To the Editor: Perminder Sachdev, M.B.B.S., M.D., Ph.D., F.R.A.N.Z.C.P., and colleagues R6215511CHDEACAE reported in a case-control study that ECT was a risk factor for neuroleptic malignant syndrome in their patient group. Since ECT has been used effectively to treat neuroleptic malignant syndrome R6215511CHDEAEEG, it seems quite unlikely that it could be causally related to subsequent development of the syndrome. I offer the following speculation about a correlational (as opposed to a causal) relationship between ECT and neuroleptic malignant syndrome. The patients in the group of Sachdev et al. who developed neuroleptic malignant syndrome were agitated and received large early parenteral neuroleptic doses. Even though the authors indicated that the comparison group was matched for diagnoses, it is possible, and even likely, that patients with a broad range of psychopathology were clumped into the same diagnostic groups. Thus, a group of "schizoaffective" patients may have vastly different current or past symptom pictures. Since patients with manic symptoms respond well to ECT, it could be that the patients in the group with neuroleptic malignant syndrome were thus more likely at some point in their illness course to have had ECT.