ARTICLE IN PRESS
JID: JJBE
[m5G;July 18, 2015;10:12]
Medical Engineering and Physics 000 (2015) 1–6
Contents lists available at ScienceDirect
Medical Engineering and Physics journal homepage: www.elsevier.com/locate/medengphy
Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests Kao-Shang Shih a,b,c, Ching-Chi Hsu d,∗, Sheng-Mou Hou a,b,c, Shan-Chuen Yu e, Chen-Kun Liaw a,b a
Department of Orthopedic Surgery, Shin Kong Wu Ho-Su Memorial Hospital, Taipei 111, Taiwan, ROC College of Medicine, Fu Jen Catholic University, Taipei 242, Taiwan, ROC School of Medicine, Taipei Medical University, Taipei 110, Taiwan, ROC d Graduate Institute of Applied Science and Technology, National Taiwan University of Science and Technology, No. 43, Sec. 4, Keelung Rd, Taipei 106, Taiwan, ROC e Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei 106, Taiwan, ROC b c
a r t i c l e
i n f o
Article history: Received 10 June 2014 Revised 21 May 2015 Accepted 24 June 2015 Available online xxx Keywords: Bending performance Breakage Cannulated pedicle screw Finite element analysis Biomechanical test
a b s t r a c t Spinal pedicle screw fixations have been used extensively to treat fracture, tumor, infection, or degeneration of the spine. Cannulated spinal pedicle screws with bone cement augmentation might be a useful method to ameliorate screw loosening. However, cannulated spinal pedicle screws might also increase the risk of screw breakage. Thus, the purpose of this study was to investigate the bending performance of different spinal pedicle screws with either solid design or cannulated design. Three-dimensional finite element models, which consisted of the spinal pedicle screw and the screw’s hosting material, were first constructed. Next, monotonic and cyclic cantilever bending tests were both applied to validate the results of the finite element analyses. Finally, both the numerical and experimental approaches were evaluated and compared. The results indicated that the cylindrical spinal pedicle screws with a cannulated design had significantly poorer bending performance. In addition, conical spinal pedicle screws maintained the original bending performance, whether they were solid or of cannulated design. This study may provide useful recommendations to orthopedic surgeons before surgery, and it may also provide design rationales to biomechanical engineers during the development of spinal pedicle screws. © 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Posterior pedicle screw and rod fixation systems have been widely used for the treatment of degenerative diseases of human spines and the stabilization of fractured vertebrae [1–3]. In clinical applications, spinal pedicle screws are the key components of pedicle screw and rod systems [4]. However, patients might have non-union or other complications when breakage, loosening, or pulling out of spinal pedicle screws occurs [5–7]. A breakage of spinal pedicle screws should be avoided because a broken screw fragment trapped in the vertebral body is difficult to remove and may impede subsequent revision surgeries [8]. The interfacial strength of pedicle screw fixation in an osteoporotic spine can be improved by the addition of bone cement [9]. One of the injection techniques is cannulated pedicle screws with bone cement augmentation. To reduce the risk of pedicle screw breakage, past studies investigated the bending performance of traditional solid pedicle screws using biomechanical tests
and/or finite element analyses [10–12]. However, there is little research on the bending performance of cannulated pedicle screws. To our knowledge, past research studies were mainly focused on the pullout strength of cannulated or expandable pedicle screws [13–16]. In fact, the risk of pedicle screw breakage may become more serious, especially when cannulated pedicle screws were used. In the present study, three-dimensional finite element models were first constructed to calculate the bending performance of spinal pedicle screws. Next, samples of each type of spinal pedicle screws were fabricated and tested to validate the results of the finite element simulations. Finally, both the numerical and experimental results were compared and discussed. The purpose of this study was to investigate the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests. 2. Materials and methods 2.1. Designs of the spinal pedicle screws
∗
Corresponding author. Tel.: +886 2 27303771; fax: +886 2 27303733. E-mail address: [email protected] (C.-C. Hsu).
Most spinal pedicle screws have either a cylindrical or conical core. To investigate the effects of different screw design, two groups
http://dx.doi.org/10.1016/j.medengphy.2015.06.008 1350-4533/© 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
ARTICLE IN PRESS
JID: JJBE 2
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
Fig. 2. Loading and boundary conditions of the finite element model.
Fig. 1. Six designs of the spinal pedicle screws.
of spinal pedicle screws were considered in the present study, including the cylindrical design group and the conical design group. For each design group, solid and cannulated spinal pedicle screws were considered. In addition, a cannulated spinal pedicle screw with an additional 2-mm pin was also considered to restore its biomechanical performances. Thus, six types of spinal pedicle screws were designed and considered: cylindrical and solid design (Cy-SD), cylindrical and cannulated design (Cy-CD), cylindrical and cannulated design with a pin (Cy-CDP), conical and solid design (Co-SD), conical and cannulated design (Co-CD), and conical and cannulated design with a pin (Co-CDP) (Fig. 1). All of the spinal pedicle screws have the same screw thread and their geometry and dimension were referred to the design ranges of commercial products. The area moment of inertia at the centroid for each spinal pedicle screw was calculated. 2.2. Finite element models The solid models of the spinal pedicle screws and the screw’s hosting materials were constructed and assembled using SolidWorks 2013 (SolidWorks Corporation, Concord, MA, USA). The geometry and dimension of the spinal pedicle screws were as follows: outer diameter of 6.5 mm; inner diameter of 3.3 mm; cannulation diameter of 2 mm; pitch of 2.83 mm; proximal root radius of 0.4 mm; distal root radius of 1.2 mm; thread width of 0.1 mm; proximal half angle of 5°; distal half angle of 25°; conical angle of 2.04°; and screw length of 45 mm. The screw’s hosting material was simplified as a cylinder with a diameter of 20 mm and a length of 45 mm [10]. The solid models of the spinal pedicle screw and the bone were converted into parasolid format and transferred to the ANSYS Workbench 14.5 (ANSYS, Inc., Canonsburg, PA, USA). The spinal pedicle screws were made from titanium and their material properties were set as 110 GPa for the elastic modulus and 0.3 for the Poisson’s ratio. For the material properties of the bone, the elastic modulus was set as 2.6 GPa and Poisson’s ratio was set as 0.3 [4]. The material properties of the bone were referred to the material of the testing specimens used in this study. All solid models were free-meshed using 10-node tetrahedral elements (SOLID 187). A convergent study was conducted by decreasing the element size of the local mesh regions. The interfacial condition
Fig. 3. (A) The experimental setup. (B) A specimen after the monotonic tests. (C) A specimen after the cyclic cantilever bending tests.
between the spinal pedicle screw and the screw’s hosting material was assumed to be in contact without any frictional force. In addition, the interfacial condition between the metallic pin and the screw was also assumed to be in contact. The boundary condition was fully constrained at the surfaces of the pedicle screw head. In the loading condition, a vertical load of 220 N was applied on the bone (Fig. 2). This loading was determined by averaging the cyclic loading of 40∼400 N. During post-processing, the maximum displacement of the numerical model was used to represent the yielding strength, and the lower maximum displacement reveals the higher yielding strength. In addition, the maximum tensile stress of the spinal pedicle screw was chosen to represent the fatigue strength, and the lower maximum tensile stress reveals the better fatigue strength. 2.3. Monotonic tests of the spinal pedicle screws The testing specimens used in this study consisted of the spinal pedicle screw and the bone. High-molecular-weight polyethylene cylinders were used instead of real bones in the present study. The cylinders had a diameter of 20 mm and a length of 45 mm. The diameter of the predrilled holes was 3.3 mm. The spinal pedicle screw was inserted into the predrilled hole, and the insertion length of the pedicle screws was 40 mm. In this study, six types of the spinal pedicle screws were manufactured and tested. The pedicle screw head of the testing specimens were gripped in a specially designed jig (Fig. 3A). Monotonic tests were conducted by applying a static load at the distal part of the spinal pedicle screws using a materials
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
ARTICLE IN PRESS
JID: JJBE
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
3
Table 1 Numerical and experimental results of six types of the spinal pedicle screws. Cylindrical design group Cylindrical and solid design (Cy-SD) Finite element analyses Maximum displacement 2.13 (mm) Maximum tensile stress 2726 (MPa) Yielding tests Stiffness (N/mm) 56 ± 1 Yielding load (N) 87 ± 2 Fatigue tests (6∼60 N) Fatigue life (cycles) 1,000,000 Multi-cycle stiffness 70 ± 2 (N/mm) Fatigue tests (40∼400 N) Fatigue life (cycles) – Multi-cycle stiffness – (N/mm) Area moment of inertia at the centroid 36.48 X-direction (mm4 ) 14.12 Y-direction (mm4 ) 4 22.36 Z-direction (mm )
Conical design group
Cylindrical and cannulated design (Cy-CD)
Cylindrical and cannulated design with a pin (Cy-CDP)
Conical and solid design (Co-SD)
0.56
Conical and cannulated design (Co-CD)
0.57
Conical and cannulated design with a pin (Co-CDP)
2.40
2.20
0.57
3272
2839
409
423
414
49 ± 3 68 ± 2
50 ± 1 68 ± 4
305 ± 9 677 ± 4
299 ± 4 676 ± 2
300 ± 4 677 ± 4
134,816 ± 34,077 62 ± 3
139,043 ± 42,932 64 ± 4
1,000,000 338 ± 10
1,000,000 335 ± 5
1,000,000 333 ± 3
– –
– –
1,000,000 348 ± 9
1,000,000 349 ± 8
1,000,000 347 ± 11
32.44 12.60 19.83
36.48 14.12 22.36
170.01 85.21 84.80
168.44 84.42 84.01
170.01 85.21 84.80
testing machine (Instron 8872, Instron Corporation, Canton, MA, USA). A loading rate of 2.5 mm/min in displacement control mode was applied in the monotonic tests, and the loading continued until the displacement reached 6 mm to ensure plastic deformation of the pedicle screw. The load-displacement curves were recorded at a data acquisition rate of 100 Hz. The monotonic tests for each type of spinal pedicle screw were repeated six times, and the stiffness and the yielding load were obtained. One-way ANOVA with Bonferroni post-hoc tests was used. All P values less than 0.05 were considered to indicate statistically significant differences. A correlation study between the maximum displacement calculated by the finite element models and the yielding load obtained by the monotonic tests was conducted. 2.4. Cyclic cantilever bending tests of the spinal pedicle screws The testing specimens and the experimental jig used in the cyclic cantilever bending tests were the same as those used in the monotonic tests. A 10-Hz cyclic loading with sinusoidal waveform in load control mode was applied. Two loading conditions, 60 and 400 N, were tested with a stress ratio of 10%. The cyclic cantilever bending tests were terminated when the number of testing cycles was more than one million or when the displacement of the actuator was beyond 8 mm. All experimental data were continuously monitored at a data acquisition rate of 50 Hz. The cyclic cantilever bending tests for each type of spinal pedicle screw were also repeated six times, and the multi-cycle stiffness and the fatigue life were obtained. One-way ANOVA was also used. The correlation study between the maximum tensile stress calculated by the finite element models and the logarithm of the fatigue life obtained by the cyclic cantilever bending tests was conducted. 3. Results 3.1. Finite element analyses The total number of elements varied from 155,000 to 205,000, and the solution time varied from 5 to 8 h. In the convergent analysis, both the maximum displacement and the maximum tensile stress were found to converge as expected, and the variation was less than 10%. For the results of the maximum displacement, the conical design group (Co-SD, Co-CD, and Co-CDP) revealed the lower maximum
Fig. 4. The stress distribution of the spinal pedicle screws.
displacement compared to the cylindrical design group (Cy-SD, CyCD, and Co-CDP) (Table 1), with the average percentage decrease being approximately 75%. In addition, the Co-SD had the lowest maximum displacement of the conical design group. Similarly, the CySD had the lowest maximum displacement of the cylindrical design group. For the results of the maximum tensile stress, the maximum tensile stress was located in the proximal part of the spinal pedicle screw (Fig. 4). The maximum tensile stress of the conical design group was significantly lower than that of the cylindrical design group (Table 1), and the average percentage decrease was approximately 86%. In addition, the Co-CD and the Cy-CD had the highest maximum tensile stress of the spinal pedicle screws for the conical design group and the cylindrical design group, respectively. 3.2. Monotonic tests of the spinal pedicle screws All of the spinal pedicle screws were found to undergo plastic deformation at the proximal part (Fig. 3B). The load-displacement curve was linear initially and had an obvious yielding (Fig. 5A).
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE 4
ARTICLE IN PRESS
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
Fig. 5. (A) The load-displacement curve in the monotonic tests. (B) The displacement-cycle curve in the cyclic cantilever bending tests.
The stiffness of the conical design group was significantly higher than that of the cylindrical design group (p < 0.05). However, the solid design, the cannulated design, and the cannulated design with a pin did not exhibit a statistically significantly difference for both the conical design group and the cylindrical design group (Table 1). The yielding load was defined by the 0.2% offset method. In the present study, the yielding load of the conical design group was significantly higher than that of the cylindrical design group (p < 0.05). In addition, the yielding load of the Cy-SD was significantly higher than that of the Cy-CD and the Cy-CDP (p < 0.05). However, the yielding loads of the Co-SD, the Co-CD, and the Co-CDP were quite similar and had no statistically significantly difference (Table 1). In the correlation study, the maximum displacement was closely related to the yielding load with a correlation coefficient of –0.99. In addition, the area moment of inertia was also closely related to the yielding load with a correlation coefficient of 0.99. 3.3. Cyclic cantilever bending tests of the spinal pedicle screws Two loading conditions, 60 and 400 N, with a stress ratio of 10% were applied to test each spinal pedicle screw. The loading condition of 60 N was determined by 90% of the lowest yielding load of the spinal pedicle screws (the Cy-CD and the Cy-CDP). In addition, the loading condition of 400 N was also applied to obtain additional outcomes. For the results of the cyclic cantilever bending tests at 60 N, fatigue failure occurred in the designs of the Cy-CD and the Cy-CDP, and the crack initiated in the proximal part of the spinal pedicle screws (Fig. 3C) (Fig. 5B). Other pedicle screw designs remained intact when the number of cycles to failure reached one million. The multi-cycle stiffness of the conical design group was significantly higher than that of the cylindrical design group (p < 0.05) (Table 1). In the correlation study, the maximum tensile stress was related to the logarithm of the fatigue life with a correlation coefficient of –0.76. In addition, the area moment of inertia was also closely related to the logarithm of the fatigue life with a correlation coefficient of 0.71. For the results of the cyclic cantilever bending tests at 400 N, none of the conical design group failed when the number of cycles to failure reached one million. However, the fatigue life of the cylindrical design group was unable to be obtained because the fatigue loading of 400 N was significantly larger than their yielding load. The multi-cycle stiffness of
the Co-SD, the Co-CD, and the Co-CDP were quite similar and had no statistically significant difference (Table 1). 4. Discussion Spinal pedicle screws are usually the weakest part of the posterior pedicle screw and rod fixation systems in clinical applications [17,18]. Thus, improving the biomechanical performances of spinal pedicle screws would also improve the clinical performances of the pedicle screw and rod fixation systems [10]. In the present study, the results of the finite element models indicated that the maximum tensile stress occurred in the proximal part of the spinal pedicle screws. Compared to the results of the cyclic cantilever bending tests, the point with the maximum tensile stress corresponded to the failure site of the spinal pedicle screws. In addition, the correlation study revealed that the numerical results were closely related to the experimental outcomes. Although the correlation study showed the feasibility of the numerical models, the implants had a high stress level of more than 2,700 MPa. No material of the titanium-bone complex will ever withstand this high stress level. Thus, a nonlinear material condition might need to be considered. Bending performances in terms of the pedicle screw designs have been discussed in past studies. Cho et al. [19] mentioned that the outer diameter of the pedicle screw determines the pullout strength, while the inner diameter determines the fatigue strength. In addition, Chao et al. [4] compared different designs of pedicle screws and concluded that conical pedicle screws yield a significantly higher bending performance than cylindrical pedicle screws. The above studies implied that pedicle screws with larger inner diameter or of conical core design effectively improved their bending performance. In the present study, the spinal pedicle screws from the conical design group (Co-SD, Co-CD, and Co-CDP) had smaller maximum tensile stress, higher yielding load, and longer fatigue life compared to the cylindrical design group (Cy-SD, Cy-CD, and Cy-CDP). Thus, the spinal pedicle screws with conical core design significantly improved their bending performances. This finding was the same as the results of the past studies mentioned above. Brown et al. [20] concluded that simple-beam theory is capable of providing enough evidence to support the optimization of the design of cylindrical bone screws. In the present study, the area moment of inertia for each spinal pedicle
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE
ARTICLE IN PRESS
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
screw was calculated to validate this finding. According to the results of the correlation study, this theory was also capable to support the design of conical bone screws. Although the numerical outcomes of the current study were validated according to the past studies, the numerical models of the present study were unable to provide the absolute data. Additionally, maybe using the relationship between the maximum tensile stress and the fatigue life is not the best way, but this is one of possible ways to dig the biomechanical performances of cannulated pedicle screws. Cannulated or expandable spinal pedicle screws with bone cement augmentation had been used to increase their anchoring strength in osteoporotic spine [21–23]. However, spinal pedicle screws with cannulated design might dramatically reduce their bending or fatigue strength. Reese et al. [24] tested cannulated and noncannulated screws of different screw diameter using cyclic loading. They found that the largest screw possible should be used for surgical fixation of Jones fractures. Although Reese’s study investigated both cannulated and noncannulated screws for the treatment of Jones fractures, their conclusion provides useful evidence for the present study because the fatigue life of the Cy-SD (Solid design) was significantly increased compared with that of the Cy-CD (cannulated design). The fatigue life of the Cy-SD was almost eight times higher than that of the Cy-CD. However, it was interesting that there is no statistically significant difference between the cannulated screw (the Co-CD) and the solid screw (the Co-SD) for the conical design group. Thus, a spinal pedicle screw with a conical core design is recommended if it needs to have a cannulated design. Two loading conditions, 60 and 400 N, were tested with a stress ratio of 10%. The reason to apply two loading conditions was explained as follow: the value of 60 N was calculated according to the 90% of the yielding strength of the Cy-CD design (the screw with lowest yielding strength). However, four types of the spinal pedicle screws remain intact after one million cyclic loading in the fatigue tests with a loading condition of 60 N. Thus, a loading condition of 400 N was applied to get additional outcome of the spinal pedicle screws. A past study [24] and the present study demonstrated that cylindrical bone screws with cannulated design would lose their original bending performances. Thus, retrieving the loss of material in the cannulated screws might avoid the loss of the original bending performances. In the present study, the cannulated pedicle screws with an additional pin were investigated to retrieve their biomechanical performances. For the cylindrical design group, the numerical results indicated that both the maximum displacement and the maximum tensile stress of the Cy-CDP could be retrieved compared to that of the Cy-CD. Unfortunately, this finding was not proved in the monotonic and cyclic cantilever bending tests. There are two possible reasons that account for this inconsistency. First, the numerical models may have eliminated the gap between the pin and cannulated hole and make a perfect contact interface. However, the experimental specimens had a gap caused by a machining tolerance. This gap might increase the deformation and stress. Second, all of the experimental specimens would yield when the loading exceeded their yielding point. However, the ideal linear elastic material was used for the finite element models. Thus, the bending performances of the cannulated design were achieved by using the numerical approach, but the screw was found to fail using the experimental approach. For the conical design group, retrieving the loss of the original bending performances in the cannulated design was also found to fail using the experimental approach. In fact, the present study proved that there is no statistically significant difference between the cannulated screw (the Co-CD) and the solid screw (the Co-SD) for the conical design group. Thus, it was easy to understand that adding a pin to the cannulated screws did not improve the bending performances. There are limitations in the current numerical and experimental approaches that may affect the ability to completely characterize the biomechanical behaviors of the spinal pedicle screws. First, although
5
the real geometry of the spinal pedicle screws was constructed during the computer simulations, the geometry of spinal vertebrae was simplified as a cylinder. This simplification might affect the numerical models ability to predict the actual biomechanical performances. Second, a linear material was used in the present study. This assumption might also affect the numerical models ability to predict the actual biomechanical performances. Third, only the bending performance of the spinal pedicle screws with either solid design or cannulated design was evaluated. However, the bending performance was just one of the complications of the spinal pedicle screws. Other performances should also be considered together, such as the interfacial strength between the pedicle screw and the screw’s hosting material and/or the maximum stress of the vertebrae. Finally, the bending performances of different screw structures were evaluated in this study. However, other screw parameters may also play an important role, such as the inner diameter, the ratio of the outer over the inner diameter, the diameter of cannulated hole, and different insertion technique. In conclusion, cylindrical spinal pedicle screws with a cannulated design may significantly deteriorate the bending performances. In addition, conical spinal pedicle screws maintained the original bending performances, whether they were solid or of a cannulated design. Adding a pin to the cannulated spinal pedicle screws did not improve the bending performances. This study may provide useful recommendations to orthopedic surgeons before surgery, and it could also provide design rationales to biomechanical engineers during the development of spinal pedicle screws. Conflict of interest statement None of the authors have any conflicts of interest to declare. Ethical approval Not required. Acknowledgments This work was sponsored by the Shin Kong Wu Ho-Su Memorial Hospital Research Program under the Grant No. SKH-8302-102DR-14. References [1] Silbermann J, Riese F, Allam Y, Reichert T, Koeppert H, Gutberlet M. Computer tomography assessment of pedicle screw placement in lumbar and sacral spine: comparison between free-hand and O-arm based navigation techniques. Eur Spine J. 2011;20:875–81. [2] Verlaan J-J, Dhert WJA, Oner FC. Intervertebral disc viability after burst fractures of the thoracic and lumbar spine treated with pedicle screw fixation and direct end-plate restoration. The Spine J. 2013;13(3):217–21. [3] Pihlajamaki H, Myllynen P, Bostman O. Complications of transpedicular lumbosacral fixation for non-traumatic disorders. J Bone Joint Surg. 1997;79B(2):183–9. [4] Chao CK, Hsu CC, Wang JL, Lin J. Increasing bending strength and pullout strength in conical pedicle screws: biomechanical tests and finite element analyses. J Spinal Disord Tech. 2008;21(2):130–8. [5] Hicks J, Singla A, Shen FH, Arlet V. Complications of pedicle screw fixation in scoliosis surgery: a systematic review. Spine. 2010;35(11):E465–70. [6] Okuyama K, Abe E, Suzuki T, Tamura Y, Chiba M, Sato K. Influence of bone mineral density on pedicle screw fixation: a study of pedicle screw fixation augmenting posterior lumbar interbody fusion in elderly patients. The Spine J. 2001;1(6):402–7. [7] Chatzistergos PE, Magnissalis EA, Kourkoulis SK. A parametric study of cylindrical pedicle screw design implications on the pullout performance using an experimentally validated finite-element model. Med Eng Phy. 2010;32(2):145–54. [8] Vanichkachorn JS, Vaccaro AR, Cohen MJ, Cotler J. Potential large vessel Injury during thoracolumbar pedicle screw removal: a case report. Spine. 1997;22(1):110–13. [9] Paré PE, Chappuis JL, Rampersaud R, Agarwala AO, Perra JH, Erkan S, et al. Biomechanical evaluation of a novel fenestrated pedicle screw augmented with bone cement in osteoporotic spines. Spine. 2011;36(18):E1210–14.
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE 6
ARTICLE IN PRESS
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
[10] Chao CK, Lin J, Putra ST, Hsu CC. A neuro-genetic approach to a multiobjective design optimization of spinal pedicle screws. J Biomech Eng-T ASME. 2010;132(9):091006. [11] Stanford RE, Loefler AH, Stanford PM, Walsh WR. Multiaxial pedicle screw designs: static and dynamic mechanical testing. Spine. 2004;29:367–75. [12] Chen CS, Chen WJ, Cheng CK, Jao SHE, Chueh SC, Wang CC. Failure analysis of broken pedicle screws on spinal instrumentation. Med Eng Phy. 2005;27(6):487– 96. [13] Cook SD, Salkeld SL, Stanley T, Faciane A, Miller SD. Biomechanical study of pedicle screw fixation in severely osteoporotic bone. The Spine J. 2004;4(4):402–8. [14] Chen LH, Tai CL, Lai PL, Lee DM, Tsai TT, Fu TS, et al. Pullout strength for cannulated pedicle screws with bone cement augmentation in severely osteoporotic bone: influences of radial hole and pilot hole tapping. Clin Biomech. 2009;24(8):613– 18. [15] Frankel BM, D’Agostino S, Wang C. A biomechanical cadaveric analysis of polymethylmethacrylate-augmented pedicle screw fixation. J Neurosurg: Spine. 2007;7(1):47–53. [16] Chao KH, Lai YS, Chen WC, Chang CM, McClean CJ, Fan CY, et al. Biomechanical analysis of different types of pedicle screw augmentation: a cadaveric and synthetic bone sample study of instrumented vertebral specimens. Med Eng Phy. 2013;35(10):1506–12.
[17] Jutte P, Castelein R. Complications of pedicle screws in lumbar and lumbosacral fusions in 105 consecutive primary operations. Eur Spine J. 2002;11(6):594–8. [18] Stoffel M, Behr M, Reinke A, Stüer C, Ringel F, Meyer B. Pedicle screw-based dynamic stabilization of the thoracolumbar spine with the Cosmic®-system: a prospective observation. Acta Neurochir. 2010;152(5):835–43. [19] Cho W, Cho SK, Wu C. The biomechanics of pedicle screw-based instrumentation. J Bone Joint Surg. 2010;98B(8):1061–5. [20] Brown GA, McCarthy T, Bourgeault CA, Callahan DJ. Mechanical performance of standard and cannulated 4.0-mm cancellous bone screws. J Orthop Res. 2000;111:307–12. [21] McKoy BE, An YH. An injectable cementing screw for fixation in osteoporotic bone. J Biomed Mater Res. 2000;53(3):216–20. [22] Fransen P. Increasing pedicle screw anchoring in the osteoporotic spine by cement injection through the implant: technical note and report of three cases. J Neurosurg: Spine. 2007;7(3):366–9. [23] Bullmann V, Schmoelz W, Richter M, Grathwohl C, Schulte TL. Revision of cannulated and perforated cement-augmented pedicle screws: a biomechanical study in human cadavers. Spine. 2010;35(19):E932–9. [24] Reese K, Litsky A, Kaeding C, Pedroza A, Shah N. Cannulated screw fixation of jones fractures: a clinical and biomechanical study. Am J Sport Med. 2004;32(7):1736–42.
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE
[m5G;July 18, 2015;10:12]
Medical Engineering and Physics 000 (2015) 1–6
Contents lists available at ScienceDirect
Medical Engineering and Physics journal homepage: www.elsevier.com/locate/medengphy
Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests Kao-Shang Shih a,b,c, Ching-Chi Hsu d,∗, Sheng-Mou Hou a,b,c, Shan-Chuen Yu e, Chen-Kun Liaw a,b a
Department of Orthopedic Surgery, Shin Kong Wu Ho-Su Memorial Hospital, Taipei 111, Taiwan, ROC College of Medicine, Fu Jen Catholic University, Taipei 242, Taiwan, ROC School of Medicine, Taipei Medical University, Taipei 110, Taiwan, ROC d Graduate Institute of Applied Science and Technology, National Taiwan University of Science and Technology, No. 43, Sec. 4, Keelung Rd, Taipei 106, Taiwan, ROC e Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei 106, Taiwan, ROC b c
a r t i c l e
i n f o
Article history: Received 10 June 2014 Revised 21 May 2015 Accepted 24 June 2015 Available online xxx Keywords: Bending performance Breakage Cannulated pedicle screw Finite element analysis Biomechanical test
a b s t r a c t Spinal pedicle screw fixations have been used extensively to treat fracture, tumor, infection, or degeneration of the spine. Cannulated spinal pedicle screws with bone cement augmentation might be a useful method to ameliorate screw loosening. However, cannulated spinal pedicle screws might also increase the risk of screw breakage. Thus, the purpose of this study was to investigate the bending performance of different spinal pedicle screws with either solid design or cannulated design. Three-dimensional finite element models, which consisted of the spinal pedicle screw and the screw’s hosting material, were first constructed. Next, monotonic and cyclic cantilever bending tests were both applied to validate the results of the finite element analyses. Finally, both the numerical and experimental approaches were evaluated and compared. The results indicated that the cylindrical spinal pedicle screws with a cannulated design had significantly poorer bending performance. In addition, conical spinal pedicle screws maintained the original bending performance, whether they were solid or of cannulated design. This study may provide useful recommendations to orthopedic surgeons before surgery, and it may also provide design rationales to biomechanical engineers during the development of spinal pedicle screws. © 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Posterior pedicle screw and rod fixation systems have been widely used for the treatment of degenerative diseases of human spines and the stabilization of fractured vertebrae [1–3]. In clinical applications, spinal pedicle screws are the key components of pedicle screw and rod systems [4]. However, patients might have non-union or other complications when breakage, loosening, or pulling out of spinal pedicle screws occurs [5–7]. A breakage of spinal pedicle screws should be avoided because a broken screw fragment trapped in the vertebral body is difficult to remove and may impede subsequent revision surgeries [8]. The interfacial strength of pedicle screw fixation in an osteoporotic spine can be improved by the addition of bone cement [9]. One of the injection techniques is cannulated pedicle screws with bone cement augmentation. To reduce the risk of pedicle screw breakage, past studies investigated the bending performance of traditional solid pedicle screws using biomechanical tests
and/or finite element analyses [10–12]. However, there is little research on the bending performance of cannulated pedicle screws. To our knowledge, past research studies were mainly focused on the pullout strength of cannulated or expandable pedicle screws [13–16]. In fact, the risk of pedicle screw breakage may become more serious, especially when cannulated pedicle screws were used. In the present study, three-dimensional finite element models were first constructed to calculate the bending performance of spinal pedicle screws. Next, samples of each type of spinal pedicle screws were fabricated and tested to validate the results of the finite element simulations. Finally, both the numerical and experimental results were compared and discussed. The purpose of this study was to investigate the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests. 2. Materials and methods 2.1. Designs of the spinal pedicle screws
∗
Corresponding author. Tel.: +886 2 27303771; fax: +886 2 27303733. E-mail address: [email protected] (C.-C. Hsu).
Most spinal pedicle screws have either a cylindrical or conical core. To investigate the effects of different screw design, two groups
http://dx.doi.org/10.1016/j.medengphy.2015.06.008 1350-4533/© 2015 IPEM. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
ARTICLE IN PRESS
JID: JJBE 2
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
Fig. 2. Loading and boundary conditions of the finite element model.
Fig. 1. Six designs of the spinal pedicle screws.
of spinal pedicle screws were considered in the present study, including the cylindrical design group and the conical design group. For each design group, solid and cannulated spinal pedicle screws were considered. In addition, a cannulated spinal pedicle screw with an additional 2-mm pin was also considered to restore its biomechanical performances. Thus, six types of spinal pedicle screws were designed and considered: cylindrical and solid design (Cy-SD), cylindrical and cannulated design (Cy-CD), cylindrical and cannulated design with a pin (Cy-CDP), conical and solid design (Co-SD), conical and cannulated design (Co-CD), and conical and cannulated design with a pin (Co-CDP) (Fig. 1). All of the spinal pedicle screws have the same screw thread and their geometry and dimension were referred to the design ranges of commercial products. The area moment of inertia at the centroid for each spinal pedicle screw was calculated. 2.2. Finite element models The solid models of the spinal pedicle screws and the screw’s hosting materials were constructed and assembled using SolidWorks 2013 (SolidWorks Corporation, Concord, MA, USA). The geometry and dimension of the spinal pedicle screws were as follows: outer diameter of 6.5 mm; inner diameter of 3.3 mm; cannulation diameter of 2 mm; pitch of 2.83 mm; proximal root radius of 0.4 mm; distal root radius of 1.2 mm; thread width of 0.1 mm; proximal half angle of 5°; distal half angle of 25°; conical angle of 2.04°; and screw length of 45 mm. The screw’s hosting material was simplified as a cylinder with a diameter of 20 mm and a length of 45 mm [10]. The solid models of the spinal pedicle screw and the bone were converted into parasolid format and transferred to the ANSYS Workbench 14.5 (ANSYS, Inc., Canonsburg, PA, USA). The spinal pedicle screws were made from titanium and their material properties were set as 110 GPa for the elastic modulus and 0.3 for the Poisson’s ratio. For the material properties of the bone, the elastic modulus was set as 2.6 GPa and Poisson’s ratio was set as 0.3 [4]. The material properties of the bone were referred to the material of the testing specimens used in this study. All solid models were free-meshed using 10-node tetrahedral elements (SOLID 187). A convergent study was conducted by decreasing the element size of the local mesh regions. The interfacial condition
Fig. 3. (A) The experimental setup. (B) A specimen after the monotonic tests. (C) A specimen after the cyclic cantilever bending tests.
between the spinal pedicle screw and the screw’s hosting material was assumed to be in contact without any frictional force. In addition, the interfacial condition between the metallic pin and the screw was also assumed to be in contact. The boundary condition was fully constrained at the surfaces of the pedicle screw head. In the loading condition, a vertical load of 220 N was applied on the bone (Fig. 2). This loading was determined by averaging the cyclic loading of 40∼400 N. During post-processing, the maximum displacement of the numerical model was used to represent the yielding strength, and the lower maximum displacement reveals the higher yielding strength. In addition, the maximum tensile stress of the spinal pedicle screw was chosen to represent the fatigue strength, and the lower maximum tensile stress reveals the better fatigue strength. 2.3. Monotonic tests of the spinal pedicle screws The testing specimens used in this study consisted of the spinal pedicle screw and the bone. High-molecular-weight polyethylene cylinders were used instead of real bones in the present study. The cylinders had a diameter of 20 mm and a length of 45 mm. The diameter of the predrilled holes was 3.3 mm. The spinal pedicle screw was inserted into the predrilled hole, and the insertion length of the pedicle screws was 40 mm. In this study, six types of the spinal pedicle screws were manufactured and tested. The pedicle screw head of the testing specimens were gripped in a specially designed jig (Fig. 3A). Monotonic tests were conducted by applying a static load at the distal part of the spinal pedicle screws using a materials
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
ARTICLE IN PRESS
JID: JJBE
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
3
Table 1 Numerical and experimental results of six types of the spinal pedicle screws. Cylindrical design group Cylindrical and solid design (Cy-SD) Finite element analyses Maximum displacement 2.13 (mm) Maximum tensile stress 2726 (MPa) Yielding tests Stiffness (N/mm) 56 ± 1 Yielding load (N) 87 ± 2 Fatigue tests (6∼60 N) Fatigue life (cycles) 1,000,000 Multi-cycle stiffness 70 ± 2 (N/mm) Fatigue tests (40∼400 N) Fatigue life (cycles) – Multi-cycle stiffness – (N/mm) Area moment of inertia at the centroid 36.48 X-direction (mm4 ) 14.12 Y-direction (mm4 ) 4 22.36 Z-direction (mm )
Conical design group
Cylindrical and cannulated design (Cy-CD)
Cylindrical and cannulated design with a pin (Cy-CDP)
Conical and solid design (Co-SD)
0.56
Conical and cannulated design (Co-CD)
0.57
Conical and cannulated design with a pin (Co-CDP)
2.40
2.20
0.57
3272
2839
409
423
414
49 ± 3 68 ± 2
50 ± 1 68 ± 4
305 ± 9 677 ± 4
299 ± 4 676 ± 2
300 ± 4 677 ± 4
134,816 ± 34,077 62 ± 3
139,043 ± 42,932 64 ± 4
1,000,000 338 ± 10
1,000,000 335 ± 5
1,000,000 333 ± 3
– –
– –
1,000,000 348 ± 9
1,000,000 349 ± 8
1,000,000 347 ± 11
32.44 12.60 19.83
36.48 14.12 22.36
170.01 85.21 84.80
168.44 84.42 84.01
170.01 85.21 84.80
testing machine (Instron 8872, Instron Corporation, Canton, MA, USA). A loading rate of 2.5 mm/min in displacement control mode was applied in the monotonic tests, and the loading continued until the displacement reached 6 mm to ensure plastic deformation of the pedicle screw. The load-displacement curves were recorded at a data acquisition rate of 100 Hz. The monotonic tests for each type of spinal pedicle screw were repeated six times, and the stiffness and the yielding load were obtained. One-way ANOVA with Bonferroni post-hoc tests was used. All P values less than 0.05 were considered to indicate statistically significant differences. A correlation study between the maximum displacement calculated by the finite element models and the yielding load obtained by the monotonic tests was conducted. 2.4. Cyclic cantilever bending tests of the spinal pedicle screws The testing specimens and the experimental jig used in the cyclic cantilever bending tests were the same as those used in the monotonic tests. A 10-Hz cyclic loading with sinusoidal waveform in load control mode was applied. Two loading conditions, 60 and 400 N, were tested with a stress ratio of 10%. The cyclic cantilever bending tests were terminated when the number of testing cycles was more than one million or when the displacement of the actuator was beyond 8 mm. All experimental data were continuously monitored at a data acquisition rate of 50 Hz. The cyclic cantilever bending tests for each type of spinal pedicle screw were also repeated six times, and the multi-cycle stiffness and the fatigue life were obtained. One-way ANOVA was also used. The correlation study between the maximum tensile stress calculated by the finite element models and the logarithm of the fatigue life obtained by the cyclic cantilever bending tests was conducted. 3. Results 3.1. Finite element analyses The total number of elements varied from 155,000 to 205,000, and the solution time varied from 5 to 8 h. In the convergent analysis, both the maximum displacement and the maximum tensile stress were found to converge as expected, and the variation was less than 10%. For the results of the maximum displacement, the conical design group (Co-SD, Co-CD, and Co-CDP) revealed the lower maximum
Fig. 4. The stress distribution of the spinal pedicle screws.
displacement compared to the cylindrical design group (Cy-SD, CyCD, and Co-CDP) (Table 1), with the average percentage decrease being approximately 75%. In addition, the Co-SD had the lowest maximum displacement of the conical design group. Similarly, the CySD had the lowest maximum displacement of the cylindrical design group. For the results of the maximum tensile stress, the maximum tensile stress was located in the proximal part of the spinal pedicle screw (Fig. 4). The maximum tensile stress of the conical design group was significantly lower than that of the cylindrical design group (Table 1), and the average percentage decrease was approximately 86%. In addition, the Co-CD and the Cy-CD had the highest maximum tensile stress of the spinal pedicle screws for the conical design group and the cylindrical design group, respectively. 3.2. Monotonic tests of the spinal pedicle screws All of the spinal pedicle screws were found to undergo plastic deformation at the proximal part (Fig. 3B). The load-displacement curve was linear initially and had an obvious yielding (Fig. 5A).
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE 4
ARTICLE IN PRESS
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
Fig. 5. (A) The load-displacement curve in the monotonic tests. (B) The displacement-cycle curve in the cyclic cantilever bending tests.
The stiffness of the conical design group was significantly higher than that of the cylindrical design group (p < 0.05). However, the solid design, the cannulated design, and the cannulated design with a pin did not exhibit a statistically significantly difference for both the conical design group and the cylindrical design group (Table 1). The yielding load was defined by the 0.2% offset method. In the present study, the yielding load of the conical design group was significantly higher than that of the cylindrical design group (p < 0.05). In addition, the yielding load of the Cy-SD was significantly higher than that of the Cy-CD and the Cy-CDP (p < 0.05). However, the yielding loads of the Co-SD, the Co-CD, and the Co-CDP were quite similar and had no statistically significantly difference (Table 1). In the correlation study, the maximum displacement was closely related to the yielding load with a correlation coefficient of –0.99. In addition, the area moment of inertia was also closely related to the yielding load with a correlation coefficient of 0.99. 3.3. Cyclic cantilever bending tests of the spinal pedicle screws Two loading conditions, 60 and 400 N, with a stress ratio of 10% were applied to test each spinal pedicle screw. The loading condition of 60 N was determined by 90% of the lowest yielding load of the spinal pedicle screws (the Cy-CD and the Cy-CDP). In addition, the loading condition of 400 N was also applied to obtain additional outcomes. For the results of the cyclic cantilever bending tests at 60 N, fatigue failure occurred in the designs of the Cy-CD and the Cy-CDP, and the crack initiated in the proximal part of the spinal pedicle screws (Fig. 3C) (Fig. 5B). Other pedicle screw designs remained intact when the number of cycles to failure reached one million. The multi-cycle stiffness of the conical design group was significantly higher than that of the cylindrical design group (p < 0.05) (Table 1). In the correlation study, the maximum tensile stress was related to the logarithm of the fatigue life with a correlation coefficient of –0.76. In addition, the area moment of inertia was also closely related to the logarithm of the fatigue life with a correlation coefficient of 0.71. For the results of the cyclic cantilever bending tests at 400 N, none of the conical design group failed when the number of cycles to failure reached one million. However, the fatigue life of the cylindrical design group was unable to be obtained because the fatigue loading of 400 N was significantly larger than their yielding load. The multi-cycle stiffness of
the Co-SD, the Co-CD, and the Co-CDP were quite similar and had no statistically significant difference (Table 1). 4. Discussion Spinal pedicle screws are usually the weakest part of the posterior pedicle screw and rod fixation systems in clinical applications [17,18]. Thus, improving the biomechanical performances of spinal pedicle screws would also improve the clinical performances of the pedicle screw and rod fixation systems [10]. In the present study, the results of the finite element models indicated that the maximum tensile stress occurred in the proximal part of the spinal pedicle screws. Compared to the results of the cyclic cantilever bending tests, the point with the maximum tensile stress corresponded to the failure site of the spinal pedicle screws. In addition, the correlation study revealed that the numerical results were closely related to the experimental outcomes. Although the correlation study showed the feasibility of the numerical models, the implants had a high stress level of more than 2,700 MPa. No material of the titanium-bone complex will ever withstand this high stress level. Thus, a nonlinear material condition might need to be considered. Bending performances in terms of the pedicle screw designs have been discussed in past studies. Cho et al. [19] mentioned that the outer diameter of the pedicle screw determines the pullout strength, while the inner diameter determines the fatigue strength. In addition, Chao et al. [4] compared different designs of pedicle screws and concluded that conical pedicle screws yield a significantly higher bending performance than cylindrical pedicle screws. The above studies implied that pedicle screws with larger inner diameter or of conical core design effectively improved their bending performance. In the present study, the spinal pedicle screws from the conical design group (Co-SD, Co-CD, and Co-CDP) had smaller maximum tensile stress, higher yielding load, and longer fatigue life compared to the cylindrical design group (Cy-SD, Cy-CD, and Cy-CDP). Thus, the spinal pedicle screws with conical core design significantly improved their bending performances. This finding was the same as the results of the past studies mentioned above. Brown et al. [20] concluded that simple-beam theory is capable of providing enough evidence to support the optimization of the design of cylindrical bone screws. In the present study, the area moment of inertia for each spinal pedicle
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE
ARTICLE IN PRESS
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
screw was calculated to validate this finding. According to the results of the correlation study, this theory was also capable to support the design of conical bone screws. Although the numerical outcomes of the current study were validated according to the past studies, the numerical models of the present study were unable to provide the absolute data. Additionally, maybe using the relationship between the maximum tensile stress and the fatigue life is not the best way, but this is one of possible ways to dig the biomechanical performances of cannulated pedicle screws. Cannulated or expandable spinal pedicle screws with bone cement augmentation had been used to increase their anchoring strength in osteoporotic spine [21–23]. However, spinal pedicle screws with cannulated design might dramatically reduce their bending or fatigue strength. Reese et al. [24] tested cannulated and noncannulated screws of different screw diameter using cyclic loading. They found that the largest screw possible should be used for surgical fixation of Jones fractures. Although Reese’s study investigated both cannulated and noncannulated screws for the treatment of Jones fractures, their conclusion provides useful evidence for the present study because the fatigue life of the Cy-SD (Solid design) was significantly increased compared with that of the Cy-CD (cannulated design). The fatigue life of the Cy-SD was almost eight times higher than that of the Cy-CD. However, it was interesting that there is no statistically significant difference between the cannulated screw (the Co-CD) and the solid screw (the Co-SD) for the conical design group. Thus, a spinal pedicle screw with a conical core design is recommended if it needs to have a cannulated design. Two loading conditions, 60 and 400 N, were tested with a stress ratio of 10%. The reason to apply two loading conditions was explained as follow: the value of 60 N was calculated according to the 90% of the yielding strength of the Cy-CD design (the screw with lowest yielding strength). However, four types of the spinal pedicle screws remain intact after one million cyclic loading in the fatigue tests with a loading condition of 60 N. Thus, a loading condition of 400 N was applied to get additional outcome of the spinal pedicle screws. A past study [24] and the present study demonstrated that cylindrical bone screws with cannulated design would lose their original bending performances. Thus, retrieving the loss of material in the cannulated screws might avoid the loss of the original bending performances. In the present study, the cannulated pedicle screws with an additional pin were investigated to retrieve their biomechanical performances. For the cylindrical design group, the numerical results indicated that both the maximum displacement and the maximum tensile stress of the Cy-CDP could be retrieved compared to that of the Cy-CD. Unfortunately, this finding was not proved in the monotonic and cyclic cantilever bending tests. There are two possible reasons that account for this inconsistency. First, the numerical models may have eliminated the gap between the pin and cannulated hole and make a perfect contact interface. However, the experimental specimens had a gap caused by a machining tolerance. This gap might increase the deformation and stress. Second, all of the experimental specimens would yield when the loading exceeded their yielding point. However, the ideal linear elastic material was used for the finite element models. Thus, the bending performances of the cannulated design were achieved by using the numerical approach, but the screw was found to fail using the experimental approach. For the conical design group, retrieving the loss of the original bending performances in the cannulated design was also found to fail using the experimental approach. In fact, the present study proved that there is no statistically significant difference between the cannulated screw (the Co-CD) and the solid screw (the Co-SD) for the conical design group. Thus, it was easy to understand that adding a pin to the cannulated screws did not improve the bending performances. There are limitations in the current numerical and experimental approaches that may affect the ability to completely characterize the biomechanical behaviors of the spinal pedicle screws. First, although
5
the real geometry of the spinal pedicle screws was constructed during the computer simulations, the geometry of spinal vertebrae was simplified as a cylinder. This simplification might affect the numerical models ability to predict the actual biomechanical performances. Second, a linear material was used in the present study. This assumption might also affect the numerical models ability to predict the actual biomechanical performances. Third, only the bending performance of the spinal pedicle screws with either solid design or cannulated design was evaluated. However, the bending performance was just one of the complications of the spinal pedicle screws. Other performances should also be considered together, such as the interfacial strength between the pedicle screw and the screw’s hosting material and/or the maximum stress of the vertebrae. Finally, the bending performances of different screw structures were evaluated in this study. However, other screw parameters may also play an important role, such as the inner diameter, the ratio of the outer over the inner diameter, the diameter of cannulated hole, and different insertion technique. In conclusion, cylindrical spinal pedicle screws with a cannulated design may significantly deteriorate the bending performances. In addition, conical spinal pedicle screws maintained the original bending performances, whether they were solid or of a cannulated design. Adding a pin to the cannulated spinal pedicle screws did not improve the bending performances. This study may provide useful recommendations to orthopedic surgeons before surgery, and it could also provide design rationales to biomechanical engineers during the development of spinal pedicle screws. Conflict of interest statement None of the authors have any conflicts of interest to declare. Ethical approval Not required. Acknowledgments This work was sponsored by the Shin Kong Wu Ho-Su Memorial Hospital Research Program under the Grant No. SKH-8302-102DR-14. References [1] Silbermann J, Riese F, Allam Y, Reichert T, Koeppert H, Gutberlet M. Computer tomography assessment of pedicle screw placement in lumbar and sacral spine: comparison between free-hand and O-arm based navigation techniques. Eur Spine J. 2011;20:875–81. [2] Verlaan J-J, Dhert WJA, Oner FC. Intervertebral disc viability after burst fractures of the thoracic and lumbar spine treated with pedicle screw fixation and direct end-plate restoration. The Spine J. 2013;13(3):217–21. [3] Pihlajamaki H, Myllynen P, Bostman O. Complications of transpedicular lumbosacral fixation for non-traumatic disorders. J Bone Joint Surg. 1997;79B(2):183–9. [4] Chao CK, Hsu CC, Wang JL, Lin J. Increasing bending strength and pullout strength in conical pedicle screws: biomechanical tests and finite element analyses. J Spinal Disord Tech. 2008;21(2):130–8. [5] Hicks J, Singla A, Shen FH, Arlet V. Complications of pedicle screw fixation in scoliosis surgery: a systematic review. Spine. 2010;35(11):E465–70. [6] Okuyama K, Abe E, Suzuki T, Tamura Y, Chiba M, Sato K. Influence of bone mineral density on pedicle screw fixation: a study of pedicle screw fixation augmenting posterior lumbar interbody fusion in elderly patients. The Spine J. 2001;1(6):402–7. [7] Chatzistergos PE, Magnissalis EA, Kourkoulis SK. A parametric study of cylindrical pedicle screw design implications on the pullout performance using an experimentally validated finite-element model. Med Eng Phy. 2010;32(2):145–54. [8] Vanichkachorn JS, Vaccaro AR, Cohen MJ, Cotler J. Potential large vessel Injury during thoracolumbar pedicle screw removal: a case report. Spine. 1997;22(1):110–13. [9] Paré PE, Chappuis JL, Rampersaud R, Agarwala AO, Perra JH, Erkan S, et al. Biomechanical evaluation of a novel fenestrated pedicle screw augmented with bone cement in osteoporotic spines. Spine. 2011;36(18):E1210–14.
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
JID: JJBE 6
ARTICLE IN PRESS
[m5G;July 18, 2015;10:12]
K.-S. Shih et al. / Medical Engineering and Physics 000 (2015) 1–6
[10] Chao CK, Lin J, Putra ST, Hsu CC. A neuro-genetic approach to a multiobjective design optimization of spinal pedicle screws. J Biomech Eng-T ASME. 2010;132(9):091006. [11] Stanford RE, Loefler AH, Stanford PM, Walsh WR. Multiaxial pedicle screw designs: static and dynamic mechanical testing. Spine. 2004;29:367–75. [12] Chen CS, Chen WJ, Cheng CK, Jao SHE, Chueh SC, Wang CC. Failure analysis of broken pedicle screws on spinal instrumentation. Med Eng Phy. 2005;27(6):487– 96. [13] Cook SD, Salkeld SL, Stanley T, Faciane A, Miller SD. Biomechanical study of pedicle screw fixation in severely osteoporotic bone. The Spine J. 2004;4(4):402–8. [14] Chen LH, Tai CL, Lai PL, Lee DM, Tsai TT, Fu TS, et al. Pullout strength for cannulated pedicle screws with bone cement augmentation in severely osteoporotic bone: influences of radial hole and pilot hole tapping. Clin Biomech. 2009;24(8):613– 18. [15] Frankel BM, D’Agostino S, Wang C. A biomechanical cadaveric analysis of polymethylmethacrylate-augmented pedicle screw fixation. J Neurosurg: Spine. 2007;7(1):47–53. [16] Chao KH, Lai YS, Chen WC, Chang CM, McClean CJ, Fan CY, et al. Biomechanical analysis of different types of pedicle screw augmentation: a cadaveric and synthetic bone sample study of instrumented vertebral specimens. Med Eng Phy. 2013;35(10):1506–12.
[17] Jutte P, Castelein R. Complications of pedicle screws in lumbar and lumbosacral fusions in 105 consecutive primary operations. Eur Spine J. 2002;11(6):594–8. [18] Stoffel M, Behr M, Reinke A, Stüer C, Ringel F, Meyer B. Pedicle screw-based dynamic stabilization of the thoracolumbar spine with the Cosmic®-system: a prospective observation. Acta Neurochir. 2010;152(5):835–43. [19] Cho W, Cho SK, Wu C. The biomechanics of pedicle screw-based instrumentation. J Bone Joint Surg. 2010;98B(8):1061–5. [20] Brown GA, McCarthy T, Bourgeault CA, Callahan DJ. Mechanical performance of standard and cannulated 4.0-mm cancellous bone screws. J Orthop Res. 2000;111:307–12. [21] McKoy BE, An YH. An injectable cementing screw for fixation in osteoporotic bone. J Biomed Mater Res. 2000;53(3):216–20. [22] Fransen P. Increasing pedicle screw anchoring in the osteoporotic spine by cement injection through the implant: technical note and report of three cases. J Neurosurg: Spine. 2007;7(3):366–9. [23] Bullmann V, Schmoelz W, Richter M, Grathwohl C, Schulte TL. Revision of cannulated and perforated cement-augmented pedicle screws: a biomechanical study in human cadavers. Spine. 2010;35(19):E932–9. [24] Reese K, Litsky A, Kaeding C, Pedroza A, Shah N. Cannulated screw fixation of jones fractures: a clinical and biomechanical study. Am J Sport Med. 2004;32(7):1736–42.
Please cite this article as: K.-S. Shih et al., Comparison of the bending performance of solid and cannulated spinal pedicle screws using finite element analyses and biomechanical tests, Medical Engineering and Physics (2015), http://dx.doi.org/10.1016/j.medengphy.2015.06.008
