Juliette-Agathe Mirtil, Jeremy Deverdun, Liesjet Eh van Dokkum, Olivia Portalier, Bénédicte Maréchal, Tommaso Di Noto, Nicolas Menjot de Champfleur, Anne Ducros, Vincent Costalat, Federico Cagnazzo, Emmanuelle Le Bars
Targeted CSF leak closure induced a rapid, subtle reorganization of intracranial compartments, primarily characterized by ventricular expansion alongside gray matter and brainstem volume contraction. Gray matter volumetric modifications were detectable even in people with low-intermediate Bern scores. This establishes quantitative volumetry as an objective monitoring tool for treatment evaluation across the full spectrum of SIH severity.
BACKGROUND AND PURPOSE: To characterize brain morphometric adaptations following targeted CSF leak closure using automated volumetry.
MATERIALS AND METHODS: In this retrospective, single-center, longitudinal study, 61 patients with confirmed spontaneous intracranial hypotension (SIH) who underwent targeted treatment (neurosurgical repair for dural tears; endovascular embolization for CSF-venous fistulas) were included. Brain MRI was performed at baseline, 24 hours post-treatment (n=41), and at 3-month follow-up. Bern scores and automated volumetric Z-scores of CSF, gray matter, white matter, ventricular compartments, and brainstem subregions were derived using the DL-MorphoBox research application and compared longitudinally with Wilcoxon signed-rank tests and Bonferroni correction for multiple comparisons.
RESULTS: Marked radio-clinical improvement was observed at 3 months post-treatment, with increased total CSF volume (ΔZ=+0.66; p<0.001) across all ventricular compartments (p<0.001). In parallel, brain parenchymal volume decreased (ΔZ=-0.60; p<0.001), presumably driven by gray matter contraction (ΔZ = -0.57; p<0.001), involving the temporal (ΔZ =-0.46; p < 0.001), frontal (ΔZ=-0.64; p=0.01), and prefrontal (ΔZ=-0.37; p=0.007) cortices. Finaly, brainstem volume decreased as well (ΔZ=-0.27; p=0.04). The MRI-based volumetric changes were detectable as early as 24 hours post-treatment, with lateral ventricular volume expansion continuing to 3 months. While the Bern score correlated with ventricular volume variations (p<0.001), it was unrelated to gray matter contraction.
CONCLUSIONS: Targeted CSF leak closure induced a rapid, subtle reorganization of intracranial compartments, primarily characterized by ventricular expansion alongside gray matter and brainstem volume contraction. Gray matter volumetric modifications were detectable even in people with low-intermediate Bern scores. This establishes quantitative volumetry as an objective monitoring tool for treatment evaluation across the full spectrum of SIH severity.