Cases
Home Cases / Design of personalised titanium membrane for bone reconstruction in tumour surgery (Phidias newsletter Nr.4 2000)
Design of personalised titanium membrane for bone reconstruction in tumour surgery (Phidias newsletter Nr.4 2000)
Pattijn, V., Samson, I., Vander Sloten, J., Van Audekercke, R., De Buck, V. and Swaelens, B.
Introduction
Large or locally aggressive benign bone tumours frequently occur in the extremities just beneath the joint surface. The surgical treatment of most tumours consists of two stages: first the tumour is removed and controlled and second the defect is reconstructed. The reconstruction consists of the filling of the defect with bone cement or bone graft and in case of large defects adding internal fixation plates and/or nails. The major drawbacks of these techniques are the high fracture risk caused by early loading of the bone and the insufficient fixation of the filling material in the cavity.
Therefore we introduce a custom made preformed titanium "membrane" (i.e. a thin metal foil) produced pre-operatively to be attached at the periosteal surface of the bone around the cavity and fixed to the bone with small screws, during surgery. The membrane has a dual function, i.e. containing the filling material in the cavity and restoring the strength and stiffness of the bone to make an early normal functioning of the patient possible.
The aim of this study is first to look to the process of designing such a custom made titanium membrane for the specific case of a giant cell tumour in the proximal tibia and second to evaluate the designed membrane biomechanically.
Materials and methods
Design process
Titanium is chosen as membrane material because of the following reasons: titanium is highly biocompatible, relatively transparent on plain X-rays and it leads to minimal artefacts on CT and MR imaging. So titanium is an adequate material for implantation and radiographic follow-up.
The first step in the design process is the CT-scanning of the affected region of the tibia. Based on these medical images a virtual 3D-model of the tibia with tumour is created. Then the geometry of the membrane is determined so as to reconstruct the cortical surface. Therefore the desired shape of the cortical surface is drawn as a contour in each CT-image and a virtual 3D-model is made based on these corrected images. The surgeon then indicates interactively on the computer screen the outer contours of the membrane (figure 1).
Using rapid prototyping a physical 3D-model is built of the reconstructed tibia. Then a replica of last mentioned model is made in synthetic hard plaster, which is used as a mould during the hydroforming process for pressing the titanium sheet on (figure 2).
Biomechanical evaluation
Before using this new reconstruction technique on patients, a biomechanical evaluation of the design is necessary. Therefore the case of a giant cell tumour in the proximal tibia was studied because of its relevance.
First of all a finite element analysis was performed on a simplified model of only a membrane under several loading conditions simulating different physical loads and for different thickness (0.2, 0.3 and 0.5 mm). These models showed that the maximal calculated Von Mises stress does not exceed the yield stress of titanium when excluding the jumping load (figure 3).
A wave pattern with ribs aligned along the longitudinal axis was introduced to improve the strength during compression loading of the membrane. Even introducing such a wave pattern in the membrane was not sufficient to allow the jumping load.
The risk for buckling of the membrane was not included in the finite element analysis. An experimental approach was chosen to determine the critical pressure load that starts the buckling process of the membrane. The critical buckling forces are given in Table 1.
|
Membrane
thickness (mm) |
Buckling
force (N) |
|
|
No wave pattern
|
Wave pattern
|
|
| 0.2 | 1600 | 3200 |
| 0.3 | 4700 | - |
| 0.5 | 11200 | 13300 |
Table 1: Critical buckling forces.
Further experiments showed that the mean force causing rupture of a perforation hole of 4.2 mm diameter in the titanium membrane was 210 N, 430 N or 1070 N for a membrane of respectively 0.2, 0.3 and 0.5 mm thick.
The maximal shear load a screw can withstand without failing, i.e. breaking of the surrounding bone or breaking of the screw itself was also experimentally determined (figure 4).
The failure of the screw-bone fixation was always due to a fracture of the surrounding bone. The mean ultimate force measured was 305 N using a cortical 2.7 mm diameter screw of 16 mm length; 420 N for a 3.5 mm cortical screw of 18 mm and finally 640 N for a trabecular screw with a diameter of 4.0 mm and a length of 28 mm. For the two cortical screws a good correlation was found between the force at which the bonding fails and the thickness of the cortex at the site of the screw.
A second finite element analysis was made based on the results of the former tests. The loads during two main daily activities (standing and walking) were simulated on a tibia, on which a 0.3 mm thick non wave-formed titanium membrane was fixed with 7 screws. The number of screws is based on the comparison of the load the screw bone interface can withstand with the load transmitted through the screws to the membrane. The 7 screws were in reality not drawn, but simulated by defining a fixed contact between the membrane and the tibia at those sites. The Von Mises stress and the deformation were calculated under the two mentioned loading conditions (figure 5).
Discussion and conclusions
The proximal tibia was chosen as study case because of its relevance for the design task of the titanium membrane. First a giant cell tumour frequently occurs in the proximal tibia just beneath the joint surface. Second the tibia is highly loaded during daily activities; hence the titanium membrane must come up to high requirements.
From the performed tests and analyses it appears that a titanium membrane without wave pattern, of 0.3 mm thickness, fixed to the bone with 7 trabecular screws (4 mm diameter and 28 mm length) is capable of carrying the anticipated mechanical loads. This means that for a patient with a giant cell tumour in the proximal tibia treated conventionally but reconstructed with such a membrane the period of non-weight bearing or using crutches is reduced to a period of limited activity (no jumping). This membrane is designed and produced pre-operatively, hence not complicating the actions in the surgical theatre. Moreover there are no restrictions to the surgeon regarding the removal of the tumour, because a per-operative fine-tuning of the shape and punching of the screw holes is possible. The membrane provides also a containment and pressurisation of the filling material. Hence a better contact of the filling material to the walls of the cavity is achieved. This leads to a better ingrowth of the bone graft or to an increased load sharing of the bone cement.
Acknowledgements
This research was performed in the Brite Euram project PISA (NR. BRPR CT97 0378).