- Email: [email protected]
Effects of bone stiffness, cement volumes and vertebral body height loss on the load transfer change of adjacent vertebrae in percutaneous vertebroplasty
S584
Poster 105, Poster Session 1/Orthopaedics. 14:45-15:45, Room 103 & Alley Area
Effects of bone stiffness, cement volumes and vertebral body height loss on the load transfer change of adjacent vertebrae in percutaneous vertebroplasty C-Y Lin1, S-Y Chuang2, D-T Ju3 , W-P Chen1 1 Department of Biomedical Engineering, Chung Yuan Christian University 2 Department of Orthopaedics, Tri-Service General Hospital 3 Department of Neurosurgery, Tri-Service General Hospital processed on a local personal computer. All analysis results INTRODUCTION for the models with interbody spacers were compared with Osteoporotic vertebral compression fracture is one of the those of the corresponding intact ones. most common diseases that attacks the elder population especially in postmenopausal women [1,2]. Percutaneous vertebroplasty (PVP) has been reported to be an effective treatment for indicated compression fracture patients. However, new vertebral fractures in adjacent vertebrae after PVP were often reported in literature [3]. The detailed mechanism of the altered biomechanics in the treated vertebra and untreated adjacent vertebrae were still unclear. Therefore, the optimal cement volume in terms of the percent fill of the vertebral body volume is lacking and no Figure 1: Finite element models of the intact thoracolumbar literature discusses on the effects of bone stiffness and the spine vertebral body height loss. The objective of the current study was to use a computational finite element method to investigate the effects of bone quality, cement volume and body height loss on the overall stiffness of the treated vertebra and the load transfer changes of adjacent vertebrae following percutaneous vertebroplasty. METHODS A finite element model of the vertebral body was generated from the digitized CT scans using the image softwareAmira (Mercury Computer Systems, Inc., Massachusetts, U.S.A.). The surface models of the vertebral bodies and discs were transferred to a finite element pre-processing program – Mentat (MSC Software Corp., Los Angeles, U.S.A.) and the finite element mesh of the intact T10-L2 vertebrae was generated with 78,000 8-node hexahedral elements as shown in Fig. 1. To investigate the effects of different bone cement volumes injected during vertebroplasty, the T12 vertebral body was modified with four different cement volumes (2, 4, 6 and 8 mL) delievered into the medial and lateral portions of the cancellous bone regions. Fig. 2 shows the four different configurations of bone cement volumes for the T12 vertebral body in the transverse cross-sectional view. The elastic modulus of the cancellous bone region was varied with three different values (100, 500 and 1000 MPa) to simulate three different bone qualities. In order to simulate the collapsed vertebral boby with different levels of body height loss, the intact T10-L2 finite element model was modified with four different anterior body height loss (20%, 40%, 60% and 80%) as shown in Fig.3. Where the percentage of body height loss was defined as the percent reduction of anterior body height compared to the original intact body height. Four different volumes of bone cement were considered for each model with different body height loss. A 300 N axial, compressive load was applied evenly on the upper endplate nodes of T12 (or T10 for the five-segment model) while the bottom surface nodes of the T12 (or L2 for the five-segment model) were constrained. A commercially available FEA software – MARC (MSC Software) was used as a solver for these analyses. Analyses were performed using the computing facilities at the National Center for High-Performance Computing (NCHC, Taiwan) via internet connection. Analysis results were retrieved back and Journal of Biomechanics 40(S2)
Figure 2: Finite element model was simulated bipedicular vertebroplasty. Orthocomp cement injected into T12 vertebra with 2, 4, 6 and 8ml (left to right).
Figure 3: Finite element meshes of T12 vertebra. Compared to the intact vertebra, the anterior vertebral body height was reduced by 20%, 40%, 60% and 80% (left to right) in the vertebroplasty model. RESULTS Single level model According to our analysis, vertebral stiffness was strongly influenced by more cement volume and higher cancellous bone density. By increase cement volume with 2ml., 4ml, 6ml and 8ml, the percentage of incremental stiffness of poor bone density were 5.3%, 36.5%, 85.6% and 106.5% respectively. However, the percentage of increased stiffness of healthy bone density were 5%, 15.7%, 26.2% and 33.5% respectively. There is a noticeable change of vertebral stiffness by increasing cement volume in poor bone density. In poor bone density (100MPa), thirty percent fill (6ml) restored the vertebral stiffness to healthy one value (Fig.4). In largest body height loss, the vertebral stiffness decreases 64% with no cement injected. When the body height loss is 20%, the least amount of bone cement (10% fill) restored the vertebral stiffness to its initial value. Twenty percent fill (4c.c.) restored the vertebral stiffness to its initial value in the 40% and 60% body height loss cases. In the highest body height loss case, thirty percent fill (6c.c.) restored the vertebral stiffness to its initial value. (Fig.5) Five level model The load shifting to the adjacent vertebras had positive relationship to the collapse angle, the percentage changes of adjacent vertebras stress increased several times. The load XXI ISB Congress, Poster Sessions, Wednesday 4 July 2007
Poster Session 1/Orthopaedics. 14:45-15:45, Room 103 & Alley Area,
Poster 105 (cont’d)
S585
shifting was reduced by cement injection while the fractured vertebra has higher stress accumulation, the percentage changes of adjacent vertebras stress decreased only 2%-3%. On the severity of the collapse, bone stress was reduced 25% after inject cement. The cement amount did not showed significant effect on the alteration of load shifting. Furthermore inferior level (L1) was shared more stress than superior one (T10). (Fig.6)
Figure 6: (a) Maximum von Mises stresses in the cancellous bone of the T11 for different body height loss and different cement volume. (b) Maximum von Mises stresses in the cancellous bone of the L1 for different body height loss and different cement volume.
Figure 4: (a) Vertebral stiffness of the T12 for different cancellous bone Young’s modulus and different cement volume. Maximum von Mises stresses in the cortical bone (b) and cancellous bone (c) of the T12 for different cancellous bone Young’s modulus and different cement volume
DISCUSSION We found that only small change of vertebral stiffness by increasing cement volume in good bone density. Our results suggest that if the goal is to restore vertebral stiffness to its predamaged level in poor cancellous bone quality. It needs to inject cement volume of 6ml, the stiffness increase is 85.6%. Belkoff et al. [4] also found that only between 5 and 6 ml of bone cement in a bipedicular vertebroplasty approach is needed to restore stiffness and strength after a compression fracture in the thoracolumbar spine. A wedge-shaped fracture of a vertebral body decreases vertebral stiffness and increases maximum von Mises stress of cortical bone and cancellous bone. The vertebral body may be fractured due to hight stress concentration. However, to inject more cement volume will not restore the vertebral stiffness. We also found that inferior level (L1) was shared more stress than superior one (T10) due to the loading position was changed. CONCLUSIONS It is more sensitive for vertebral stiffness and bone stress by increasing cement volume in poor cancellous bone quality. For the wedge-shaped fracture case, not only bone cement injection but also reconstruction of the lordotic curvature of spine is necessary. REFERENCES 1. Lukert BP, et al. Geriatrics 49, 22, 1994. 2. Lyles KW, et al. Am J Med 94, 595, 1993. 3. Mehbod A, et al. Eur Spine J 12, S155-S162, 2003. 4. Belkoff S, et al. Trans Orthop Res Soc 46, 356, 2000.
Figure 5: (a) Vertebral stiffness of the T12 for different body height loss and different cement volume. Maximum von Mises stresses in the cortical bone (b) and cancellous bone (c) of the T12 for different body height loss and different cement volume. XXI ISB Congress, Poster Sessions, Wednesday 4 July 2007
ACKNOWLEDGEMENTS The computing facitities provided by the National Center for High – Performance Computing are also greatly appreciated.
Journal of Biomechanics 40(S2)
Poster 105, Poster Session 1/Orthopaedics. 14:45-15:45, Room 103 & Alley Area
Effects of bone stiffness, cement volumes and vertebral body height loss on the load transfer change of adjacent vertebrae in percutaneous vertebroplasty C-Y Lin1, S-Y Chuang2, D-T Ju3 , W-P Chen1 1 Department of Biomedical Engineering, Chung Yuan Christian University 2 Department of Orthopaedics, Tri-Service General Hospital 3 Department of Neurosurgery, Tri-Service General Hospital processed on a local personal computer. All analysis results INTRODUCTION for the models with interbody spacers were compared with Osteoporotic vertebral compression fracture is one of the those of the corresponding intact ones. most common diseases that attacks the elder population especially in postmenopausal women [1,2]. Percutaneous vertebroplasty (PVP) has been reported to be an effective treatment for indicated compression fracture patients. However, new vertebral fractures in adjacent vertebrae after PVP were often reported in literature [3]. The detailed mechanism of the altered biomechanics in the treated vertebra and untreated adjacent vertebrae were still unclear. Therefore, the optimal cement volume in terms of the percent fill of the vertebral body volume is lacking and no Figure 1: Finite element models of the intact thoracolumbar literature discusses on the effects of bone stiffness and the spine vertebral body height loss. The objective of the current study was to use a computational finite element method to investigate the effects of bone quality, cement volume and body height loss on the overall stiffness of the treated vertebra and the load transfer changes of adjacent vertebrae following percutaneous vertebroplasty. METHODS A finite element model of the vertebral body was generated from the digitized CT scans using the image softwareAmira (Mercury Computer Systems, Inc., Massachusetts, U.S.A.). The surface models of the vertebral bodies and discs were transferred to a finite element pre-processing program – Mentat (MSC Software Corp., Los Angeles, U.S.A.) and the finite element mesh of the intact T10-L2 vertebrae was generated with 78,000 8-node hexahedral elements as shown in Fig. 1. To investigate the effects of different bone cement volumes injected during vertebroplasty, the T12 vertebral body was modified with four different cement volumes (2, 4, 6 and 8 mL) delievered into the medial and lateral portions of the cancellous bone regions. Fig. 2 shows the four different configurations of bone cement volumes for the T12 vertebral body in the transverse cross-sectional view. The elastic modulus of the cancellous bone region was varied with three different values (100, 500 and 1000 MPa) to simulate three different bone qualities. In order to simulate the collapsed vertebral boby with different levels of body height loss, the intact T10-L2 finite element model was modified with four different anterior body height loss (20%, 40%, 60% and 80%) as shown in Fig.3. Where the percentage of body height loss was defined as the percent reduction of anterior body height compared to the original intact body height. Four different volumes of bone cement were considered for each model with different body height loss. A 300 N axial, compressive load was applied evenly on the upper endplate nodes of T12 (or T10 for the five-segment model) while the bottom surface nodes of the T12 (or L2 for the five-segment model) were constrained. A commercially available FEA software – MARC (MSC Software) was used as a solver for these analyses. Analyses were performed using the computing facilities at the National Center for High-Performance Computing (NCHC, Taiwan) via internet connection. Analysis results were retrieved back and Journal of Biomechanics 40(S2)
Figure 2: Finite element model was simulated bipedicular vertebroplasty. Orthocomp cement injected into T12 vertebra with 2, 4, 6 and 8ml (left to right).
Figure 3: Finite element meshes of T12 vertebra. Compared to the intact vertebra, the anterior vertebral body height was reduced by 20%, 40%, 60% and 80% (left to right) in the vertebroplasty model. RESULTS Single level model According to our analysis, vertebral stiffness was strongly influenced by more cement volume and higher cancellous bone density. By increase cement volume with 2ml., 4ml, 6ml and 8ml, the percentage of incremental stiffness of poor bone density were 5.3%, 36.5%, 85.6% and 106.5% respectively. However, the percentage of increased stiffness of healthy bone density were 5%, 15.7%, 26.2% and 33.5% respectively. There is a noticeable change of vertebral stiffness by increasing cement volume in poor bone density. In poor bone density (100MPa), thirty percent fill (6ml) restored the vertebral stiffness to healthy one value (Fig.4). In largest body height loss, the vertebral stiffness decreases 64% with no cement injected. When the body height loss is 20%, the least amount of bone cement (10% fill) restored the vertebral stiffness to its initial value. Twenty percent fill (4c.c.) restored the vertebral stiffness to its initial value in the 40% and 60% body height loss cases. In the highest body height loss case, thirty percent fill (6c.c.) restored the vertebral stiffness to its initial value. (Fig.5) Five level model The load shifting to the adjacent vertebras had positive relationship to the collapse angle, the percentage changes of adjacent vertebras stress increased several times. The load XXI ISB Congress, Poster Sessions, Wednesday 4 July 2007
Poster Session 1/Orthopaedics. 14:45-15:45, Room 103 & Alley Area,
Poster 105 (cont’d)
S585
shifting was reduced by cement injection while the fractured vertebra has higher stress accumulation, the percentage changes of adjacent vertebras stress decreased only 2%-3%. On the severity of the collapse, bone stress was reduced 25% after inject cement. The cement amount did not showed significant effect on the alteration of load shifting. Furthermore inferior level (L1) was shared more stress than superior one (T10). (Fig.6)
Figure 6: (a) Maximum von Mises stresses in the cancellous bone of the T11 for different body height loss and different cement volume. (b) Maximum von Mises stresses in the cancellous bone of the L1 for different body height loss and different cement volume.
Figure 4: (a) Vertebral stiffness of the T12 for different cancellous bone Young’s modulus and different cement volume. Maximum von Mises stresses in the cortical bone (b) and cancellous bone (c) of the T12 for different cancellous bone Young’s modulus and different cement volume
DISCUSSION We found that only small change of vertebral stiffness by increasing cement volume in good bone density. Our results suggest that if the goal is to restore vertebral stiffness to its predamaged level in poor cancellous bone quality. It needs to inject cement volume of 6ml, the stiffness increase is 85.6%. Belkoff et al. [4] also found that only between 5 and 6 ml of bone cement in a bipedicular vertebroplasty approach is needed to restore stiffness and strength after a compression fracture in the thoracolumbar spine. A wedge-shaped fracture of a vertebral body decreases vertebral stiffness and increases maximum von Mises stress of cortical bone and cancellous bone. The vertebral body may be fractured due to hight stress concentration. However, to inject more cement volume will not restore the vertebral stiffness. We also found that inferior level (L1) was shared more stress than superior one (T10) due to the loading position was changed. CONCLUSIONS It is more sensitive for vertebral stiffness and bone stress by increasing cement volume in poor cancellous bone quality. For the wedge-shaped fracture case, not only bone cement injection but also reconstruction of the lordotic curvature of spine is necessary. REFERENCES 1. Lukert BP, et al. Geriatrics 49, 22, 1994. 2. Lyles KW, et al. Am J Med 94, 595, 1993. 3. Mehbod A, et al. Eur Spine J 12, S155-S162, 2003. 4. Belkoff S, et al. Trans Orthop Res Soc 46, 356, 2000.
Figure 5: (a) Vertebral stiffness of the T12 for different body height loss and different cement volume. Maximum von Mises stresses in the cortical bone (b) and cancellous bone (c) of the T12 for different body height loss and different cement volume. XXI ISB Congress, Poster Sessions, Wednesday 4 July 2007
ACKNOWLEDGEMENTS The computing facitities provided by the National Center for High – Performance Computing are also greatly appreciated.
Journal of Biomechanics 40(S2)