Association Between Cardiovascular Risk Factors and Late-Onset Epilepsy

ABSTRACT Introduction Late-onset epilepsy (LOE) is an emerging public health concern in the elderly population. In addition to established acquired etiologies, particularly stroke, recent evidence suggests that subclinical cerebrovascular disease and neurodegenerative processes may play a relevant pathophysiological role. This study aimed to assess the association between vascular risk factors, dementia-related biomarkers, and cerebral small vessel disease (cSVD) burden, estimated by white matter hyperintensities (WMH) and the Fazekas score, in patients with LOE compared to healthy controls. Methods This prospective, single-centre study (2020–2026) included 47 patients with temporal lobe epilepsy of unknown etiology with onset after age 50 (LO-TLE) and 32 healthy controls. All participants underwent 3T brain MRI with a dedicated protocol including 3D FLAIR sequences; white matter hyperintensity burden was systematically assessed using the Fazekas score, and patients were stratified into three groups according to their score: Fazekas 0 (absent or minimal WMH), Fazekas 1 (mild WMH), and Fazekas 2 (moderate WMH). Clinical, demographic, vascular risk factor, and lifestyle data were collected. In patients, cerebrospinal fluid (CSF) Alzheimer's disease biomarkers were measured (Aβ1-40, Aβ1-42, Aβ1-42 / Aβ1-40 ratio, t-Tau, p-Tau181, p-Tau181 / t-Tau ratio); all participants underwent plasma biomarker assessment (p-Tau217, NfL, Aβ1-42, p-Tau217 / Aβ1-42 ratio). Statistical analyses were performed using parametric or non-parametric tests as appropriate; significance threshold was set at p < 0.05. Results Patients were older than controls (68.1 ± 8.8 vs. 63.3 ± 8.7 years; p = 0.021), with comparable sex distribution. WMH burden was significantly greater in patients than in controls. Specifically, Fazekas score 0 was observed in 29.8% of patients versus 65.6% of controls, Fazekas score 1 in 51.1% versus 31.2%, and Fazekas score 2 in 19.1% versus 3.1%, respectively (χ² = 11.12; p = 0.004). Among vascular risk factors, hyperlipidemia was associated with a higher prevalence of Fazekas score 1 in patients (p = 0.009), while no significant associations were found with hypertension, diabetes, smoking, TIA/stroke, or other comorbidities. Plasma NfL levels were significantly higher in patients than in controls (p < 0.001). Among CSF biomarkers in patients, Aβ1-42 (p = 0.011) and Aβ1-40 (p = 0.019) differed across Fazekas subgroups, both with elevated levels in Fazekas 1 versus Fazekas 0. In contrast, no significant differences in the others CSF biomarkers and in all the evaluated plasma biomarker levels were observed across Fazekas score subgroups. Conclusions WMH burden is significantly greater in patients with LO-TLE compared to healthy controls, suggesting a relevant contribution of cSVD to the pathogenesis of late-onset epilepsy even in the absence of clinically manifest cerebrovascular events. The association between hyperlipidemia and greater WMH supports the involvement of modifiable vascular risk factors. Elevated plasma NfL in patients indicate a possible link between neurodegeneration, white matter integrity, and cortical hyperexcitability. These findings reinforce the hypothesis of a convergence between subclinical vascular pathology and neurodegenerative processes in LO-TLE and suggest that targeting vascular risk factors in middle-to-late adulthood may help reduce its incidence. Larger longitudinal studies are needed to clarify causal relationships and the prognostic relevance of the identified biomarkers.