From a mechanical standpoint, an osteoporotic vertebral fracture represents a structural failure of bone. Vertebral bone in the aging spine fails because it cannot support the internal stresses and strains that result from loads applied to the spine. The ability of any structure to carry load depends on several characteristics: the matter that makes up the structure and the corresponding mechanical properties of that substance, the design of the structure, and the loading conditions. In vertebrae (Figure 1), the material is predominately trabecular bone; the structural design is dictated by the shape, size, and organization of the vertebral body; and the loading conditions arise from activities of daily living and from more severe loading conditions (e.g., falls or motor vehicle trauma). With aging and osteoporosis, there are compromises in the strength of the trabecular bone and in the structural capacity of the vertebral body. In addition, loads applied to the spine may be influenced by aging of the musculo skeletal system. This article outlines the mechanical properties of vertebral bone, the structural load-carrying capacity of vertebral bodies, and the biomechanics of loads applied to the spine, each in relation to osteoporosis and risk of vertebral fracture.
Mechanical Properties of Vertebral Bone: Factors Contributing to the Strength of Trabecular Bone
Compressive forces in the spine are transferred from the intervertebral discs to the vertebral endplates and are then distributed between the trabecular centrum and the thin shell of condensed bone that make up the vertebral body. The majority of axial force is carried by the trabecular bone.25,44,45,51 Axial forces external to the vertebra, therefore, result in stresses and strains in the trabecular centrum. The ultimate strength of the trabecular bone is determined by the maximum stress that it can sustain within the overall structure of the vertebral body. The trabecular bone will fail if the strength is not greater than the working stresses in the vertebra under physiologic or traumatic conditions. Such local failures or cracks can lead to fracture of the whole bone. Strength, therefore, is an important mechanical property when considering risk of fracture.
Reductions in the apparent density of trabecular bone typify aging and osteoporosis.37 For example, direct measurements of trabecular bone density between ages 20 and 80 have shown reductions in apparent density of approximately 50%.32 In addition, results of cross-sectional studies of bone mineral density (BMD), assessed by the noninvasive technique of dual energy x-ray absorptiometry (DXA) show losses of approximately 1% per year in the spine of women more than 65 years old14 (Figure 2). It should be noted that BMD of the spine reflects apparent density of the trabecular bone and of the condensed shell and is adjusted for area rather than for volume. However, such changes in adult lumbar BMD detected by DXA reflect, in part, decreases in apparent volumetric density of the trabecular centrum.
The compressive mechanical strength of a specimen of trabecular bone is related to the apparent density squared.6,7,13,20,39 Consequently, a decrease in apparent density of the bone results in a disproportionate reduction in strength. Conversely, an increase in apparent density, as is possible with treatments for osteoporosis, could result in a beneficial effect on trabecular bone strength.
The strength of trabecular bone is also a function of the trabecular architecture, which is described by orientation, connectivity, thickness, number, and spacing of trabeculae.20 The primary architecture in the lumbar spine is characterized by horizontal and vertical trabeculae (Figure 1A). There are more vertical than horizontal trabeculae at any given density.46 Changes in the architecture of trabecular bone have been noted with the decrease in density that accompanies aging and osteoporosis. The number and thickness of the trabeculae decrease as bone density decreases in women.29,31,38,46 If the vertical trabeculae of the vertebral body represent columns supporting compressive loads and the horizontal trabeculae act as cross-struts, then thinning and loss of trabeculae would decrease the stability of the vertical columns and result in trabecular buckling.46 These alterations in trabecular architecture and reductions in density with aging and osteoporosis weaken vertebral trabecular bone substantially.
Structural Design of Vertebral Bodies: Failure Load of Human Vertebrae
Apart from the density and morphology of the trabecular bone inside a vertebra, the overall geometry of the vertebral body has an influence on the load that can be carried by the bone (Figure 1B). For instance, under compressive force, the area perpendicular to the compression axis is an important factor in the mechanical response of the structure.
In many experimental studies, the maximum force during compression of vertebral specimens from human cadavers has been assessed and the failure force related to apparent density and geometric properties of the vertebrae.5,8,25,28,30,33,47 From ages 25 to 75, the average compressive failure force changes from approximately 8000 to 2000 N in vertebrae from the thoracolumbar spine in men and women.4 The compressive failure force measured has been as low as approximately 500 N in elderly cadaveric thoracic vertebrae when the vertebra is tested with intact intervertebral discs to transfer the load.28 When measures of density and geometry are compared with failure force, strong correlation coefficients are found for human vertebrae. Bone density measured by quantitative computed tomography (QCT) times endplate area was found by Brinckmann et al5 to associate significantly with compressive failure force of thoracolumbar specimens (r = 0.8). Cody et al8 found QCT regional density values multiplied by the minimum cross-sectional area of the vertebral body to correlate strongly with lumbar failure load (multiple R2 = 0.9). More recently, DXA assessments have been used to predict vertebral failure force.28,33,47 Bone mineral density measured by DXA is the bone mineral content of a region of interest (a whole vertebral body or a subregion) divided by the projected area in the view of the DXA scan. Therefore, apparent density and geometric bone properties influence the magnitude of standard DXA measurements. Correlation coefficients for the association between lumbar BMD and failure force have ranged in results of various studies from 0.7 to 0.9, indicating that approximately 50% to 80% of the variance in failure force is explained by BMD. In results from the study by Moro et al,28 BMD of the lumbar region correlated significantly with failure force of the same lumbar vertebra tested in axial compression (r = 0.89) as well as in remote thoracic sites (r = 0.94; Figure 3).
According to these results, BMD is a convenient and relatively specific method for predicting vertebral failure force under compressive loads. Furthermore, the correlation between failure force and BMD appears to be a continuum, so that there is no single "fracture threshold" value of BMD that can be used in general to identify people at risk. These studies of specimens from cadavers did not examine the influence of race on the correlation between BMD and failure force, and future work is needed to confirm correlations with race.
Under repetitive loading, a vertebra fails at lower load levels than those required to cause failure during a single application of force.4,17 This phenomenon is know as fatigue. The mechanisms for damage caused by fatigue in cortical and trabecular bone are thought to be crack initiation, crack growth, and final failure.20 Such microcracks have been observed in human vertebral bone and may contribute to decreased resistance of vertebrae to fracture.48
The strength and structural properties of the bone are only part of the biomechanics of osteoporotic fracture. Knowledge of the loading state of the bone is necessary to complete the picture. Fractures occur only when the loading on the bone results in internal stresses that exceed the strength of the bone. Therefore, it is necessary not only to know the strength of the bone and geometry of the vertebra but also to know the loads applied to the vertebral bodies.
Every activity, from sitting to walking to lifting a heavy object, creates loads on the spine. It is important to understand the activities during which vertebral fractures commonly occur to understand the loading environment. Unlike hip fractures, for which a fall has been identified as the associated event in approximately 90% of cases10,11,16,27 and which are readily identified as discrete fractures, the osteoporotic spine fracture is much more difficult to characterize. Difficulties arise when trying to assess the activities that characterize osteoporotic spine fractures, because vertebral fractures are classified by subtle vertebral deformities and reductions in height, making it difficult to recognize fractures consistently.3,12,26 There have been only a few observational studies that have assessed the events at the time of vertebral fracture in the elderly. Cooper et al9 used data extracted from a review of medical records from a 5-year study period to determine activities in a population-based sample of 341 patients with vertebral fracture. Sixteen percent of fractures were diagnosed incidentally during examination of the radiographs for other problems. Among the reviewed records, 113 of 341 (33%) indicated a fall at the time of fracture and another 29 (9%) indicated lifting. In a small hospital-based sample that used interviews with patients after diagnosis of fracture, preliminary results showed that almost 50% of acute, symptomatic vertebral fractures in people 60 years old and older occurred with a fall and approximately 20% occurred with controlled activities, including reaching, bending over, and lifting.35 Most of the remaining patients with fracture could not identify an activity or event at the time of the fracture. Thus, forces from falls and controlled activities should be analyzed to understand the biomechanics of a large portion of vertebral fractures.
Forces applied to the spine that result in osteoporotic fracture during controlled activities have only recently been investigated. Helpful tools for evaluating the magnitude of compressive loads on the spine are the lumbar models of Schultz, Andersson, and others, which were designed to evaluate development of low back pain in working adults engaged in bending and lifting actions.1,42,43 These models use optimization techniques to estimate trunk muscle forces and the compressive force on the spine. Wilson and Myers49,50 have adapted such lumbar models to the thoracic region and incorporated the geometry of the thorax of older people to estimate the forces on the vertebra during controlled activities. The forces applied to the spine at T8, T11, and L2 during various activities were calculated for a woman of 65 kg weight and 1.59 m height (values were mean values for a cohort of 120 women more than 65 years old).14,34 These activities were modeled as static events and are illustrated schematically in Figure 4. The estimated forces ranged from approximately 400 to 2100 N for the body habitus typical of an older women.
The loading of the spine during falls is under investigation. There has been significant work on the relation of falls to age-related hip fractures.10,15,18,36 The impact force on the hip has been estimated at between 2900 and 4300 N for a sideways fall from standing height using rigid-body models and experimental fall data.21,22 These estimations depend on weight and height of the person falling and are derived from the impact velocity of the hip, the effective mass of that part of the body that is moving before impact, and the properties of the soft tissues overlying the hip. Using data from soft tissue property experiments40 and results of experimental studies of falling backward,23 the impact force on the pelvis can be estimated at between 2000 and 2500 N during a fall backward. The energy from such an impact would be expected to dissipate somewhat before it reached the thoracolumbar spine where fractures typically occur. Current research efforts are focused on improving impact models and examining the conduction of impact forces along the spine.
Another important loading mode on the spine, which may account for the large portion of vertebral fractures not associated with a specific single loading event, is repetitive loading. Based on the review of records by Cooper et al,9 approximately 50% of vertebral fractures are reported as spontaneous or are detected incidentally, and these may be related to such fatigue processes.
Factor of Risk
Low BMD is often used to indicate risk of fracture. However, BMD and the corresponding structural capacity of bone should always be considered in relation to the forces applied to the structure during physiologic conditions or during traumatic loading. Hayes et al19 suggested the use of a factor of risk to characterize the ratio of the force applied to a bone during operation divided by the force at which the bone fractures. A factor of risk higher than 1 indicates that the bone is overloaded and failure is likely. Conversely, a factor of risk much less than 1 indicates that fracture would not be likely to occur.
Factors of risk for vertebral fracture during controlled actions can exceed 1 for low levels of BMD in the spine (Figure 4). Factors of risk are the ratio of the estimated forces applied to the spine over the vertebral failure load on the basis of lumbar BMD. The denominator of the factor of risk was determined for a range of BMD values by using the linear correlations between BMD and failure load from results of tests in cadavers. Because the optimization of muscle forces and the static modeling of the activities give conservative estimates of the forces on the spine, risk factors near 1, as well as those higher than 1, can be considered dangerous. One interesting indication from the estimates of factor of risk in Figure 4 is that people with low BMD operate near a factor of risk of 1 during many activities of daily living. In women with below average BMD, factors of risk approach 1 for such activities as lifting a portable television or a toddler. For women in the very low range of BMD values, even such simple activities as tying shoes may place the spine at risk for fracture.
Biomechanical Indications for Prevention of Vertebral Fractures
Considering the correlation between bone strength and apparent density, therapies that maintain or increase apparent density but do not have an adverse effect on intrinsic tissue properties should have a positive effect on trabecular bone strength. In addition to enhancement of apparent density, treatments that affect the morphologic or geometric profile of bone should have a beneficial effect on vertebral structural capacity and consequently reduce the number of vertebral fractures.
An example of a therapy that affects strength is the bisphosphonate alendronate, which has been shown in nonhuman primates that have had ovariectomy to increase vertebral trabecular bone strength. With a regimen of 0.25 mg/kg given every 2 weeks for 2 years, the strength of lumbar vertebral cores increased compared with those in untreated controls and compared with those in placebo-treated animals with ovariectomy2 (Figure 5). The bone strength in the treated specimens depended on BMD in a manner very similar to the correlation in untreated bone, indicating that the accretion of bone in the vertebra produced by alendronate treatment contributed to bone strength comparable to that in normal bone.2 Clinical trials of oral alendronate in post-menopausal women have shown reductions in vertebral fractures from 6.2% to 3.2%.24 The beneficial effect on bone strength observed in mechanical testing in an animal model, therefore, seems to translate into reduced vertebral fracture rates in humans.
Another approach to reducing the number of vertebral fractures comes from the observation that some fractures (approximately 10-20%) arise from activities is under the control of the person and occur because the person with low BMD operating near a factor of risk of 1 during certain actions. If preventable actions that load the vertebrae beyond safe levels are identified and those at risk are educated regarding avoidance of such actions, then such fractures may be prevented. As suggested by the outcome of the analysis of compressive loads on the spine and the resulting factors of risk (Figure 4), people with low bone density in the spine should be cautioned against lifting moderate to heavy objects or other actions - opening "stuck" windows, for example.
The biomechanics of vertebral fractures associated with falls are only now under investigation. Models of fall impact and loads to the spine must be developed to assess the loads on the spine during a fall. Future studies of falls and vertebral fracture may indicate future preventive measures to decrease the number of fractures associated with these accidental events.
Vertebral bone in the aging spine fails because it cannot support the internal stresses that result from loads applied to the spine. The strength of the trabecular bone in the vertebral centrum is determined by apparent density and by trabecular architecture. Reductions in density and thinning and loss of trabeculae result in corresponding reductions in strength.
The failure load of a whole vertebral body is determined by the strength of the trabecular bone in the centrum and by the geometric shape and organization of the vertebral body. Bone mineral density values, determined using such noninvasive techniques as DXA, have a direct, positive correlation with the compressive failure load of the vertebral body in white women and white men. Results of biomechanical studies show that assessment of BMD in vivo should indicate the peak compressive load that can be carried by the vertebral body. This correlation should be determined according to race.
The strength of bone and the load-bearing capacity should be considered in relation to the forces applied to the spine during normal activities or trauma. The ratio of loads applied to the spine during specified activities divided by the failure load of the vertebral body is the factor of risk. Factors of risk higher than 1 indicate that the load on the vertebra exceeds the structural capacity and suggest that fracture is likely to occur. To create a model of the loading conditions and determine the loads on the spine, it is necessary to understand the events that are associated with vertebral fractures. Vertebral fractures occur in a heterogeneous set of circumstances. Activities associated with some fractures include falling and controlled activities (e.g., lifting a moderate weight), but approximately 50% of vertebral fractures are not attributed to a known loading activity. For fractures resulting from controlled actions, it has been shown that everyday activities can result in factors of risk higher than 1 in people with very low vertebral BMD.
The authors thank Jeanine Dulong for help in preparing the manuscript and Ralph Müller for providing the image of trabecular bone in Figure 1A.
1. Andersson G, Ortengren R, Schultz A. Analysis and measurement of the loads on the lumbar spine during work at a table. J Biomech 1980;13:513-20.
2. Balena R, Toolan BC, Shea M, et al. The effects of 2-year treatment with the aminobisphosphonate alendronate on bone metabolism, bone histomorphometry, and bone strength
in ovariectomized nonhuman primates. J Clin Invest 1993;92:2577-86.
3. Banks LM, van Kuijk C, Genant HK. Radiographic technique for assessing osteoporotic vertebral deformity. In: Genant HK, Jergas M, van Kuijk C, eds. Vertebral Fracture
. San Francisco: Osteoporosis
Research Group, University of California, 1995:131-47.
4. Biggemann M, Brinckmann P. Biomechanics
of osteoporotic vertebral fractures. In: Genant HK, Jergas M, van Kuijk C, eds. Vertebral Fracture
. San Francisco: Osteoporosis
Research Group, University of California, 1995:21-34.
5. Brinckmann P, Biggemann M, Hilweg D. Prediction of the compressive strength of human lumbar vertebrae. Spine 1989;14:606-10.
6. Carter DR, Hayes WC. Bone compressive strength: The influence of density and strain rate. Science 1976;194:1174-6.
7. Carter DR, Hayes WC. The compressive behavior of bone as a two-phase porous structure. J Bone Joint Surg [Am] 1977;59:954-62.
8. Cody DD, Goldstein SA, Flynn MJ, Brown EB. Correlations between vertebral regional bone mineral density
(rBMD) and whole bone fracture load. Spine 1991;16:146-54.
9. Cooper C, Atkinson EJ, O'Fallon WM, Melton LJ. Incidence of clinically diagnosed vertebral fractures: A population based study in Rochester, Minnesota, 1985-1989. J Bone Miner Res 1992;7:221-77.
10. Cumming R, Klineberg R. Fall frequency and characteristics and the risk of hip fractures. J Am Geriatr Soc 1994;42:774-8.
11. Cummings SR, Black DM, Nevitt MC, et al. Appendicular bone density and age predict hip fracture in women. JAMA 1990;263:665-8.
12. Cummings SR, Melton LJ III, Felsenberg D, et al. Report: Assessing vertebral fractures. J Bone Miner Res 1995;10:518-23.
13. Goldstein SA. The mechanical properties of trabecular bone: Dependence on anatomic location and function. J Biomech 1987;20:1055-61.
14. Greenspan SL, Maitland-Ramsey L, Myers E. Classification of osteoporosis
in the elderly is dependent on site-specific analysis. Calcif Tissue Int 1996;58:409-14.
15. Greenspan S, Myers E, Maitland L, Resnick N, Hayes W. Fall severity and bone mineral density
as risk factors for hip fracture in ambulatory elderly. JAMA 1994;271:128-33.
16. Grisso JA, Kelsey JL, Strom BL, et al. Risk factors for falls as a cause of hip fracture in women. N Engl J Med 1991;324:1326-31.
17. Hansson TH, Keller TS, Spengler DM. Mechanical behavior of the human lumbar spine. II. Fatigue strength during dynamic compressive loading. J Orthop Res 1987;5:479-87.
18. Hayes WC, Myers ER, Morris JN, Gerhart TN, Yett HS, Lipsitz LA. Impact near the hip dominates fracture risk in elderly nursing home residents who fall. Calcif Tissue Int 1993;52:192-8.
19. Hayes W, Piazza S, Zysset P. Biomechanics
of fracture risk prediction of the hip and spine by quantitative computed tomography. In: Rosenthal D, ed. Radiol Clin North Am. Philadelphia: WB Saunders Company, 1991:1-18.
20. Keaveny TM, Hayes WC. Mechanical properties of cortical and trabecular bone. In: Hall BK, ed. Bone. Vol. 7. Bone Growth-B. Boca Raton: CRC Press, 1992:285-344.
21. Kroonenberg A, Hayes W, McMahon T. Dynamic models for sideways falls from standing height. J Biomech Eng 1995;117:309-18.
22. Kroonenberg A, Hayes WC, McMahon TA. Hip impact velocities and body configurations for voluntary falls from standing height. J Biomech 1996;29:807-11.
23. Kroonenberg, A van den, Wilson SE, Myers ER, Hayes WC, McMahon TA. Impact velocities and body configurations for backward falls from standing height. J Biomech (submitted)
24. Liberman UA, Weiss SR, Broll J, et al. Effect of oral alendronate on bone mineral density
and the incidence of fractures in postmenopausal osteoporosis
. N Engl J Med 1995;333:1437-43.
25. McBroom RJ, Hayes WC, Edwards WT, Goldberg RP, White AA III. Prediction of vertebral body compressive fracture using quantitative computed tomography. J Bone Joint Surg [Am] 1985;67:1206-14.
26. Melton LJ, Kan SH, Frye MA, Wahner HW, O'Fallon WM, Riggs BL. Epidemiology of vertebral fractures in women. Am J Epidemiol 1989;129:1000-11.
27. Michelson J, Myers A, Jinnah R, Cox Q, Van Natta M. Epidemiology of hip fractures among the elderly. Clin Orthop 1995;311:129-35.
28. Moro M, Hecker A, Bouxsein M, Myers E. Failure load of thoracic vertebrae correlates with lumbar bone mineral density
measured by DXA. Calcif Tissue Int 1995;56:206-9.
29. Mosekilde L. Age-related changes in vertebral trabecular bone architecture-assessed by a new method. Bone 1988;9:247-50.
30. Mosekilde L, Bentzen SM, Ortoft G, Jorgensen J. The predictive value of quantitative computed tomography for vertebral body compressive strength and ash density. Bone 1989;10:465-70.
31. Mosekilde L, Mosekilde L. Sex differences in age-related changes in vertebral body size, density and biomechanical competence in normal individuals. Bone 1990;11:67-73.
32. Mosekilde L, Mosekilde L, Danielsen CC. Biomechanical competence of vertebral trabecular bone in relation to ash density and age in normal individuals. Bone 1987;8:79-85.
33. Myers B, Arbogast K, Lobaugh B, Harper K, Richardson W, Drezner M. Improved assessment of lumbar vertebral body strength using supine lateral dual-energy x-ray absorptiometry. J Bone Miner Res 1994;9:687-93.
34. Myers ER, Wilson SE. Beyond bone density: Risk assessment and design of programs to prevent osteoporotic fracture. Proceedings of the Pennington Nutrition Series Symposium on Women's Health: Prevention is the Best Medicine. Vol 7. Baton Rouge: Louisiana State University Press 1997; in press.
35. Myers ER, Wilson SE, Greenspan SL. Vertebral fractures in the elderly occur with falling and bending. J Bone Miner Res 1996;11(Suppl):S355.
36. Nevitt M, Cummings S. Type of fall and risk of hip and wrist fractures: The study of osteoporotic fractures. J Am Geriatr Soc 1993;41:1226-34.
37. Peck W. Consensus development conference: Diagnosis, prophylaxis, and treatment. Am J Med 1993;94:646-50.
38. Preteux F, Bergot C, Laval-Jeantet AM. Automatic quantification of vertebral cancellous bone remodeling during aging. Anat Clin 1985;7:203-8.
39. Rice JC, Cowin SC, Bowman JA. On the dependence of the elasticity and strength of cancellous bone on apparent density. J Biomech 1988;21:155-68.
40. Robinovitch SN, Hayes WC, McMahon TA. Prediction of femoral impact forces in falls on the hip. J Biomech Eng 1991;113:366-74.
41. Ruegsegger P, Koller B, Muller R. A microtomographic system for the nondestructive evaluation of bone architecture. Calcif Tissue Int 1996;58:24-9.
42. Schultz A, Andersson G. Analysis of loads on the lumbar spine. Spine 1981;6:76-82.
43. Schultz A, Andersson G, Ortengren R, Haderspeck K, Nachemson A. Loads on the lumbar spine. Validation of a biomechanical analysis by measurements of intradiscal pressures and myoelectric signals. J Bone Joint Surg [Am] 1982;64:713-20.
44. Silva MJ, Keaveny TM, Hayes WC. Load sharing between the shell and centrum in the lumbar vertebral body. Spine 1997;22:140-50.
45. Silva MJ, Wang C, Keaveny TM, Hayes WC. Direct and computed tomography thickness measurements of the human, lumbar vertebral shell and endplate. Bone 1994;15:409-14.
46. Snyder BD, Piazza S, Edwards WT, Hayes WC. Role of trabecular morphology in the etiology of age-related vertebral fractures. Calcif Tissue Int 1993;53:S14-S22.
47. Tabensky AD, Williams J, DeLuca V, Briganti E, Seeman E. Bone mass, areal, and volumetric bone density are equally accurate, sensitive, and specific surrogates of the breaking strength of the vertebral body: An in vitro study. J Bone Miner Res 1996;11:1981-8.
48. Wenzel TE, Schaffler MB, Fyhrie DP. In vivo trabecular microcracks in human vertebral bone. Bone 1996;19:89-95.
49. Wilson SE. Development of a model to predict the compressive forces on the spine associated with age-related vertebral fractures. Cambridge: Massachusetts Institute of Technology, 1994. Thesis.
50. Wilson S, Myers E. A model to predict the compressive forces associated with age-related vertebral fractures of thoracolumbar vertebrae. Trans Orthop Res Soc 1996;21:252.
51. Yoganandan N, Myklebust JB, Cusick JF, Wilson CR, Sances A. Functional biomechanics
of the thoracolumbar vertebral cortex. Clin Biomech 1988;3:11-8.