In fetal echocardiography, conventional parameters for assessing cardiac function are limited because of limited echocardiographic windows or the fetus' position. We aimed to evaluate the feasibility and reproducibility of fetal left ventricular (LV) twist by two-dimensional, speckle-tracking echocardiography (2DSTE) in a Japanese population.We included 55 normal fetuses at gestational ages between 21 and 36 weeks. Subjects with adverse maternal health issues were excluded. LV twist was calculated as the net difference between LV basal and apical rotation at end-systole estimated with 2DSTE.We were able to analyze the 2DSTE images in 44 cases (80%). The mean (±SE) apical rotation, basal rotation, and LV twist were 7.88 ± 0.77, −3.68 ± 0.50, and 11.1 ± 0.75 degrees, respectively. We could not analyze 11 cases (20%) because of poor image quality due to fetal position in five cases (45.5%), failure to track the endocardium because of blurred images in five cases (45.5%), and failure to obtain images of the heart due to the presence of the placenta in front of the fetus in one case (9.1%). There were no significant differences in the demographic data between pregnant women in whom LV twist analysis was feasible and not feasible. The intra-and interobserver intraclass correlation coefficients were 0.67 and 0.64, respectively.LV twist analysis by 2DSTE in the fetus was feasible in a substantial population and may provide new insight into cardiac function during the prenatal period. On the other hand, its reproducibility was moderate and needs to be improved.(Int Heart J 2019; 60: 671-678)