Non-Destructive Measurement of Fatigue Microdamage Accumulation using Contrast-Enhanced
Micro-Computed Tomography: Validation by Conventional Histology
1
Landrigan, M D; 2Li, J; 2Burr, D B; 1Niebur, G L; +1Roeder, R K
+1University of Notre Dame, Notre Dame, IN, 2Indiana University School of Medicine, Indianapolis, IN
[email protected]
INTRODUCTION:
Conventional methods used to image and quantify microdamage
accumulation in bone are limited to thin histological sections, which are
inherently invasive, destructive, tedious and two-dimensional [1].
Recent studies have begun to investigate methods for non-destructive,
three-dimensional (3-D) detection and imaging of microdamage in bone
tissue. Micro-computed tomography (micro-CT) has been used to detect
and quantify microdamage accumulation in vitro using iodinated [1,2],
lead sulfide [3,4] and barium sulfate (BaSO4) [5-7] contrast agents.
Therefore, the objective of this study was to validate measurement of
microdamage accumulation in human cortical bone with micro-CT using
a BaSO4 contrast agent by conventional histology using basic fuchsin.
METHODS:
Ninety specimens were sectioned from the femoral mid-diaphyses of
three adult men (63 ± 1.7 years of age), turned down to a 2.5 mm
diameter by 5 mm gauge length using a CNC mini-lathe, and randomly
divided into six groups (n = 15), comprising an unloaded control and
two mechanically loaded groups, for both micro-CT and conventional
histology. Specimens were wrapped in gauze, hydrated in PBS, and
stored at –20°C in airtight containers during interim periods.
Mechanically loaded specimens were fatigued in cyclic uniaxial
tension at 2 Hz under load-control (R = 0) on an electromagnetic test
instrument (Bose ELF3300) while hydrated with a water drip at ambient
temperature until a 5 or 10% reduction in secant modulus was achieved.
Specimens were preloaded at 60 MPa for 20 cycles to measure the initial
secant modulus and the fatigue load was normalized to an initial
maximum strain of 6400 ± 300 microstrain.
For micro-CT, specimens were stained by BaSO4 precipitation
soaking in an equal parts mixture of 0.5 M BaCl2 in DI water, buffered
saline, and acetone solution for three days, followed by equal parts 0.5
M NaSO4 in DI water, buffered saline, and acetone for three days, both
under vacuum (~50 mm Hg), and rinsed with de-ionized water. The
entire gauge length of each specimen was imaged by micro-CT (Scanco
µCT-80) at 10 µm resolution, 70 kVp voltage, 113 µA current, and 200
ms integration time with slices taken perpendicular to the longitudinal
axis of the specimen. Images were thresholded to determine the total
bone volume (BV) and BaSO4 stained volume (SV).
For conventional histology, blinded specimens were stained en bloc
by basic fuchsin, sectioned, and imaged by transmitted light microscopy
using standard methods [8]. The number of cracks (Cr.N), total crack
length, bone area (Bn.Ar), mean crack length (Cr.Le), crack density
(Cr.Dn), and crack surface density (Cr.S.Dn) were measured.
RESULTS:
The presence and spatial location of fatigue microdamage was nondestructively detected using micro-CT after staining with BaSO4 (Fig.
1). All specimens loaded in cyclic uniaxial tension exhibited at least one
distinct region of concentrated BaSO4 stain which appeared
characteristic of fatigue damage and/or propagating cracks. The amount
of damage increased continuously from the unloaded control group to
groups loaded to a 5 and 10% reduction in secant modulus for either
micro-CT (p < 0.0005, ANOVA) or conventional histology (p < 0.0001,
ANOVA). The accumulation of microdamage measured by micro-CT
(SV/BV) scaled linearly with the microrack density (Cr.Dn.) measured
by conventional histology (Fig. 2).
DISCUSSION:
Quantification of the accumulation of fatigue microdamage in vitro
using contrast-enhanced micro-CT was validated against measurements
using conventional histology (Fig. 2). BaSO4 staining provided
enhanced contrast for the detection of damage that was not able to be
detected by micro-CT (10 µm resolution) prior to staining (Fig. 1).
Previous studies verified BaSO4 labeling of microcracks and diffuse
damage using electron microscopy [5-7]. Micro-CT also enabled nondestructive imaging and 3-D spatial information (Fig. 1b) which are not
possible using conventional histological methods.
The technique demonstrated in this study was not without limitations.
Micro-CT measurements of SV/BV were slightly less sensitive to
detecting differences in group means compared to conventional
histological measurements of Cr.Dn. For example, the difference in
SV/BV between the unloaded group and a 5% degradation of secant
modulus was not statistically significant (p = 0.14, t-test) (Fig. 2). This
was largely attributed to non-specific BaSO4 staining within void space
and on free surfaces of specimens. Staining by BaSO4 precipitation is
also limited to in vitro studies since the staining solutions are not
biocompatible. Nonetheless, these new methods are immediately useful
for scientific studies investigating the etiology of fatigue and fragility
fractures in bone.
Fig. 1. (a) Grayscale and segmented micro-CT images of the same
specimen cross-section (2.5 mm diameter) after loading to a 10%
reduction in secant modulus but prior to staining (post-fatigue), and after
staining with BaSO4 (post-staining), showing enhanced contrast (bright
voxels) for the detection of damage that was not able to be detected prior
to staining. (b) Segmented, 3-D reconstruction of the entire gauge
section (2.5 mm in diameter by 5 mm in length) showing the ability of
micro-CT to detect spatial variation in damage accumulation.
Fig. 2. The ratio of BaSO4 stain volume to bone volume (SV/BV)
measured in segmented micro-CT images scaled linearly with the
microcrack density (Cr.Dn) measured by conventional histology. Data
points show the mean and one standard deviation (n = 15) for the
unloaded control group and specimens loaded to a 5 and 10%
degradation in secant modulus.
ACKNOWLEDGEMENTS:
This research was supported by the U.S. Army Medical Research and
Materiel Command (W81XWH-06-1-0196) through the Peer Reviewed
Medical Research Program (PR054672).
REFERENCES:
[1] Lee TC, et al., J. Anat., 203:161-172, 2003. [2] Parkesh R, et al., J.
Biomechanics, 39:1552-1556, 2006. [3] Leng H, et al., Trans. Orthop.
Res. Soc., 30:665, 2005. [4] Tang SY, Vashishth D, Bone, 40:12591264, 2007. [5] Wang, et al., J. Biomechanics, 40:3397-3403, 2007. [6]
Leng H, et al., J. Mech. Behav. Biomed. Mater., 1:68-75, 2008. [7]
Landrigan MD, et al., Trans. Orthop. Res. Soc., 34:2143, 2009. [8] Burr
DB, Hooser M, Bone, 17:431-433, 1995.
Paper No. 336 • 56th Annual Meeting of the Orthopaedic Research Society