Prediction of cement volume for vertebroplasty based on imaging and biomechanical results
Control of bone cement volume (PMMA) may be critical for preventing complications in vertebroplasty, the percutaneous injection of PMMA into vertebra. The purpose of this study was to predict the optimal volume of PMMA injection based on CT images. For this, correlation between PMMA volume and textural features of CT images was examined before and after surgery to evaluate the appropriate PMMA amount. The gray level run length analysis was used to determine the textural features of the trabecular bone. Estimation of PMMA volume was done using 3D visualization with semi-automatic segmentation on postoperative CT images. Then, finite element (FE) models were constructed based on the CT image data of patients and PMMA volume. Appropriate material properties for the trabecular bone were assigned by converting BMD to elastic modulus. Structural reinforcement due to the changes in PMMA volume and BMD was assessed in terms of axial displacement of the superior endplate. A strong correlation was found between the injected PMMA volume and the area of the intertrabecular space and that of trabecular bone calculated from the CT images (r-0.90 and −0.90, respectively). FE results suggested that vertebroplasty could effectively reinforce the osteoporotic vertebra regardless of BMD or PMMA volume. Effectiveness of additional PMMA injection tended to decrease. For patients with BMD well lower than 50mg/ml, injection of up to 30% volume of the vertebral body is recommended. However, less than 30% is recommended otherwise to avoid any complications from excessive PMMA because the strength has already reached the normal level.
Key WordsVertebroplasty Bone Cement Image Analysis Biomechanics FEM
Coefficient of determination
The level of BMD (Bone Mineral Density) by quantitative computed tomography measurement
Unable to display preview. Download preview PDF.
- Grosland N, 1998, “Spinal Adaptations in Response to Interbody Fusion Systems: A Theoretical Investigation. Ph. D. Thesis,” Graduate College The University of Iowa, Iowa City Iowa, pp. 33–41.Google Scholar
- Ito M., Ohki M., Hayashi K., Yamada M., Uetani M., and Nakamura T., 1995, “Trabecular Texture Analysis of CT Images in the Relationship with Spinal Fracture,”Radiology, 194, pp. 55–59.Google Scholar
- Mir A. H., Hanmandlu M., and Tandon S. N., 1995, “Texture Analysis of CT Images,”IEEE Eng. in Med. & Biol., pp. 781–786.Google Scholar
- Nachemson A., 1966, “The Load on Lumbar Discs in Different Positions of the Body,”Clin. Orthop., 45: 107.Google Scholar
- Sternberg S. R., 1983, “Biomedical Image Processing,”IEEE Computer, pp. 22–34.Google Scholar
- Suk S. I., 1997,Spinal Surgery, Choishineuihaksa, Seoul, Korea.Google Scholar