To the Editor: We support the view of Michael Alan Taylor, M.D., and Max Fink, M.D. (1), that catatonia should be considered as an individual category in psychiatric diagnostic systems.

First, we would like to comment on the proposed categories for the diagnostic classification of catatonia. The category "delirious catatonia" is confusing, since it is difficult to distinguish catatonic excitement from the excited states of bipolar disorder. As a consequence, the category "delirious catatonia" shows an important overlap with the DSM category of the manic episode. An alternative classification could be composed of two categories, "nonmalignant catatonia" and "malignant catatonia," with an additional specifier of "retarded" or "excited." In this classification, the Kahlbaum syndrome would be regarded as a retarded form of nonmalignant catatonia, and delirious mania would be classified as an excited form of nonmalignant catatonia. The classic description of "lethal catatonia" would correspond with the excited malignant catatonia, whereas neuroleptic malignant syndrome could be considered a retarded variant.

Second, the authors argued that exposure to an atypical antipsychotic drug usually worsens catatonia, but the scientific evidence to which they refer for this statement is poor. Several cases in which exposure to an atypical antipsychotic drug led to an improvement or to the remission of nonmalignant catatonia have been reported. In some of these cases, a causal relationship is probable. For instance, in the case studies by Cook et al. (2) and by Hesslinger et al. (3), a decrease in the dose of risperidone was followed by a recurrence of symptoms and the subsequent increase in dose by remission.

Third, we would like to nuance the therapeutic effects of benzodiazepines in catatonia. According to Rosebush and Masurek (4), patients with schizophrenia are the least likely to respond to benzodiazepines; the response rates range from 40% to 50%.

Finally, Drs. Taylor and Fink (1) use a broad definition of catatonia, stating that neuroleptic malignant syndrome and the toxic serotonin syndrome are most likely severe forms of catatonia. Their evidence for this statement is based on the clinical similarities and on the responses to similar treatment strategies of these syndromes. However, in our view, the current pathophysiological knowledge of catatonia should be considered when establishing the diagnostic validity of the syndrome. In a description of a final common pathway of catatonia, one should consider, for instance, the work of Northoff (5). This author assumed that there is a "bottom-up" deregulation of the motor circuit in neuroleptic malignant syndrome as a result of the antipsychotic blockade of striatal dopamine D₂ receptors, which is in contrast to the "top-down" modulation as a result of a cortical γ-aminobutyric acid (GABA)-ergic alternation in catatonia.