The prognostic value of seizure etiology, neurologic examination, EEG, and neuroimaging in the neurodevelopmental outcome of 89 term infants with neonatal seizures was determined at the Children’s Hospital and Harvard Medical School, Boston, MA. The seizure etiologies were global cerebral hypoxia-ischemia (HI) in 40%, focal HI in 18%, intracranial hemorrhage (17%), cerebral dysgenesis (5%), transient hypoglycemia or hypocalcemia (3%), meningitis or encephalitis (3%), and pyridoxine dependency (1%). Neurologic outcome at 1 year was favorable in 72%, and poor in 28%. Neurologic examination was abnormal in 54% (mild in 26% and severe in 22%) with motor impairment in 53%, mental impairment in 48%, and seizures after NICU discharge in 21%. Long-term outcome was poor in 28% of survivors; neonatal mortality was 7%. Risk factors for a poor outcome were seizures associated with cerebral dysgenesis or global HI, an abnormal EEG background activity, and multifocal cortical or deep gray matter neuroimaging lesions in the neonate. A favorable outcome at 12-18 month follow-up was predicted by a normal neurologic examination in the neonatal and early infancy period, and a normal/mildly abnormal neonatal EEG. An abnormal neurologic exam in the neonate was an unreliable predictor of outcome. 
COMMENT. Mortality is relatively low but long-term neurodevelopmental outcome is poor in 28% of infants with a history of neonatal seizures. Global cerebral hypoxicischemia is the most frequent cause of neonatal seizures and a strong predictor of poor longterm outcome. The effect of neonatal seizures on the developing brain is controversial, some finding them harmful  and some, harmless . Studies in experimental animals have shown that immature rats less than 20 days old respond to electroshock with hyperkinesias, breast stroke swimming movements, tremors, catatonia, and clonic movements, and a major tonic-clonic seizure with post-ictal depression could not be induced until the animal was older . Neonatal seizures in infants (Volpe JJ, 1977) are described as subtle and include the swimming movements resembling those seen in animals. The absence of generalized tonic-clonic seizures in the neonate reflects the lack of cortical organization required to propagate the electrical discharge (Aicardi J, 1986). These findings support those of Camfield and others who discount adverse effects of neonatal seizures per se, attributing the poor prognosis to the cause, especially cerebral hypoxiaischemia. Subtle and focal post-ictal cerebral ischemia affecting subcortical and limbic regions cannot be ruled out.
Relation of pregnancy and neonatal factors to development of childhood epilepsy. Prenatal factors contributed to the subsequent development of childhood epilepsy in a study in Nova Scotia, Canada . Risk factors for epilepsy included eclampsia, CNS anomalies, placenta abruptio, infection in pregnancy, and unmarried mother. Other factors associated with increased risk of epilepsy were neonatal seizures, neonatal metabolic disorders, low birth weight, and small for gestational age.