To the Editor: G. Bondolfi, M.D., and colleagues (1) recently reported that risperidone was as effective as clozapine in reducing psychopathology in neuroleptic-resistant schizophrenia on the basis of an 8-week, double-blind, multicenter trial conducted in Europe with 86 chronically ill patients with schizophrenia. Other open studies have also suggested the effectiveness of risperidone for some patients with schizophrenia whose positive symptoms did not respond to typical neuroleptics (2–5). Risperidone and clozapine have also been reported to be equally effective with regard to the reduction of psychopathology in neuroleptic-responsive patients with schizophrenia (6).

It is important to carefully examine studies of the relative ability of these two drugs in these two groups of patients with schizophrenia, since definitions and degrees of neuroleptic resistance may vary widely (7) and the psychotic symptoms of neuroleptic-intolerant patients may be equally responsive to the antipsychotic effects of the antipsychotic agents under study, while they may differ markedly in their effectiveness for neuroleptic-resistant patients. Thus, the fact that Dr. Bondolfi et al. (1) included both neuroleptic-resistant and neuroleptic-intolerant patients in their study is a potential confound. Data on the response of neuroleptic-resistant patients should have been reported separately but were not. Even had they been, the overall group size was small and probably lacked sufficient power to find a difference between the two drugs for either or both subgroups. Second, the determination of neuroleptic resistance included in this study was made retrospectively and included trials of only 4 weeks’ duration—too brief a period. A third concern is the rapid titration schedule for clozapine and the low final dose (300 mg/day) achieved, both of which should decrease its efficacy. The method of use and dose of risperidone (2–6 mg/day) may have been optimal. The low dose of risperidone no doubt contributed to the finding that motor side effects were less with risperidone than with clozapine. Fourth, the duration of the study may have been too short to find a difference between the two drugs. Clozapine has been shown to require up to 6 months to achieve its full benefits (8, 9). And finally, it is important to note the difficulty of maintaining a blind study with clozapine and risperidone because of their differences in side effects.

It can be safely concluded from this study that a significant proportion of the patients, some of whom may have been neuroleptic-resistant, did respond well to risperidone and had very few side effects. It would be important to know specifically how many of the patients were neuroleptic-resistant. A larger 6-month study of only neuroleptic-resistant patients with schizophrenia—shown to be so with a final run-in trial with typical neuroleptic drugs, a slower titration of clozapine, and multiple, fixed doses of risperidone or clozapine—is needed to confirm that risperidone is as effective as clozapine in treating neuroleptic-resistant patients.