Available data on antipsychotic-induced metabolic risks are often constrained by potential confounding effects due to prior antipsychotic treatment. In this study, we assessed the baseline prevalence of metabolic abnormalities and changes following treatment with five commonly-used antipsychotic drugs (haloperidol, amisulpride, olanzapine, quetiapine or ziprasidone) in first-episode, partially antipsychotic-naive patients with schizophrenia in the European first-episode schizophrenia trial (EUFEST). Overall baseline prevalence of metabolic syndrome (MetS) was 6.0%, with similar rates observed in the antipsychotic-naive patients (5.7%, 9/157) and in the other patients with only a brief prior exposure to antipsychotics (6.1%, 20/326). These results are consistent with the MetS prevalence rate estimated in a general population of similar age. Examination of individual risk factors showed 58.5% of subjects had one or more elevated metabolic risks at baseline: 28.5% demonstrated suboptimal HDL; 24.2% hypertension; 17.7% hypertriglyceridemia; 8.2% abdominal obesity; 7.3% hyperglycaemia. Increase in body weight (kg/month) occurred in patients treated with haloperidol (0.62 S.E. 0.11), amisulpride (0.76 S.E. 0.08), olanzapine (0.98 S.E. 0.07) and quetiapine (0.58 S.E. 0.09), which was significantly greater than that in the ziprasidone group (0.18 S.E. 0.10). The incidence rate of new diabetes cases over a 52-wk follow-up period was 0.82% (4/488). More patients experienced worsening rather than improvement of hypertriglyceridemia or hyperglycaemia in all treatment groups. Our findings suggest that in first-episode, partially antipsychotic-naive patients, the baseline prevalence rate of MetS appears to be no higher than that in the general population, but serious underlying individual risk factors nevertheless existed.