Risk of excessive weight gain in epileptic children treated with valproate.

We sought to identify factors associated with excessive weight gain in children treated with valproate, excluding patients fed by gastrostomy or treated with medications known to affect appetite (eg, stimulants). Weight and height were recorded before treatment and at the time of follow-up; a measure of adiposity, body mass index, was computed and expressed in kg/m2, and weight and height for age were converted to Z-score. Putative risk factors included sex, age at start of treatment, monotherapy at start of treatment, duration of follow-up, mental retardation, seizure type (generalized or partial), etiology (idiopathic or cryptogenic versus remote symptomatic), and dose of valproate. Fifty-five children (30 girls, 25 boys), ranging in age at the start of therapy from 1.8 to 16.9 years were followed for 8.6 to 33.8 months. Forty-three patients had primarily generalized seizures, 34 had idiopathic or cryptogenic epilepsy (including 30 with generalized idiopathic epilepsy), and 18 had mental retardation. Valproate was the first antiepileptic drug for 21 patients, and 43 were on monotherapy at the time of follow-up. Height Z-score decreased significantly in girls but was stable in boys. There was a significant increase in body mass index and in weight Z-score. Body mass index was greater than the 90th percentile for age in 14 patients at the start of treatment and in 20 patients at follow-up. Decrease in height Z-score was significantly correlated with female sex and duration of follow-up. Changes in weight Z-score and body mass index were significantly correlated with initial weight Z-score and initial body mass index, respectively, but not with age at start of therapy, duration of follow-up, sex, seizure type, etiology, dose of valproate, or monotherapy.

Risk factors for status epilepticus in children with symptomatic epilepsy.

The purpose of this study was to determine risk factors for status epilepticus (SE) in children with symptomatic epilepsy through a retrospective case-control study. Patients (44 children with a prior diagnosis of symptomatic epilepsy experiencing one or more episodes of SE between January 1, 1991, and June 1, 1995) were matched for age at follow-up to controls (88 children with symptomatic epilepsy without SE during that interval) and medical records were reviewed. Patients and controls did not differ in etiology or in age at epilepsy onset (1 year 5 months [SD, 2 years 3 months] versus 1 year 3 months [SD, 1 year 5 months]). Univariate analysis revealed significant associations between SE and several factors, including history of a first seizure as SE, partial epilepsy, partial seizures with secondary generalization, focal paroxysmal abnormalities and focal background abnormalities on EEG, and generalized abnormalities on neuroimaging. Groups did not differ in the proportion of patients on polypharmacy or with subtherapeutic serum antiepileptic drug levels, but there was a trend for greater use of phenobarbital in patients (57% versus 38%, odds ratio (OR) = 2.6, p = 0.057). With multiple logistic regression, four factors emerged as independent predictors of SE risk: focal background EEG abnormalities (OR = 6.51, p = 0.0005), partial seizures with secondary generalization (OR = 4.61, p = 0.0021), first seizure as SE (OR = 3.99, p = 0.034), and generalized abnormalities on neuroimaging (OR = 2.85, p = 0.034). These four factors are indicators of a higher risk of SE in children with symptomatic epilepsy.