Background: Chronic subdural hematoma (cSDH) is a common neurosurgical condition with high recurrence rates following surgical evacuation, largely due to persistent inflammation and neovascularized membrane formation. Middle meningeal artery embolization (MMAE) is a minimally invasive adjunctive therapy that targets membrane vascularity to reduce re-bleeding and promote hematoma resolution. However, the effect of membrane presence on outcomes following MMAE remains unclear. This study aimed to assess whether membrane presence on pre-procedural imaging is associated with improved hematoma resolution, reduced recurrence, and decreased need for repeat surgical evacuation in patients undergoing combined surgical evacuation and adjunctive MMAE. Materials and Methods: In this retrospective cohort study, 691 patients (849 cSDHs) from the international MESH registry underwent surgery plus MMAE between January 2018 and December 2024. Membranous hematomas were identified on CT by neuroradiologists blinded to outcomes. Primary outcomes included hematoma resolution (>60% and >80% reduction in thickness), recurrence, and repeat surgical evacuation. Multivariable logistic regression adjusted for clinical and imaging variables. Results: Of 849 hematomas, 460 were membranous and 389 non-membranous. Membranous cSDHs were initially thicker but showed greater resolution (>60%: 71% vs. 59%, p80%: 51% vs. 41%, p=.004) and lower recurrence (1.8% vs. 4%, p=.05). Repeat evacuation was less frequent in the membranous group (4.9% vs. 7.4%, p=.18). Membrane presence independently predicted higher resolution (aOR 1.54; 95% CI: 1.25-1.89; p<.001) and lower recurrence (aOR 0.50; 95% CI: 0.31-0.81; p=.005). Conclusion: Membrane presence on imaging is associated with improved radiographic outcomes and reduced recurrence following surgery with adjunctive MMAE. A non-significant trend toward fewer repeat surgeries was also observed, suggesting membrane status may serve as a prognostic imaging biomarker in cSDH management.
Musmar et al. (Thu,) studied this question.