A prospective study of the efficacy of seizure cessation by phenobarbital versus phenytoin administration utilized both clinical and electroencephalographic expressions of seizure behaviors. The phenomenon of uncoupling was defined as the persistence of electrographic seizures despite the suppression of >or=50% clinical seizures after either one or both antiepileptic drugs use. Fifty-nine neonates (25 to 43 weeks estimated gestational age) with electrically-confirmed seizures were assigned to either of two drugs and continuously monitored over a 24-hour period. Nine of the fifty-nine patients had only electrographic seizure expression both before and after drug administration. Of the remaining 50 patients who had both electrical and clinical seizure expression before treatment, 24 infants responded to the first choice of an antiepileptic drug with no further seizures. Fifteen of the remaining 26 infants (58%) with persistent seizures after treatment had uncoupling of electrical and clinical expressions of seizures; no difference in the uncoupling effect was noted for neonates who were treated with either antiepileptic drug or based on prematurity or gender. Serial electroencephalographic monitoring helps document continued electrographic seizure expression after antiepileptic drug use, following complete or partial suppression of clinical seizure behaviors.