Impact of Reversible Visual Impairment on Cognition and Blood-Brain Barrier Integrity

Abstract

IMPACT OF REVERSIBLE VISUAL IMPAIRMENT ON COGNITION AND BLOOD-BRAIN BARRIER INTEGRITY

Background: Cataract is a leading cause of reversible visual impairment in later life and is increasingly recognised as a potentially modifiable sensory risk factor for cognitive decline. Cataract surgery offers a clinically relevant model of sensory restoration for examining subsequent changes in cognitive function and neurovascular integrity.

Aims: This thesis examined 1) baseline associations between visual status, cognition, and blood-brain barrier (BBB) permeability indices, and 2) longitudinal change following cataract surgery at 3 months (T1) and 9 months (T2) relative to baseline (T0), integrating cognitive, neuroimaging, brain-age, and plasma proteomic outcomes.

Methods: 20 adults scheduled for cataract surgery were enrolled; 18 completed longitudinal follow-up (dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) n = 17). At T0, T1, and T2, participants underwent best-corrected visual acuity (VA) assessment, cataract grading using the Lens Opacities Classification System II (LOCS II), cognitive testing with the Repeatable Battery for the Assessment of Neuropsychological Status—Update (RBANS-U; age-adjusted percentile scores), and affective assessment using the Beck Depression Inventory-II (BDI-II) and State–Trait Anxiety Inventory (STAI-Y1/Y2). MRI included DCE-MRI-derived BBB indices (whole-brain Ktrans, whole-brain percentage BBB disruption, and whole-brain 95th percentile linear leakage (WB 95th Linear)), structural volumetry and cortical thickness, and brain-age modelling. Longitudinal outcomes were analysed primarily using linear mixed-effects models with Tukey-adjusted pairwise comparisons; paired non-parametric tests were applied when distributional assumptions were not met. Plasma proteomics was performed using label-free liquid chromatography–tandem mass spectrometry (LC–MS/MS), with differential abundance testing in limma and prespecified thresholds of false discovery rate (FDR)–adjusted p < 0.05 and |log₂ fold change (log₂FC)| ≥ 0.3.

Results: VA improved robustly and remained improved at T2 (median logMAR 0.50 to 0.10; p < 0.0001). RBANS-U Total percentile increased across follow-up (median 24th at T0 to 39th at T1 and 46th at T2; p < 0.001), while state anxiety remained mild and stable (STAI-Y1: 38 to 36 to 37; p = 0.57), supporting that cognitive gains were unlikely to be driven primarily by reduced test anxiety. Threshold-based BBB indices (whole-brain percentage BBB disruption; WB 95th Linear) showed no significant T0–T2 change (all p > 0.05), whereas whole-brain Ktrans decreased from 0.0036 to 0.0027 min⁻¹ (p = 0.041). Global and lobar structural MRI measures (volumes and cortical thickness) remained tightly clustered over follow-up (all p > 0.05). Brain-age analysis demonstrated a reduction in the rate of brain ageing from T0 to T2 (p = 0.009). Proteomics quantified 664 proteins; a small subset met prespecified differential-abundance thresholds, including consistent directionality for nucleoplasmin-2 (NPM2; increased) and UDP-glucose 4-epimerase (GALE; decreased), alongside additional T2 changes. No proteomic evidence of change was identified for canonical circulating BBB serum marker candidates (e.g., glial fibrillary acidic protein (GFAP) and vascular endothelial growth factor (VEGF)).

Conclusions: In this older cohort, cataract surgery was associated with sustained visual restoration and clinically meaningful improvement in global cognitive performance over 9 months, accompanied by a modest reduction in BBB permeability (Ktrans) with relative stability of threshold-based leakage indices and structural brain measures. Together with deceleration of brain-ageing indices and selective, non-inflammatory-appearing plasma proteomic shifts, the findings support sensory integrity as a candidate contributor to cognitive and neurovascular resilience in ageing, while emphasising the metric-dependent sensitivity of BBB measures and the exploratory nature of proteomic inferences.