Objective This study attempts to analyse the biomechanical effect of internal fixation (plated in parallel or plated vertically) on the basis of distal humeral fractures on musculoskeletal multibody dynamics using AnyBody in Finite Element Method. Method Humeral 3D models were reconstructed by MIMICS after volunteers' CT image input in *.dicom format, and processed by Geomagic Studio for surfaces, while locking plates and screws were then designed by Pro-E. A humeral model of T-type fracture was created and assembled in Hypermesh, to integrate fixtures (e.g., MPL/PML/ML), to grid the mesh and then assign materials. A musculoskeletal model of the upper limb was established by AnyBody to simulate elbow flexion and extension. They were finally imported to Abaqus for boundary conditions and dynamic analysis. Result In terms of Von Mises stress, its maximum increased and then decreased gradually during the joint motion, but p > 0.05 in SPSS suggests no significant difference for all three fixtures. In terms of displacement, when the elbow was at 90 degrees, each motional pattern reached its peak as follows: ML180 degrees = 0.28 mm, MPL90 degrees = 0.49 mm & PML90 degrees = 0.54 mm during flexion; ML180 degrees = 0.073 mm, MPL90 degrees = 0.10 mm & PML90 degrees = 0.12 mm during extension. p p = 0.007 < 0.01667 suggests the significant difference between the two fixations, for example, PML90 degrees and ML180 degrees, indicating that the peak displacement of ML180 degrees is less than that of PML90 degrees. Conclusion After generally analysed in musculoskeletal dynamics, the biomechanical property of the fixtures was presented as follows: the displacement of the parallel plate was less than that of the vertical, and the parallel plate may optimise the clinical reduction anatomically.