Lumbar total disc replacement: A numerical study

Transcription

1 Journal of Applied Biomaterials & Biomechanics 2010; Vol. 8 no. 2, pp original article Lumbar total disc replacement: A numerical study Valeriano Di Mascio 1,2, Chiara M. Bellini 2, Fabio Galbusera 2, Manuela T. Raimondi 1, Marco Brayda-Bruno 2,3, Roberto Assietti 4 1 LaBS, Department of Structural Engineering, Politecnico di Milano, Milan - Italy 2 IRCCS Istituto Ortopedico Galeazzi, Milan - Italy 3 Spine Care Group, Milan - Italy 4 Fatebenefratelli and Oftalmico Hospital, Milan - Italy ABSTRACT Aim: The aim of this study was to evaluate the biomechanical effects of the Maverick disc prosthesis at the implanted and adjacent level by the finite element (FE) method. Materials and Methods: A 3D FE model of the L3-L5 segment was built. To simulate the different physiological movements (flexion, extension, lateral bending, axial rotation) pure moments of 10 Nm were applied. To evaluate the effect of the prosthesis, a 3D model of the device was built and inserted in the L3-L5 model. The ROMs obtained with the intact model were imposed as maximal rotations to the instrumented model, therefore implementing the Panjabi hybrid protocol. Results: Increased ROMs at the implanted level and reduced ROMs at the adjacent level were predicted. A similar momentrotation behavior was calculated after simulation of prosthesis insertion. No significant effect was predicted in terms of von Mises stress at the adjacent level after implantation of the prosthesis. Conclusions: Within the limitations of the models, the numerical results of this study predicted a preserved kinematics and stress at the adjacent segment, after insertion of the prosthesis. Key words: Lumbar disc prosthesis, Hybrid protocol, Finite element, Biomechanics Accepted: 09/06/10 INTRODUCTION Spinal fusion has been identified as the gold standard in the treatment of degenerative disc disease (DDD) (1). Literature data indicated a correlation between fusion and the development of radiographic adjacent segment degeneration and symptomatic adjacent segment disease (2). The use of total disc prostheses in the treatment of DDD in the lumbar spine has led to satisfactory clinical outcomes as well as complications, therefore pointing out the need for further investigations (3). The Maverick Total Disc Replacement (Medtronic Sofamor Danek, Inc., Memphis, TN, USA) was developed and first implanted in This is a so called ball-and-socket disc prosthesis made of cobalt-chromium-molybdenum. Tournier et al (4) set up experimental studies to evaluate the effect of this prosthesis for sagittal balance and lumbar spine movement. The motion characteristics of the intact lumbar spine after anterior L4 5 discectomy and after implantation of the Maverick disc prosthesis were investigated in human cadaveric spines under quasistatic loading conditions by Hitchon et al (5). The aim of this study was to evaluate the biomechanical effects of the Maverick disc prosthesis at the implanted and adjacent level by the finite element (FE) method. Motion characteristics, facet forces, von Mises stress of an intact and an instrumented L3-L5 segment were predicted following the hybrid approach. MATERIALS AND METHODS A 3D FE model of the L3-L5 segment was built (Fig. 1a). The geometry was reconstructed from computed tomography scans taken from the Visible Human Dataset. The 3D reconstruction was performed using commercial software (Amira, TGS, San Diego, CA, USA). The geometry was exported to an FE pre-processor software (Gambit, Fluent Inc, Lebanon, NH, USA). The intervertebral discs were represented as continuum structures occupying the intervertebral spaces. Each structure was subdivided in an internal and an external volume representing the nucleus pulposus and the annulus fibrosus, respectively. Each volume was discretized using linear tetrahedral elements. The cortical bone was modeled using triangular shell elements of 0.35 mm thickness at the endplates and 0.5 mm at the lateral side of the vertebral bodies (6). Table I gives the material properties assumed for each structure. In order to include the contribution of the facet joints, a contact condition was defined between the surfaces of the articular facets, with a friction coef- 97

2 Lumbar total disc replacement: a numerical study TABLE I - MATERIAL PROPERTIES ASSUMED FOR THE FE MODELS Material Moduli [MPa] Poisson s ratio References Cortical bone E = ν = 0.3 [13] Posterior bone E = 3500 ν = 0.25 [13] Cancellous bone E xx = 112 ν xy = 0.3 E yy = 112 ν xy = 0.3 [14] E zz = 340 ν zx = 0.1 G xy =52 ν zy = 0.1 G zx = 53 G zy = 53 Annulus ground substance E = 0.7 ν = 0.45 [9] (calibration) Nucleus pulposus E = [9] (calibration) Annulus collagen fibers Force-deflection curves (calibration) [15] Ligaments Force-deflection curves (calibration) [16] Implant Loading beam Fig. 1-3D FE models: (a) intact model, (b) instrumented model. Fig. 2-3D FE model of the Maverick disc prosthesis. ficient of (7). The annulus fibrosus was modeled by three criss-cross layers representing the collagen fibers embedded in a homogeneous volume representing the ground substance (8, 9). The collagen fiber network was modeled by monodimensional, non-linear, tension only spring elements. The same elements were used to model the ligaments. The ligaments included in the model were: anterior longitudinal, posterior longitudinal, flavum, intertransverse, interspinous, supraspinous and capsular. Material properties of the annulus ground substance, the collagen fibers and the ligaments were obtained by a calibration process. For the intact model, the nodes belonging to the inferior endplate of the L5 vertebra were kept fixed. A rigid beam connection was defined between the superior endplate of the L3 vertebra and a node located at the superior space. Pure moments of 10 Nm were applied to this node to simulate the different physiological movements (flexion, extension, lateral bending, axial rotation). To evaluate the effect of the prosthesis, a 3D model of the device was built (Fig. 2) and inserted in the L3-L5 model (Fig. 1b). To simulate the surgical procedure, the nucleus, the anterior longitudinal ligament, the anterior portion of the annulus and part of the posterior portion of the annulus were removed. A contact condition with a friction coefficient equal to 0.04 (10) was assumed between the metal components of the prosthesis, while a fully bonded contact condition was defined between the device and the endplates, therefore, assuming a complete osteointegration. Rotation-controlled simulation was conducted for each loading condition (flexion, extension, lateral ben- 98

3 Di Mascio et al ding, axial rotation) with the instrumented model. The ROMs obtained with the intact model were imposed as maximal rotations to the instrumented model, therefore, implementing the Panjabi hybrid protocol (11). The analyses were conducted using the commercial software Abaqus 6.6 (Simulia, Providence, RI, USA). RESULTS Figure 3 shows the calculated ROMs for the intact model and the instrumented model, in all loading conditions. Insertion of the prosthesis caused an increment in ROM at the implanted level (L4-L5) and a reduction at the adjacent intact level (L3-L4) during flexion, extension and lateral bending, while a significant difference was not detected in axial rotation. Figures 4-6 show the calculated moment as a function of the applied rotation, for the whole segment, in all loading conditions. A similar trend was predicted as comparing the models, but the final moment resulted in a decrease of about 2 Nm in flexion, 4.3 Nm in extension, 2.4 Nm in lateral bending and 1 Nm in axial rotation. Figures 7 and 8 show von Mises stress calculated at the Fig. 3 - Maximal ROM values for the intact model and the instrumented model in flexion, extension, lateral bending and axial rotation. Fig. 4 - Moment-rotation curves for the intact model and the instrumented model during flexion and extension. Fig. 5 - Moment-rotation curves for the intact model and the instrumented model during lateral bending. Fig. 6 - Moment-rotation curves for the intact model and the instrumented model during axial rotation. 99

4 Lumbar total disc replacement: A numerical study Fig. 7 - Von Mises stress at the adjacent intervertebral disc (L3-L4) for the intact model and the instrumented model: (a) flexion; (b) extension. Fig. 9 - Relative facet force (the ratio between the facet force of the instrumented model and the facet force of the intact model) during extension. adjacent intervertebral disc (L3-L4) for both the models. No significant difference was predicted for all loading conditions comparing the intact model to the instrumented model. Figure 9 shows the relative facet force (the ratio between the facet force of the instrumented model and the facet force of the intact model), in extension. A reduction was detected at both the adjacent level (L3-L4) and the implanted level (L4-L5), but the reduction was more significant at the implanted level. DISCUSSION Fig. 8 - Von Mises stress at the adjacent intervertebral disc (L3-L4) for the intact model and the instrumented model: (a) lateral bending; (b) axial rotation. This study was conducted to evaluate the biomechanical effects of the implantation of the Maverick disc prosthesis at the L4-L5 level by an FE model which included the adjacent L3-L4 level. Increased ROMs were predicted at the implanted level, while a reduction was calculated at the adjacent level. At the implanted level, the range of variation was estimated to be between 3.8% and 20% (axial rotation and lateral bending, respectively). The increase in ROM at the implanted level during flexion and extension may be related to the lack of the anterior portion of the annulus and of the ALL, as previously observed (5). In the literature, an increase of flexibility in all directions at the implanted level was predicted after insertion of the Maverick disc prosthesis, compared to the intact condition (5). Different values of variation could be related to the different loading protocol and to the insertion of the prosthesis at a different level. Specifically, to test the prosthesis a load-control protocol was used. This protocol involving the application of pure moments to the intact and instrumented spine is not suitable to evaluate the effects of disc prostheses at the adjacent level (12). This investigation was conducted by using the hybrid protocol which is believed to be more clinically relevant (12) allowing the investigation of the effects of prosthesis implantation at the instrumented and adjacent level. In axial rotation, the non-significant variation in flexibility (3.8%) at the implanted level may be related to the mechanical role of the articular facets, which shift the instantaneous axis of rotation posteriorly, therefore, limiting the movement. A similar load-displacement behavior was predicted as comparing the instrumented model to the intact model (Figs. 4-6). This was particularly accentuated in axial rotation due to the role of the mechanical facet force, as stated above. Generally speaking, under all loading conditions a decrease in moment values after implantation of the prosthesis was predicted, therefore, indicating a reduced resistance to rotation. No significant effect was predicted at the adjacent level after implantation of the prosthesis in terms of von 100

5 Di Mascio et al Mises stress, indicating that the prosthesis does not induce overloading in the adjacent segments. A significant reduction in facet force was calculated at the implanted level during extension. This could be related to posterior location of the center of rotation imposed by the prosthesis, which unloads the facet joints. The numerical results of this study have to be regarded as theoretical predictions of the biomechanical effects of the prosthesis because an FE model represents a simplification of the real structure. Limitations of this study pertain to the physiological model of the biological structures; therefore, neglecting any degenerative condition of the remaining portion of the annulus fibrosus, the adjacent disc and the ligaments. At the implant/bone interface a full osteointegrated condition was assumed by modeling a bonded contact condition, neglecting any possible micromotion and subsidence of the device. Changes in the sagittal balance could be induced by implanting a total disc prosthesis. In this study, due to the lack of muscle tissues these changes could not be evaluated. However, sagittal balance after implantation of the Maverick disc prosthesis was investigated by Tournier et al (4) by in vivo biomechanical studies. They found that after implantation of the prosthesis, patients were able to maintain the preoperative sagittal balance. The current FE model was calibrated to evaluate the intervertebral disc and ligaments mechanical behavior, but no experimental tests on human cadavers were performed. Further developments of the study could concern the evaluation of clinically relevant variables such as prosthesis position, prosthesis size and type of anulotomy. Conflict of interest: There is no conflict of interest with any organization regarding the material discussed in the manuscript. Address for correspondence: Fabio Galbusera IRCCS Istituto Ortopedico Galeazzi Via R. Galeazzi Milano - Italy fabio.galbusera@polimi.it REFERENCES 1. Nockels RP. Dynamic stabilization in the surgical management of painful lumbar spinal disorders. Spine 2005; 15: S Harrop JS, Youssef JA, Maltenfort M, et al. Lumbar adjacent segment degeneration and disease after arthrodesis and total disc arthroplasty. Spine 2008; 33: Galbusera F, Bellini CM, Zweig T, et al. Design concepts in lumbar total disc arthroplasty. Eur Spine J 2008; 17: Tournier C, Aunoble S, Le Huec JC, et al. Total disc arthroplasty: consequences for sagittal balance and lumbar spine movement. Eur Spine J 2007; 16: Hitchon PW, Eichholz K, Barry C, et al. Biomechanical studies of an artificial disc implant in the human cadaveric spine. J Neurosurg Spine 2005; 2: Silva MJ, Wang C, Keaveny TM, Hayes WC. Direct and computed tomography thickness measurements of the human, lumbar vertebral shell and endplate. Bone 1994; 15: Forster H, Fisher J. The influence of continuous sliding and subsequent surface wear on the friction of articular cartilage. Proc Inst Mech Eng H 1999; 213: Cheung JT, Zhang M, Chow DH. Biomechanical responses of the intervertebral joints to static and vibrational loading: a finite element study. Clin Biomech (Bristol, Avon) 2003; 18: Schmidt H, Heuer F, Simon U, Kettler A, Rohlmann A, Claes L, Wilke HJ. Application of a new calibration method for a threedimensional finite element model of a human lumbar annulus fibrosus. Clin Biomech (Bristol, Avon) 2006; 21: Unsworth A, Hall RM, Burgess IC, Wroblewski BM, Streicher RM, Semlitsch M. Frictional resistance of new and explanted artificial hip joints. Wear 1995; 190: Panjabi MM. Hybrid multidirectional test method to evaluate spinal adjacent-level effects. Clin Biomech (Bristol, Avon) 2007; 22: Goel VK, Grauer JN, Patel TCh, et al. Effects of charité artificial disc on the implanted and adjacent spinal segments mechanics using a hybrid testing protocol. Spine 2005; 30: Lee KK, Teo EC, Fuss FK, et al. Finite-element analysis for lumbar interbody fusion under axial loading. IEEE Trans Biomed Eng 2004; 51: Crawford RP, Cann CE, Keaveny TM. Finite element models predict in vitro vertebral body compressive strength better than quantitative computed tomography. Bone 2003; 33: Shirazi-Adl A, Ahmed AM, Shrivastava SC. Mechanical response of a lumbar motion segment in axial torque alone and combined with compression. Spine 1986; 11: Eberlein R, Holzapfel GA, Frölich M. Multi-segment FEA of the human lumbar spine including the heterogeneity of the annulus fibrosus. Comput Mech 2004; 34: