A randomized prospective clinical trial of carbamazepine or phenytoin for 6 or 24 months including a control no-treatment group is reported in 276 post-craniotomy patients at the Walton Hospital, Liverpool, England. 103 (37%) patients suffered at least 1 seizure. Of those who developed status epilepticus in the first week after operation, 8% were treated with carbamazepine, 5% with phenytoin, and 2% had no treatment. Serum anticonvulsant levels were not monitored in all patients. In those who were monitored the levels were never in the optimal range, particularly in the phenytoin treated group. Early anticonvulsant treatment did not affect the long-term response of epilepsy to anticonvulsant drugs. The occurrence of seizures within the first post-operative week did not increase the likelihood of late epilepsy. Acute allergic skin rashes occurred in 28 (13%) patients. The authors concluded that prophylactic anticonvulsants should not be recommended routinely following supratentorial craniotomy. [1]

COMMENT. The authors admit that the small and uncertain effect of antiepileptic drug treatment in this study might be related to non-compliance and failure to achieve optimal blood levels. Of those blood levels available a significant proportion of phenytoin levels were suboptimal, and it could be argued that the more rigorous monitoring might have improved results. In contrast, satisfactory carbamazepine levels were achieved and yet, there was no difference in outcome. These findings suggest that phenytoin has a greater prophylactic action than carbamazepine in postcraniotomy seizure control. According to Drs. Wingkun and Awad at the Cleveland Clinic, changes in antiepileptic drugs, dosages or poor compliance may be important factors in postoperative recurrence of seizures (see Ped Neurol Briefs Jan 1992; 6:8). The unusually high incidence of skin rashes associated with anticonvulsant treatment in this study suggests that postcraniotomy patients are more susceptible than non-surgical patients with epilepsy. The respective frequency of skin rash with the two anti-convulsants used in this study was not stated. It would be of interest to know if the antibiotic Cefotaxim was administered to these surgical patients in conjunction with the anticonvulsant, since this combination of drugs is known to increase the risk of serious skin reactions (see Ped Neurol Briefs May 1992; 5:37).