SEEG negative motor response sites occurred at the expected rate in the clinical epileptogenic zone

We asked whether stereo-electroencephalography (SEEG) sites that produce a negative motor response (NMR) to stimulation occur less often in the clinical epileptogenic zone (EZ) or seizure-onset zone (SOZ), or more often in white matter (WM), than adjacent electrode pairs from the same patient session. We analyzed metadata only from OpenNeuro ds006519 v1.0.0: stimulation events and clinical electrode flags. The pre-specified comparison used each patient's adjacent same-shaft pair rate as the null and exact Poisson-binomial tests, Holm correction across three tests, and patient-level sign-flip checks. The electrode table omitted one or both stimulated contacts for 13 of 35 events; this exclusion was not specified in the plan. Among events with listed contacts, the EZ analysis included 21 events from 14 patients: 4 were in the EZ versus 5.62 expected (p = 0.28; Holm p = 0.84; patient sign-flip p = 0.77). The SOZ analysis also included 21 events from 14 patients: 2 were in the SOZ versus 2.03 expected (p = 0.67; Holm p = 0.84; sign-flip p = 0.80). The WM analysis included 22 events from 15 patients: 6 were in WM versus 3.96 expected (two-sided p = 0.37; Holm p = 0.84; sign-flip p = 0.27). These data do not show that NMR sites avoid epileptogenic tissue. Four EZ-positive events occurred in four patients, so complete EZ avoidance is not compatible with these observations. Only effective stimulations were available, the sample was small, and clinical zone labels were supplied by the dataset authors. The analysis cannot compare effective with ineffective stimulation sites.

stereo-electroencephalography · negative motor response · epilepsy · epileptogenic zone · seizure-onset zone · white matter · electrical stimulation · OpenNeuro · metadata analysis