Minimally invasive treatment of displaced femoral shaft fractures with a teleoperated robot-assisted surgical system.

BACKGROUND Minimally invasive surgical operation of intramedullary (IM) nailing is a standard technique for treating diaphyseal fractures. However, in addition to its advantages, there are some drawbacks such as the frequent occurrence of malalignment, physical fatigue and high radiation exposure to medical staff. The use of robotic and navigation techniques is promising treatments for femoral fractures. MATERIALS AND METHODS This paper presents a novel robot-assisted manipulator for femoral shaft fracture reduction with indirect contact with the femur. An alternative clinical testing model was proposed for orthopedic surgeons to practice femoral fracture reduction. This model imitates the human musculoskeletal system in shape and functional performance. The rubber tube simulate muscles providing contraction forces, and the silicone simulates passive elasticity of muscles. Two-group experiments were performed for studying feasibility of the teleoperated manipulator. RESULTS The average operative time was about 7min. In the first group experiments, the femur axial, antero-posterior (AP) and lateral views mean errors were 2.2mm, 0.7mm and 1.1mm, respectively, and their maximums were 3.0mm, 0.9mm and 1.5mm; the mean errors of rotation were 0.8° around x-axis, 1.6° around y-axis, 2.0° around z-axis, and their maximums were 1.1°, 2.2°, 2.9°, respectively. For the second group experiments, the femur axial, AP and lateral views mean errors were 1.8mm, 0.4mm and 0.8mm, respectively, and their maximums were 2.2mm, 0.7mm and 1.1mm; the mean errors of rotation were 1.2° around x-axis, 1.6° around y-axis, 1.9° around z-axis, and their maximums were 2.4°, 1.8°, 2.7°, respectively. Reduction for AP view displacement is easier than lateral (p<0.05) because of the tube-shaped anatomy and the muscle contraction forces. Errors around x-axis are smaller than those around y-, and z- axes (p<0.05), i.e., electro-mechanical actuator is easier to control than pneumatic. CONCLUSION An experimental model for simulating human femoral characteristics was proposed. Experiments conducted on the artificial lower limb model demonstrated high reduction accuracy, safety, sufficient working space, and low radiation exposure of the proposed robot-assisted system. Thus, the minimally invasive teleoperated manipulator would have greater development prospect.