVA in patients with neovascular AMD, so new diagnostic tools are needed.

“There is a lot of variation and disconnect between BCVA and CST values,” she said. “CST may be the same for different BCVA values.”
Researchers sought a simpler method for building a disease activity scale (DAS) for neovascular AMD based entirely on spectral domain OCT and key markers of subretinal fluid, intraretinal fluid, hyperreflective material and pigment epithelial detachment. Each marker is scored 0 to 4 for severity, from none to severe, with a maximum overall score of 16.
An independent grader assessed for DAS, BCVA and CST at baseline, month 4 and month 12 in a sample of 52 treatment-naive eyes managed with anti-VEGF.
The median DAS was 9 at baseline and 4 at month 4 and month 12.
Chakravarthy said DAS and CST showed strong correlations throughout the study period. While the correlations between DAS and BCVA were weak at baseline and month 4, they were stronger at month 12. The correlations between CST and BCVA were moderate at baseline and weak during the maintenance phase, she said.
A regression model showed that change in DAS was a stronger predictor of BCVA (P < .001) compared with change in CST.
“Clearly, some refinement of the scale is needed to improve the correlations at baseline,” Chakravarthy said. “Its reproducibility and reliability require testing in other data sets.”