When a patient undergoes hip replacement surgery, the ultimate aim is for the implant to outlive them and give them a pain-free lifestyle. Increasingly, younger people are receiving the surgery, which puts more demand on the longevity of the implant and the stresses likely to be placed on it.
A number of factors influence the performance of hip replacements. To date, experimental investigations have been limited to analysing a single situation.
A project by the School of Engineering Sciences at the University of Southampton will apply engineering principles to create a computer model to identify which implants will perform best. It will take into account variations in parameters, such as the properties of bone, the loading conditions and the surgical technique in a single analysis.
It will also study cementless implants, which rely on the implant attaching itself to the bone biologically, and which are more suitable for younger patients.
According to Dr Martin Browne, reader in biomaterials science at Southampton, applying engineering reliability algorithms could improve the outlook for implants.
‘In a previous project that looked at the structural integrity of knee replacements, it became evident that because of the safety-critical nature of implants, they tend to be over-designed,’ Browne said. ‘In an aircraft, you might have a safety factor on the load that a wing can take, but there was no real justification as to what safety factors are appropriate for implants.
‘As hip and knee implants are safety critical, and could cause pain and discomfort to a patient if they fail, I identified the potential for what are known as reliability methods that use probability as a way of identifying a particular safety factor.
‘In the body, there are tens of variables that could apply, such as the varying strengths and densities of the bone and tissues. To take all these into account, rather than look at every possible combination, we use probabilistic methods.’
These techniques were originally used in the civil engineering, aerospace and maritime industries. This project will apply them in the biomedical and bioengineering field, taking into account more of the biological factors.
‘The main reason why a hip replacement fails is aseptic loosening, where loosening occurs without any infection,’ said Browne. ‘This project relates mechanical factors and biological factors to loosening, to find out what factors cause it and how to reduce the incidence.’
A misaligned implant can cause increased stress in the bone, which can eventually lead to loosening. The tool will identify what kind of implant is more robust and able to tolerate a misalignment.
‘It will determine if a certain implant is fitted to any group of patients and there were variations in its alignment, the probability is less that it would fail,’ said Browne.
To build the model, the team will test an implant with a known, successful clinical history against a less successful one. It will then introduce biological factors by looking at patients’ bones with different geometries from CT scans.
‘The more CT scans we have, the more variables we can look at,’ said Browne. ‘It’s very difficult to do an experimental model for every different scenario, so we can do a selected group of tests on standard materials. There are artificial bones that we can use as a baseline just to verify results, but eventually it will all have to be computationally driven.’
The initial three-year programme aims to have an experimental model validated and running. ‘We are hypothesising that successful designs are those which are tolerant to variables in patient and surgical factors, so we want to find an approach for assessing this robustness,’ said Browne. ‘In the shorter term we want to show we can demonstrate robustness for certain implants. If that is successful, we can provide prosthesis manufacturers with data that can help them with design strategies for their implants.
‘In the long run, it would be great for the surgeon to have this system look at a patient’s geometry and find that this implant would best fit.’
The methods to be applied are being used by partner DePuy International, a manufacturer of orthopaedic devices. Finsbury Orthopaedics will provide data to increase the accuracy of the model, to be based on software developed at Southampton, which converts CT scans into finite element models.
‘If we can demonstrate our model on the hip, it should work on any of the other load-bearing joints,’ said Browne. ‘Once we’ve completed the algorithm and proved it works on the hip, there’s no reason not to apply it to other implants.’
A computer modelling system may soon identify which hip implants are most likely to succeed for the long term