- Email: [email protected]
Radiographic Calibration for Analysis of Bone Mineral Density of the Equine Third Metacarpal Bone
Journal of Equine Veterinary Science 33 (2013) 1131–1135
Contents lists available at SciVerse ScienceDirect
Journal of Equine Veterinary Science journal homepage: www.j-evs.com
Original Research
Radiographic Calibration for Analysis of Bone Mineral Density of the Equine Third Metacarpal Bone Amber J. Bowen MS a, *, Mathew A. Burd MS, DVM a, John J. Craig PhD b, Monique Craig b a b
Department of Animal Science, California Polytechnic State University, San Luis Obispo, CA EponaTech Inc., Creston, CA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 January 2013 Received in revised form 16 April 2013 Accepted 24 April 2013 Available online 22 June 2013
Digital radiography represents the primary diagnostic tool the veterinarian uses to diagnose skeletal injuries in the horse. Advances in digital radiography have provided the veterinarian with opportunities to make simple radiographic assessments from calibrated digital radiographs such as dimensional analyses; however, more complex variables such as radiographic opacity have yet to be standardized. Therefore, we investigated the quantification of bone mineral density (BMD) via computed radiographic absorptiometry at various radiographic exposure intensities (kV), times (sec), and milliamps (mA) in the third metacarpal in the horse. By developing a brightness/darkness index (BDI), the grayscale of radiographs, calibrated with an aluminum (Al) marker of various known thicknesses and uniform densities, can be compared to the average BMD of a region of interest at various radiographic exposures. Al BDI was a significant predictor of bone BDI (r2 ¼ 0.960, P < .001) and BMD (r2 ¼ 0.971, P < .001). This method of calibration can be used for quantitative noninvasive bone mineral analysis and allows direct comparison of radiographs taken under different exposure settings. Ó 2013 Elsevier Inc. All rights reserved.
Keywords: Radiographic absorptiometry Digital Bone mineral Density
1. Introduction Dual energy x-ray absorptiometry (DEXA) is the most precise, accurate, and widely accepted method of measuring bone mineral density (BMD) for assessment of skeletal integrity and fracture risk in humans [1]. In horses, DEXA analyses show BMD, measured in g/cm2, to be a valid predictor of bone strength [2], exercise-induced bone remodeling in young horses [3-5], and risk for fracture development [6-8]. Unfortunately, the generalized use of DEXA has limited use for equine practitioners as it is expensive, time consuming, and not portable. Radiography and particularly digital and computed radiography represent the primary diagnostic tools the veterinarian uses to diagnose skeletal injuries in the horse. Radiographic * Corresponding author at: Amber J. Bowen, MS, University of California, Davis, School of Veterinary Medicine, Davis, CA 95616. E-mail address: [email protected] (A.J. Bowen). 0737-0806/$ – see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jevs.2013.04.016
absorptiometry and computed digital absorptiometry (CDA) using a single-energy X-ray system has been shown to accurately assess BMD in humans [9,10] and horses [11]. Inclusion of aluminum (Al) markers of various known thicknesses and uniform densities, the grayscale of a region of interest (ROI), termed the brightness/darkness index (BDI), may be calibrated from image to image. The BDI of bone measured from a radiographic ROI can then be related to average BMD, dependent on thickness. However, opacity determined from radiographs taken with different radiographic techniques cannot be directly compared. Calibration of radiographic opacity with various radiographic exposure intensities (kV), times (sec), and milliamps (mA) would allow for quantification of BMD, assuming that the area measured includes minimal soft tissue and has consistent thickness and attenuation properties. Digital radiography has the advantage of being portable, relatively inexpensive, and practical in a field setting. By using CDA techniques, digital radiography may be used to
1132
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135
determine BMD. Therefore, our objective was to develop a method to calibrate radiographic images such that BMD determined from radiographs taken with different techniques could be accurately compared. By calibrating the BDI of digital radiographs, we formed a standardized relationship allowing comparison of computer-determined BDI values with physical properties of BMD regardless of radiographic settings of exposure intensity (kV) and time (sec). This would allow us to account for changes in the opacity of radiographs due to changes in exposure, affording standardization of any radiograph. By including an Al wedge of known density and thickness (g/cm2) in the radiograph, BMD can be determined from BDI by creating a standard curve. Next, BMD can be correlated with BDI by creating a bone step-wedge of known density and thickness. We believe that calibration of digital radiographs to determine BMD from an Al wedge standard will provide a tool for quantification of BMD from common digital radiographs, regardless of radiographic technique. 2. Materials and Methods Radiographs of a single third metacarpal equine cadaver bone and Al wedge taken at various exposures were used to determine the relationship between BDI and radiographic exposure for both bone and Al. Each radiograph contained an Auto-Scaler,1 an Al wedge and cadaver bone. The Al wedge was simply machined from type 6061 Al and was 20.36 cm in length and 0.2-31.8 mm in thickness, with a constant density of 2.70 g/cm3. The Auto-Scaler contained metal markers that standardized the dimensions of the radiographic image when imported into the software program Metron-DVM.1 All radiographs were taken with a portable X-ray system2 and X-ray sensor.3 All radiographs were taken at 15 mA, the only setting available for this system. The cadaver bone was positioned perpendicular to the ground in the center of the line of exposure with the Al wedge and Auto-Scaler positioned on either side. The bone and X-ray system was placed to produce a standard dorsal palmar radiograph. The focal distance was 26 cm, with the plane of interest containing the cadaver bone, Al wedge, and Auto-Scaler positioned against the face of the sensor. Radiographs were taken at all available combinations of preset exposure intensities ranging from 55-80 kV in 5-kV intervals and exposure times from 0.02-0.14 seconds every 0.02 second. The BDI of the Al wedge and cadaver bone was measured on each radiographic image (Fig. 1). To determine BDI, a unitless value was assigned to each 16-bit pixel on a grayscale from 0-65,500, with zero being completely black and 65,500 being completely white. A predetermined ROI on either bone or Al was generated by forming a rectangle using Metron-DVM software.1 BDI was determined by averaging the grayscale value of each pixel in the ROI. In determining BDI, we did not include background pixels. BDI of the Al wedge was measured by creating a rectangle
1 2 3
EponaTech Inc., Creston, CA. MinXray HF80, Northbrook, IL. Thales FS23, Vetel Diagnostics, San Luis Obispo, CA.
Fig. 1. Dorsal palmar radiograph of third metacarpal cadaver bone with Auto-Scaler (left) and Al wedge (right), taken at 60 kV, 0.08 seconds, and 15 mA. ROIs (red) were selected for measuring bone BDI at the nutrient foramen and Al BDI from an Al wedge.
encompassing the entire wedge. BDI of the cadaver bone was characterized by an ROI with the width larger than the cross-sectional width of the bone perpendicular to the sagittal plane and the height equal to the diameter of the nutrient foramen. To ensure change in BDI was due solely to effects of radiographic exposure, the same ROI was used in each radiographic image. BDI of the Al wedge was compared to BDI of the cadaver bone in the same radiographic image at various X-ray intensities and exposure times to ensure the relationship between BDI and radiographic exposure was similar for both materials. BMD was correlated with BDI by measuring BDI from a radiograph of cortical bone of known density and thickness, determined as g/cm2. A bone step-wedge of known BMD was made from cortical bone obtained from a equine cadaver third metacarpal bone in the mid-diaphysis region. The cortical bone was cut into slices using a diamond blade tile saw. The average dimension of each slice was 15.28 3.3 mm square and 4.23 1.0 mm thick. A bone step-wedge was created by stacking each bone slice face-to-face in increasing thickness from one slice to 6 slices thick, consecutively, from step 1-10. The thickness, weight, and volume corresponding to each bone slice was determined
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135
using 6-inch digital calipers4 with 0.01-mm accuracy and a digital balance5 with 0.1-mg accuracy. BMD for each slice of bone was determined from weight (g) multiplied by thickness (cm) and divided by volume (cm3). Aluminum BMD equivalent (Al BMDE) was determined from the bone wedge as density (g/cm3) multiplied by thickness (cm). BDI was determined from the digital radiograph for each step of bone and for the Al wedge (Fig. 2). Digital radiographs of the bone step-wedge and Al wedge on the same image were taken. The bone step-wedge was positioned to intersect the center of the line of exposure, with the Al wedge on one side. The focal distance was 21 cm, with the plane of interest containing the bone step-wedge and Al wedge positioned against the face of the sensor. Radiographs were taken at 60 kV and 0.08 second. Standardized curves were created for bone BDI versus BMD (Fig. 4) and Al BDI versus Al BMDE (Fig. 5). Where Al BDI equaled that of bone, the correlation between bone and Al thickness was determined.
1133
Fig. 2. Bone step-wedge (center) made of 10 cortical bone steps of decreasing thickness (left to right) on a standardized hoof block and Al wedge (right), from a digital radiograph taken at 60 kV, 0.08 seconds, and 15 mA. ROIs (red) were selected for measuring Al BDI (right) and bone BDI (left) from the tenth and thickest step of the bone step-wedge.
2.1. Statistical Analysis Regression analysis was performed using the least squares method. Significance of linear correlation between bone and Al BDI and bone and Al BMDE and bone and Al BDI versus exposure intensity and time using an ANOVA and associated F-test. Statistical significance was set at P < .05. Statistical analysis was performed with R software.6 3. Results and Discussion Veterinary clinicians routinely use radiographs to assess bone health and diagnose abnormalities such as fractures and osteoarthritis, subjectively and qualitatively assessing the opacity or optical density of ROIs. A tool to quantitatively evaluate bone strength or radiographic opacity in ROIs from digital radiographs would be valuable. Therefore, we developed a method based on standardized digital radiographs that affords accurate measurement of BMD. The third metacarpal in the mid-diaphysis region is composed of dense cortical bone and very little trabecular bone and soft tissue and is an ideal choice for investigation of the parameters associated with radiographic standardization. In this study, we calibrated BDI over a range of exposure settings using an Al wedge standard of known density and increasing thickness or an Al BMD equivalent. For each combination of intensity and exposure time, Al BDI showed a linear correlation with bone BDI (R2 ¼ 0.9599, F [1,40] ¼ 957.3, P < .001). As Al BDI increased per unit, there was an increase in bone BDI of 1.12 0.04 (Fig. 3). These results indicate that calibration of BDI with an Al wedge standard curve for a range of exposure settings allows for accurate assessment of BMD from any region of a digital radiograph in units of grams per square centimeter of Al. In this study, we determined BMD from an Al wedge by creating a bone step-wedge of known density and
4 5 6
Fisherbrand, Fisher Scientific, Pittsburg, PA. Accu-224, Fisher Scientific, Pittsburg, PA. R Foundation for Statistical Computing, Vienna, Austria.
thickness. From steps 1-10 of the bone step-wedge, BDI, density, thickness, and BMD were determined (Table 1). Bone BDI was shown to increase linearly with increasing BMD (R2 ¼ 0.9716, F[1,1193] ¼ 40,850, P < .001) (Fig. 4). Similarly, Al BDI increased with increasing Al BMDE, ranging from a BDI of 0-53,810 with a constant density of 2.70 g/cm3. Aluminum thickness increased linearly along the length of the Al wedge from 0-4.26 cm (R2 ¼ 0.9973, F [1,631] ¼ 232,800, P < .001) (Fig. 5). This method permits accurate evaluation of BDI and, therefore, BMD from an Al wedge standard independent of the exposure setting. The quantification of BMD via measurement of BDI from standardized digital radiographs allows for the opacities of radiographs to be compared when taken under different
Fig. 3. Bone versus Al brightness/darkness index (BDI) with linear regression, determined from a digital radiograph of a third metacarpal cadaver bone and Al wedge taken at all available combinations of exposure intensity (55-80 kV) and exposure time (0.02-0.14 seconds).
1134
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135 Table 1 BDI, thickness, density, and BMD determined for 10 steps of the cortical bone step-wedge Step
Mean Bone BDI SD*
1 2 3 4 5 6 7 8 9 10
3,648 4,591 15,962 17,187 22,964 27,536 33,969 46,201 47,870 54,151
723 757 129 268 133 280 127 115 952 691
Thickness (cm)
Density (g/cm3)
BMD (g/cm2)
0.271 0.400 0.754 0.929 1.047 1.325 1.601 2.140 2.498 3.004
1.76 2.02 1.92 1.86 1.95 1.90 1.87 1.96 1.81 1.87
0.48 0.81 1.45 1.79 2.04 2.50 3.00 4.12 4.53 5.61
BDI, brightness/darkness index; BMD, bone mineral density. * Determined from a digital radiograph taken at 60 kV, 0.08 seconds, and 15 mA.
4. Conclusions Fig. 4. Bone brightness/darkness index (BDI) versus bone mineral density (BMD) with linear regression determined from a radiograph of the cortical bone step-wedge.
circumstances. This may have practical applications as BMD is dependent only on thickness and density of the material; therefore, a given area of bone with lower BMD than another area of bone of similar thickness and soft tissue thickness can be concluded to have a lower density. Estimation of bone cross-sectional area or thickness may be a valuable adjunct to this technique. Effect of beam angle, offset, gain, and digital image processing may affect the measurement of BMD and need to be investigated further for assessment of BMD to be clinically applicable [12]. Other imaging modalities available to the equine veterinarian may also accurately measure BMD, such as standard radiography and computed radiography, which provides correct optical density on film independent of the incident radiation exposure [13].
It has been suggested that bone fractures in racehorses are caused by areas of weakened bone or fatigue fractures that occur during training [8,14]. Furthermore, stress fractures induced by exercise are well correlated with decreased BMD in humans, and this technique has been shown to be an accurate predictor of BMD in the horse, using single-energy X-ray systems [11]. This technique may be an asset in the study of factors which influence changes in bone density, in determination of possible risk cases related to increases or decreases in bone density and may find application in the study of osteopenia and other bone diseases of horses. CDA measurement of BMD provides an inexpensive, noninvasive and accurate method of estimating BMD of the third metacarpal in the horse [15,16,17].
Acknowledgement Financial support was provided by the US Department of the Navy, Office of Naval Research, award N00014-08-11209, and by the Agriculture Research Initiative of California State University, under Award # SC350019-09-01.
References
Fig. 5. Al brightness/darkness index (BDI) versus Al bone mineral density equivalents (Al BMDE) with linear regression, determined from a radiograph taken at 66 kV, 0.08 seconds, and 15 mA.
[1] World Health Organization Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Geneva, Switzerland: WHO; 1994. [2] Nielsen D, McEvoy FJ, Madsen MT, Jensen JB, Svalastoga E. Relationship between bone strength and dual-energy x-ray absorptiometry measurements in pigs. J Anim Sci 2007;85:667–72. [3] Jeffcott LB, Buckingham SH, McCarthey RN. Noninvasive measurement of bone quality in horses and changes associated with exercise. Equine Vet J Suppl 1988;6:71–9. [4] Raub RH, Jackson SG, Baker JP. The effect of exercise on bone growth and development of weanling horses. J Anim Sci 1989;67:2508–14. [5] Riggs CM, Boyde A. Effect of exercise on bone density in distal regions of the equine third metacarpal bone in 2-year-old thoroughbreds. Equine Vet J Suppl 1999;30:555–60. [6] Bell GH, Dunbar O, Beck JS, Gibb A. Variations in strength of vertebrae with age and their relation to osteoporosis. Calcif Tissue Res 1967;1:75–86. [7] Järvinen TL, Sievänen H, Kannus P, Järvinen M. Dual-energy X-ray absorptiometry in predicting mechanical characteristics of rat femur. Bone 1998;22:551–8.
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135 [8] Nunamaker DM, Butterweck DM, Provost MT. Fatigue fractures in thoroughbred racehorses: relationships with age, peak bone strain, and training. J Orthop Res 1990;8:604–11. [9] Bouxsein ML, Michaeli DA, Plass DB, Schick A, Melton ME. precision and accuracy of computed digital absorptiometry for assessment of bone density of the hand. Osteoporos Int 1997;7:444–9. [10] Nolla JM, Gomez-Vaquero C, Fiter J, Roig-Escofet D. Computed digital absorptiometry of the hand: screening method of bone loss in postmenopausal women with RA. Ann Rheum Dis 2000; 59:490. [11] Vaccaro C, Busetto R, Bernardini D, Anselmi C, Zotti A. Accuracy and precision of computer-assisted analysis of bone density via conventional and digital radiography in relation to dual-energy xray absorptiometry. Am J Vet Res 2012;73:381–4. [12] Secombe CJ, Firth EC, Perkins NR, Anderson BH. Pathophysiology and diagnosis of third carpal bone disease in horses: a review. N Z Vet J 2002;50:2–8.
1135
[13] Seibert JA, Shelton DK, Moore EH. Computed radiography X-ray exposure trends. Acad Radiol 1996;3:313–8. [14] Warden SJ, Hurst JA, Sanders MS, Turner CH, Burr DB, Li J. Bone adaptation to a mechanical loading program significantly increases skeletal fatigue resistance. J Bone Miner Res 2005;20:809–16. [15] Leichter I, Margulies JY, Weinreb A, Mizrahi J, Robin GC, Conforty B, Makin M, Bloch B. The relationship between bone density, mineral content, and mechanical strength in the femoral neck. Clin Orthop Relat Res 1982;163:272–81. [16] Nagamine R, Hanada Y, Masakazu K, Fukumoto S, Shuto T, Nakashima Y, Hirata G, Katayama A, Iwamoto Y. Quantification of bone volume on radiographs using NIH image. Mod Rheumatol 2000;10:220–5. [17] Tabensky AD, Williams J, DeLuca V, Briganti E, Seeman E. Bone mass, areal, and volumetric bone density are equally accurate, sensitive, and specific surrogates of the breaking strength of the vertebral body: an in vitro study. J Bone Miner Res 1996;11:1981–8.
Contents lists available at SciVerse ScienceDirect
Journal of Equine Veterinary Science journal homepage: www.j-evs.com
Original Research
Radiographic Calibration for Analysis of Bone Mineral Density of the Equine Third Metacarpal Bone Amber J. Bowen MS a, *, Mathew A. Burd MS, DVM a, John J. Craig PhD b, Monique Craig b a b
Department of Animal Science, California Polytechnic State University, San Luis Obispo, CA EponaTech Inc., Creston, CA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 January 2013 Received in revised form 16 April 2013 Accepted 24 April 2013 Available online 22 June 2013
Digital radiography represents the primary diagnostic tool the veterinarian uses to diagnose skeletal injuries in the horse. Advances in digital radiography have provided the veterinarian with opportunities to make simple radiographic assessments from calibrated digital radiographs such as dimensional analyses; however, more complex variables such as radiographic opacity have yet to be standardized. Therefore, we investigated the quantification of bone mineral density (BMD) via computed radiographic absorptiometry at various radiographic exposure intensities (kV), times (sec), and milliamps (mA) in the third metacarpal in the horse. By developing a brightness/darkness index (BDI), the grayscale of radiographs, calibrated with an aluminum (Al) marker of various known thicknesses and uniform densities, can be compared to the average BMD of a region of interest at various radiographic exposures. Al BDI was a significant predictor of bone BDI (r2 ¼ 0.960, P < .001) and BMD (r2 ¼ 0.971, P < .001). This method of calibration can be used for quantitative noninvasive bone mineral analysis and allows direct comparison of radiographs taken under different exposure settings. Ó 2013 Elsevier Inc. All rights reserved.
Keywords: Radiographic absorptiometry Digital Bone mineral Density
1. Introduction Dual energy x-ray absorptiometry (DEXA) is the most precise, accurate, and widely accepted method of measuring bone mineral density (BMD) for assessment of skeletal integrity and fracture risk in humans [1]. In horses, DEXA analyses show BMD, measured in g/cm2, to be a valid predictor of bone strength [2], exercise-induced bone remodeling in young horses [3-5], and risk for fracture development [6-8]. Unfortunately, the generalized use of DEXA has limited use for equine practitioners as it is expensive, time consuming, and not portable. Radiography and particularly digital and computed radiography represent the primary diagnostic tools the veterinarian uses to diagnose skeletal injuries in the horse. Radiographic * Corresponding author at: Amber J. Bowen, MS, University of California, Davis, School of Veterinary Medicine, Davis, CA 95616. E-mail address: [email protected] (A.J. Bowen). 0737-0806/$ – see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jevs.2013.04.016
absorptiometry and computed digital absorptiometry (CDA) using a single-energy X-ray system has been shown to accurately assess BMD in humans [9,10] and horses [11]. Inclusion of aluminum (Al) markers of various known thicknesses and uniform densities, the grayscale of a region of interest (ROI), termed the brightness/darkness index (BDI), may be calibrated from image to image. The BDI of bone measured from a radiographic ROI can then be related to average BMD, dependent on thickness. However, opacity determined from radiographs taken with different radiographic techniques cannot be directly compared. Calibration of radiographic opacity with various radiographic exposure intensities (kV), times (sec), and milliamps (mA) would allow for quantification of BMD, assuming that the area measured includes minimal soft tissue and has consistent thickness and attenuation properties. Digital radiography has the advantage of being portable, relatively inexpensive, and practical in a field setting. By using CDA techniques, digital radiography may be used to
1132
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135
determine BMD. Therefore, our objective was to develop a method to calibrate radiographic images such that BMD determined from radiographs taken with different techniques could be accurately compared. By calibrating the BDI of digital radiographs, we formed a standardized relationship allowing comparison of computer-determined BDI values with physical properties of BMD regardless of radiographic settings of exposure intensity (kV) and time (sec). This would allow us to account for changes in the opacity of radiographs due to changes in exposure, affording standardization of any radiograph. By including an Al wedge of known density and thickness (g/cm2) in the radiograph, BMD can be determined from BDI by creating a standard curve. Next, BMD can be correlated with BDI by creating a bone step-wedge of known density and thickness. We believe that calibration of digital radiographs to determine BMD from an Al wedge standard will provide a tool for quantification of BMD from common digital radiographs, regardless of radiographic technique. 2. Materials and Methods Radiographs of a single third metacarpal equine cadaver bone and Al wedge taken at various exposures were used to determine the relationship between BDI and radiographic exposure for both bone and Al. Each radiograph contained an Auto-Scaler,1 an Al wedge and cadaver bone. The Al wedge was simply machined from type 6061 Al and was 20.36 cm in length and 0.2-31.8 mm in thickness, with a constant density of 2.70 g/cm3. The Auto-Scaler contained metal markers that standardized the dimensions of the radiographic image when imported into the software program Metron-DVM.1 All radiographs were taken with a portable X-ray system2 and X-ray sensor.3 All radiographs were taken at 15 mA, the only setting available for this system. The cadaver bone was positioned perpendicular to the ground in the center of the line of exposure with the Al wedge and Auto-Scaler positioned on either side. The bone and X-ray system was placed to produce a standard dorsal palmar radiograph. The focal distance was 26 cm, with the plane of interest containing the cadaver bone, Al wedge, and Auto-Scaler positioned against the face of the sensor. Radiographs were taken at all available combinations of preset exposure intensities ranging from 55-80 kV in 5-kV intervals and exposure times from 0.02-0.14 seconds every 0.02 second. The BDI of the Al wedge and cadaver bone was measured on each radiographic image (Fig. 1). To determine BDI, a unitless value was assigned to each 16-bit pixel on a grayscale from 0-65,500, with zero being completely black and 65,500 being completely white. A predetermined ROI on either bone or Al was generated by forming a rectangle using Metron-DVM software.1 BDI was determined by averaging the grayscale value of each pixel in the ROI. In determining BDI, we did not include background pixels. BDI of the Al wedge was measured by creating a rectangle
1 2 3
EponaTech Inc., Creston, CA. MinXray HF80, Northbrook, IL. Thales FS23, Vetel Diagnostics, San Luis Obispo, CA.
Fig. 1. Dorsal palmar radiograph of third metacarpal cadaver bone with Auto-Scaler (left) and Al wedge (right), taken at 60 kV, 0.08 seconds, and 15 mA. ROIs (red) were selected for measuring bone BDI at the nutrient foramen and Al BDI from an Al wedge.
encompassing the entire wedge. BDI of the cadaver bone was characterized by an ROI with the width larger than the cross-sectional width of the bone perpendicular to the sagittal plane and the height equal to the diameter of the nutrient foramen. To ensure change in BDI was due solely to effects of radiographic exposure, the same ROI was used in each radiographic image. BDI of the Al wedge was compared to BDI of the cadaver bone in the same radiographic image at various X-ray intensities and exposure times to ensure the relationship between BDI and radiographic exposure was similar for both materials. BMD was correlated with BDI by measuring BDI from a radiograph of cortical bone of known density and thickness, determined as g/cm2. A bone step-wedge of known BMD was made from cortical bone obtained from a equine cadaver third metacarpal bone in the mid-diaphysis region. The cortical bone was cut into slices using a diamond blade tile saw. The average dimension of each slice was 15.28 3.3 mm square and 4.23 1.0 mm thick. A bone step-wedge was created by stacking each bone slice face-to-face in increasing thickness from one slice to 6 slices thick, consecutively, from step 1-10. The thickness, weight, and volume corresponding to each bone slice was determined
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135
using 6-inch digital calipers4 with 0.01-mm accuracy and a digital balance5 with 0.1-mg accuracy. BMD for each slice of bone was determined from weight (g) multiplied by thickness (cm) and divided by volume (cm3). Aluminum BMD equivalent (Al BMDE) was determined from the bone wedge as density (g/cm3) multiplied by thickness (cm). BDI was determined from the digital radiograph for each step of bone and for the Al wedge (Fig. 2). Digital radiographs of the bone step-wedge and Al wedge on the same image were taken. The bone step-wedge was positioned to intersect the center of the line of exposure, with the Al wedge on one side. The focal distance was 21 cm, with the plane of interest containing the bone step-wedge and Al wedge positioned against the face of the sensor. Radiographs were taken at 60 kV and 0.08 second. Standardized curves were created for bone BDI versus BMD (Fig. 4) and Al BDI versus Al BMDE (Fig. 5). Where Al BDI equaled that of bone, the correlation between bone and Al thickness was determined.
1133
Fig. 2. Bone step-wedge (center) made of 10 cortical bone steps of decreasing thickness (left to right) on a standardized hoof block and Al wedge (right), from a digital radiograph taken at 60 kV, 0.08 seconds, and 15 mA. ROIs (red) were selected for measuring Al BDI (right) and bone BDI (left) from the tenth and thickest step of the bone step-wedge.
2.1. Statistical Analysis Regression analysis was performed using the least squares method. Significance of linear correlation between bone and Al BDI and bone and Al BMDE and bone and Al BDI versus exposure intensity and time using an ANOVA and associated F-test. Statistical significance was set at P < .05. Statistical analysis was performed with R software.6 3. Results and Discussion Veterinary clinicians routinely use radiographs to assess bone health and diagnose abnormalities such as fractures and osteoarthritis, subjectively and qualitatively assessing the opacity or optical density of ROIs. A tool to quantitatively evaluate bone strength or radiographic opacity in ROIs from digital radiographs would be valuable. Therefore, we developed a method based on standardized digital radiographs that affords accurate measurement of BMD. The third metacarpal in the mid-diaphysis region is composed of dense cortical bone and very little trabecular bone and soft tissue and is an ideal choice for investigation of the parameters associated with radiographic standardization. In this study, we calibrated BDI over a range of exposure settings using an Al wedge standard of known density and increasing thickness or an Al BMD equivalent. For each combination of intensity and exposure time, Al BDI showed a linear correlation with bone BDI (R2 ¼ 0.9599, F [1,40] ¼ 957.3, P < .001). As Al BDI increased per unit, there was an increase in bone BDI of 1.12 0.04 (Fig. 3). These results indicate that calibration of BDI with an Al wedge standard curve for a range of exposure settings allows for accurate assessment of BMD from any region of a digital radiograph in units of grams per square centimeter of Al. In this study, we determined BMD from an Al wedge by creating a bone step-wedge of known density and
4 5 6
Fisherbrand, Fisher Scientific, Pittsburg, PA. Accu-224, Fisher Scientific, Pittsburg, PA. R Foundation for Statistical Computing, Vienna, Austria.
thickness. From steps 1-10 of the bone step-wedge, BDI, density, thickness, and BMD were determined (Table 1). Bone BDI was shown to increase linearly with increasing BMD (R2 ¼ 0.9716, F[1,1193] ¼ 40,850, P < .001) (Fig. 4). Similarly, Al BDI increased with increasing Al BMDE, ranging from a BDI of 0-53,810 with a constant density of 2.70 g/cm3. Aluminum thickness increased linearly along the length of the Al wedge from 0-4.26 cm (R2 ¼ 0.9973, F [1,631] ¼ 232,800, P < .001) (Fig. 5). This method permits accurate evaluation of BDI and, therefore, BMD from an Al wedge standard independent of the exposure setting. The quantification of BMD via measurement of BDI from standardized digital radiographs allows for the opacities of radiographs to be compared when taken under different
Fig. 3. Bone versus Al brightness/darkness index (BDI) with linear regression, determined from a digital radiograph of a third metacarpal cadaver bone and Al wedge taken at all available combinations of exposure intensity (55-80 kV) and exposure time (0.02-0.14 seconds).
1134
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135 Table 1 BDI, thickness, density, and BMD determined for 10 steps of the cortical bone step-wedge Step
Mean Bone BDI SD*
1 2 3 4 5 6 7 8 9 10
3,648 4,591 15,962 17,187 22,964 27,536 33,969 46,201 47,870 54,151
723 757 129 268 133 280 127 115 952 691
Thickness (cm)
Density (g/cm3)
BMD (g/cm2)
0.271 0.400 0.754 0.929 1.047 1.325 1.601 2.140 2.498 3.004
1.76 2.02 1.92 1.86 1.95 1.90 1.87 1.96 1.81 1.87
0.48 0.81 1.45 1.79 2.04 2.50 3.00 4.12 4.53 5.61
BDI, brightness/darkness index; BMD, bone mineral density. * Determined from a digital radiograph taken at 60 kV, 0.08 seconds, and 15 mA.
4. Conclusions Fig. 4. Bone brightness/darkness index (BDI) versus bone mineral density (BMD) with linear regression determined from a radiograph of the cortical bone step-wedge.
circumstances. This may have practical applications as BMD is dependent only on thickness and density of the material; therefore, a given area of bone with lower BMD than another area of bone of similar thickness and soft tissue thickness can be concluded to have a lower density. Estimation of bone cross-sectional area or thickness may be a valuable adjunct to this technique. Effect of beam angle, offset, gain, and digital image processing may affect the measurement of BMD and need to be investigated further for assessment of BMD to be clinically applicable [12]. Other imaging modalities available to the equine veterinarian may also accurately measure BMD, such as standard radiography and computed radiography, which provides correct optical density on film independent of the incident radiation exposure [13].
It has been suggested that bone fractures in racehorses are caused by areas of weakened bone or fatigue fractures that occur during training [8,14]. Furthermore, stress fractures induced by exercise are well correlated with decreased BMD in humans, and this technique has been shown to be an accurate predictor of BMD in the horse, using single-energy X-ray systems [11]. This technique may be an asset in the study of factors which influence changes in bone density, in determination of possible risk cases related to increases or decreases in bone density and may find application in the study of osteopenia and other bone diseases of horses. CDA measurement of BMD provides an inexpensive, noninvasive and accurate method of estimating BMD of the third metacarpal in the horse [15,16,17].
Acknowledgement Financial support was provided by the US Department of the Navy, Office of Naval Research, award N00014-08-11209, and by the Agriculture Research Initiative of California State University, under Award # SC350019-09-01.
References
Fig. 5. Al brightness/darkness index (BDI) versus Al bone mineral density equivalents (Al BMDE) with linear regression, determined from a radiograph taken at 66 kV, 0.08 seconds, and 15 mA.
[1] World Health Organization Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Geneva, Switzerland: WHO; 1994. [2] Nielsen D, McEvoy FJ, Madsen MT, Jensen JB, Svalastoga E. Relationship between bone strength and dual-energy x-ray absorptiometry measurements in pigs. J Anim Sci 2007;85:667–72. [3] Jeffcott LB, Buckingham SH, McCarthey RN. Noninvasive measurement of bone quality in horses and changes associated with exercise. Equine Vet J Suppl 1988;6:71–9. [4] Raub RH, Jackson SG, Baker JP. The effect of exercise on bone growth and development of weanling horses. J Anim Sci 1989;67:2508–14. [5] Riggs CM, Boyde A. Effect of exercise on bone density in distal regions of the equine third metacarpal bone in 2-year-old thoroughbreds. Equine Vet J Suppl 1999;30:555–60. [6] Bell GH, Dunbar O, Beck JS, Gibb A. Variations in strength of vertebrae with age and their relation to osteoporosis. Calcif Tissue Res 1967;1:75–86. [7] Järvinen TL, Sievänen H, Kannus P, Järvinen M. Dual-energy X-ray absorptiometry in predicting mechanical characteristics of rat femur. Bone 1998;22:551–8.
A.J. Bowen et al. / Journal of Equine Veterinary Science 33 (2013) 1131–1135 [8] Nunamaker DM, Butterweck DM, Provost MT. Fatigue fractures in thoroughbred racehorses: relationships with age, peak bone strain, and training. J Orthop Res 1990;8:604–11. [9] Bouxsein ML, Michaeli DA, Plass DB, Schick A, Melton ME. precision and accuracy of computed digital absorptiometry for assessment of bone density of the hand. Osteoporos Int 1997;7:444–9. [10] Nolla JM, Gomez-Vaquero C, Fiter J, Roig-Escofet D. Computed digital absorptiometry of the hand: screening method of bone loss in postmenopausal women with RA. Ann Rheum Dis 2000; 59:490. [11] Vaccaro C, Busetto R, Bernardini D, Anselmi C, Zotti A. Accuracy and precision of computer-assisted analysis of bone density via conventional and digital radiography in relation to dual-energy xray absorptiometry. Am J Vet Res 2012;73:381–4. [12] Secombe CJ, Firth EC, Perkins NR, Anderson BH. Pathophysiology and diagnosis of third carpal bone disease in horses: a review. N Z Vet J 2002;50:2–8.
1135
[13] Seibert JA, Shelton DK, Moore EH. Computed radiography X-ray exposure trends. Acad Radiol 1996;3:313–8. [14] Warden SJ, Hurst JA, Sanders MS, Turner CH, Burr DB, Li J. Bone adaptation to a mechanical loading program significantly increases skeletal fatigue resistance. J Bone Miner Res 2005;20:809–16. [15] Leichter I, Margulies JY, Weinreb A, Mizrahi J, Robin GC, Conforty B, Makin M, Bloch B. The relationship between bone density, mineral content, and mechanical strength in the femoral neck. Clin Orthop Relat Res 1982;163:272–81. [16] Nagamine R, Hanada Y, Masakazu K, Fukumoto S, Shuto T, Nakashima Y, Hirata G, Katayama A, Iwamoto Y. Quantification of bone volume on radiographs using NIH image. Mod Rheumatol 2000;10:220–5. [17] Tabensky AD, Williams J, DeLuca V, Briganti E, Seeman E. Bone mass, areal, and volumetric bone density are equally accurate, sensitive, and specific surrogates of the breaking strength of the vertebral body: an in vitro study. J Bone Miner Res 1996;11:1981–8.