Ultrasound-Enhanced Assessment of Vitreous Status in Exudative AMD: Associations with Neovascular Phenotypes, Treatment Burden, and Functional Outcomes.

Background/Objectives: The influence of the vitreoretinal interface on neovascular age-related macular degeneration (nAMD) remains poorly characterized. Most previous studies relied solely on macular optical coherence tomography (OCT), which provides limited information about global posterior vitreous detachment (PVD). This study evaluated (1) whether ultrasonography-defined PVD status differs between nAMD eyes and healthy controls, and (2) whether baseline PVD influences macular neovascularization (MNV) phenotype and functional outcomes following anti-vascular endothelial growth factor (anti-VEGF) therapy. Methods: In this prospective longitudinal study, treatment-naïve nAMD eyes and population-based healthy controls underwent dynamic B-scan ultrasonography and spectral-domain OCT. PVD was categorized as absent, partial, or complete. nAMD eyes received intravitreal aflibercept according to a treat-and-extend protocol and were followed for 12 months. Structural parameters-including subretinal fluid (SRF), intraretinal fluid (IRF), and central foveal thickness-along with best-corrected visual acuity (BCVA) were recorded. A multivariable linear regression model was performed to assess whether PVD independently predicted BCVA gain after adjusting for age, baseline BCVA, MNV subtype, SRF, atrophy, and number of injections. Results: Absence of PVD was significantly more frequent in nAMD eyes than in controls (p < 0.001), whereas complete PVD prevalence was comparable. In nAMD, absence of PVD was associated with a higher prevalence of MNV type 2 (p = 0.032), while partial/complete PVD correlated with type 1 lesions. After 12 months, eyes without PVD achieved the greatest visual improvement (mean BCVA gain +0.34 ± 0.26), outperforming eyes with complete PVD (p = 0.026). A multivariable model confirmed that absence of PVD was an independent predictor of greater BCVA gain (β = -0.27; 95% CI -0.42 to -0.12; p = 0.0008). Eyes with complete PVD required more injections (p = 0.046). SRF and foveal-thickness reductions occurred across groups, whereas IRF changes were similar. Conclusions: Ultrasonography-defined PVD status differs markedly between nAMD and healthy eyes and independently influences neovascular phenotype and functional response to anti-VEGF therapy. These findings underscore the physiological importance of the vitreoretinal interface and support the use of ocular ultrasonography as an adjunct tool for assessing global vitreous status in selected nAMD settings.