Blurring of the grey–white matter boundary in the ipsilateral temporal pole is frequently reported in patients with hippocampal sclerosis (HS) but its cause has been uncertain. Competing explanations include seizure-driven disruption of cortical myelination during development (developmental disruption hypothesis), degeneration related to chronic seizures (seizure-driven degeneration hypothesis), or spread/extension of the primary HS pathology such as focal cortical dysplasia (shared pathology hypothesis). Prior work relied on reader-dependent visual classification in small, single-age cohorts. The present study quantified temporopolar blurring across a multi-centre cross-sectional cohort to adjudicate these hypotheses.
The study analysed T1-weighted MRI from 154 patients with histopathologically confirmed HS (median age 27.5 years; IQR 18.4–38.0 years) and 118 healthy controls (median age 15.3 years; IQR 12.0–24.8 years) collected at four centres. The investigators computed surface-based grey–white matter contrast and used depth-dependent intensity sampling to localize where signal changes occurred relative to the grey–white boundary.
Three explicit models produced distinct subject-level and spatial predictions:
The authors combined four topographic predictors to test these models: regional myelination timing, geodesic proximity to the hippocampus, molecular similarity to hippocampal tissue, and functional connectivity to the hippocampus.
Grey–white matter contrast measures on T1-weighted images were compared between patients and controls. Depth-dependent sampling assessed signal change at multiple distances from the cortical boundary, allowing localization of abnormalities to cortex or superficial white matter layers.
The four topographic predictors were entered into a regression framework to explain regional variation in temporopolar blurring. The model's overall explained variance was evaluated and statistical significance assessed using a spatial spin permutation approach (reported as Pspin). The relative contribution of each predictor was quantified using dominance analysis to estimate percentage contribution to the explained variance.
Grey–white matter contrast was reduced in the ipsilateral temporal pole and entorhinal cortex in patients with HS. Ninety percent of patients were below the 5th centile of control values for temporopolar contrast, indicating a common and substantial imaging abnormality.
Depth-dependent sampling localized the effect primarily to a white matter hypointensity approximately 1 mm below the grey–white boundary (U = 1768, P < 0.001), implicating superficial white matter rather than cortical grey matter as the principal locus of signal change.
Blurring showed a positive association with earlier epilepsy onset (r = 0.336, P < 0.001). There was no significant relationship between temporopolar blurring and epilepsy duration (r = −0.117, P = 1.000). Correlation with hippocampal atrophy did not reach conventional significance (r = 0.206, P = 0.071). The presence of FCD type IIIa did not associate with increased blurring (U = 2953, P = 0.981).
The combined topographic model explained 36% of the variance in temporopolar blurring (Pspin = 0.007). Dominance analysis attributed the largest proportion of explained variance to regional myelination timing (45.1%), followed by geodesic proximity to the hippocampus (25.6%). Molecular similarity and functional connectivity contributed the remaining portion of explained variance.
The spatial localization to superficial white matter and the stronger association with earlier epilepsy onset — but not with epilepsy duration, hippocampal atrophy, or FCD IIIa — argue against a primarily progressive degenerative process or simple extension of hippocampal pathology. Instead, the findings support the developmental disruption hypothesis: early seizures occurring during periods of ongoing cortical myelination are the best explanation for the observed temporal pole blurring in HS, particularly in cortex near the affected hippocampus.
Temporopolar blurring on routine T1-weighted MRI appears to be a common marker in HS, driven by superficial white matter changes consistent with disrupted myelination. This characterization may inform interpretation of MRI in people with early-onset epilepsy and HS and suggests a developmental window during which seizures can alter cortical myelination. The topographic analysis framework combining myelination timing and proximity to hippocampus may guide further mechanistic and longitudinal work.
Code for the analyses is available at the project GitHub repository cited by the authors. Requests for access to extracted surface-based features can be made via the study contact email provided in the source. Ethical approvals were obtained from the UK Health Research Authority and institutional review boards at participating centres; informed consent was waived for this retrospective secondary analysis of anonymised MRI data. Funders declared include Epilepsy Research UK and the Wellcome Trust. The authors declared no competing interests.
In a multi-centre cohort of patients with histopathologically confirmed HS, temporal pole blurring is common and driven by a superficial white matter hypointensity. The pattern and clinical associations are best explained by early seizure-related disruption of cortical myelination in regions proximal to the affected hippocampus, rather than by progressive degeneration or extension of primary hippocampal pathology.