This article is a US Government work and, as such, is in the public domain in the United States of America.
Research has shown that there is a dramatic increase in the fractional area of calcified fibrocartilage from tendon and capsular insertions on the human femoral neck (Vajda and Bloebaum, 1999; Shea et al., 2001b). Additional information regarding the properties of the proximal femur's cortical shell, gained from the use of an animal model, may result in a better understanding of elderly hip fracture since the cortical shell is a significant contributor to the strength of the proximal femur. The objective of the present study was to determine if the greater trochanter's tendon insertions of the human, rat, and sheep differ in terms of morphology and mineralization. The tendons of the greater trochanter of the human, rat, and sheep were observed to insert via a fibrocartilage insertion. The mineral content of the human and sheep calcified fibrocartilage was significantly higher than that of the rat calcified fibrocartilage (P < 0.01). Additionally, the mineral content of the rat cortical bone was significantly higher than that of the human cortical bone (P < 0.01). The mineral content of the calcified fibrocartilage and bone of the human and sheep were not statistically different from each other. There were also more similarities between the bone structure and lacunae density of the human and sheep than between the human and the rat. This suggests that the tendon insertions of the sheep are a better model than the tendon insertions of the rat for the investigation of calcified fibrocartilage in elderly hip fractures. Anat Rec 266:177–183, 2002. Published 2002 Wiley-Liss, Inc.
There has been a heightened focus on changes with age in the cortical shell of the human proximal femur, with the goal of better characterizing how these changes influence the exponential increase in hip fractures with age (Mazess, 1990; Vega et al., 1991; Mautalen et al., 1996). These changes include a decrease in cortical thickness, an increase in porosity, and an increase in the fractional area of a potentially brittle hypermineralized tissue (Boyce and Bloebaum, 1993; Vajda and Bloebaum, 1999). A recent investigation by Shea et al. (2001b) identified the hypermineralized tissue on the human femoral neck and trochanteric region as calcified fibrocartilage associated with tendon and capsular insertions. Despite the lack of concrete evidence associating calcified fibrocartilage with hip fractures, the tissue can comprise up to 60% of the cortical shell in some elderly individuals (Vajda and Bloebaum, 1999; Shea et al., 2001b). Additionally, finite element research and experimental evidence suggests that the cortical shell is a significant contributor to the strength of the proximal femur (Lotz et al., 1995; Cheung et al., 1997; Michelotti and Clark, 1999). Any tissue that affects the properties of the metaphyseal cortex could greatly change the susceptibility of this region to fracture during a fall. Therefore, additional information regarding the healing potential and mechanical properties of calcified fibrocartilage may lead to further insight into the causes of the dramatic increase in hip fractures with age.
The use of animal models is one important way to gain information and learn about the biological mechanisms that influence the aging changes in a tissue. An ideal animal model of calcified fibrocartilage, like all models, would ideally resemble the properties of human tissue. One important step in justifying an animal model is to compare the structure of the proposed animal tissue with that of human tissue. Several qualitative comparative anatomy studies have investigated the structure of the human Achilles tendon (Benjamin et al., 2000) and patellar tendon (Clark and Stechschulte, 1998) compared to different animal models. Yet, there have been no quantitative investigations. Additionally, the mineral content of the calcified fibrocartilage zone of tendon insertions has mostly been ignored, even though it is known that mineralization is one of the key factors in determining the mechanical properties of bone (Mather, 1968; Currey, 1988; Currey et al., 1996). Mineralization of calcified fibrocartilage may also be important, since data collected in this laboratory indicated that the mineralization of the femoral neck cortical bone did not increase with age, but the mineralization of the calcified fibrocartilage increased by 5%. This suggests that the mineralization of calcified fibrocartilage may play a greater role in the aging changes to the proximal femur than the mineralization of the cortical bone (Shea et al., 2001a).
Recent advances in backscattered electron (BSE) technology have enabled researchers to examine the mineral content of bone independent of porosity on a microscopic scale (Vajda et al., 1995; Bloebaum et al., 1997; Roschger et al., 1998). The main objective of the present study was to begin the process of validating an animal model for examining the healing and mechanical properties of calcified fibrocartilage. This was accomplished by determining whether the tendon insertions of the greater trochanter of the human, rat, and sheep differ in terms of morphology, mineral content of the calcified fibrocartilage, and mineral content of the underlying cortical bone. The rat and sheep were chosen because both are common animal models used in orthopedic and soft-tissue insertion research (Frost and Jee, 1992; Newman et al., 1995; Gao et al., 1996; Aerssens et al., 1998; Clark and Stechschulte, 1998).
MATERIALS AND METHODS
Femurs from five human females (59–64 years old), five female rats (6–8 months old), and five female sheep (3–5 years old) were obtained with the tendon insertions still intact. The specimens were fixed in 70% alcohol for a minimum of 2 weeks and embedded in methylmethacrylate following previously published protocols (Sanderson et al., 1990; Sanderson and Kitabayashi, 1994). The greater trochanter was isolated and coronally divided into two equal sections (Fig. 1). The greater trochanters of the sheep and humans were sectioned with a high-speed, water-cooled, slow-feed bone saw with a diamond-impregnated blade (Rockazona, Peoria, AZ). The greater trochanters of the rats were sectioned with a smaller diamond-impregnated blade using a water-cooled bone saw (Isomet; Beuler, Lake Bluff, IL). One coronal section from each specimen was used for light histology of the tendon insertion, and the other coronal section was used for quantitative BSE imaging of the calcified fibrocartilage and cortical bone (Skedros et al., 1993a; Bloebaum et al., 1997; Roschger et al., 1998).
One section of each greater trochanter was ground to a thickness of 50–100 μm and surface-stained with modified Giemsa stain (de Leon and Armstrong, 1991). Each specimen was analyzed with a light microscope at a magnification of 50× for the presence of the four zones of the tendon insertion: 1) tendon, 2) uncalcified fibrocartilage, 3) calcified fibrocartilage, and 4) cortical bone (Benjamin and Ralphs, 1997). The arrangement and shape of the cells in each zone were also analyzed.
Composite Block Preparation
One section of the greater trochanter from each specimen was used to make a composite block for BSE imaging and analysis (Bloebaum et al., 1997). The composite block was diamond-micromilled and coated with carbon (Vajda et al., 1999). Two mineral standards were also placed in the composite block.
Mineral Standards/Ash Measurements
The mineral standards used in the composite blocks were chosen for their mineral contents. A fin whale tympanic bulla and a Rocky Mountain mule deer antler were selected because they represent the highest and lowest mineral content, respectively, that is typically observed in mammalian bones (Bloebaum et al., 1997). A section of each mineral standard was divided into three similar-sized segments. The outer two segments were used for ash measurements, and the inner segment was embedded in methylmethacrylate and used in the composite block.
The bone mineral standard segments used for ash percent measurements were defatted by soaking the specimens for 20 days in a large volume of reagent-grade chloroform (Omnisolv; EM Industries, Inc., Gibbstown, NJ) under vacuum with constant stirring (Folch et al., 1957). Residual chloroform was removed by placing the specimens in an 80°C oven for 5 days. Following defatting and drying, each bone segment was weighed using a precision analytical balance (Mettler H51; Mettler Instruments Corp., Hightstown, NJ) and then placed in a furnace at 550°C for 24 hr to remove organic constituents (Gong et al., 1964; Skedros et al., 1993b). The bone specimens were weighed again after ashing. Ash percent was calculated as 100 times the ratio of the weight of ashed bone (WAB) to the weight of dry defatted bone (WDB) [Ash % = (WAB/WDB) × 100]. Percent ash is a measurement of mineral content independent of porosity. The value used in this study is the mean value of the two segments used from each bone.
Microscope calibration, imaging, and analysis were performed using a previously published protocol (Bloebaum et al., 1997). The carbon-coated composite block was placed on the stage of a JEOL 6100 SEM (JEOL USA, Inc., Peabody, MA). BSEs were collected with a solid-state annular BSE detector (Tetra, Oxford Instruments Ltd., Buckingshire, UK). Operating conditions of the scanning electron microscope included a 20-kV accelerating voltage, 0.7-nA probe current, 50-μm aperture size, and 15-mm working distance. Additionally, a probe current detector (model 6100; JEOL USA, Inc., Peabody, MA) attached to an external picometer (model 485 autoranging picometer; Keithly Instruments, Inc., Cleveland, OH) was monitored to ensure that the probe current remained consistent to within ± 0.02 nA. Probe current was measured before each image capture and any deviations in probe current were manually corrected for by fine adjustments in the condenser lens strength. Two 50× images were taken of each specimen in the block. The first image was taken at the apex of the trochanter, and the second image was taken adjacent and lateral to the first image. The specimens were aligned so that the periosteal surface was parallel to the top of the image. Additionally, three random 200× images were taken from this same region of the calcified fibrocartilage to quantify lacunae density (lacunae/mm2). Images were captured as 512 × 512-pixel digital images with a graylevel depth of 8 bits (256 distinct graylevels) using a computer-controlled image capture and retrieval system (ISIS 300 series; Oxford Instruments, Ltd., Buckingshire, UK).
The weighted mean graylevel (WMGL) of the calcified fibrocartilage and the underlying cortical bone was calculated using a previously published equation (Bloebaum et al., 1997):
where: Ai = area of ith graylevel; GLi = ith graylevel; and At = total area imaged. This provides a mean value for the backscattered signal independent of porosity (black pixels 0–5 in the BSE image). BSE-image WMGLs were calibrated at 20-min intervals using pure aluminum and carbon as calibration standards (Bloebaum et al., 1997; Roschger et al., 1998). The use of pure elements provides a simple method of confirming beam stability during the course of an experiment (Grynpas et al., 1994; Roschger et al., 1995; Vajda et al., 1995). While the mineral standards were used to convert WMGLs to percent ash, a mechanized stage was utilized to ensure that the same microscopic regions of the metal and mineral calibration standards were analyzed throughout the course of the experiment. Quantitative, experimental studies have demonstrated the applicability of this technique (Boyce et al., 1990; Vajda et al., 1995).
WMGLs are linearly correlated to the ash content of bone (Bloebaum et al., 1997). The two mineral standards of known percent ash were used to create a linear regression line correlating WMGL to percent ash. This equation was then used to convert the WMGL of the images from each specimen to percent ash. The percent ash of calcified fibrocartilage and cortical bone of the three species was compared using a Kruskal-Wallis test. A Mann-Whitney U test was then used to analyze individual differences. Statistical significance was set at a P-value < 0.05.
The area of calcified fibrocartilage from the 200× images was analyzed using image analysis software (NIH-Image 1.61). The number of cells, identified as lacunae spaces in the BSE image, were then quantified (Boyce and Bloebaum, 1993). The statistical difference of lacunae density between the three species was compared using a Kruskal-Wallis test. A Mann-Whitney U test was used to analyze individual differences. Statistical significance was set at a P-value < 0.05.
The general structure of the tendon insertions of the greater trochanter did not differ among the three species. The tendons of the human, rat, and sheep were observed to insert into the cortical bone through the known four distinct zones: tendon, uncalcified fibrocartilage, calcified fibrocartilage, and cortical bone. Additionally, the cells of the tendon, uncalcified fibrocartilage, and calcified fibrocartilage from the three species were arranged parallel to and between the tendon fibers. The cells of the uncalcified and calcified fibrocartilage were characteristically spherical, in contrast to the elongated fibroblasts of the tendon.
Despite these similarities in the general structure of the tendon insertions in the three species analyzed, minor differences were noted. For instance, the junction between the calcified fibrocartilage and cortical bone appeared to be more interdigitated in the human and sheep compared to the straighter boundary between the calcified fibrocartilage and cortical bone of the rat (Fig. 2). Also, the human cortical bone appeared to have more secondary osteons than the cortical bone of the rat and sheep. Primary osteons were present at the junction of the calcified fibrocartilage and cortical bone of the sheep, but most of the cortical bone was plexiform in structure. However, there were a few secondary osteons in the sheep's cortical bone near the tendon insertions. The cortical bone of the rat did not appear to have any osteons at the junction of the calcified fibrocartilage and cortical bone. The rat cortical bone was observed to be composed entirely of lamellar bone.
The ash percent values of the human and sheep calcified fibrocartilage were not statistically different from each other (P > 0.05). The ash percent values of the human calcified fibrocartilage (72.2% ± 1.6%; mean ± S.D.) and the sheep calcified fibrocartilage (69.4% ± 3.5%) were significantly higher than those of the rat calcified fibrocartilage (62.9% ± 1.7%; P < 0.01) (Fig. 3). Conversely, the ash percent of the rat cortical bone (69.0% ± 1.6%) was significantly higher than the ash percent of the human cortical bone (66.2% ± 0.32%; P < 0.05) but not the sheep cortical bone (66.4% ± 1.8%; P > 0.05) (Fig. 4). These observations were also qualitatively observed in the BSE images (Fig. 5). The calcified fibrocartilage of the human and the sheep appeared brighter than the underlying cortical bone, while the calcified fibrocartilage of the rat appeared darker than the underlying cortical bone (Fig. 5).
The lacunae density of the human calcified fibrocartilage (121.9 ± 63.0 lacunae/mm2) and the lacunae density of the sheep calcified fibrocartilage (152.1 ± 32.3 lacunae/mm2) were significantly lower than the lacunae density of the rat calcified fibrocartilage (315.7 ± 71.1 lacunae/mm2; P < 0.01) (Fig. 6). There was no statistical difference between the lacunae density of the calcified fibrocartilage of the human and the sheep (P > 0.05).
The tendon insertions of the greater trochanter of the human, rat, and sheep were observed to insert through four distinct zones: tendon, uncalcified fibrocartilage, calcified fibrocartilage, and cortical bone. This general structure has been shown to exist in numerous species, including human (Cooper and Misol, 1970; Benjamin et al., 1986; Evans et al., 1991; Shea et al., 2001b), rat (Benjamin et al., 1991; Bland and Ashhurst, 1997; Benjamin et al., 2000), rabbit (Gao and Messner, 1996; Clark and Stechschulte, 1998), dog (Clark and Stechschulte, 1998), and sheep (Clark and Stechschulte, 1998). However, this is the first investigation to compare the structure of the tendon insertions of the greater trochanter between different species. Because of their uniqueness, it is necessary to individually characterize each tendon and ligament insertion. This is exemplified by the dramatic differences in the morphology of the medial collateral ligament insertions compared to the Achilles tendon insertion (Gao et al., 1996; Rufai et al., 1996).
In terms of the cortical bone, there were some differences in morphological structure between the three species. The human had both primary and secondary osteons throughout the cortical shell and in the cortical bone underlying the calcified fibrocartilage. The sheep lacked secondary osteons in the majority of its cortical bone. However, near the junction of the calcified fibrocartilage and cortical bone there were primary as well as occasional secondary osteons. Other investigators have reported the presence of primary and secondary osteons in the cortical bone of sheep (Newman et al., 1995; Skedros et al., 1997). The cortical bone of the rat lacked the primary and secondary osteons that were present in both the human and the sheep. These findings are supported by the observations of other investigators (Frost and Jee, 1992; Newman et al., 1995; Bentolila et al., 1998), who found that the rat normally lacks Haversian remodeling, except in extreme mechanical loading situations and in the occasional older rat.
This is the first investigation to compare the percent ash of calcified fibrocartilage from tendon insertions between different species. In this study the percent ash values of the calcified fibrocartilage of the human and sheep were 14% and 9% higher, respectively, than that of the rat, and the difference was statistically significant. However, the percent ash values of the human and sheep calcified fibrocartilage were not statistically different from each other. The calcified fibrocartilage of the rat was less mineralized than the cortical bone, whereas the calcified fibrocartilage of the human and sheep was more mineralized than the underlying cortical bone. A limitation of this investigation was that the rats and sheep were not necessarily from equivalent age groups. However, it has been shown that the calcified fibrocartilage in young humans (17–30 years of age) is more mineralized than the underlying cortical bone (Shea et al., 2001a). The fact that the calcified fibrocartilage of the rat in this investigation had a lower mineralization than the underlying cortical bone suggests that even if the representative ages are not equivalent, the tendon insertion of the rat in terms of mineralization is different from that of the human. Although there were limitations to this investigation, the data indicate that a sheep model would be useful for investigations concerning calcified fibrocartilage.
For the cortical bone, the percent ash of the rat was statistically higher than that of the human, and there was no statistical difference between the percent ash of the sheep and human cortical bone. The data on percent ash of the cortical bone of the human, rat, and sheep from this investigation were comparable with data obtained by traditional ashing techniques (Trotter and Peterson, 1955; Blitz and Pellegrino, 1969; Aerssens et al., 1998). However, those studies were limited to the mid-diaphysis. The mineral content and morphology of cortical bone can vary within a cross section (Skedros et al., 1996) and between sites distant from each other (Vajda and Bloebaum, 1999; Shea et al., 2001a). Therefore, each site under investigation needs to be analyzed individually. The present investigation is unique in that it compared the percent ash of the proximal femur's greater trochanter cortical bone between different commonly used animal models.
The differences in measured percent ash of the calcified fibrocartilage of the human and sheep compared to the rat could have mechanical consequences. Mineral content, as measured by percent ash, is strongly correlated to elastic modulus, strength, and fracture toughness of cortical bone (Martens et al., 1983; Currey, 1988; McCalden et al., 1993; Currey et al., 1996). The strong influence of mineral content on mechanical properties is expected to be true for calcified fibrocartilage, since in this investigation it was shown that mineral constituted 72% of the mass fraction of calcified fibrocartilage in the human. Any differences in the mineralization of the calcified fibrocartilage and cortical bone between the animal model and the human could limit the application of the animal model when investigating the aging changes and mechanical properties of the tissue. Therefore, in terms of mineral content the sheep appears to be a better model for the calcified fibrocartilage and cortical bone interface than the rat. Currently, the relationship between mineral content and the mechanical properties of calcified fibrocartilage are unknown, but it is now being investigated in our laboratory.
In addition to the differences in mineralization between the rat and human, there were also statistical differences in lacunae density. However, there were no statistical differences in the lacunae density between the sheep and human. The similarities between the human and sheep measured in this study indicate that the sheep may be a preferential model in future studies examining the remodeling dynamics of bone and calcified fibrocartilage, since it has been shown in bone that a decrease in lacunae density is associated with an increase in microdamage and porosity (Vashishth et al., 2000; Power et al., 2001). Microdamage accumulation and porosity are both negative influencers on mechanical properties (Currey, 1988; Burr et al., 1998). Although the role of fibrocartilage cells is unknown, a reduced lacunae density and vascular supply (Cooper and Misol, 1970; Benjamin and Ralphs, 1997; Petersen and Tillmann, 1999) could limit the ability of this region to sense and repair damage.
The differences between the morphology of the calcified fibrocartilage and cortical bone of the rat and human do not invalidate the usefulness of this animal model for tendon and ligament research. The rat is inexpensive, easy to house, and well researched in terms of development and morphology (Gao et al., 1996; Bagi et al., 1997; Miller, 1997; Ralphs et al., 1998; Benjamin et al., 2000). However, the rat was not morphologically similar to the human in terms of the properties measured in the present investigation.
The present investigation observed more similarities between the morphology of the tendon insertions of the greater trochanter of the sheep and the human compared to the rat and the human. The morphological commonalities between the human and sheep tendon insertions include: percent ash of the calcified fibrocartilage, percent ash of the cortical bone, lacunae density, and presence of primary and secondary osteons in the cortical bone. These greater similarities between the human and the sheep suggest that the sheep is superior to the rat as a model for investigating the role of calcified fibrocartilage in the dramatic increase in hip fractures with age. However, the present investigation is only a small stepping stone in validating an animal model for calcified fibrocartilage. Future research should more fully elucidate the role of calcified fibrocartilage in hip fractures, as well as the influences of hormonal changes on the properties of calcified fibrocartilage.
This material is based on work supported by the Office of Research and Development, Medical Research Service, Department of Veterans Affairs Salt Lake City Health Care System; and the Department of Orthopedics, University of Utah School of Medicine. We thank Dr. Kent Bachus, Erin Whitaker, and Trina VanAusdal for their technical support, and Gwenevere Shaw for her assistance in manuscript preparation.