BACKGROUND: Recent regulatory and clinical concerns have brought into sharp focus antipsychotic drug-induced QTc interval prolongation, torsades de pointes, and sudden cardiac death. Several new generation (atypical) antipsychotic drugs have either been withdrawn from clinical use or delayed in reaching the marketplace due to these concerns. Because torsades de pointes is rarely found, QTc interval prolongation serves as a surrogate marker for this potentially life-threatening arrhythmia. Current methods of calculating this electrocardiographic parameter have limitations. The primary care physician is a key member of the team managing a patient who requires administration of antipsychotic drugs. This article focuses on new generation antipsychotic drugs and principles useful to both the primary care physician and the psychiatrist. METHOD: PubMed was searched in September 2002 using the terms antipsychotic drug and QT interval. References were examined from review articles describing antipsychotic drugs and the QT interval. The author's files gathered over the past 20 years on the QT interval were also reviewed. RESULTS: Nine cases were available in which drug-induced QTc interval prolongation was associated with new generation antipsychotic drug administration. Eight cases were taken from the literature, and the author added one additional report. The newer agents involved were risperidone, quetiapine, and ziprasidone. In at least 8 cases, there was evidence of other risk factors associated with QTc interval prolongation. In one case frequently referenced in the literature, the authors misunderstood their own data showing that QTc interval prolongation did not relate to delayed ventricular repolarization. In another instance, 2 authors reported on the same patient, with important information missing from both articles. No evidence of torsades de pointes appeared in any of the 9 cases. CONCLUSIONS: No evidence is currently available in the literature implicating new generation antipsychotic drugs in the production of torsades de pointes. However, the absence of such evidence does not prove that newer antipsychotic drugs do not cause torsades de pointes. Among patients free of risk factors for QTc interval prolongation and torsades de pointes, current literature does not dictate any specific consultative or laboratory intervention before administering new generation antipsychotic drugs. When risk factors are present, evaluation and intervention specific to those risk factors should dictate the clinician's course of action. More specific guidelines for monitoring the QT interval and risk of torsades de pointes await improved methods of measuring the QTc interval relevant to each patient.