Visual acuity changes in foveal sparing geographic atrophy due to AMD

Abstract

 

Purpose

 

To investigate the impact of atrophy progression in patients exhibiting a “foveal-sparing” pattern geographic atrophy (GA) on visual acuity

 

Methods

 

Patients from the prospective natural history “Fundus-Autofluorescence imaging in Age-related Macular Degeneration” study (FAM, NCT00393692) were examined longitudinally with fundus autofluorescence (FAF) and near-infrared reflectance (IR) imaging (Spectralis HRA+OCT or HRA2, Heidelberg Engineering). Eyes with a contiguous well-demarcated area of GA either in a complete ring around the spared fovea or in a horseshoe pattern >270° encompassing the fovea were included into the analysis. Areas of foveal sparing and GA were measured using a semi-automated software tool allowing for combined IR- and FAF-image grading (RegionFinder^TM v. 2.5.5.0, Heidelberg Engineering). Best corrected visual acuity (BCVA) was acquired using ETDRS-charts. Changes in foveal sparing area and BCVA were analyzed using a linear mixed-model approach. To investigate minimal areas of spared fovea that were required to hold certain visual acuity, maximum likehood estimations (MLE) were performed for undercutting defined BCVA levels.

 

Results

 

A total of 42 eyes (35 patients, 29 female, mean age at baseline: 72.79±7.89 y) were examined longitudinally over 25.8±18.7 months. During the observation period the size of the intact foveal island decreased from 1.54±0.86 mm² at baseline to 1.01±0.69 mm², and BCVA decreased from 0.36 ±0.29 logMAR to 0.50±0.31 logMAR units. In nine eyes, BCVA remained stable over the whole observational period. Modelling changes in BCVA in all patients with a linear mixed-effects model using time and sparing size as fixed effects revealed a decrease of BCVA by 0.08 logMAR per year and 0.05 logMAR per mm² loss of spared fovea. Highest values for MLE were obtained for undercutting a BCVA of 0.2 logMAR occurring when sparing size falls below 1.2 mm². Odds-ratio to decrease below a BCVA of 0.2 was 3.3 for a sparing sizes of <1.2 mm².

 

Conclusions

 

The results demonstrate a dependency of BCVA on the size of the residual fovea island. A foveal sparing area of 1.2 mm² appears to be necessary to maintain good visual acuities of <0.2 logMAR. There seems to be no clear critical size for the maintenance of lower BCVA, possibly because other factors may be of higher relevance at these low BCVA levels. These natural history data may be helpful in the design of future interventional clinical trials.