New plating techniques, such as non-contact plates, have been introduced in acknowledgment of the importance of biological factors in internal fixation. Knowledge of the fixation stability provided by these new plates is very limited and clarification is still necessary to determine how the mechanical stability, e.g. fracture motion, and the risk of implant failure can best be controlled. The results of a study based on in vitro experiments with composite bone cylinders and finite element analysis using the Locking Compression Plate (LCP) for diaphyseal fractures are presented and recommendations for clinical practice are given. Several factors were shown to influence stability both in compression and torsion. Axial stiffness and torsional rigidity was mainly influenced by the working length, e.g. the distance of the first screw to the fracture site. By omitting one screw hole on either side of the fracture, the construct became almost twice as flexible in both compression and torsion. The number of screws also significantly affected the stability, however, more than three screws per fragment did little to increase axial stiffness; nor did four screws increase torsional rigidity. The position of the third screw in the fragment significantly affected axial stiffness, but not torsional rigidity. The closer an additional screw is positioned towards the fracture gap, the stiffer the construct becomes under compression. The rigidity under torsional load was determined by the number of screws only. Another factor affecting construct stability was the distance of the plate to the bone. Increasing this distance resulted in decreased construct stability. Finally, a shorter plate with an equal number of screws caused a reduction in axial stiffness but not in torsional rigidity. Static compression tests showed that increasing the working length, e.g. omitting the screws immediately adjacent to the fracture on both sides, significantly diminished the load causing plastic deformation of the plate. If bone contact was not present at the fracture site due to comminution, a greater working length also led to earlier failure in dynamic loading tests. For simple fractures with a small fracture gap and bone contact under dynamic load, the number of cycles until failure was greater than one million for all tested constructs. Plate failures invariably occurred through the DCP hole where the highest von Mises stresses were found in the finite element analysis (FEA). This stress was reduced in constructions with bone contact by increasing the bridging length. On the other hand, additional screws increased the implant stress since higher loads were needed to achieve bone contact. Based on the present results, the following clinical recommendations can be made for the locked internal fixator in bridging technique as part of a minimally invasive percutaneous osteosynthesis (MIPO): for fractures of the lower extremity, two or three screws on either side of the fracture should be sufficient. For fractures of the humerus or forearm, three to four screws on either side should be used as rotational forces predominate in these bones. In simple fractures with a small interfragmentary gap, one or two holes should be omitted on each side of the fracture to initiate spontaneous fracture healing, including the generation of callus formations. In fractures with a large fracture gap such as comminuted fractures, we advise placement of the innermost screws as close as practicable to the fracture. Furthermore, the distance between the plate and the bone ought to be kept small and long plates should be used to provide sufficient axial stiffness.