A finite element strategy applied to intramedullary nailing of the proximal femur

Abstract

An intramedullary nail is a trauma treatment device used for fracture fixation of long bones. These devices are subject to failure, including lag screw cut-out and failure at the lag screw insertion hole from high stress concentrations in that region. Clinical developments for such devices are frequently based on a trial and error method, which often results in failure before improvement. However, the finite element method can be used for the development of trauma treatment devices, and their interaction with bone, by providing a large data set at a relatively low cost. Also, parameters can be changed to assess the relative benefits of one device to another. A novel finite element model has been developed that can be used for the analysis of intramedullary nails inserted into long bones. A commercially available finite element package, ANSYS, has been used to implement the modelling strategy. The finite element modelling technique has been applied to fractures of the proximal femur, but the model is generic, and can be developed to deal with any form of intramedullary device where contact between the bone and implant is important. The finite element strategy can be used in pre-clinical trials to test a new device, or for the design optimisation of existing devices. The finite element model consists of the device surrounded by a thin layer of bone, which forms a 'base' model component that is re-usable. This 'base' component can be mathematically connected to any long bone model, forming an integrated implant and bone construct. The construct can be used to assess which device is best suited to a particular fracture, for example. Contact elements have been used to allow stresses to develop as contact is achieved within the implant and bone construct. Pre-assignment of contact points is not required. Verification of the finite element model is achieved by comparison to available data from experiments carried out on constructs of bone and device that use intramedullary femoral nails. In this thesis the finite element model has been applied to two areas of proximal femoral nailing. The finite element model is used to analyse the distal end of a Gamma nail, and shows that analyses that do not consider contact may not lead to accurate predictions of stresses. The model has been developed for using configurations with one and two distal locking screws. The most distal locking screw is more critical under axial loading, and the more proximal screw is more important for bending loads. The use of 'softer' screws distributes the load more evenly between them. The finite element model has been used to investigate the mechanical environment of a fracture callus for a femoral neck fracture, and a subtrochanteric fracture. The use of one and two lag screws, fracture gap size and material properties of the nail have been investigated for a stiffening callus. Results show that the use of two lag screws for a neck fracture provides a more rigid support at the early stages of fracture healing, and minimises stress-shielding once the callus has healed. For subtrochanteric fractures there is a critical point at which the fracture callus is able to carry any load. A Titanium nail significantly reduces the peak stress at the lag screw insertion hole, and titanium lag screws share the load more evenly between them. Each two-lag-screw configuration used transfers a similar load into the fracture callus. A configuration using a larger lag screw above a smaller has a significantly higher stress at the upper lag screw insertion hole. Critically, the load shared between two lag screws changes as the fracture callus stiffens and an assessment should be made at different stages of fracture healing to optimise the use of a device.