Sixty-two first-episode psychotic patients who were neuroleptic-naive were studied to examine predictors of acute dystonia after treatment with haloperidol. Twenty-three patients developed dystonia, two of them despite being treated with biperiden. Biperiden significantly prevented dystonic reactions. Dystonia development was significantly related to younger age, severity of illness, and negative symptoms at baseline and showed a trend to be related to positive symptoms as well. No significant effect of gender or diagnosis was found. The authors suggest that young, severely ill patients in their first psychotic episode who have never been treated with neuroleptics might be at higher risk to develop dystonia.