In this study, finite element analysis (FEA) was used to evaluate nine rigid internal fixation techniques for sagittal split ramus osteotomy. To achieve this, a computed tomography (CT) scan of a healthy patient was obtained and used to generate the geometry of a half-mandible. The geometries of bicortical screws, miniplates, and monocortical screws were designed and combined with the mandible in nine models simulating various techniques. Four models used bicortical screws in various arrangements and four used miniplates of various designs. One model represented a hybrid technique. A load of 500 N was applied to the posterior teeth and FEA was applied. The most stable techniques were the hybrid technique and a single straight miniplate, presenting the least displacement among all models. Bicortical screws, while presenting reasonable stability, showed high strain areas near the anterior ramus ridge, superoposterior to the screws, implying a risk of bone fracture in this area. On the other hand, the T-shaped and double Y-shaped miniplates were associated with high von Mises stresses that would impair their rigidity, especially where angles appeared in their designs. We recommend the use of a single straight miniplate because it provides sufficient stable fixation with minimal risks or disadvantages.
Securing stable fixation is a primary concern in orthognathic surgery. Inadequate stability results in non-union or malunion of the fragments and compromises the outcome of the surgery, leading to relapse. For this reason, the fixation technique is considered a key factor that influences the risk of relapse. Rigid internal fixation (RIF) is now used widely in orthognathic surgery as a stable and reliable technique that eliminates the need for intermaxillary fixation. This technique is especially useful and is required when mandibular surgery is combined with another osteotomy of the maxilla. Many techniques for RIF that utilize miniplates or bicortical screws to provide fixation in sagittal split ramus osteotomy (SSRO), have been the subject of studies that have adopted various methods, such as mechanical and photoelastic tests, animal models, and follow-up of patients. However, to date there is no agreement on the best RIF technique in terms of stability of the components.
The finite element method (FEM) is a numerical technique used in engineering to simulate mechanical problems and improve the designs of constructions. It has been employed in biomechanical studies and found to be a reliable technique in simulating bone behaviour. It can be used to predict bone fracture or failure under increased loads. Because bone is very sensitive to applied loads and responds by remodelling to adapt to the new distribution of strains inside it, this technique can serve as a useful diagnostic tool to provide insight into strain distribution under various loads and designs.
Some studies have used the FEM to assess fixation techniques in SSRO and found differences between them. However, these studies mainly focused on the distribution of stress around fixation units and depended on differences in stress in bone to state their recommendations. No information is currently available regarding strain around fixation units or displacement of the osteotomy fragments.
The purpose of this study was to compare rigidity and strain distribution in the bone in nine combinations of fixation units when used in rigid fixation of an SSRO.
Materials and methods
A computed tomography (CT) image of a 24-year-old healthy patient was obtained. This was used to generate the geometry of a half-mandible in Mimics v. 10.01 (Materialise, Leuven, Belgium). The osteotomy was applied to the mandible using Rapidform (INUS Technology, Seoul, Korea). The fixation units were designed in Rapidform and combined with the mandible geometry. Nine models were prepared for simulation. In four of these models, 2-mm bicortical screws (positional screws) were used, the threads of which engaged both cortices. They were arranged in a straight line at an angulation of 60 degrees (model 1), in a straight line at an angulation of 90 degrees (model 2), in an L arrangement (model 3), or in a reverse L arrangement (model 4). In four additional models, miniplates with monocortical screws were used. Models 5 and 6 comprised straight and double straight miniplates, respectively. Model 7 had a T-shaped miniplate, and model 8 had a double Y-shaped miniplate. Model 9 simulated a hybrid technique, utilizing one bicortical screw and one straight miniplate with monocortical screws. A reference model, in which no osteotomy was simulated, was used to compare results between the various techniques and the normal mandible.
The software ANSYS v. 13 (ANSYS, Inc., Canonsburg, PA, USA) was used for the finite element analysis. The elastic modulus and Poisson ratio for bone and titanium alloy were assigned according to previous studies. Bone was given the properties of cancellous bone ( E = 1500 MPa, Poisson ratio = 0.3), and a shell of 2-mm thickness was added to the surface of the mandible to simulate the cortical bone ( E = 14 GPa, Poisson ratio = 0.3). Fixation units were given the properties of titanium alloy ( E = 114 GPa, Poisson ratio = 0.34).
To simulate non-osseointegration and the nature of the interface between bone fragments immediately after surgery, a gap element was added to the contact surfaces between the bone fragments, and contact surface between bone and fixation units. A load of 500 N was applied to the occlusal surfaces of the posterior teeth, and constraints were assigned to the articular surface of the condyle and the lateral surface of the ramus at the masseter attachment. The load coincided with previous studies on bite force in patients with dentofacial deformities. Static linear analysis was applied.
The stability of fixation was evaluated by calculating the maximum displacement of the anterior fragment of the mandible ( Fig. 1 ).
The reference model had a strain of around 0.3% in both cortical and cancellous bone. This was used as a threshold for the other models, and the elements that had strain greater than this threshold were selected. The average strain values of these elements were calculated separately for cortical and cancellous bone ( Figs 2 and 3 ).
To predict the failure of fixation units, the maximum von Mises stress was calculated ( Fig. 4 ).
Correlation tests were done to investigate the relationship between displacement and the other calculated parameters ( Table 1 ).
|Variable||By variable||Spearman ρ||Prob > | ρ ||
|Displacement, μm||MaxPN (%) cortical bone||0.7406||0.0225 a|
|Displacement, μm||MaxPN (%) cancellous bone||0.3782||0.3156|
|Displacement, μm||ES (MPa) fixation units||−0.4167||0.2646|
Maximum principal strain in bone
Quantitatively, all models predicted strain higher than that in the normal mandible (0.3% in cortical and cancellous bone). Strain in cortical bone was lowest in model 9 and highest in model 3 ( Fig. 2 ). In cancellous bone, it was lowest in model 1 and highest in model 3 ( Fig. 3 ). With regard to the strain distribution contours, the bicortical screws showed focally concentrated strain near the anterior ridge of the ramus ( Fig. 5 ). The miniplate and hybrid models predicted more equally distributed strain, with high strains only at the contact area between bone and fixation units ( Fig. 5 ). The strain contours of cancellous bone predicted concentrated strain around monocortical or bicortical screws, but no specific difference was predicted between the models ( Fig. 6 ).