Risperidone, Tardive Dyskinesia, and the Elderly

To the Editor: Dilip V. Jeste, M.D., et al. (1) provided good evidence for the use of risperidone to reduce psychotic symptoms and to prevent and treat tardive dyskinesia in elderly patients with moderate to severe dementia. However, there are problems with their methods, most notably with their analysis of the patients who did not complete the study. They state that of the 330 patients enrolled in an open-label trial of risperidone, 133 (40.3%) completed the 12-month trial. Few reasons were given for patients dropping out, except that nine patients stopped taking risperidone because of extrapyramidal symptoms. The remaining 188 (57.0%) remained unaccounted for at the end of the study.

It is unclear if they dropped out from the study because of other toxic side effects, withdrawal of consent, worsening cognitive impairment, or intercurrent physical illness or if they simply were not deemed to need risperidone any more. The mortality rate of such an elderly group (mean age=82.5 years) with advanced dementia could have been predicted to be high but was not stated by the authors. Hence, the low risk of tardive dyskinesia is presented in isolation from the overall tolerability of risperidone, and a judgment cannot be made about its overall safety.

It is also unclear exactly how many patients the authors were able to follow up in order to assess the presence of tardive dyskinesia. One must assume that all 255 patients who did not initially have tardive dyskinesia must have been assessed for it because the authors did not explicitly state how many did not participate. Therefore, at the 1-year follow-up, the authors concluded that six patients developed emergent tardive dyskinesia during the study and gave an incidence rate of 2.6% of the patients treated for a year. However, only a minority actually received treatment for this long.

Alternatively, the incidence of tardive dyskinesia could be calculated in the 133 patients receiving risperidone at the end of the study. It would not be possible to give an exact rate of tardive dyskinesia on the basis of these figures, as the relative completion rates of those with and without tardive dyskinesia at the beginning of the study were not provided. However, the number would be in excess of 4.5%. Such a distinction is important, as the length of antipsychotic treatment is associated with the development of tardive dyskinesia in the first few years of treatment (2).