摘要 : Objective In patients with medically refractory epilepsy (MRE), interictal ripples (80–250Hz) are observed in large brain areas whose resection may be unnecessary for seizure freedom. This limits their utility as epilepsy biomarke... 展开 Objective In patients with medically refractory epilepsy (MRE), interictal ripples (80–250Hz) are observed in large brain areas whose resection may be unnecessary for seizure freedom. This limits their utility as epilepsy biomarkers for surgery. We assessed the spatiotemporal propagation of interictal ripples on intracranial electroencephalography (iEEG) in children with MRE, compared it with the propagation of spikes, identified ripples that initiated propagation ( onset‐ripples ), and evaluated their clinical value as epilepsy biomarkers. Methods Twenty‐seven children who underwent epilepsy surgery were studied. We identified propagation sequences of ripples and spikes across multiple iEEG contacts and calculated each ripple or spike latency from the propagation onset. We classified ripples and spikes into categories (ie, onset , spread , and isolated ) based on their spatiotemporal characteristics and correlated their mean rate inside and outside resection with outcome (good outcome, Engel 1 versus poor outcome, Engel≥2). We determined, as onset‐zone , spread‐zone , and isolated‐zone , the areas generating the corresponding ripple or spike category and evaluated the predictive value of their resection. Results We observed ripple propagation in all patients and spike propagation in 25 patients. Mean rate of onset‐ripples inside resection predicted the outcome (odds ratio = 5.37; p = 0.02) and correlated with Engel class (rho = ?0.55; p = 0.003). Resection of the onset‐ripple‐zone was associated with good outcome ( p = 0.047). No association was found for the spread‐ripple‐zone , isolated‐ripple‐zone , or any spike‐zone. Interpretation Interictal ripples propagate across iEEG contacts in children with MRE. The association between the onset‐ripple‐zone resection and good outcome indicates that onset‐ripple s are promising epilepsy biomarkers, which estimate the epileptogenic tissue better than spread‐ripples or onset‐spikes. Ann Neurol 2018;84:331–346 收起