Background: Complex cardiac deformation caused by simultaneous contraction and twist cannot be exactly evaluated with 2D technique due to out of plain motion. Recently, 3D-STE has been introduced to assess regional left ventricular function. The aim of study was to evaluate the 3D-STE by comparing the regional and global longitudinal function measurement against the 2D-STE values in a wide range of subjects Methods: Simultaneous 2D and 3D echocardiography using GE Vivid E9 system were performed in 26 normal subjects and 13 patients with RWMAs. 2D images were acquired in the apical 4, 2 and 3 chamber views. Volumetric data sets were subsequently recorded at a frame rate of 25짹8, by ECG gating over 4 cardiac cycles. Results: In overall subjects, global longitudinal strain (LS) determined by 2D and 3D were well correlated (-17.31짹4.0 vs -18.28짹4.6, R=0.84, p<0.001), and Bland-Altman analysis concurred with correlation analysis. However, only weak correlation was found in each segment of apical walls, but the correlation coefficients of mid wall segments were modest. Moreover, each segment of basal walls showed poor correlation and agreement between LS determined by 2D and 3D STE. LS of basal 6 segments were not statistically correlated between 2D and 3D STE. However, correlation between 2D and 3D STE in terms of mid wall 6 segments (R=0.83, p<0.001) and apical 4 segments (R=0.69, p<0.001) were excellent Conclusion Although 3D strain agreed well with 2D strain in the measurement of global longitudinal function, agreement between 2D and 3D in term of LS value of each segment was not good. Especially in basal segments, the correlation between 2D and 3D strain was poor. Therefore, our findings indicate that care must be taken when interpreting or comparing each regional strain value obtained using different (2D or 3D) methods. However, the value of global LS averaged from all segments can be interchangeable.