This art icle was downloaded by: [ Universit y of Zaragoza] On: 09 July 2015, At : 01: 16 Publisher: Taylor & Francis I nform a Lt d Regist ered in England and Wales Regist ered Num ber: 1072954 Regist ered office: 5 Howick Place, London, SW1P 1WG Computer Methods in Biomechanics and Biomedical Engineering Publicat ion det ails, including inst ruct ions f or aut hors and subscript ion inf ormat ion: ht t p: / / www. t andf online. com/ loi/ gcmb20 A Comparative Analysis of Different Treatments for Distal Femur Fractures using the Finite Element Method J. Cegoñino a , J. M. García Aznar a , M. Doblaré a , D. Palanca b , B. Seral b & F. Seral b a Aragón Inst it ut e of Engineering Research , Universit y of Zaragoza , María de Luna, 3. 50018, Zaragoza, Spain b Depart ment of Surgery and Traumat ology, Facult y of Medicine , Universit y of Zaragoza, Ciudad Universit aria. , 50010, Zaragoza, Spain Published online: 21 Aug 2006. To cite this article: J. Cegoñino , J. M. García Aznar , M. Doblaré , D. Palanca , B. Seral & F. Seral (2004) A Comparat ive Analysis of Dif f erent Treat ment s f or Dist al Femur Fract ures using t he Finit e Element Met hod, Comput er Met hods in Biomechanics and Biomedical Engineering, 7: 5, 245-256, DOI: 10. 1080/ 10255840412331307182 To link to this article: ht t p: / / dx. doi. org/ 10. 1080/ 10255840412331307182 PLEASE SCROLL DOWN FOR ARTI CLE Taylor & Francis m akes every effort t o ensure t he accuracy of all t he inform at ion ( t he “ Cont ent ” ) cont ained in t he publicat ions on our plat form . However, Taylor & Francis, our agent s, and our licensors m ake no represent at ions or warrant ies what soever as t o t he accuracy, com plet eness, or suit abilit y for any purpose of t he Cont ent . Any opinions and views expressed in t his publicat ion are t he opinions and views of t he aut hors, and are not t he views of or endorsed by Taylor & Francis. The accuracy of t he Cont ent should not be relied upon and should be independent ly verified wit h prim ary sources of inform at ion. Taylor and Francis shall not be liable for any losses, act ions, claim s, proceedings, dem ands, cost s, expenses, dam ages, and ot her liabilit ies what soever or howsoever caused arising direct ly or indirect ly in connect ion wit h, in relat ion t o or arising out of t he use of t he Cont ent . This art icle m ay be used for research, t eaching, and privat e st udy purposes. Any subst ant ial or syst em at ic reproduct ion, redist ribut ion, reselling, loan, sub- licensing, syst em at ic supply, or dist ribut ion in any form t o anyone is expressly forbidden. Term s & Condit ions of access and use can be found at ht t p: / / www.t andfonline.com / page/ t erm s- and- condit ions Computer Methods in Biomechanics and Biomedical Engineering Vol. 7, No. 5, October 2004, pp. 245–256 A Comparative Analysis of Different Treatments for Distal Femur Fractures using the Finite Element Method J. CEGOÑINOa, J.M. GARCÍA AZNARa, M. DOBLARÉa,*, D. PALANCAb, B. SERALb and F. SERALb a Aragón Institute of Engineering Research, University of Zaragoza, Marı́a de Luna, 3. 50018 Zaragoza, Spain; Department of Surgery and Traumatology, Faculty of Medicine, University of Zaragoza, Ciudad Universitaria. 50010 Zaragoza, Spain b Downloaded by [University of Zaragoza] at 01:16 09 July 2015 (Received 23 July 2003; In final form 26 July 2004) The main objective of this work is the evaluation, by means of the finite element method (FEM) of the mechanical stability and long-term microstructural modifications in bone induced to three different kinds of fractures of the distal femur by three types of implants: the Condyle Plate, the less invasive stabilization system plate (LISS) and the distal femur nail (DFN). The displacement and the stress distributions both in bone and implants and the internal bone remodelling process after fracture and fixation are obtained and analysed by computational simulation. The main conclusions of this work are that distal femoral fractures can be treated correctly with the Condyle Plate, the LISS plate and the DFN. The stresses both in LISS and DFN implant are high especially around the screws. When respect to remodelling, the LISS produces an important resorption in the fractured region, while the other two implants do not strongly modify bone tissue microstructure. Keywords: Internal bone remodelling; Finite element simulation; Distal femur fractures; Distal femur nail; Condyle plate; LISS plate INTRODUCTION The femur is a long bone that constitutes the skeletal of the thigh. It is articulated at its upper part with the coxal bone and at its lower with the knee complex (tibia, patella and knee tendon system). Two important parts can be distinguished in the femur: the diaphysis or central body and two epiphysis or extremities. The thick cortical walls of the diaphysis, mainly formed by dense cortical bone, become thinner as they form the metaphysis, being the epiphysis mainly conformed by a complex heterogeneous mixture of cortical and cancellous bone [1]. The thick, dense, tubular cortical structure of the diaphysis provides maximum resistance to bending and torsion, while the thinner cortices and cancellous bone in the metaphysis and epiphysis form an articular surface, designed to absorb impact and directionally dependent loads. The main role of the musculoskeletal system of supplying a rigid structure to the rest of the body is affected when one of its elements is fractured, becoming essential to know the main factors that affect the correct treatment of these fractures. In fact, the worrying increment of fractures of the musculoskeletal system is *Corresponding author. E-mail: mdoblare@unizar.es ISSN 1025-5842 print/ISSN 1476-8259 online q 2004 Taylor & Francis Ltd DOI: 10.1080/10255840412331307182 generating prevention campaigns and promoting research that have lead to new therapeutic methods [2,3]. One of the most important factors to achieve an adequate fracture healing is the appropriate stabilisation of the fractured zone after the implantation of a specific fixation [4]. Five independent variables can be considered, in general, to evaluate the efficacy of an implant [5]: . . . . . Bone quality and its evolution (remodelling); Geometry and position of the fractured fragments; The reduction of the fracture; The design of the implant employed; The location of the implant; The only variable that can be chosen by the surgeon is the implant and its position. Moreover, these strongly affect the other three after surgery. Despite the low number of distal femoral fractures, a 3.59% of the total, they are very important due to their difficult in the treatment. Distal femoral fractures affecting the lower epiphysis may be classified according to the specific location of the fracture zone as shown in Table I based on the AO/ASIF 246 J. CEGOÑINO et al. TABLE I AO classification [5] Type A Extra-articular fracture A1—simple A2—metaphyseal wedge A3—metaphyseal complex Type B Articular parcial fracture B1—lateral condyle, sagittal B2—medial condyle, sagittal B3 –frontal condyle, sagittal Downloaded by [University of Zaragoza] at 01:16 09 July 2015 Type C Articular fracture, articular simple C1—metaphyseal simple C2—metaphyseal multifragmentary C3—articular multifragmentary classification [5]. The fractures here considered are of the following types: extra-articular metaphyseal wedge (A2.3), articular complete metaphyseal multifragmentary (C2.1) and articular complete multifragmentary (C3.1). Some of these fractures can be treated by plate-type fixations achieving an appropriate reduction of the fracture. In other cases however, it is necessary to use a nail fixation. Techniques of operative treatment of supra and intercondylar fractures have changed in recent years. These changes refer to reduction techniques and implant selection. Operative approach concepts, which have remained unchanged for several decades, were critically evaluated and modified leading to the so-called minimal invasive osteosynthesis concept [6]. This included for intraarticular fractures a transarticular joint reconstruction and a retrograde plate osteosynthesis. For extraarticular fractures a minimally invasive percutaneous plate osteosynthesis via stab incisions only or retrograde intramedullary nailing is available. Many authors have performed biomechanical studies of the distal femur fractures. Most of them are based on clinical studies [7 – 12], experimental essays [13 –15] and finite element simulations [16]. The advantage of computer simulations is the possibility of performing parametric analyses and personalised virtual tests, reducing the economic and social cost in comparison to experimental techniques and allowing in addition to test different situations impossible to simulate in real practice. This paper is focussed in the application of the finite element method (FEM) to the study of the femur functional performance, after fracture and fixation by three types of implants: the Condyle Plate [17 – 18], the less invasive stabilization system plate (LISS) [6,19,20] and the distal femur nail (DFN) [21 –23]. Its main conclusions are that distal femoral fractures can be treated correctly with all types of fixations. With respect to the stresses on the implant, the highest values appear in the LISS plate in the zone surrounding the screws and in the DFN on the central nail, particularly in the zone around the transversal screws. Regarding the bone remodelling performance, the LISS plate produces a high resorption in the fractured region due to the bridge effect induced by the position of the fixation screws, while the other two implants do not strongly modify bone tissue microstructure. MATERIAL AND METHODS A femur, tibia and fibula of a 76 years old woman donor were scanned to obtain a set of slices by computed tomography (CT). The distance chosen between each two slices was in the epiphysis about 2 and 4 mm, and in the diaphysis about 5 and 9 mm. From these slices, the geometry and associated mesh was reconstructed with the help of the CAD package I-DEAS (SDRC, Milford, Ohio). The resulting mesh was composed by 17,631 brick elements and 18,291 nodes (Fig. 1) A simplified constitutive behaviour was established for bone tissue, considering it as a homogeneous isotropic linear elastic material, although distinguishing between cortical and trabecular bone. This is in fact a very simplified model, since bone is actually heterogeneous, non-linear and anisotropic [24 – 26]. However, for comparative purposes and taking into account that our main goal is the analysis of the stabilisation performance, that is clearly a non-local problem, the use of average properties is accurate enough. The mean mechanical properties considered for each type of bone tissue [1,26] and soft tissue [27] are shown in Table II. The loading conditions on the femur are multiple and variable along time, depending on its position [27 –29]. Moreover, due to the amount of ligaments and muscles inserted in the femur, it is necessary to perform some simplifications. Following different other authors [30], only one loading condition was considered in this work, corresponding to the stance phase of the gait cycle. This is one of the most important and unfavourable loading stages appearing in the femur. In the model we considered a load of 2460 N acting on the femoral head with angles of 238 with the frontal plane and 68 with the sagittal plane. A second load acting on the greater trochanter that corresponds to the reaction of three muscles (gluteus minimus, medius and maximus) was also included. This load had a modulus of 1700 N, with angles of 248 with the frontal plane and 158 with the sagittal plane. Finally, a third load was applied in the lesser trochanter, due to the effect of the psoas-iliac muscle, with a value of 771 N and forming angles of 418 with the frontal plane and 268 with the sagittal plane. With respect to the boundary conditions, as a first approach, we considered the tibia constrained at its proximal part. In particular, several nodes were constrained at its proximal epiphysis. Again, this situation is not real, since there is a complex interaction between the femur, the knee system and the ankle system. However, this model has shown to be useful to perform a comparative analysis between different distal femur fixations, studying the stabilisation and the stress state of the fractured distal femur. Downloaded by [University of Zaragoza] at 01:16 09 July 2015 DIFFERENT TREATMENTS FOR DISTAL FEMUR FRACTURES 247 FIGURE 1 FE mesh of the femur and knee. (a) Slices obtained by computed tomography. (b) Evolving surface (c) Final FE mesh. The most important knee ligaments were also considered as non-compression uniaxial truss elements with average elastic modulus [31]. Finally, fractures were modelled with gap contact elements avoiding the interpenetration between the surfaces, that is, establishing a node-to-node unilateral contact. This is possible in this case, since fractures were created by separating face elements of the original mesh of the intact femur, and the displacements are small, leading to consistent meshes for both contacting surfaces. All of these analyses were performed by using the commercial FE package ABAQUS [32]. TABLE II Averaged mechanical properties (Evans, 1973; Jacobs, 1994; Fithian, 1990) Bone Tissue Cortical Trabecular Medular Meniscus Young modulus (N/mm2) Poisson coeff. 14,217 100 1 50 0.32 0.3 0.3 0.3 In this work we used three of the most common fixations employed: the Condyle Plate, the LISS plate and the DFN Treatment with Condyle Plate The Condyle Plate [17 –18] is composed by a long and narrow plate that adjusts its shape to the distal part of the femur (transition zone between diaphysis and epiphysis and condyles). The distal part of the plate has shape of petals of clover, being these petals curved for a better adjustment to the faces of the condyles. The plate is attached to the diaphysis and to the condyles through several screws inserted along a set of holes that exist all along the plate, allowing a perfect joint between fixation and bone. These holes are oval and without thread. The screws, placed along the plate, cross over the two cortical layers having different lengths depending of the zone of the bone. The Condyle Plate is available in a wide range of different measures in order to adapt to each different femur. We chose the geometrical parameters appropriate for the proposed femur, following the surgeons advice. Downloaded by [University of Zaragoza] at 01:16 09 July 2015 248 J. CEGOÑINO et al. FIGURE 2 (a) FE mesh of the Condyle Plate. (b) Position of the implant in the fractured femur. The finite element model was composed by 27,096 brick elements (31,691 nodes). The screws were modelled by brick elements with a circular section of 6.5 mm of diameter. The FE mesh of the plate and screws and the positioning of the Condyle Plate after fixation along the femur as shown in Fig. 2. The material used is stainless steel with the following mechanical properties: Young’s modulus of 2,00,000 N/mm2 and a Poisson’s ratio of 0.28. The fixation was considered in perfect contact with bone (same displacements) by connecting the respective nodes of the screws. The plate is not in contact with the periosteum (different displacements) mimicking the clinical situation. Treatment using the Less Invasive Stabilization System The LISS is a new type of fixation [6,19,20] designed following the principles of “minimally invasive percutaneous osteosynthesis” (MIPO) [31], giving precedence to biological aspects to stabilization. It consists of a plate and screws joined to the plate by means of threaded holes. The implant is positioned close to the bone but without contact and the screws only cross one cortical layer, promoting a better vascularisation. The screws are placed along the implant avoiding the fractured zone. The LISS plate is also available in a wide range of sizes. In this case, we used a plate with 16 cm of length and with five holes. The FE mesh was composed by 28,896 brick elements (33,821 nodes) as shown in Fig. 3, which also includes its location along the fractured femur. The LISS is composed by an alloy of titanium, aluminium and niobium, with the following average mechanical properties: Young’s modulus of 1,05,000 N/mm2 and Poisson’s ratio of 0.28. With respect to the interaction conditions between the implant and bone, we considered a perfect union between the screws and the bone, while the contact between the plate and the cortical external surface of the femur was avoided. Treatment using the Distal Femur Nail The DFN [21 –23] consists of a long nail and four transversal screws. The nail can be smooth or grooved. The nail is placed along the marrow cavity of the femur perforating an orifice between the condyles. It is attached to the femur by means of transversal screws at the begin and at the end of the nail (see Fig. 4). The FE mesh of the DF nail in our case (included in Fig. 4a) was composed by 18,437 brick elements (20,745 nodes). The placement of the nail along the fractured femur is also shown in the Fig. 4. FIGURE 3 (a) FE mesh of the LISS plate. (b) Position of the implant in the fractured femur. DIFFERENT TREATMENTS FOR DISTAL FEMUR FRACTURES 249 TABLE III Values and orientations of the three load cases used in the 3D case acting on the femoral head. Orientations are referred to the frontal (FP) and sagittal planes (SP) Force acting on the head Load case Downloaded by [University of Zaragoza] at 01:16 09 July 2015 1 2 3 FIGURE 4 (a) FE mesh of the DFN nail. (b) Position of the implant in the fractured femur. The mechanical properties are the ones of the LISS plate and the screws were considered in perfect contact with bone, avoiding contact between the nail and bone. Remodelling Behaviour As it is well-known, bone is a living material that optimally modifies its structure according to its specific mechanical environment [25,26,35 –38]. It is therefore very important to establish the main trends of the bone density distribution (both directional and in average) after modification of the natural stress distribution due, for instance, to the inclusion of a specific implant. This will allow therefore, not only to analyse the short-term behaviour after fracture (stabilisation) but also the evolution of bone tissue (longterm behaviour) with important clinical consequences like implant-induced osteoporosis and the associated weakening of the bone structure. Bone remodelling theories allow us to simulate the evolution of bone microstructure and its mechanical behaviour due to the acting loads [26,35 – 38]. Some of these models have been used to perform biomechanical studies after implantation, trying to predict the alteration suffered by bone when a fixation is implanted, in order to optimise its design [16,30,35]. In this work we have employed a new bone remodelling theory, based on the principles of damage mechanics, and able to predict the evolution of the heterogeneity and anisotropy of bone. This theory is described in detail in [38]. This model, uses two independent variables that can be quantified experimentally: the apparent density (a measure of the porosity) and the Cowin’s fabric tensor [39] (a measure of the directionality of bone trabeculae). Since remodelling is induced by cyclic loads, we considered now the loads corresponding to the gait cycle (instant when the foot touches the floor and the other Cycles per block Value (N) Orientation (8FP) (8SP) 6000 2000 2000 2317 1158 1548 24 215 56 6 35 220 two alternative moments of abduction and adduction respectively) [27,28], that are representative of the daily forces producing remodelling. The three loads in Table III were applied to the femoral head sequentially in blocks of 10,000 cycles with different frequencies as shown in Table III. In the stress analysis, a simplified constitutive behaviour was considered: a homogeneous isotropic linear elastic material. In the remodelling analysis, it was initially started from an ideal homogeneous and isotropic material with an average apparent density of 0.5 g/cm3. The above forces were applied 300 blocks until convergence in the constitutive behaviour and microstructure, that is, in the apparent density and anisotropy, obtaining a density distribution very close to the actual (Fig. 5). After fracture and implantation and starting from this density and anisotropy distributions, the same load blocks were applied again until new convergence. The corresponding results are presented in the next section. RESULTS In this section we present the comparison between the results obtained both for the vertical displacements and stresses that are considered in this case the most important for comparison. These were analysed for each type of implant and three types of fractures: an extra-articular metaphyseal wedge fracture (A2.3), an articular complete metaphyseal multifragmentary fracture (C2.1) and an articular complete multifragmentary fracture (C3.1). These distributions were also compared to the ones of the intact femur under the same loads and boundary conditions. The stresses on the implant were also computed and discussed. Finally, a study of the modification of the bone tissue average density due to remodelling in the case of the extraarticular metaphyseal wedge fracture was conducted for the three implants using the bone remodelling approach presented in [38]. Although this model had several limitations (i.e the femoral head constraints with the pelvic acetabulum and the tibia constraints were not exhaustively analysed, the healing process was not considered, etc.), we obtained important qualitative comparative conclusions as will be discussed later on. Downloaded by [University of Zaragoza] at 01:16 09 July 2015 250 J. CEGOÑINO et al. FIGURE 7 Vertical distribution of stress in the intact femur. Results using the Condyle Plate FIGURE 5 Density distribution in a frontal section of the intact femur. Figures 6 and 7 show respectively the vertical displacements and the distribution of vertical stresses in the intact femur under the loads described in the previous section. The maximum level of stresses was about 90 MPa in compression and 45 in tension, while the maximum compression and tension strengths are about 150 MPa [40] and 60 MPa [41], respectively. Tension stresses appeared mainly in the concave part of the femur, while compressive stresses were in the convex part of the diaphysis. The results of displacements for the three types of fractures after the implantation of the Condyle Plate are shown in Fig. 8. It is clear that this type of implant achieves a good stability of the fractured zone (the relative displacement between the fractured surfaces is less than 0.06 mm), which benefits fracture healing. The inclusion of the plate, although modifies the stress distribution reducing the stress level, did not lead to important alterations of the maximum values in comparison with the intact femur (Fig. 9). The stress level was higher for the fracture A2.3 than for the other two. In all cases, the fractured zone had very low stresses (an average value of about 10 MPa was obtained). Finally, in Fig. 10 the von Mises stress distribution along the implant is presented, being clear the bending effect in the fixation. The maximum stress was about 1000 MPa close to the yield stress of the material of 1200 MPa. This stress in the plate was however very much concentrated along the fractured zone, specifically near the holes and the screws, being these regions the ones with the highest probability of failure. The fracture that induced the greatest stress level is the A2.3. Results using the LISS Plate FIGURE 6 Vertical displacements in the intact femur. Figure 11 shows the results of vertical displacements for the three types of fractures using the LISS plate. From them it is clear that this implant also achieves a good stability of the fractured zone in the three cases, being the relative displacements of the two fractured fragments close to zero. The vertical stress distribution was slightly modified in comparison with the one of the intact femur (the maximum stress is about 10 MPa lower than in the intact femur), being very similar to the one of the previous implant DIFFERENT TREATMENTS FOR DISTAL FEMUR FRACTURES 251 Downloaded by [University of Zaragoza] at 01:16 09 July 2015 FIGURE 8 Distributions of vertical displacements in mm in the fractured femur, after treatment with a Condyle Plate of fractures A2.3, C2.1 and C3.1. (Fig. 12). The stress in the fractured region was very low due to the “bridge effect” that produces this implant, since there is no connection with bone in this zone. This is probably the cause of a stronger resorption in this part that will be detected in the remodelling analysis. The von Mises stress distribution along the implant is presented in Fig. 13. The maximum stress was about 1000 MPa slightly higher than the yield stress of the material of 900 MPa. The zone closest to the holes had the highest stress level. In fact, the rigid connections of the screws with the plate induced stress concentrations in those joints as is clearly seen in that figure. Results using the DFN Implant As shown in Fig. 14, the relative displacements in the region around the fracture were very low (an average value of 0.2 mm for the vertical displacements) demonstrating the correct stabilisation achieved by this implant in all the three different fractures very much like the other types of implants. Again, the distribution of vertical stresses was modified, like in the previous cases, reducing the stresses in the fractured zone with respect to the intact case (an average value of about 15 MPa appeared also in this region). This value increased when going up along the diaphysis as shown Fig. 15. Figure 16 shows the von Mises stress distribution along the implant, being clear the bending effect on the nail. The maximum stress in the nail was about 850 MPa clearly lower than the yield stress of the material 1200 MPa. In the transversal screws the average stress value was of 500 MPa. Bone Remodelling As explained in the Material and Methods section, the aim of this study is the prediction of the evolution of the bone FIGURE 9 Distributions of vertical stresses in N/mm2 in the fractured femur, after treatment with a Condyle Plate of fractures A2.3, C2.1 and C3.1. Downloaded by [University of Zaragoza] at 01:16 09 July 2015 252 J. CEGOÑINO et al. FIGURE 10 von Mises stress distribution in the Condyle Plate in N/mm2 for the three types of fractures A2.3, C2.1 and C3.1. microstructure after fracture and treatment with the three different implants. In order to get this evolution, an adequate initial density distribution has to be obtained. This may be done using results of a densitometry or after a correlation between the grey levels in a set of tomographies with bone density. However, we have used the same simulation approach later used after fracture as explained in the previous section. Although the obtained results were only qualitative (i.e. fracture healing and other biological processes were not considered although they strongly modify the loading transmission) they gave us information about the global mechanical environment and therefore the most probable regions in which problems could appear. Taking into account that the characteristic time for the remodelling process is much longer than the one of healing, we also analysed the remodelling evolution without fracture, that is, in the intact femur with the implants, that leaded to more realistic results. FIGURE 11 Distributions of vertical displacements in mm in the fractured femur, after treatment with a LISS Plate of fractures A2.3, C2.1 and C3.1. FIGURE 12 Distributions of vertical stresses in N/mm2 in the fractured femur, after treatment with a LISS Plate of fractures A2.3, C2.1 and C3.1. Downloaded by [University of Zaragoza] at 01:16 09 July 2015 DIFFERENT TREATMENTS FOR DISTAL FEMUR FRACTURES FIGURE 13 von Mises stress distribution in the LISS Plate in N/mm2 for the three types of fractures A2.3, C2.1 and C3.1. 253 Only the A2.3 fracture was considered and all the next figures show the different density distributions obtained after convergence. The density distributions after 150 and 300 load blocks with the fractured femur and with implantation of the Condyle plate are shown in Fig. 17. In the same figure the density distribution after 150 load blocks is shown for the intact femur after implantation of the Condyle plate. A slight resorption in the fractured region can be observed, caused by the higher stiffness of the implant that causes the load to be mainly transferred through the implant, decreasing consequently the stress in bone. On the contrary, far away of the fractured zone, bone mass is formed increasing the cortical layer. In the case of the LISS plate the density distribution after convergence was different (Fig. 18). An important resorption is observed in the cortical layer just around the fractured zone. The bridge effect produced by the way the plate is inserted, strongly modifies the way in which load is transferred, reducing the stress in this part and increasing the reaction and the associated stresses down FIGURE 14 Distributions of vertical displacements in mm in the fractured femur, after treatment with a DFN implant of fractures A2.3, C2.1 and C3.1. FIGURE 15 Distributions of vertical stresses in N/mm2 in the fractured femur, after treatment with a DFN implant of fractures A2.3, C2.1 and C3.1. 254 J. CEGOÑINO et al. Downloaded by [University of Zaragoza] at 01:16 09 July 2015 FIGURE 16 von Mises stress distribution in the DFN implant in N/mm2 for the three types of fractures A2.3, C2.1 and C3.1. FIGURE 17 Density distribution in a frontal section after inclusion of a Condyle Plate in a type A2.3 fracture (150 and 300 time steps in the fractured femur and 150 steps in the intact femur). FIGURE 18 Density distribution in a frontal section after inclusion of a LISS Plate in a type A2.3 fracture (150 and 300 time steps in the fractured femur and 150 steps in the intact femur). Downloaded by [University of Zaragoza] at 01:16 09 July 2015 DIFFERENT TREATMENTS FOR DISTAL FEMUR FRACTURES 255 FIGURE 19 Density distribution in a frontal section after inclusion of a DFN implant in a type A2.3 fracture (150 and 300 time steps in the fractured femur and 150 steps in the intact femur). the fracture, which produces bone formation that is easily noticed in the same figure. Of course, this effect is much more important if no healing is considered, but is also detected in the intact case, predicting bone weakening near fractures treated with the LISS system. Finally, Fig. 19 shows the density distribution in the same three situations for the DFN, being very similar to the case of the Condyle plate. A small resorption in the fractured region is observed, caused again by the higher stiffness of the implant in comparison to bone, although this is not very important, specially in the intact case, in contrast to the LISS implant. DISCUSSION Distal femoral fractures are an important problem, risk and difficult treatment that imply important economical and social costs. The choice of the appropriate implant is one of the key factors to achieve an appropriate stabilisation and an adequate stress distribution on the femoral bone in order to produce a correct heal and small changes in its density distribution during the subsequent remodelling process and, therefore, a low probability of appearance of undesired effects like induced osteoporosis. The use of simulation tools like the FEM to analyse the global functional performance of different implants in the treatment of distal fractures allows us to carry out a cheap and extensive qualitative comparison between them, becoming more and more an essential help in preoperative planning and clinical decisions assessment. Although these models still present important limitations, it is possible to get important qualitative conclusions useful from the clinical point of view. In this case, for instance, the use of an isotropic behaviour of bone, the establishment of simplified contacts between prosthesis and bone, the consideration of a perfect union between the screws and bone or the use of a simplified single load case and boundary conditions are some of these limitations. In distal fractures (A2.3, C2.1 and C3.1), the Condyle plate, the LISS plate and the DFN give rise to similar results regarding stability (the three treatments get the appropriate stabilisation of these types of fractures) and the overall stress distribution on the femur. Only in the LISS plate stress values higher than the yield stress of the material appeared around the threaded joints of the screws and therefore a more likely failure of the implant, although very much concentrated. This could lead maybe not to an unstable fracture but to a fatigue one. Bone remodelling leads to different results depending of the type of fixation. With the Condyle plate and the DFN a small resorption in the fractured region appears, whereas an increase of the cortical layer is clearly detected far from this zone. On the contrary, with the LISS plate, an important resorption is observed in the cortical layer just on the fractured zone, due to a strong alteration of the way in which load is transferred. Although these last conclusions are only tentative since some important effects have not been considered in the model like a correct fracture healing process and other biological and metabolic effects like implant-induced necrosis, they are in agreement with well-known clinical tests, validating in a qualitative sense this simulation approach [42 –43]. Acknowledgements This work has been partially funded by the Diputación General de Aragón as a part of the project D.G.A. P-79/96. References [1] Evans, F.G. (1973) In: Evans, F.G., ed., Mechanical Properties of Bone (Charles C. Thomas, Springfield, Illinois). Downloaded by [University of Zaragoza] at 01:16 09 July 2015 256 J. CEGOÑINO et al. [2] Lips, P. and Cooper, C. (1998) “Osteoporosis 2000-2010”, Acta Orthop Scand 69(281), 21–27. [3] Connolly, J.F. (1999) “The next 10 years. The bone and joint decade”, Clin. Orthop. 358, 255–256. [4] McKibbin, B. (1978) “The biology of fracture healing in long bones”, J. Bone Joint Surg. 60B, 150–162. [5] Müller, M.E. (1991) “Femur”, In: Müller, M.E., Allgöwer, M., Schneider, R. and Willenegger, H., eds, Manual of Internal Fixation. Teniques Recommended by the AO-ASIF Group (Springer, Berlin), pp 485– 500. [6] Krettek, C., Schandelmaier, P. and Tscherne, H. (1999) “Liss: less invasive stabilization system”, Dialogue I, 7. [7] Rizzo, E., et al. (1998) “Biomechanical behaviour at the distal third of the femur: possible use of a medial metaphyseal plate”, Injury 29(6), 451–456. [8] Sanders Roy, et al. (1991) “Double-plating of comminuted, unstable fractures of the distal part of the femur”, J. Bone Joint Surg. 73-A(3), 341– 346. [9] Siliski John, M., et al. (1989) “Supracondylar-intercondylar fractures of the femur”, J. Bone Joint Surg. 71-A(1), 95–104. [10] Johnson Kenneth and Hicken Greg (1987) “Distal femoral fractures”, Orthop. Clin. North Am. 18(1), 115 –131. [11] Schatzker Joseph (1998) “Fractures of the distal femur revisited”, Clin. Orthop. 347, 43–56. [12] Krettek, C., Schandelmaier, P., Richter, M. and Tscherne, H. (1998) “Distal femoral fractures”, Swiss surg.(6), 263 –278. [13] Firoozbakhsh, K., et al. (1995) “Mechanics of retrograde nail versus plate fixation for supracondylar femur fractures”, J. Orthop. Trauma 9(2), 152–157. [14] Harder, Y., et al. (1999) “The mechanics of internal fixation of fractures of the distal femur: a comparison of the condlar screw (DCS) with the condlar plate (CP)”, Injury(30 suppl.), A31–A39. [15] Krettek, C., Schandelmaier, P. and Tscherne, H. (1996) “Distal femoral fractures. Transarticular reconstruction, percutaneous plate osteosnthesis and retrograde nailing”, Unfallchirurg 99(1), 2–10. [16] Reiter, J.T., Böhm, H.J., Kratch, W. and Rammerstorfer, F.G. (1994) “Some applications of the finite-element method in biomechanical stress analyses”, Int. J. Comput. Appl. Technol. 7, 233–241. [17] Böstman, O.M. (1990) “Refracture after removal of a condylar plate from the distal third of the femur”, J. Bone Joint Surg. 72-A(7), 1013–1018. [18] Simon, J.A., Hale, J., Kummer, F. and Koval, K.J. (1997) “Improved fixation in osteoporotic bone: a novel locked supracondylar buttress plate”. OTA Posters, Scientific Basis for Fracture Care. [19] Krettek, C., Schandelmaier, P. and Tscherne, H. (1997) “New developments in stabilization of dia- and metaphyseal fractures of long tublar bones” 26(5), 408–421. [20] Helfet, D.L., Shonnard, P.Y., Levine, D. and Borrelli, J., Jr. (1997) “Minimally invasive plate osteosynthesis of distal fractures of the tibia”, Injury 28(supp 1), 42–48. [21] Iannacone William M., et al. (1994) “Initial experience with the treatment supracondylar femoral fractures using the supracondylar intramedullary nail: a preliminary report”, J. Orthop. Trauma 8(4), 322–327. [22] Helfet David, L. and Lorich Dean, G. (1998) “Retrograde intramedullary nailing of supracondylar femoral fractures”, Clin. Orthop.(350), 80–84. [23] Hora, N. (1999) “Biomechanical analysis of supracondylar femoral fractures ficed with modern retrograde intramedullary nails”, J. Orthop. Trauma 13(8), 539 –544. [24] Buckwalter, J.A., Glimcher, M.J., Cooper, R.R. and Recker, R. (1995) “Bone biology. Part I: structure, blood supply, cells, matrix and mineralization”, J. Bone Joint Surg. 45, 1256–1275. [25] Fung, Y.C. (1993) Biomechanics. Mechanical Properties of Living Tissues (Springer-Verlag, Berlin). [26] Jacobs, C.R. (1994) “Numerical simulation of bone adaptation to mechanical loading”, Dissertation for the Degree of Doctor of Philosophy, Stanford University). [27] Pedersen, D.R., Brand, R.A. and Davy, D.T. (1997) “Pelvic muscle and acetabular contact forces during gait”, J. Biomech. 30(9), 959–965. [28] Bergmann, G., Graichen, F. and Rohlmann, A. (1993) “Hip joint loading during walking and running, measured in two patients”, J. Biomech. 26(8), 969–999. [29] Duda, G.N., Schneider, E. and Chao, E.Y.S. (1997) “Internal forces and moments in the femur during walking”, J. Biomech. 30(9), 933–941. [30] Verdonschot, N. and Huiskes, R. (1995) “Can polished stems reduce mechanical failures of the cement/bone interface in THA?”, Trans. Eur. Orthop. Res. Soc. 5, 42. [31] Li, G., et al. (1999) “A validated three-dimensional computational model of a human knee joint”, J. Biomech. Eng. 121, 657–662. [32] Hibbit, Karlsson, Sorensen, Inc., 2001. ABAQUS. User’s Manual. Version, 6.2. [33] Perren and Stephan, M. (2001) “Evolution and rationale of locked internal fixator technology. Introductory remarks”, Injury 32, S-B-3-9. [34] Terrier, A., Rakotomanana, R.L., Ramaniraka and Leyvraz, P.F. (1997) “Adaptation models of anisotropic bone”, Comput. Methods Biomech. Biomed. Eng. 1, 47–49. [35] Huiskes, R., et al. (1987) “Adaptive bone-remodeling theory applied to prosthetic-design analysis”, J. Biomech. 20, 1135– 1150. [36] Beaupré, G.S., Orr, T.E. and Carter, D.R. (1990) “An approach for time-dependent bone modeling and remodeling-theoretical development”, J. Orthop. Res. 8, 551–651. [37] Carter, D.R., Orr, T.E. and Pyhrie, D.P. (1989) “Relationships between loading history and femoral cancellous bone architecture”, J. Biomech. 22, 231 –244. [38] Doblaré, M. and Garcı́a, J.M. (2002) “Anisotropic bone remodelling model based on a continuum damage-repair theory”, J. Biomech. 35(1), 1–17. [39] Cowin, S.C. (1986) “Wolff’s law of trabecular architecture at remodeling equilibrium”, J. Biomech. Eng. 108, 83–88. [40] Keller, T.S. (1994) “Predicting the compressive mechanical behaviour of bone”, J. Biomech. 27, 1159–1168. [41] Keyak, J.H. and Rossi, S.A. (2000) “Prediction of femoral fracture load using finite element models: an examination of stress- and strain-based failure theories”, J. Biomech. 33, 209–214. [42] McKibbin, B. (1978) “The biology of fracture healing in long bones”, J. Bone Joint Surg 60B, 150 –162. [43] Kerner, J., Huiskes, R., van Lenthe, G.H., Weinans, H., van Rietbergen, B., Engh, C.A. and Amis, A.A. (1999) “Correlation between pre-operative periprosthetic bone density and postoperative bone loss in THA can be explained by strain-adaptive remodelling”, J. Biomech. 32, 695–703.