QT-interval duration is shorter in men than in women. Estrogens do not significantly influence the duration of repolarization. The effect of testosterone has not been studied directly in humans. The aim of this study was to assess the effects of testosterone on corrected QT duration in hypogonadic men. Eleven hypogonadic men were enrolled in this prospective interventional study. Digital electrocardiograms were recorded for each participant at 3 levels of testosterone, high, medium, and low, after a single intramuscular administration of testosterone. Heart rate-independent assessment of QT-interval duration was used. QT(1,000) (QT at 60 beats/min) was determined for each subject. Total blood testosterone and the ratio of testosterone to sex hormone-binding globulin were assessed at each visit. The median values of QT(1,000) were 352 ms (interquartile range 340 to 363), 357 ms (interquartile range 349 to 367), and 363 ms (interquartile range 357 to 384) at high, medium, and low testosterone concentrations, respectively (p <0.013 for the 3 comparisons). A maximal mean difference of 13.6 +/- 2.8 ms (p = 0.0007) was observed between high and low levels of testosterone. A negative linear relation was found between QT(1,000) and testosterone concentration (p = 0.0001) or the ratio of testosterone to sex hormone-binding globulin (p = 0.004). In conclusion, the difference in QT-interval duration between men and women might be explained by differences in testosterone levels, and testosterone shortens ventricular repolarization.