The management of condylar head fractures has been associated with considerable controversy over the years. However, the latest clinical studies have clearly shown that open reduction and internal fixation (ORIF) offers favourable functional outcomes for this type of mandibular trauma. For condylar trauma in general (i.e. base and neck fractures), Worsaae and Thorn, Neff, Ellis and others have demonstrated that surgical treatment provides better results than conservative management in adult patients.7, 34, 35Properreduction and rigid fixation of the bony fragments was impossible in most of the conservatively treated patients, resulting in functional deficits and compromised quality of life.35Current trends in maxillo-facial traumatology are associated with an increased number of patients treated with ORIF and extensive research aiming to optimize surgical techniques and fixation hardware.7
At the same time, surgical treatment for condylar head fractures remains challenging and non-surgical treatment has significant limitations and drawbacks, as reported by Hlawitschka M.2, 36When ORIF is used to treat such fractures, the appropriate fixation method as well as the type, material, number, dimension and geometric shape of fixation devices and elements are important points of discussion.
The clinical and biomechanical simulations performed by Neff et al., revealed that, for condylar head fractures, the use of two or three small-fragment positional screws is superior other methods of osteosynthesis, such as mini- or microplates. Screw fixation provides better fixation stability and avoids excessive loading of the bone tissue.13 It also preserves the attachment of the lateral pterygoid muscle and capsular ligaments, reduces articular scarification and ensures sufficient blood supply to the condyle. According to Kolk and Neff, ORIF leads to significantly reduced loss of mandibular vertical height and provides superior functional outcomes compared to the use of mini- and microplates.3
Conventional two-screw fixation is still technically challenging due to poor visualisation and limited access to the fracture area. Each surgeon handles the fractured fragments according to their own clinical experience, which frequently results in improper reduction and screw positioning, followed by development of complications and even an impaired functional prognosis. The most problematic cases are fractures with fragmentation of the posterior surface and the lateral pole of the condylar head, leading to the loss of anatomical landmarks. In these cases, repositioning of the fragments becomes less predictable. Additionally, due to screw installation, the torque values may be too high to keep the fragments in the correct position. Another problem is that, in cases when the lateral condylar segment (or its cortical layer) is not thick enough (e.g. vertical fracture patterns, Neff et al.), the positional screws cannot effectively stabilize mastication forces, resulting in bone overloading, fixation failure, and loosening or translocation of the titanium hardware.26
We have developed the concept of a patient-specific device for guided TMJ surgery with the use of advanced CAD/CAM technology in order to increase the accuracy of fragment reduction and to reinforce fixation, especially in the complex cases noted above. The proposed two-component patient-specific titanium guide and small individualized plate located at the lateral surface of the condyle, can be used both for precise fragment reduction and screw positioning and for reinforcement of the fixation system.
The use of CAD/CAM technology in condylar head fractures has been previously reported by Wang, Yang, Smolka, and Han.16, 22, 24, 25 However, there are few articles devoted to this problem; all existing literature is limited to direct measurements and virtual simulation of the surgical procedures in CAD software. Though the surgical procedure itself remains highly operator-dependent, computer modelling provides a wealth of additional information either about the trauma pattern and anatomical parameters of the injured zone or reference data about position and angulation of the screw hole.
The existing clinical research gives evidence that the precision and predictability of surgical intervention in orthognathic, bone reconstructive and traumatological surgery, including the management of base and neck fractures, is significantly increased by the additional manufacturing and application of patient-specific plates or implants, as well as the implementation of surgical guides for bony fragment reduction and installation of plates and screws.20-22, 38The introduction of a patient-specific plate facilitates virtual operation planning and reduces the deleterious impact of low operator skill and experience. However, we could not find any reference in the literature that such an approach has been used for condylar head fractures. The design proposed in this study meets the demand for such a construction. It can be applied during the conventional (e.g. retroauricular) approach, and its dimension is adapted to the limited surgical access to the mandible; its shape is determined by existing anatomical ‘safe zones’ that can be exposed without risk of intra- or postoperative complications. Finally, the plate, can be fixed in its proper position by microscrews. The small reinforcement plate is not only used for increasing strength and rigidity of the fixation system but also determines the position of the drilling holes and proper angulation of the long positional screws.
To have the necessary mechanical properties, such a thin construction should be made only of metal (titanium is preferable), which can be manufactured through direct laser metal sintering (DMLS) with a precision of 0.8 mm. Before beginning research on clinical efficacy and potential limitations for such an approach, we simulated the plate’s biomechanical behavior through finite element analysis (FEA) as presented in the article.
FEA is a computational technique used to model the mechanical behaviour of structures, including biological tissues, which can measure parameters that cannot be directly assessed in vivo (such as the internal stress and strain of bones and fixation hardware). FEA can be considered a powerful instrument for preclinical estimation as well as comparison of various fixation techniques and optimization of surgical approaches in facial fractures; it has been successfully used in numerous studies to investigate distribution stresses and strains in screw and plate ostheosynthesis.39 Recent FEA simulations demonstrated that conventional two-screw osteosynthesis of the condylar head had significant biomechanical advantages over the other fixation techniques.10, 24However, these studies analysed only typical B-type fractures (i.e. corresponding to the type p fractures of the present study) rather than addressing clinically frequent multifragmented fractures or biomechanically unfavourable fractures, such as those with a compromised cortex of the lateral condylar stump.26 We also found no experimental reports regarding the possibilities for reinforcement of the two-screw fixation system with patient-specific implants or conventional mini- or microplates.
Previous studies have demonstrated that FEA model accuracy is increased as model complexity and the number of finite elements increases.10, 22, 24, 33The number of nodes and elements in our models was sufficient for proper representation of the complex geometry of the mandibular bone and its inner structure. In previously reported models of the human mandible, the number of elements ranged from 22,98610 to 1.5 million.33 In this study, special attention was given to the accurate simulation of muscular forces in various loading conditions. In previous studies regarding the finite element simulation of mandible loading, authors assigned anywhere from two to nine mandibular muscles with simulations of different occlusal contactsor forces.10, 33, 38 In our study, the application of constraints and muscular loads was performed according to the concepts described by Korioth and Koolstra, as is common in more recent FEA studies.29-31
With these parameters, the model provided a close approximation to real clinical conditions and results, comparable with previously published data.
The material properties used in the models were taken from the literature and based on the direct measurements of the intact bones. For such a simulation one should be aware of the fact than the main mechanical constants of the cortical and spongious bone may vary significantly in different individuals and depend on the method used for their estimation in experimental studies. The reported values of the Young's modulus for cortical bone used for FE modeling is from 4 to 20-22 GPa, and for spongious bone from 0,05 to 1,5 GPa. The trauma itself and the posttraumatic bone remodeling also influence the mechanical properties of the condylar head, but there is lack of information concerning this issue due to significant limitations for the direct measurements in the human beings.
At the same time the models, created in our study, give and adequate representation of the general patterns for stress and strain distribution. They can be also used for comparison of the different fixation techniques and prediction of changes/errors associated with altered mechanical properties if lineally elastic behavior of the system is assumed.
The generated data on stress and strain distribution inside the bone and screws, as well as the total deformation values, corresponded to the previous research on this topic. The post-processing of the FEA models demonstrated that stress was transmitted through the screws to the medial fragment and that adequate rigidity was achieved at normal chewing, with no signs of stress concentration higher than the ultimate level. However, if maximal force of the muscular contracture was applied, the model revealed areas of excessive stress concentration appearing near the screw holes, predominantly at the cortical layer of the lateral stump. Von Mises stress concentration is an important indicator in the assessment of bone remodeling; Sugiura et al. indicated that stress exceeding 50 MPa in bone can lead to bone resorption.40 In our simulations, the maximum von Mises stress exceeded 50 MPa in conventional two-screw fixation at maximal loading, thus demonstrating the possible risks of this method in unfavourable biomechanical conditions. Excessive strains may lead to failure of fixation due to bone resorption, cracks, fragmentations and loosening of the screws.
In the models with patient-specific reinforcement, plate stress and strain distribution patterns were the same – however, their values were 2–10 times lower than in the conventional. Stress values with patient-specific reinforcement were less than 50 MPa even in cases with maximal muscular forces. The rigidity of fixation also increased and was much closer to normal jaw deformability.
Although application of the reinforcement plate had little, influence on the maximal stress values in the titanium hardware itself (especially inside the screws), these values were less than ultimate stress for the titanium even in maximal loading conditions, the clinical data. Excessive stress developed only at the thread edges, which can undergo plastic deformation without fixation failure.
As was reported in recent studies, trauma resulting from extensive exposure of the condylar head fracture and application of micro- or miniplates often causes scar-induced limitation of postoperative translation in the joint13, 14and even development of postoperative fibrous ankylosis in at long-term follow-up.3 In the current study, the patient-specific plate has a special design that minimizes the need to detach the capsule from the lateral pole area. This aspect, along with the low bone strain and minimum condylar head displacement demonstrated in the current FEA models, suggests that screw fixation with patient-specific reinforcement plate (PSRP) will perform best in fixation of condylar head fractures in certain biomechanically unfavourable fractures.
This study has some limitations. As with any simulation approach, simplifications and assumptions were made in the models: the bone was considered to be homogenic and isotropic within one material type; only the static loading conditions of anterior biting were reproduced; a linear elastic solution was applied, and others. In reality, the biomechanical properties and behaviour of the bone is more complicated and may significantly change in the injured areas. However, authors in many studies use such simplifications and obtain results which correlate with clinical data and experiments in cadaveric bone. In post-processing, we also ignored the regions of artificially high strain that some authors consider to be errors of numerical simulations. Additionally, it is impossible to predict the reaction of soft tissue to installation of the construction as well as clinical peculiarities and limitations for application of the proposed method.
Thus the results of a simulation should be critically judged any clinical recommendations should be made with proper caution. The further work is also needed to validate the results in both experimental and clinical studies.