The role of the direct directed transfer function in identifying the primary epileptogenic zone from generalized sharp and wave discharges in Lennox-Gastaut syndrome

Abstract

Purpose: In intractable epilepsy, the patient can be freed of seizures by removing the epileptogenic zone. However, identifying the primary epileptogenic zone is particularly difficult in Lennox-Gastaut syndrome (LGS), despite current multimodal approaches. I identified the primary epileptogenic zone via brain connectivity of the generalized sharp and wave discharge (GSW). I also reviewed the differences between LGS patients who had the epileptogenic zones removed and those who did not.Methods: I reviewed LGS patients who underwent surgical treatment from 2005 to 2013. I separated them into a group (Group A) who underwent resective surgery for epileptogenic zones (N=12) with good surgical outcome, and a group (Group B) who only underwent corpus callostomy and retained independent bilateral epileptogenesity (N=15). I analyzed the GSW in preoperative electroencephalography to identify the primary epileptogenic areas by using direct directed transfer function (dDTF)- based a multivariate autoregressive model. I compared the areas identified by direct directed transfer function with the resection areas in Group A and with the postoperative EEG in Group B.Results: The results of dDTF showed localization or lateralization in 83.3% of Group A, and while bilateral or multifocal localization in 93.3% of Group B (p<0.01). The localization shown by dDTF included resective areas in all patients and agreed with resective areas in 58.3% of Group A. Among areas identified by dDTF, frontal area was localized at 91.7% and 100%, while extra-frontal areas were localized at mean 33.3% and 24.5% in group A and group B, respectively. The diagnostic sensitivity of dDTF for lateralization and localization was 83.3% and 62.5%, respectively, and the specificity was 86.7% and 63.2%, respectively.Conclusion: Analyzing the GSW by using the dDTF might be a valuable approach for the identification of the primary epileptic zone in LGS, alongside a multimodal approach.