To measure and compare medial proximal tibial angle (MPTA) of lower limbs under different axial rotation angles(neutral position, 30° internal rotation, 30° external rotation) on the load position radiographs, and explore changes and significance of MPTA measured within and between groups of tibia at different axial rotation positions.From January 2018 to December 2018, 40 patients with knee osteoarthritis (KOA) were selected, with a total of 80 limbs, including 12 males and 28 females, aged from 29 to 73 years old with an average of (59.6±12.7) years old. Full length radiographs of the lower limbs were taken on neutral tibia position, 30° internal rotation and 30° external rotation, respectively. MPTA was measured and the results were compared between groups and within groups.MPTA measured on the left lower extremity of neutral tibia, 30° internal rotation and 30° external rotation were (86.08±2.48) °, (88.62±2.94) ° and (83.47±3.10) °, respectively. MPTA measured on the right lower limb were (86.87±1.97) °, (89.02±2.39) ° and (83.80±2.77) °, respectively, and there were no significant difference in MPTA measured between rotation angle group (P>0.05). While there were statistical difference in MPTA on the same limb between groups (P<0.05). On 30° internal rotation, MPTA of left and right lower limbs increased by (2.54±1.74) ° and (2.15±1.78) ° compared with tibia neutral position. On 30° external rotation, MPTA of left and right lower limbs decreased (2.61±2.03) ° and (3.07±1.75) ° compared with tibial neutral position.When a full-length X-ray film is taken on the weight-bearing position of both lower limbs, if there is axial rotation or external rotation of tibia, MPTA will increase or decrease compared with neutral position, which may cause a certain degree of deviation in clinical operation based on the accurate measurement of MPTA. However, the extent to which this bias affects the clinical operation effect remains to be verified. In addition, limited by the total number of samples and the number of measurement groups, whether there is a linear relationship between MPTA deviation and tibial axial rotation needs to be further studied.