Abstract
BACKGROUND: White matter hyperintensities (WMH), a key imaging marker of cerebral small vessel disease, are associated with stroke, cognitive decline, and mortality in aging populations, yet current magnetic resonance imaging (MRI)-based assessments are costly and unsuitable for large-scale monitoring. Owing to shared microvascular characteristics between the retina and brain, ocular imaging offers a non-invasive alternative, but most studies have examined single-modal retinal metrics or total WMH volume, with little attention to multimodal biomarkers and lobe-specific WMH. This study aims to explore the associations between multimodal ocular imaging biomarkers and WMH volume to enable earlier, non-invasive detection of cerebral small vessel disease. METHODS: Participants (age: 23-83 years) in the KaiLuan Study underwent comprehensive ophthalmic and neuroimaging evaluations, including optical coherence tomography (OCT) for measuring macular retinal nerve fiber layer (mRNFL), macular ganglion cell-inner plexiform layer (mGCIPL), macular inner nuclear layer (mINL), macular ganglion cell complex (mGCC), and total macular thickness. Fundus photography was used to assess central retinal arteriolar equivalent (CRAE), central retinal venular equivalent (CRVE), arteriole/venular ratio (AVR), fractal dimension (FD), global vein width, and global artery width. MRI was employed to quantify WMH volume. Generalized linear regression models were used to analyze associations between ocular biomarkers and WMH, with adjustments for demographics and vascular risk factors. RESULTS: The study included 761 participants (average age 55 years, 54.4% women). Thinner mGCIPL and mGCC were significantly associated with greater WMH volume in the total brain (P<0.02), temporal lobes (P≤0.02), and occipital lobes (P≤0.003). Reduced mINL correlated with larger WMH in the total brain and specific lobes (P<0.04). No significant associations were found for mRNFL or total macular thickness (all P>0.05). Smaller AVR, narrower CRAE, and wider CRVE were linked to higher WMH volume in the occipital lobe (P<0.02). Lower FD complexity and wider global vein width demonstrated significant associations with increased WMH volume in the total brain and individual lobes (frontal, occipital, and temporal lobes; all P<0.01). Sex-stratified analysis showed that these associations were stronger in females, with thinner mGCIPL, mINL, and wider global vein width associated with greater total and lobe-specific WMH (P<0.05). CONCLUSIONS: It is suggested that parameters from OCT and fundus photography are independently associated with WMH volume. These parameters may serve as potential biomarkers for the early detection and longitudinal monitoring of WMH, enabling precise identification and ongoing surveillance of white matter alterations.