Thirty-one children (5 months to 19 years) with medically intractable
childhood epilepsy had nitrazepam added to their antiepileptic drug (A
ED) regimen. All patients had previously received phenobarbital, pheny
toin, carbamazepine, valproate, and at least one benzodiazepine. Most
patients (22) had multiple seizure types, including atonic, tonic, myo
clonic, complex partial, and generalized tonic-clonic, whereas nine ha
d only one seizure type. Initially, 24 patients (77.4%) had a greater
than 25% decrease in seizure frequency, whereas 19 (61%) had an improv
ement of greater than 50%. Tolerance occurred in 12 of 23 (43.4%) resp
onders, at a mean period of 7 weeks. Eight of these 12 improved after
further nitrazepam dosage increases, maintaining a greater than 25% de
crease in seizure frequency as compared to baseline. On follow-up (mea
n, 5.4 months), 65% of our 31 patients maintained a seizure reduction
greater than 25%, and in 54% this reduction was greater than 50% (p le
ss-than-or-equal-to 0.001, McNemar's test). We conclude that nitrazepa
m is a useful adjunctive AED in intractable childhood epilepsy despite
the occurrence of tolerance in over one-third of the patients. Furthe
rmore, previous failure of benzodiazepine therapy does not preclude a
positive response to nitrazepam.