SPINE Volume 26, Number 12, pp E253–E260 ©2001, Lippincott Williams & Wilkins, Inc. Load Sharing Within a Human Lumbar Vertebral Body Using the Finite Element Method Khoa D. Cao, MS, Michele J. Grimm, PhD, and King-Hay Yang, PhD Study Design. A finite element parametric study was performed to investigate the structural roles of the verte- bral cortical shell and the trabecular centrum. Objectives. To address the debated issue of the rela- tive load-carrying role of the vertebral cortical shell. Summary of Background Data. Several experimental and computational studies have been aimed at quantify- ing the load-carrying roles of the human vertebral cortical shell and trabecular centrum. These studies, however, have supported no consensus. Methods. A finite element model of three lumbar ver- tebral bodies was developed to predict the fraction of the total compressive load acting on the lumbar vertebral body, under two different loading conditions, that was supported by the cortical shell. Parametric variations in vertebral material and geometric properties were exam- ined to determine how this fraction was influenced by such changes. Results. The fraction of the compressive load sup- ported by the cortical shell was found to be strongly dependent on the distance from the endplate, increasing from about 34% at either endplate to approximately 63% at the midtransverse plane. This fraction was indepen- dent of the loading characteristics, proportional to the properties of the cortex, and inversely proportional to the modulus of the centrum. Additionally, the cortical shell force fraction was affected significantly by changes in the overall vertebral geometry. Conclusions. Our findings indicate that the structural dominance of the cortical shell and centrum alternate depending on the location within the vertebral body. However, as age-related bone loss progresses, the load- carrying role of the cortical shell could increase significantly. [Key words: biomechanics, vertebrae, lum- bar spine, cortical shell, trabecular centrum, cortical cur- vature, endplate curvature] Spine 2001;26:E253–E260 Increases in fracture incidence are generally accepted as the consequence of the decrease in overall vertebral strength that occurs with aging. 14 This decrease in the overall vertebral strength is commonly attributed to the reduction in trabecular bone mass, or osteopenia; hence, most efforts to diagnose spinal osteoporosis have focused on monitoring the bone mineral density of the centrum. 9 However, the loss of bone strength has been observed to occur with age much faster than the loss of bone mass, indicating that other factors are also important to frac- ture risk assessments. 15 It has been demonstrated that individuals (both men and women) with a vertebral os- teoporotic fracture have significantly thinner cortices than those of the same age who have not sustained one. 31 Furthermore, age-related changes in both material and geometric properties of the cortical shell and endplates, such as thinning of the cortical shell and endplates, dia- metric expansion of the vertebral body (in men), and increases in endplate curvature, have been observed by many researchers. 2,14,15,28,37 Fracture occurs when the stresses generated by ap- plied loads exceed the vertebral ultimate strength. The loads experienced by the lumbar vertebrae have been shown to be more than two times body weight during routine functional activities. 17 These loads are trans- ferred from one vertebra to the inferiorly adjacent one via two paths: the major portion through the interverte- bral disc that connects the two vertebral bodies and the remainder by the facet joints at the articular process- es. 1,19,29,36 Within the vertebral body the vertebral loads are mainly distributed between the cortical shell and tra- becular centrum. In other words, the overall strength of the vertebral body depends on the structural contribu- tion of both components. Hence, understanding their relative roles and relevance to age-related changes is helpful in assessing the effect of any medical intervention on fracture risk. Despite its importance, results from sev- eral studies, both by experimental and analytic means, have supported no consensus. The cortical shell force fraction has been suggested to be as low as 10% and as high as 75% for healthy individuals and generally higher for osteoporotic persons. 7,12,13,27,32,37,38 These conflict- ing results can be partially explained by the difference in experimental protocols and variations in biologic speci- mens or analytic assumptions and approaches. Some of the results by previous studies are summarized as follows. Rockoff et al 21 reported that the cortical shell contrib- uted approximately 45–75% of the vertebral strength and that more force was transmitted via the cortical shell in old individuals than in younger ones. This was done by repeated testing of the specimens before and after the cortical shell or centrum had been removed at the midtransverse plane. Although the load used in the ex- periments was nondestructive, it was not certain if resid- ual damages had any effect on the findings. Moreover, the bone had been removed by grinding; therefore, it was difficult to determine if unintended bone removal had occurred. Using a different experimental procedure, McBroom et al 12 reported that the removal of the corti- cal shell reduced the failure load by approximately 10%, From the Bioengineering Center, Wayne State University, Detroit, Michigan. Acknowledgment date: February 21, 2000. First revision date: June 6, 2000. Acceptance date: February 13, 2001. Device status category: 1. Conflict of interest category: 12. E253