The efficacy and side effects of intermittent oral diazepam for the prevention of febrile seizure recurrence were investigated in the Departments of Clinical Pharmacology, Neurosurgery, and Biostatistics, University of Tours, France. Between May 1985 and June 1988, 185 children, eight months to three years of age, with a first febrile seizure and normal neurological development were randomly assigned in a double-blind fashion to receive oral diazepam (0.5 mg/kg, then 0.2 mg/kg every 12 hours) or placebo, whenever the rectal temperature was more than 38°C. The duration of the study was three years and eight different centers in France participated. There were 462 febrile episodes and 1000 days with prophylactic treatment. The seizure recurrence rate in the diazepam group was 16% and in the placebo group 19.5% and the difference was not significant. Those with recurrent seizures were aged 17 ± 6.9 months and those with no recurrence were 21 ± 8.5 months at the time of the first seizure. In children with seizure recurrence, diazepam had been correctly administered in only one of 15 children and the placebo had been given in seven of 18 children with recurrences. Poor compliance with the intermittent treatment was explained by 1) seizure as the first manifestation of the fever in seven cases in each group, 2) noncompliance by parents in nine cases, and 3) patient refusal to take treatment in two. Hyperactivity was more frequent as a side effect in the diazepam than in the placebo group. The findings reflect a lack of efficacy of the intermittent method of treatment rather than the diazepam itself. 
COMMENT. The poor compliance rate which might explain the lack of efficacy of the intermittent diazepam prophylaxis in this study is disappointing and dampens the growing enthusiasm for this form of therapy as a substitute for continuous phenobarbital prophylaxis. The conclusion of Dr. Rosman's controlled study of oral diazepam is eagerly awaited.
In a recent national survey of pediatric neurologists, 22% recommended the use of intermittent prophylactic therapy with diazepam in an average dose of 0.46 mg/kg/day; oral diazepam was preferred by 8% and rectal administration was used by 16%. An alternate drug, lorazepam or clonazepam, was preferred by 2%.