Glymphatic system dysfunction in epilepsy: a systematic review and meta-analysis

The glymphatic system mediates clearance of metabolic waste via astrocytic aquaporin-4-mediated cerebrospinal fluid-interstitial fluid exchange. Structural prerequisites for glymphatic dysfunction are established in epilepsy, yet no quantitative neuroimaging synthesis exists. We performed a systematic review and meta-analysis to quantify DTI-ALPS-derived proxy marker of glymphatic impairment in epilepsy and assess its clinical correlates.

Following PRISMA 2020 guidelines and PROSPERO registration (CRD420251275885), PubMed, Embase, Web of Science, and Scopus were searched through December 2025 for human observational studies reporting DTI-ALPS indices in epilepsy with healthy controls. Pooled mean differences and correlation coefficients were estimated under random-effects models with REML. Newcastle-Ottawa Scale tools were used for quality assessment. Publication bias was assessed via Egger's test and trim-and-fill analysis.

Seventeen studies were included; fifteen (n = 1,255; 675 patients with epilepsy, 580 controls) contributed to the primary meta-analysis. DTI-ALPS values were significantly reduced in epilepsy versus controls (MD -0.157, 95% CI -0.197 to -0.116; p < 0.0001, I² = 73%). Reductions were consistent across epilepsy subtypes, but subgroup analyses were underpowered and should be interpreted cautiously (Q = 3.43, p = 0.18). Disease duration showed the strongest inverse correlation (r = - 0.392; p = 0.0014), while age at epilepsy onset showed a modest association (r = - 0.148; p = 0.020). Trim-and-fill-adjusted estimates remained significant (MD -0.107, 95% CI -0.156 to -0.059).

DTI-ALPS, as an indirect proxy marker of glymphatic function, is consistently reduced in epilepsy syndromes and correlates with disease chronicity. These findings support DTI-ALPS as a candidate neuroimaging biomarker of cumulative glymphatic burden; however, its current biomarker potential is limited by substantial heterogeneity, possible medication confounding, and the exclusive East Asian representation of available evidence. Longitudinal, multicenter studies with standardized protocols across diverse populations are needed to determine its reproducibility.