AGE EFFECTS ON THE FATIGUE BEHAVIOR OF HUMAN VERTEBRAL CANCELLOUS TISSUE

Participants: T.E. Ciarelli, M.B. Schaffler, S.A. Goldstein

Keywords: cancellous bone, fatigue, aging

Introduction

Age-related bone loss in the spine is well documented, and is believed to be a significant factor in vertebral fracture [1]. However, a consistently demonstrated overlap in bone mass measurements between fracture and non-fracture groups has led to the conclusion that factors affecting bone tissue quality are likely important [2]. Measurements and estimates of cancellous hard tissue modulus have been made from a variety of mechanical testing regimes and analytical approaches [3]. However, few studies have characterized aging effects on cancellous bone tissue mechanical properties, and none have addressed age effects on fatigue behavior. The purpose of this study was to characterize the fatigue behavior of vertebral cancellous tissue and examine aging effects to explore whether or not this aspect of bone quality plays a role in vertebral collapse.

Materials and Methods

Specimen Preparation: Individual trabeculae were dissected from anterior superior regions of L1 vertebrae of seven male cadavers (ages 38, 41, 62, 64, 76, 76, and 85 yrs). A custom milling machine was used to machine the trabeculae into parallelepiped beams with base and height dimensions ranging from 70 to 160 µm and base-to-height ratios between 0.67 and 1.5.

Mechanical Testing: Specimens were cyclically tested at room temperature in four-point bending (smin-smax=10-80 MPa) under load control on a custom-designed microfatigue machine. Load was monitored with a 60 g load cell and displacement was measured using a proximity transducer. Data acquisition was performed using a microcomputer and LabView software. All specimens were loaded to failure. Between one and four microbeams from each individual were mechanically tested.

Data Analysis: For each specimen, data were analyzed to determine initial secant modulus (Esec, N/mm, defined as the slope of the load-displacement curve), initial bending modulus (Ebend, GPa, calculated from beam theory), number of cycles to failure (Nf), and modulus degradation curve (secant modulus vs. cycle number). The mean Ebend and Nf were determined for each individual, and regression analysis was used to examine linear and log10 relationships between each of these parameters and age. Modulus degradation curves were assessed qualitatively to determine if there were any age-related patterns.

Results

Secant modulus values either declined steadily with cyclic loading or remained relatively constant up until failure. These modulus degradation patterns are consistent with those described for both cortical and cancellous bone tissue as well as for other composite materials [4]. Age did not appear to affect the modulus degradation pattern. The regression between log10(fatigue life) and age was significant at the p<0.06 level (Figure 1). There was no significant relationship between Ebend and age.

Discussion

Bending modulus values for vertebral trabeculi (mean ± SD=5.36 ± 2.70 GPa) were consistent with those reported for the iliac crest [5] and proximal tibia [6]. However, vertebral bone tissue demonstrated inferior fatigue resistance compared to bone from the proximal tibia (Nf = ~104 vs. ~107 at smax =80 MPa) [7]. This suggests that bone from the vertebra may have fundamental differences in structure or composition which compromises its fatigue strength.

There was a strong trend toward increased fatigue life and age for vertebral bone tissue (Fig. 1). Changes in microstructure consequent to remodeling over time may explain this finding. Negative balances between resorption and formation would result in smaller trabecular packet size in older individuals, and consequently, increased cement line length per area of bone. Cement lines have been shown to increase fracture toughness of bone by serving as a location where energy associated with crack propagation is dissipated and cracks are thus arrested [8,9]. Therefore, specimens consisting of smaller packets would be able to accumulate greater levels of damage and the fatigue life would be expected to increase.

Another microstructural feature that could affect fatigue life is the presence of osteocyte lacunae and canaliculi, which represent voids within the matrix. It has been shown that both osteocyte density and size decrease with aging in the iliac crest [10]. Thus, if similar phenomena occurred in the vertebra, one would expect fatigue life to be increased due to the reduced presence of voids within the matrix. Further characterization of the microstructure of specimens used in this study is required for a more critical evaluation of these potential explanations.

Conclusions

Static bending modulus values from the vertebra were similar to those reported at other skeletal sites. However, fatigue life was reduced compared to the proximal tibia, suggesting that there may be anatomical variation in cancellous tissue properties and that the reduced fatigue life of the vertebra could be a factor in its higher fracture incidence.

The age-related increase in fatigue life may reflect a qualitative improvement in tissue properties that prolongs structural integrity of the vertebral body despite bone loss in males. Fractures in very elderly males may occur because bone loss is so excessive that increased tissue properties are not adequate to maintain overall mechanical function. Further studies examining gender effects are warranted. If similar aging trends are not found in females, it may explain their higher incidence and earlier onset of vertebral collapse.

References: [1] Riggs NEngJMed 1986 [2] Heaney CTI 1993 [3] Rho JBiomech 1993 [4]Jepsen JBiomech 1997 [5] Kuhn JOR 1989 [6] Choi JBiomech 1990 [7] Choi JBiomech 1992 [8] Burr JBiomech 1988 [9] Schaffler Bone 1989 [10] Mullender Bone 1996