Purchase this article with an account.
Sebastian M Waldstein, Ana-Maria Glodan, Roland Leitner, Christian Simader, Georg Langs, Bianca S. Gerendas, Ursula Schmidt-Erfurth; Three-dimensional analysis of intra- and subretinal fluid provides precise prediction of visual acuity in neovascular AMD. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):5378.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The presence of intraretinal cystoid fluid (IRC) has been demonstrated to cause a substantial loss in visual function in neovascular age-related macular degeneration (nAMD). On the other hand, recent evidence suggests that subretinal fluid (SRF) may be associated with a favorable visual acuity prognosis. However, studies up to date rely on a categorical, qualitative grading of the presence or absence of fluid and therefore preclude precise functional predictions. Thus, the aim of this study was to correlate quantitative measures of IRC and SRF with best-corrected visual acuity (BCVA) outcomes in antiangiogenic therapy of nAMD.
38 patients with treatment-naïve subfoveal choroidal neovascularization due to AMD receiving monthly ranibizumab or aflibercept injections for 12 months were included. IRC and SRF were manually delineated on each B-scan of spectral-domain optical coherence tomography volume scans at baseline and month 1, 6 and 12. A systematic automated search was conducted to detect annotation-derived variables with the best predictive value for BCVA. An exponential model for BCVA change was constructed. Correlations were computed between IRC, SRF and baseline BCVA, final BCVA and BCVA change (exponential and linear).
A total of 19,456 scans underwent complete quantification of IRC and SRF. At baseline, an area-based, centrally weighted IRC indicator showed the best correlation with BCVA (R²=0.59, p<0.0001). The same indicator could also predict BCVA at 12 months (R²=0.21, p=0.0034). BCVA gain from baseline to month 12 was significantly correlated with a decrease in the IRC predictor in the exponential model (R²=0.40, p<0.0001) and linear model (R²=0.25, p=0.0015). In contrast, SRF was not statistically significantly associated with baseline BCVA or final BCVA outcomes.
IRC-derived morphometric variables can be used to precisely predict BCVA in the treatment-naïve condition as well as BCVA outcomes in antiangiogenic therapy of nAMD. While a reduction in IRC is statistically significantly associated with BCVA gains, a proportion of IRC-mediated neurosensory damage may remain permanent. This suggests the future path for personalized visual prognosis based on retinal pathomorphology and may serve as an important endpoint in clinical trials. Conversely, SRF quantification was not significantly associated with BCVA outcomes.
This PDF is available to Subscribers Only