Changes in the mineral density distribution in human bone with age: Image analysis using backscattered electrons in the SEM (original) (raw)
Related papers
Determining mineral content variations in bone using backscattered electron imaging
Bone, 1997
The mechanical properties of bones are greatly influenced by the ratio of organic constituents to mineral. Determination of bone mineral content on a macroscopic scale is straightforward, but microscopic variations, which can yield new insights into remodelling activities, mechanical strength, and integrity, are profoundly more difficult to measure. Measurement of microscopic mineral content variations in bone material has traditionally been performed using microradiography. Backscattered electron (BSE) imaging is a technique with significantly better resolution than microradiography with demonstrated consistency, and it does not suffer from projection-effect errors. We report results demonstrating the applicability of quantitative BSE imaging as a tool for measuring microscopic mineral content variations in bones representing a broad range of mineralization. Bones from ten species were analyzed with Fourier-transformed infrared spectroscopy, X-ray diffraction, energy dispersive X-ray spectrometry, ash measurements, and BSE imaging. BSE image intensity (graylevel) had a very strong positive correlation to mineral (ash) content. Compositional and crystallographic variations among bones had negligible influence on backscattered electron graylevels. The present study confirms the use of BSE imaging as a tool to measure the microscopic mineral variability in a broad range of mineralized tissues. (Bone 20:485-490; 1997) © 1997 by Elsevier Science Inc. All rights reserved.
Influence of mineral content and composition on graylevels in backscattered electron images of bone
Journal of Biomedical Materials Research, 1993
To determine the meaning of graylevels in backscattered electron (BSE) images of actual bone tissues, the influence of mineral content and mineral composition on BSE image graylevels was studied using chick bone tissue representing a broad age range. These tissues were analyzed for BSE image graylevels,. Ca/P molar ratios, mineral composition, mineral content (v/v), ash fraction (w/w), and density (g/cm3). Linear regression analyses showed that the weighted mean graylevels (WMGLs) in BSE images were positively correhted to ash fraction (r 2 = 0.711), mineral content (r 2 = 0.720), and density (Y' = 0.843). Although the Ca/P ratio increased from 1.65 in embryos to 1.80 in 2-year olds, the compositional changes corresponding to this Ca/P molar ratio were estimated to produce a relatively minor (< 4.0%) change in BSE image graylevel. These results demonstrate that graylevels in BSE images of actual bone tissue can be attributed to mineral content and density, but only as a coincidence of their association with atomic number.
The meaning of graylevels in backscattered electron images of bone
Journal of Biomedical Materials Research, 1993
Backscattered electron (BSE) imaging is considered to be a useful technique for determining relative differences in bone tissue density. However, it is not clear how graylevel variations seen in BSE images of bone tissue, which are primarily dependent on the tissue's average atomic number, correlate to tissue density (g/cm3) and mineral content. Simulated bone tissues, ranging from 32-500/0 mineral by volume, were made by mixing synthetic hydroxyapatite with a simulated organic matrix. This technique allowed mineral content to be varied while mineral composition and crystallography remained constant. The densities of the simulated tissues were determined using Archimedes' principle. Average atomic numbers of the simulated tissues were interpolated from a regression of BSE graylevel against average atomic numbers of pure standard materials. A strong positive correlation was found to exist between mineral content and density (r 2 = 0.978) as well as between mineral content and atomic number (r2 = 0.965). The average graylevel in the BSE image also exhibited a positive correlation to mineral content (r2 = 0.965) and density (r 2 = 0.923). Graylevel variations in BSE images of simulated bone tissue were shown to be strongly correlated to density and mineral content, but only as a coincidence of their association with atomic number.
Mineralisation density of human mandibular bone: quantitative backscattered electron image analysis
Journal of Anatomy, 1998
This study examined the tissue level mineralisation density distribution in mandibles from 88 adult humans. Mandibles (19-96 y) were sectioned vertically in midline (MID), mental foramen (MF), and third molar (M3) regions. Surgical fragments from M3 were obtained from individuals aged 16-38 y. All specimens were cleaned, embedded in PMMA, micromilled and examined by digital 20 kV backscattered electron (BSE) stereology. Quantitation was based on rescaling image histograms to the signal range between a monobrominated (0) and a monoiodinated (255) dimethacrylate resin standard. Mineralisation density increased with age (r l 0.70 ; P 0.0001) : the mean for 39 individuals aged between 16 and 50 y was significantly lower (P 0.0001) than for 35 individuals over 51 y (mean (p...) : 158.20 (1.63) and 174.71 (1.27) normalised grey level units respectively). There was good correlation in mean mineralisation density between different sites in the same mandible, but MID was significantly less highly mineralised than the other sites : MID 173.90, MF 177.34, M3 177.11 (P 0.002 and 0.01 for MF and M3 respectively ; paired t test), as was the alveolar bone density when compared with the bone of the inferior cortex (e.g. MID : 171.13 (1.53) and 174.46 (1.14) P 0.0001). No sex difference was found. Partially dentate mandibles generally had regions of higher mineralisation than fully dentate and edentulous mandibles. The lowest density bone occurred at the alveolar crest anteriorly and superolingually at M3, matching sites of net resorption following tooth loss. Highest densities were found inferolingually at MID, inferiorly at MF and buccally at M3, matching the sites thought to experience the highest functional strains. This stresses the importance that local factors may have in the remodelling of the edentulous mandible. Morphology showed that there is a preponderance of highly mineralised cement lines, and of packets containing dead, mineralised, osteocytes.
Journal of Anatomy, 2013
The shape and structure of bones is a topic that has been studied for a long time by morphologists and biologists with the goal of explaining the laws governing their development, aging and pathology. The osteonal architecture of tibial and femoral mid-diaphyses was examined morphometrically with scanning electron microscopy in four healthy young male subjects. In transverse sections of the mid-diaphysis, the total area of the anterior, posterior, lateral and medial cortex sectors was measured and analysed for osteonal parameters including osteon number and density, osteon total and bone area and vascular space area. Osteons were grouped into four classes including cutting heads (A), transversely cut osteons (B), longitudinally cut osteons (C) and sealed osteons (D). The morphometric parameters were compared between the inner (endosteal) and outer (periosteal) half of the cortex. Of 5927 examined osteons, 24.4% cutting heads, 71.1% transversely cut osteons, 2.3% longitudinally cut osteons and 2.2% sealed osteons were found. The interosteonic bone (measured as the area in a lamellar system that has lost contact with its own central canal) corresponded to 51.2% of the endosteal and 52.4% of the periosteal half-cortex. The mean number of class A cutting heads and class B osteons was significantly higher in the periosteal than in the endosteal half-cortex (P < 0.001 and P < 0.05, respectively), whereas there was no significant difference in density. The mean osteon total area, osteon bone area and vascular space area of both classes A and B were significantly higher (P < 0.001 for all three parameters) in the endosteal than in the periosteal half-cortex. The significant differences between the two layers of the cortex suggest that the osteoclast activity is distributed throughout the whole cortical thickness, with more numerous excavations in the external layer, but larger resorption lacunae closer to the marrow canal. A randomly selected population of 109 intact class B osteons was examined at higher magnification (3509) to count osteocyte lacuna and to analyse their relationship with osteon size parameters. The distribution frequency of the mean number of osteocyte lacunae increased with the increment in the sub-classes of osteon bone area, whereas the density did not show significant differences. The number of osteocyte lacunae had a direct correlation with the osteon bone area and the mean osteon wall thickness, as well as the mean number of lamellae. The osteocyte lacunae density showed an inverse relationship. These data suggest a biological regulation of osteoblast activity with a limit to the volume of matrix produced by each cell and proportionality with the number of available cells in the space of the cutting cone (total osteon area). The collected data can be useful as a set of control parameters in healthy human bone for studies on bone aging and metabolic bone diseases.
Morphologic detail of aging bone in human vertebrae
Endocrine, 2002
To investigate aging bone structure of humans--here, in the lumbar vertebral bodies-requires methodologies that have sufficiently high resolving power yet still have sufficient width and depth of field. No clinical imaging method can come close to meeting the first requirement, leading to the disadvantage of being limited to postmortem studies. Few microscopic methods meet the second and third requisites. The three-dimensional (3D) images of bone in this article were obtained using deep-field 3D optical imaging, X-ray imaging, and scanning electron microscopy (SEM) of macerated plane parallel slices. The study of bone as a 3D object provides a different perspective from conventional two-dimensional images, and enriches our understanding of how modeling and remodeling processes regulate bone structure and connectivity. The study of ultraflat block surfaces by quantitative back-scattered electron imaging permits acquisition of data on mineral distributions and densities within a very ...
Scanning, 2006
Backscattered electron (BSE) imaging has proven to be a useful method for analyzing the mineral distribution in microscopic regions of bone. However, an accepted method of standardization has not been developed, limiting the utility of BSE imaging for truly quantitative analysis. Previous work has suggested that BSE images can be standardized by energy-dispersive x-ray spectrometry (EDX). Unfortunately, EDX-standardized BSE images tend to underestimate the mineral content of bone when compared with traditional ash measurements. The goal of this study is to investigate the nature of the deficit between EDX-standardized BSE images and ash measurements. A series of analytical standards, ashed bone specimens, and unembedded bone specimens were investigated to determine the source of the deficit previously reported. The primary source of error was found to be inaccurate ZAF corrections to account for the organic phase of the bone matrix. Conductive coatings, methyl-methacrylate embedding media, and minor elemental constituents in bone mineral introduced negligible errors. It is suggested that the errors would remain constant and an empirical correction could be used to account for the deficit. However, extensive preliminary testing of the analysis equipment is essential.
Bone, 1995
Bone samples from the rim of the iliac crest were obtained at autopsy from 59 patients aged 23 to 75 years, of whom 10 men and 10 women aged 50-75 years had osteoarthritis diagnosed by hand X-rays. An equal number in the same age group and 10 men and 9 women aged less than 50 years were without osteoarthritis. After embedding the bone in PMMA, the blocks were cut, polished, and coated with carbon. The fractions of bone falling within four consecutive bands of signal level were derived from digital backscattered electron imaging. Normal males had more low and medium density bone and normal females more very high density phase tissue proportionately. In both male and female osteoarthritis cases, low and medium fractions were low. The very high density fraction was mainly calcified fibrocartilage; when it was excluded from the calculations, the low, medium, and high phases occurred equally in normal males but increased stepwise in normal females and in osteoarthritis cases of both sexes. The results suggest a lower rate of bone renewal in females than males, and in male osteoarthritis subjects than normal males. An increased proportion of bone of high density would reduce the quality of the bone by increasing its stiffness.
Back-scattered electron imaging of skeletal tissues
Metabolic bone disease & related research
The use of solid-state back-scattered electron (BSE) detectors in the scanning electron microscopic study of skeletal tissues has been investigated. To minimize the topographic element in the image, flat samples and a ring detector configuration with the sample at normal incidence to the beam and the detector are used. Very flat samples are prepared by diamond micromilling or diamond polishing plastic-embedded tissue. Density discrimination in the image is so good that different density phases within mineralized bone can be imaged. For unembedded spongy bone, cut surfaces can be discriminated from natural surfaces by a topographic contrast mechanism. BSE imaging also presents advantages for unembedded samples with rough topography, such as anorganic preparations of the mineralization zone in cartilage, which give rise to severe charging problems with conventional secondary electron imaging.