Objective To evaluate the application of 1H-MRS in chronic liver diseases related with hepatitis B, and to investigate the method of eliminating breathing influence to proton MR spectroscopy (1H-MRS) on liver. Methods Liver spectroscopy of 46 subjects were obtained by single-voxel PRESS sequence with respiratory gating with GE 1.5T scanner. The subjects were divided into 4 groups: the normal male group (10 cases), the normal female group (10 cases), the male hepatitis B cirrhosis group (14 cases) and the male liver cancer after hepatitis B group (12 cases) according to clinical data, MR manifestations and pathologic findings. Glutamine and glutamate complex (Glx) peak height served as a reference standard to judge the choline (Cho) peak height. The differences of Cho peak height of each group were compared. Results The peaks of Cho were lower than those of Glx in normal male group. The peaks of Cho were higher than those of Glx in 8 women of normal female group. The peaks of Cho were higher than those of Glx in 11 patients of the male hepatitis B cirrhosis group. The peaks of Cho were higher than those of Glx in 10 patients of the male liver cancer after hepatitis B group. The differences of Cho peak were statistically significant between normal male group and normal female group, the male hepatitis B cirrhosis group and the normal male group, the male liver cancer after hepatitis B group and the normal male group. Conclusion The peaks of Cho in normal males are lower than those of Glx. The peaks of Cho in the majority of normal females are higher than those of Glx. The peaks of Cho increase in male hepatitis B cirrhosis and liver cancer after hepatitis B. Respiratory gating can effectively eliminate the impact of breathing movement on proton MR spectroscopy of liver.