Francesco M C Lioi, Benedetta Giordano, Mauro Palmieri, Alessandro Pesce, Ludovica Chessa, Alessandro Al Giabri, Davide Lucantoni, Lorenzo De Luca, Maurizio Salvati, Antonio Santoro, Alessandro Frati
PTBE was associated with seizure-related rather than other symptomatic presentation. The retrospective design, binary edema assessment, incomplete volumetric data, and limited number of seizure events preclude causal interpretation and precise estimation of effect magnitude.
PURPOSE: Peritumoural brain edema (PTBE) is associated with preoperative seizures in meningioma, but its association with seizure-related rather than other symptomatic presentation is less clear. We examined this relationship and associations of PTBE with tumour volume and anatomical location.
METHODS: We retrospectively analysed 213 patients with surgically treated intracranial meningiomas. PTBE was classified on preoperative T2-weighted MRI by two independent neurosurgical reviewers, with disagreements resolved by consensus. The primary analysis compared seizure-related with other symptomatic presentation using Firth penalized logistic regression adjusted for skull-base status. Models incorporating tumour volume were treated as sensitivity analyses.
RESULTS: PTBE was present in 113 of 202 evaluable tumours (55.9%). Its frequency increased from 25.0% in tumours <10 cc to 82.9% in those >40 cc, and each doubling in tumour volume was associated with approximately twice the odds of PTBE (OR 2.00, 95% CI 1.51-2.65; p < 0.001). Among 153 symptomatic patients, seizure-related presentation occurred in 5 of 69 PTBE-negative (7.2%) and 22 of 84 PTBE-positive cases (26.2%). After adjustment for skull-base status, PTBE remained associated with seizure-related presentation (Firth OR 3.65, 95% profile-likelihood CI 1.39-11.13; p = 0.008). The association persisted when incidentally detected lesions were retained (OR 3.53, 95% CI 1.38-10.57; p = 0.007).
CONCLUSION: PTBE was associated with seizure-related rather than other symptomatic presentation. The retrospective design, binary edema assessment, incomplete volumetric data, and limited number of seizure events preclude causal interpretation and precise estimation of effect magnitude.