3D patient specific instrumentation for high tibial osteotomy

Abstract

High tibial osteotomy (HTO) is a surgical intervention for knee osteoarthritis (OA) by correcting the alignment of the knees. This is carried out by cutting the upper tibia at an angle to help redirect the load from the damaged cartilage, preventing further deterioration. This treatment helps to delay the patients from going through a total knee arthroplasty (TKA) or a knee replacement which may pose a higher risk of complications. As the medical industry gets more advanced with the use of patient specific instruments, patient can have tailor-made instruments which is precise and accurate while greatly improving the surgical outcome of each operation. The aim of this project is to experiment the methods to create the femur, patella, fibula, tibia cortical and tibia cancellous as accurate as possible through the 3D Slicer. The features of the bones, such as the tibial plateau datum, femur head center and the ankle joints, have to be preserved as best as possible as it is crucial for the measurements for HTO cuts. The bone models will be created using three-dimensional (3D) Slicer, meshing will be completed in MeshLab and the HTO measurements and cuts will be completed using SolidWorks. Bone models are created from computed tomography (CT) scans which are specimens provided by the Singapore General Hospital (SGH). The model is then meshed in MeshLab to reduce the number of faces and HTO cuts are then simulated in SolidWorks. The most time-intensive part of this project is creating the bone models as there are many alternatives to build it, with many unknown variables, like how the detection algorithm for the Hounsfield Units (HU) works, which helps to build the bone model. After a vast number of experiments with 3D Slicer, “Grow from Seeds” is proven to be more time efficient and accurate along with the “Logical Operators”, especially for a thin and uneven layer like the cortical region. With SolidWorks used preoperatively to simulate the HTO cuts and alignment, there are still limitations present. The creation of the locking plates and the wedges were unsuccessful due to the complexity of having it contoured to the anatomy of the patient. In conclusion, the 3D bone models produced are accurate for measurements with simulated OA conditions to be carried out in SolidWorks. The wedges and locking plates need further improvement to be used in simulation.