LQTS (long QT syndrome) is caused by mutations in cardiac ion channel genes; however, the prevalence of LQTS in the general population is not well known. In the present study, we prospectively estimated the prevalence of LQTS and analysed the associated mutation carriers in Japanese children. ECGs were recorded from 7961 Japanese school children (4044 males; mean age, 9.9+/-3.0 years). ECGs were examined again for children who had prolonged QTc (corrected QT) intervals in the initial ECGs, and their QT intervals were measured manually. An LQTS score was determined according to Schwartz's criteria, and ion channel genes were analysed. In vitro characterization of the identified mutants was performed by heterologous expression experiments. Three subjects were assigned to a high probability of LQTS (3.5< or = LQTS score), and eight subjects to an intermediate probability (1.0< LQTS score < or =3.0). Genetic analysis of these II subjects identified three KCNH2 mutations (M124T, 547-553 del GGCGGCG and 2311-2332 del/ins TC). In contrast, no mutations were identified in the 15 subjects with a low probability of LQTS. Electrophysiological studies showed that both the M124T and the 547-553 del GGCGGCG KCNH2 did not suppress the wild-type KCNH2 channel in a dominant-negative manner. These results demonstrate that, in a random sample of healthy Japanese children, the prevalence of a high probability of LQTS is 0.038% (three in 7961), and that LQTS mutation carriers can be identified in at least 0.038% (one in 2653). Furthermore, large-scale genetic studies will be needed to clarify the real prevalence of LQTS by gene-carrier status, as it may have been underestimated in the present study.