Phenobarbital and phenytoin are equally but incompletely effective as anticonvulsants in neonates. With either drug given alone, the seizures were controlled in fewer than half of the neonates.

No abstract

A double-blind crossover study with imipramine was conducted in 10 patients with absence and myoclonic-astatic seizures who had not responded to conventional medications. Imipramine produced a significant initial decrease in seizure frequency in 5 of the 10 patients, and in 2 patients the beneficial effect was maintained for more than 1 year. An open trial of imipramine in another 16 patients showed an initial reduction in seizure frequency in 10 patients (63 percent), and this decrease persisted for more than 1 year in 4 patients (25 percent). The effect of imipramine on the EEG did not always correlate with the clinical response. Serum content of imipramine in the patients who showed a long-term response was 40 to 120 ng per milliliter, on a total daily dose of 0.7 to 3.5 mg per kilogram. These results suggest that imipramine is a valuable addition to the treatment of seizures.

Neonatal seizures are frequently manifested by subtle movements that are referable to brain stem structure, ie, nystagmus, conjugate eye movements, posturing, sucking movements, and so forth. Electroencephalogram (EEG) confirmation of abnormal movements is essential in diagnosing seizures in the neonate. Clinical seizure signs are often a clue to etiology. Metabolic abnormalities must always be considered, and blood gases, calcium, magnesium, glucose, and ammonia obtained. Neonatal seizures are an indication for cerebrospinal fluid examination. Specific metabolic abnormalities are treated with metabolic approaches, not conventional anticonvulsant drugs. Hypertensive encephalopathy is treated by controlling blood pressure, and not through anticonvulsant drugs. Critically ill infants bind anticonvulsants in an unpredictable fashion, and unbound or free anticonvulsant drug concentrations should be used to guide therapy. Phenobarbital is the most commonly used drug in treating nonmetabolic seizures. Doses to achieve free concentrations of at least 35 mg/L should be used. Use in vitro binding determinations with this formula to calculate loading doses. Dose is 25 mg/kg multiplied by volume and distribution (1 L/kg) divided by % free. Phenytoin is the second most commonly used agent, and doses should be calculated to achieve, but not exceed, 3 mg/L. Dose is 3 mg/kg multiplied by volume and distribution (1 L/kg) divided by % free. Benzodiazepines, notably lorazepam and diazepam are used at doses of 0.15 mg/kg and 0.3 mg/kg, respectively.

Phenobarbital and phenytoin binding profiles were determined in 27 neonates. Binding of both drugs decreased compared with that in older subjects. In vitro binding of both agents correlated significantly with total protein and albumin concentrations. In vivo binding at 0.5 hours correlated significantly with birthweight and gestational age. Phenobarbital, but not phenytoin, binding decreased when three other therapeutic agents were concomitantly administered. Bilirubin concentrations, free fatty-acid concentrations, and pH values encountered in this population did not significantly influence binding. An in vitro binding profile accurately predicted in vivo free fractions (percent drug unbound) and plasma concentrations of both drugs.