To the Editor: The issues raised by Drs. Duggal and Nizamie are too numerous and diffuse to address within the limitations of a letter, so I will direct my remarks to the principal focal point of their commentary, which concerns the role of dopamine antagonism in the pathophysiology of neuroleptic malignant syndrome.

First, Drs. Duggal and Nizamie are unimpressed by the data linking hypothermia to dopamine antagonism and wonder (if only rhetorically) why hypothermia is not observed more frequently in neuroleptic malignant syndrome. A good deal of controversy in the literature of neuroleptic malignant syndrome centers on which clinical features ought to be considered essential for diagnosis; hyperthermia—the most dramatic and prognostically important sign—is usually accorded this status. This prevailing view has almost certainly reduced the likelihood that any episode resembling neuroleptic malignant syndrome that involves normal or subnormal body temperatures will be recognized or reported. Without intending to, Drs. Duggal and Nizamie have asked an incisive question, which I attempted to answer by reexamining our original database (1).

Of the 65 suspected episodes of neuroleptic malignant syndrome involving 45 patients, 22 episodes (33.8%)—involving 20 patients (44.4%)—had at least one subnormal body temperature nadir recorded within 7 days of the date that the episode reached maximal intensity. Nadirs (lowest temperatures recorded on any given day) ranged from 93.3°F to 98.2°F. Diaphoresis was observed in 11 (50%) of the episodes and more frequently in episodes with the lowest temperature nadirs, although this difference was not statistically significant (median split: 63.6% and 36.4%, respectively; Pearson’s χ²=1.64, df=1, p=0.20). Episodes with observed diaphoresis had a lower mean temperature nadir than episodes without observed diaphoresis, but once again this difference failed to reach statistical significance (95.8°F, SD=0.81; 96.7˚F, SD=1.44, respectively) (p=0.13, n.s.). These data should be treated cautiously because temperatures were taken orally, so patients’ inability to cooperate would tend to bias measurements toward lower temperatures. However, these data suggest that hypothermia may be more common in neuroleptic malignant syndrome than has been appreciated and that excessive diaphoresis may play a role. As Drs. Duggal and Nizamie imply, oscillations between hypo- and hyperthermia in neuroleptic malignant syndrome would be consistent with the model that I have proposed.

Second, they disapprove of the emphasis given in that model to dopaminergic modulation of sympathetic activity at the level of the spinal cord, but in support of their view they offer only a dogmatic assertion that the "antipsychotic effect of neuroleptics is due to dopamine antagonism at the level of brain and not at the level of spinal cord." I am unaware of any data that would validate this arbitrary distinction, and it seems premature to exclude involvement of the lower central and/or peripheral nervous system from the phenomenology of psychotic disorders.

In any event, Drs. Duggal and Nizamie have misconstrued one of the main points of my article, which is that whatever contribution is made by antagonism of the hypothalamic dopamine receptors to the pathophysiology of neuroleptic malignant syndrome, it is insufficient in itself to account for most of the observed clinical phenomena.