COMPARISON OF SCANNING ACOUSTIC MICROSCOPY AND NANOINDENTATION MEASURES OF THE ELASTIC PROPERTIES

OF HUMAN BONE LAMELLAE

Participants: C. E. Hoffler, N. Zhang*, K. M. Kozloff, M. J. Grimm*, S. A. Goldstein

*Bioengineering Center, Wayne State University, Detroit, MI

Keywords: bone mechanics, scanning acoustic microscopy, nanoindentation

Introduction

In order to test hypotheses concerning the mechanisms of mechanotransduction in bone, it is necessary to fully characterize the properties of the bone extracellular matrix through which the mechanical signals are transmitted. Since the receipt and response to mechanical signals occurs at the level of individual cells, the mechanical behavior of the tissue must also be determined at that scale. The purpose of this study was to compare two methods of microscopic (lamellar) property measures in bone: scanning acoustic microscopy and nanoindentation.

Materials and Methods

A cortical bone specimen was obtained from the femoral diaphysis of a 41 year old male. A tissue sample (36 mm2 x 3 mm thick) was isolated from the anterior octant and transverse to the femoral long axis with a diamond wafering blade. The cranial surface of the specimen was polished with progressive grades of SiC paper, finished with a 0.25 mm diamond slurry and cleansed ultrasonically.

Acoustic measures of cortical bone elasticity were performed with a scanning acoustic microscope fitted with a 150 MHz spherically focused transducer. Using saline solution with 0.5 mg/ml gentamicin as a coupling fluid, acoustic reflections were collected and converted into a gray scale image (scaled 0 to 255) with 10 mm resolution. An advantage of the acoustic technique is the breadth of the measured information since each image pixel represents elastic property data. Quartz and Plexiglas were scanned with the bone simultaneously to serve as calibration materials. See Figure 1a. Based on the known acoustic reflectivity of quartz and Plexiglas, a linear relationship was established between grayscale value and acoustic reflectivity. Given a known coupling fluid acoustic impedance and bone density, the acoustic reflectivity was related to the C33 elasticity coefficient [1]. The coupling fluid was assigned the properties of water. Density was measured independently using a specific gravity technique and assumed to be uniform throughout the tissue sample. Acoustic images were recorded over a 2 mm x 3 mm area for subsequent comparison with nanoindentation.

The NanoIndenter II was used to measure the lamellar elastic modulus. The nanoindentation technique offers force and displacement resolutions of 0.3 mN and 0.16 nm, respectively, and has recently been validated for use on bone tissue [2]. The practical spatial resolution in bone is about 5 mm. A custom irrigation system using a 0.5 mg/ml gentamicin solution was designed to maintain tissue moisture during the test. Under light microscopy, 28 regions within the previously tested area were selected for modulus measurements and comparison with acoustic microscopy. At each location, a 30 mm square array of 4 indents was placed. Indents were made at a loading rate of 10 nm/s to a maximum depth of 500 nm. An isotropic elastic modulus was computed from the unloading segment of each load-displacement curve based on an analytical indentation model and an assumed Poisson ratio of 0.3 [3].

Bone tissue C33 elasticity coefficients and elastic moduli were compared using a linear regression analysis with significance attributed to p < 0.05.

Comparisons were also made by measuring the properties of silica to avoid the effects of heterogeneity and anisotropy. Nine acoustic and fourteen indentation measurements were taken in the same region. Silica has uniform material properties that allow elastic modulus to be computed directly from C33 using isotropic elasticity. Elastic moduli were calculated and compared with a single factor ANOVA followed by Tukey HSD tests.

Results

Figure 1a illustrates a typical acoustic microscopy scan at 10 mm resolution. Note that each pixel of the image contains elastic property data.

The regression analysis in Figure 1b revealed a weak but significant correlation between C33 and the elastic modulus (p = 0.001, r=0.581). C33 was consistently greater than the elastic modulus as expected, but there was considerable variation in the relationship. Acoustic measures of silica yielded an elastic modulus of 67.43±4.74 GPa (c.v.=7.0%) which was significantly less than the nanoindentation value of 74.98±2.46 GPa (c.v.=3.2%; p<0.001). The published elastic modulus for silica is 72 GPa which produced 6.4 and 4.1 percent errors for acoustic and indentation techniques, respectively.

Discussion

A cursory examination of the regression reveals that only 34 % of the variation in elastic modulus can be explained by the C33 elasticity constant. Possible reasons for the discrepancy include the material homogeneity and symmetry assumptions engendered in the two techniques. Comparing the results in silica should circumvent the errors associated with heterogeneity and anisotropy. However, it is clear that the contrast is more complex. Elastic moduli measured in the silica sample are quite different suggesting that acoustic microscopy and nanoindentation may be characterizing distinct material properties that cannot be related theoretically.

Based on the silica results, nanoindentation measures of elastic modulus are slightly more accurate and reproducible than scanning acoustic microscopy measures. However, the broad and continuous field of elastic property data produced by scanning acoustic microscopy, relatively inexpensive equipment and decreased testing time still make it a technique to be explored in order to quantify lamellar level mechanical properties adjacent to bone cells.

a) b)

Figure 1. a) Acoustic microscopy scan of bone tissue with quartz and Plexiglas calibration materials. Length scale at the bottom of the image is in mm. b) Linear regression between the C33 elasticity coefficient measured by SAM and elastic modulus measured by nanoindentation.

References

1.  
2. Hasegawa K., Turner C.H., Recker R.R., Wu E., and Burr D.B., 1995, "Elastic Properties of Osteoporotic Bone Measured by Scanning Acoustic Microscopy," Bone , Vol. 16, No. 1, pp. 85-90.