To the Editor: Lamotrigine has been used successfully as an adjunctive treatment to antipsychotic medication in refractory schizophrenia (1). Topiramate is a newer antiepileptic agent with a wider spectrum of action than lamotrigine. Unlike lamotrigine, topiramate also blocks AMPA/kainate receptors, thus decreasing glutamate-mediated excitation (2). We report our experience using topiramate to treat extremely refractory schizophrenia.

We treated three men and two women (mean age=40.2 years, SD=6.1) who met DSM-IV criteria for chronic schizophrenia. We obtained oral consent from the subjects or their caregivers for their participation in the study. The mean time since their initial hospitalization for schizophrenia was 19.8 years (SD=5.8). The mean length of their current hospitalization was 8.2 years (SD=5.4). Four patients were taking clozapine. The fifth was taking quetiapine and risperidone, since he had developed agranulocytosis while taking clozapine. No patient had epilepsy. We started the patients with topiramate, 50 mg/day, and titrated their dose upward at the rate of 50 mg/day per week. Their mean maximum dose was 250 mg/day (range=200–300). We held their main medication dose constant while they were receiving topiramate. We administered the Positive and Negative Syndrome Scale for schizophrenia (3) before the patients started taking topiramate and after they had been taking the maximum dose for 1 month.

The condition of the man not taking clozapine and one woman deteriorated to the point that we could not obtain reliable posttreatment scores on the Positive and Negative Syndrome Scale. For the remaining three patients, the mean initial score on the Positive and Negative Syndrome Scale was 95 (SD=1). Their mean posttreatment score was 118 (SD=5). This represented a significant negative reaction to the addition of topiramate (paired t=8.55, df=2, p<0.007, one-tailed). Of note, both positive and negative scores on the Positive and Negative Syndrome Scale increased during posttreatment for all three of these patients.

In retrospect, there were reasons for concern about a negative response. Topiramate has a higher rate of psychotic episodes in epileptic patients than other newer antiepileptics (4). According to the glutamate model of schizophrenia, inhibition of glutamate receptors in the nucleus accumbens and prefrontal cortex should increase positive and negative symptoms, respectively (5, 6). We know of no current clinical evidence that the countertherapeutic effects we observed are mediated by this mechanism.

Topiramate has been suggested as adjunctive therapy to reduce clozapine-induced weight gain and seizures (7, 8). Our observations suggest caution when considering use of topiramate in patients with refractory schizophrenia.