Page 223 - Advances in Biomechanics and Tissue Regeneration

11.2 METHODOLOGY OF SIMULATION 219

11.2 METHODOLOGY OF SIMULATION

As already indicated, treatment of fractures of the femur between the 2 and 5 zones of Wiss is performed with the
intramedullary nail with different combinations of screws. There are multiple designs of nail, steel or titanium, reamed
and unreamed, solid or hollow, anterograde or retrograde, and various types of locking system. The use of one or other
depends on the type and location of the fracture, the characteristics of the patient, and the experience of the surgeon.
There is no scientific, clinical, or experimental evidence that demonstrates the best results and indications for each type
of nail in each type of fracture.
In view of the difficulties experienced with in vitro testing or in experiments with living subjects, FE sim-
ulation models have been developed to carry out research on biomechanical systems with high reproducibil-
ity, versatility, and limited cost. These models allow the study to be repeated as many times as desired, being
a nonaggressive investigation of modified starting conditions (loads, material properties, etc.). However, work
continues on the achievement of increasingly realistic models that allow placement of the generated results
and predictions into a clinical setting. To that end, it is necessary to use meshes suitable for every particular
problem, regarding both type of element and size. This is necessary to perform a sensitivity analysis to deter-
mine the optimal features or, alternatively, the minimum mesh necessary to achieve the required
accuracy [33].
Thus through the development of a computational model based on the FE method, it is possible to study the bio-
mechanical behavior of the same nail made of two materials (steel or titanium), introduced with or without reaming,
anterograde or retrograde, with different types of locking systems in different types of stable and unstable fractures of
the femur. This is done to find the best indication and therapeutic technique for each type of femoral fracture from the
subtrochanteric to the supracondylar region.
For this purpose, the methodology consists of the development of an FE model of a femur, on which will be sim-
ulated various types of fractures in the subtrochanteric, diaphyseal, and supracondylar areas, stabilized through var-
ious assemblies and materials of intramedullary nails. The mechanical strength of the nail against bending and
compressive loads is also studied to determine its maximum strength. Subsequently, a comparative analysis of the
different types of fixation in fractures is developed to verify what is the optimal solution in each of the analyzed cases.
The biomechanical results are evaluated in correlation with observed clinical results. Thus it is possible to observe
biomechanical needs for each type of fracture to find the optimum combination of variables (type of nail, locking sys-
tem, material, surgical approach, etc.) ideal for their treatment.
To develop an FE model, the first phase is to generate the geometry of the different parts that define the model.
One of the most significant aspects of biomechanical systems is their geometric complexity, which greatly com-
plicates the generation of accurate simulation models. Thus the use of scanners together with three-dimensional
(3D) images obtained by computed tomography (CT) generate geometric models that combine high accuracy in
theexternalformwithanexcellent definition of internal interfaces. The method requires not only appropriate soft-
ware tools, capable of processing images, but also compatibility with the programs used later to generate the FE
model [33].
Development of the model of a healthy femur is crucial to perfect the whole process of simulation, and to obtain
reliable results. A 3D FE model of the femur from a 55-year-old male donor was developed. To obtain a faithful geom-
etry, the bone was scanned using a 3D Roland Picza laser scanner (Fig. 11.4).
This device offers two sweep modes: a plane-based and a rotary sweep with a scanning resolution of 0.2mm. The
scanner provides a first approximation of the outer geometry represented by a cloud of points. By means of its own
scanning software (Roland Dr. Picza 3) [34], general and rough cleaning operations were performed and eventually
treated afterwards in Pixform software [35]. The scanned file was subjected to a specific cleaning protocol to pull the 3D
image out of scan noise (deleting spikes, cleaning abnormal and nonmanifold surfaces, fixing bad normals, etc.). Those
initial rough geometries with noisy faces and screws attached to a support to fix firmly the bone while scanning are
shown in Fig. 11.5.
Local operations could also be carried out as smoothing and bridging gaps, because a fully closed geometry was
required to continue with the process. Fig. 11.6 displays the final geometry of the femur after the cleaning and geom-
etry treatment process was accomplished.
Eventually, when a closed geometry was obtained, it was finally wrapped by an ensemble of B ezier parametric
surfaces as is exhibited in Fig. 11.7. The order and number of these surfaces were varied until a proper fidelity was
reached. Analysis tools were provided in this software to check deviation from the original geometry of the surfaces
generated. Afterward, they were exported to I-DEAS 11 NX Series software [36]. First, each volume was checked for
suitability for meshing. If not, problematic surfaces were identified and the possible source problem was referred back
to Pixform and Rhinoceros software [37].

I. BIOMECHANICS