Fluctuation of serum phenytoin concentrations during autologous bone marrow transplant for primary central nervous system tumors.

We reviewed our experience for adult patients receiving oral anticonvulsant therapy during high-dose chemotherapy and autologous bone marrow re-infusion for primary malignant tumors of the central nervous system. Nineteen patients received either iv carmustine (BCNU) 900-1050 mg/m2 and 6120 cGy cranial irradiation (N = 10), iv carmustine 900-1050 mg/m2 and iv cisplatin 200 mg/m2 (N = 8), or iv carmustine 600 mg/m2, iv cisplatin 200 mg/m2, and iv etoposide 2400 mg/m2 (N = 1). Anticonvulsant therapy consisted of phenytoin alone (N = 8), phenobarbital alone (N = 4), carbamazepine alone (N = 2), phenytoin and carbamazepine (N = 2), carbamazepine and phenobarbital (N = 1), and no anticonvulsant therapy (N = 2). Serum anticonvulsant concentrations were monitored frequently and doses adjusted to keep values in the therapeutic range. While phenobarbital and carbamazepine doses remained relatively stable, all patients required increased doses of phenytoin anticonvulsant therapy after beginning chemotherapy (mean onset 3.7 days after initiation of chemotherapy). The increase in phenytoin dose ranged from 50% to 300% above baseline (mean 134%). By the time of discharge from the hospital (approximately 3-4 weeks after the start of chemotherapy) anticonvulsant dose was decreased to near pre-therapy levels. These swings coincided with the initiation of dexamethasone therapy for antiemetic effect and were more pronounced in patients also receiving cisplatin therapy. Due to close monitoring of serum phenytoin concentrations, no instances of toxicity due to excessive drug concentration, or seizures due to subtherapeutic doses, were noted in patients with primary CNS malignancies. Serum phenytoin concentrations fluctuate markedly during high-dose chemotherapy and must be analyzed frequently during the course of therapy.