Sir: The grand rounds report of catatonia by Carey et al (Psychiatric Bulletin, February 2002, 26, 68-70) is a useful reminder that our knowledge of catatonia has progressed since its 19th-century delineations. However, the report only partially reflects this progress.

Catatonia is not rare when patients are systematically assessed for motor abnormalities. Kraepelin reported 20% of his patients with dementia praecox to be catatonic (see Reference Fink and TaylorFink & Taylor, 2003) and surveys since 1990 find that about 10% of acutely hospitalised psychiatric patients meet DSM criteria for catatonia (Reference Bush, Fink and PetridesBush et al, 1996a ).

The official linking of catatonia and schizophrenia in classification systems is a misreading of the literature. Kahlbaum described the characteristic motor signs of catatonia among patients diagnosed as suffering from both mood and general medical illnesses (see Reference Fink and TaylorFink & Taylor, 2003). Kraepelin and Bleuler incorporated the syndrome to serve their view of dementia praecox (see Reference Fink and TaylorFink & Taylor, 2003). Since 1920, however, studies of catatonic subjects find 40-60% with an underlying mood disorder, whereas only 15% have schizophrenia. The clinical information in the report suggests that the patient's 1998 condition was a psychotic depression, for which his clinicians later prescribed lithium.

The report cites the vigorous use of antipsychotic drugs leading to neuroleptic malignant syndrome (NMS). The authors are unsure as to the relation between NMS and catatonia, first suggesting NMS to be distinguishable from catatonia and later stating that ‘features common to both catatonia and NMS are increasingly recognised, with NMS felt to closely represent advanced catatonia’. We agree with the latter interpretation as attempts to demarcate the two syndromes have failed and the syndrome is induced by drugs other than neuroleptics. The more parsimonious view, based on the similarity of signs, symptoms and response to treatments, argues that NMS and malignant catatonia are best considered as one disorder (Reference FinkFink, 1996).

Successful interventions for catatonia began with the demonstration in 1930 that intravenous barbiturates resolve catatonic stupor. In 1935, patients with catatonic schizophrenia were reported to recover when treated with chemically induced seizures (now electroconvulsive therapy (ECT)). The barbiturates were replaced by the benzodiazepines that are now reported as effective in 80% of catatonic patients (Reference Bush, Fink and PetridesBush et al, 1996b ). When these drugs fail, especially in patients with malignant catatonia or delirious mania, ECT is remarkably effective (Reference FinkFink, 1999).

The use of antipsychotic drugs in patients with catatonia is problematic. A malignant syndrome is associated with antipsychotic drugs, especially among patients with a medical illness, fever and dehydration. Rising serum creatine-phosphokinase and falling serum iron levels are findings that antecede the emergence of the malignant catatonia/NMS syndrome. The present report illustrates this hazard, as high potency antipsychotic drugs prescribed during the acute psychotic episode were associated with an NMS syndrome, relieved as ‘benzodiazepines and anticholinergic medication were required on a number of occasions’. With ECT, the patient ‘improved swiftly and substantially’. The continuation treatment with antipsychotic drugs was not helpful, requiring a second course of ECT. The avoidance of potent antipsychotic drugs and the prescription of diazepam probably contributed to his ongoing well-being.