A Comparison Between Seizure Manifestations and Outcome in Frontal Lobe Epilepsy of Childhood Onset

Abstract

Purpose: The seizure manifestations of frontal lobe epilepsy (FLE) vary considerably, and the correlation between the ictal symptoms and seizure prognosis has not yet been fully elucidated. We retrospectively compared seizure manifestations with outcomes of the seizures in FLE of childhood onset. Methods: The subjects were 20 (11 male and nine female) patients diagnosed with FLE based on seizure manifestations, interictal or ictal EEG findings, and neurologic imaging, including magnetic resonance imaging (MRI), computed tomography (CT), or positron emission tomography (PET) or a combination of these. They were followed for 3 years. The age at onset of epilepsy ranged from 3 months to 11 years (mean, 3 years), and that at the last visit from 5 to 23 years (mean, 16 years). Eight patients had mental retardation. The MRI revealed structural abnormalities in six patients. Ten patients were considered to have symptomatic FLE, and the remaining 10, cryptogenic FLE. We examined the presence of the following manifestations: gestural automatism, postural seizures, secondarily generalized seizures (SGSs), vocalizations, facial expressions, eye deviation, motion arrest, staring, generalized tonic seizures, focal tonic seizures, twitching of the face, twitching of eyelids, contraction at the lip corner, abnormal sensations of the limb, swallowing, gustatory signs, atonic seizures, salivation, and other autonomic manifestations. Seizure prognosis was evaluated on the basis of seizure frequency over the last 2 years, according to the following criteria: no seizures without medication (I), no seizures with medication (11), one to several seizures in a year (III), in a month (IV), in a week (V), and in a day (VI). In addition to correlations between seizure prognosis and seizure manifestation, we examined the relation of the ictal symptoms with intelligence, the presence of seizure clustering, and etiology. The Mann‐Whitney U test was used for statistical analysis. Results: Gestural automatisms were found in 14 (70%) patients, including complex gestural automatisms in five (25%), followed by generalized tonic seizures in 10 (50%), vocalizations in nine (45%). SGSs in seven (37%), swallowing in seven (37%), facial expressions in six (30%), and contraction at lip corner in four (20%).Postural seizures were not seen in any of the patients. Fifteen patients had a history of seizure clustering of at least one per patient. Seizure prognosis was generally poor. Only three patients had remained free from seizures for 2 years at their last visit. Seizure outcome was significantly poorer in patients with clustering of seizures or with gestural automatisms (p < 0.05).The presence of other ictal manifestations, mental retardation, or the etiology of epilepsy were not associated with poor control of seizures. Conclusions: The seizure outcome of patients with FLE of childhood onset was generally poor. In particular, seizure clustering or gestural automatisms were predictive factors for poor seizure control.