A Prewarning Signalling in Motor Evoked Potentials Monitoring for Insular Glioma Surgery: A Preliminary Study
Year: 2025
Type: article
Study design: Observational Study
Abstract: OBJECTIVE: Insular gliomas are surgically challenging. Intraoperative neurophysiological monitoring assists surgical excision. Although still controversial, a reduction of Motor Evoked Potential (MEPs) signal>50% correlates with postoperative impairment. However, this reduction often occurs too late. Identifying pre-warning indicators, anticipating MEPs drop>50%, could optimize surgery of insular gliomas. METHODS: Monocentric retrospective study, collecting clinical, surgical and neurophysiological data of patients harboring insular intra-axial lesions. I) MEPs tracking included the baseline, compared with all subsequently recorded signals. Upper limb muscles were analyzed. II) The Fourier transform was used to modulate the recorded signals. III) The dynamic time warping was then applied followed by logistic regression. RESULTS: Out of 300 patients operated on for supratentorial lesions (January 2023-September 2024), 30 patients with insular tumors were finally enrolled. The efficacy of the logistic regression model resulted in high accuracy [91.45%]: it allowed us to identify a pre-warning signal (PWS), anticipating the decrease in MEPs>50% (i.e. warning sign = WS), by considering new parameters of DTW. Comparing MEPs tracks and the identified parameters, our results revealed that PWS compared to WS has the same Se[83.3%]and negative predictive value [93.8%] in identifying MEPs drop and correlating with the postoperative clinical outcome. Compared to WS, PWS has superimposable Sp and positive predictive value correlating with prediction of motor deficits (Sp:62.5%, positive predictive value: 35.7%). Therefore, PWS may be more reliable than WS to interpret intraoperative monitoring variations and predict the clinical outcome. CONCLUSIONS: Based on our preliminary data, PWS is reliably predictive of motor outcome, and more accurate than WS to identify intraoperative monitoring variations.
Source: World Neurosurgery
Authors: Camilla Bonaudo, Sara Lombardi, Giulia Sorbi, Eleonora Visocchi, Manuel Camelia, Anastasia Galletti, Alice Esposito, Agnese Pedone, Fabrizio Baldanzi, Simone Troiano, Cristiana Martinelli, Riccardo Carrai, Francesca Battista, Giovanni Muscas, Simone Orlandini, Luca Campagnaro, Andrea Boschi, Serena Tola, Antonello Grippo, Leonardo Bocchi, Alessandro Della Puppa
Language: English
Subfield: Surgery
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
Estimated APC paid: $3,943
OpenAlex ID: W4411334083