Abstract Drugs with diverse structures and from several therapeutic classes are reported to increase the risk that a patient will experience ventricular tachyarrhythmias (e.g., torsades de pointes [TdP]) during drug therapy. This review discusses the use of preclinical assays to assess the risk that a QT-prolonging drug will cause TdP. The mechanisms underlying the development of TdP and the factors that increase the risk of TdP are described and applied to the design of preclinical experimental models for detection of proarrhythmic drug actions. Recommended assays, conditions, and preparations for preclinical assessment of the drug-induced risk to TdP are given. No single preparation can simulate all conditions that cause TdP in patients. However, the assays described herein are capable of detecting the proarrhythmic effects of currently used drugs, even when these effects are reported to be extremely rare in clinical practice.