Evaluation of the Potential Clinical Application of Low-Intensity Ultrasound Stimulation for Preventing Osteoporotic Bone Fracture

Transcription

1 Annals of Biomedical Engineering, Vol. 38, No. 7, July 21 (Ó 21) pp DOI: 1.17/s Evaluation of the Potential Clinical Application of Low-Intensity Ultrasound Stimulation for Preventing Osteoporotic Bone Fracture DAE GON WOO, 1 CHANG-YONG KO, 1 HAN SUNG KIM, 1 JONG BUM SEO, 1 and DOHYUNG LIM 2 1 Department of Biomedical Engineering, Yonsei University, Wonju, Gangwon 22-71, Republic of Korea; and 2 Gerontechnology Center, Korea Institute of Industrial Technology, 35-3, Hongcheon, Ipjang, Cheonan, Chungnam , Republic of Korea (Received 2 October 29; accepted 22 February 21; published online 5 March 21) Associate Editor Eiji Tanaka oversaw the review of this article. Abstract This study evaluated the possible clinical application of low-intensity ultrasound (LI) stimulation for preventing osteoporotic bone fracture. Eight virgin 14-week-old ICR mice (weight 24. ±.7 g) were ovariectomized to induce osteoporosis. The right hind limbs ( limbs) were stimulated with LI, whereas the left hind limbs ( limbs) were not stimulated. LI was applied for 2 min a day, 5 days a week over a 6-week period using the following parameters: 1.5 MHz frequency, 1. khz pulse repetition, 3 mw/cm 2 intensity, and 2 ls pulse length. The effective structural modulus increased significantly (p <.5) in the limbs over time with the increased bone quantity, whereas that in limbs remained statistically constant (p >.5). In addition, the elastic modulus in the limbs was generally enhanced by an increased bone quality, compared with the limbs. Therefore, LI stimulation may effectively reduce the risk of osteoporotic bone fracture by increasing the mechanical characteristics of bone via improvements in both the effective structural and elastic modulus of the osteoporotic bone. In conclusion, LI may potentially prove very effective clinically for preventing osteoporotic bone fractures. Keywords Low-intensity ultrasound stimulation, Osteoporosis, Preventing osteoporotic bone fracture, Possible clinical application, Improved mechanical characteristics. INTRODUCTION Osteoporosis is a systemic skeletal disease caused by low bone mass and the microstructural deterioration of bone, resulting in increased bone fragility and the risk of fracture. Osteoporotic bone fractures are an Address correspondence to Dohyung Lim, Gerontechnology Center, Korea Institute of Industrial Technology, 35-3, Hongcheon, Ipjang, Cheonan, Chungnam , Republic of Korea. Electronic mail: dli349@gmail.com Dae Gon Woo and Chang-Yong Ko contributed to this work as First Author /1/7-2438/ Ó 21 Biomedical Engineering Society 2438 important public health concern, causing significant morbidity, mortality, and economic burden in more than 2 million people in the United States and more than 2 million worldwide. 9 It is widely accepted that the risk of osteoporotic bone fracture is generally related to a decrease in the mechanical characteristics of the bone that results from alterations in both the microstructure and the material properties of bone induced by osteoporosis. 3,13 Many studies have examined possible treatments and preventive strategies to deal with the bone loss in osteoporosis. 15 Pharmacological interventions are the mainstay of osteoporosis prevention and treatment. Parathyroid hormone (PTH), one of main pharmacological interventions, increases the concentration of calcium (Ca 2+ ) in the blood by acting on the PTH receptor in bone, stimulating bone formation in the treatment of osteoporosis. Daily self-injections of PTH reduced the risk of new osteoporotic fractures by 6% in postmenopausal women with osteoporosis. 14 This pharmacological intervention is widely accepted in clinical practice and generally improves the bone architecture, reducing the risk of osteoporotic bone fracture. However, pharmacological therapies can have various undesirable side effects (occasional severe nausea and headache), limiting their application for preventing or stabilizing osteoporosis. Therefore, we need to develop alternative therapies capable of overcoming the limitation. Ultrasound () is a potential nonpharmacological intervention for many people with an increased bone fracture risk due to osteoporosis. 27, high-frequency nonaudible acoustic energy that travels in the form of a mechanical wave, can be directed at osteoporotic sites to exert a mechanical stimulus. Recently, several studies have shown that low-intensity ultrasound (LI) with a 2 ls burst of 1.5 or.5 MHz sine

2 Evaluation of the Potential Clinical Application of LI 2439 waves, 1. khz pulse repetition and 3 mw/cm 2 intensity are capable of augmenting bone strength, particularly its irregular geometry. 2,22,26 Moreover, experimental studies have demonstrated the benefits of LI in large-bone fractures and have shown that it increases the bone mineral density at the fracture site in humans or animal models. 1,19,3 Recently, in vivo animal studies have shown that LI improves the mechanical properties of the healing callus in bone fractures associated with osteoporosis using the rabbit and sheep models, 7,16 and enhances bone bridging in rat femoral fracture. 2 Their studies evaluated the application possibility of LI to heal and prevent osteoporotic bone fracture in large animals as well as small animals. Clinical trials also showed accelerated bone-fracture healing in adults with distal radius fractures, osteotomized fibulas, and tibial fractures. 8,11,19 However, despite this indirect evidence suggesting that LI stimulates osteogenesis in bones, some conflicting results have been reported concerning potential applications. In pigeons, LI had no effect on the bone mineral loss associated with wing bandaging. 28 Similarly, Warden et al. 27 suggested that LI was not beneficial in osteoporosis and that intact bone appeared to be less sensitive to LI than were fractured bones and isolated bone cells when ovariectomized (OVX) rodents were used as an osteoporotic model. Carvalho and Cliquet 4 found no significant quantitative differences in calcium and phosphorus between -treated and nontreated bones in the proximal femurs of osteopenic rats. However, further research is required to evaluate whether LI is a possible clinical modality for preventing osteoporotic bone fractures. In the previous study, they researched LI effects at only trabecular bone except for cortical bone. Therefore, we investigated that the mechanical loading with could produce an effective compressive force at the bone surface as well as trabecular core. This study evaluated the possible clinical application of LI in osteoporotic bone-fracture prevention by examining whether LI decreases the risk of osteoporotic bone fracture via improvements in the effective structural and elastic modulus of osteoporotic bone. MATERIALS AND METHODS Animal Preparation Eight virgin 14-week-old ICR mice weighing approximately 24. ±.7 g were housed in individually ventilated and cleaned cages (room temperature 23 ± 2 C, humidity 5 ± 1%). All mice were OVX to induce osteoporosis, and the degree of osteoporosis in each mouse was confirmed morphologically in both hind limbs using in vivo microcomputed tomography (lct, Skyscan 176, Skyscan, Belgium) images with a resolution of 18 lm. All animal treatments in this study were in accordance with the National Institutes of Health (NIH) Guidelines for the care and use of laboratory animals under a protocol approved by the Yonsei University Animal Care and Ethics Committee. Low-Intensity Ultrasound Application The right hind limb of each mouse was stimulated using LI (the limbs) after confirming OVX-induced osteoporotic bone loss, whereas the left hind limb was not stimulated and served as an internal control (the limbs) (Fig. 1). LI was applied to the proximal tibia for 2 min/day, 5 days/week over a 6-week period (from 3 to 9 weeks after OVX) at a 1.5 MHz frequency, 1. khz pulse repetition on frequency, 3 mw/cm 2 intensity, and 2 ls pulse width. To apply LI, the mice were immobilized in a customized restrainer, and both tibiae were submerged in warm water in a customized tub (Fig. 1). As the mechanical waves are transmitted as a relatively focused beam, the LI was targeted over the specific regions of the right hind limb at the relatively close distance from the skin surface, compared to the left hind limb, where it exerted a local mechanical stimulus. In Vivo lct and Determination of the Volume of Interest for Analysis Both hind limbs of each mouse were scanned with the tibia placed along the longitudinal axis of the scanning area for in vivo lct with a resolution of 18 lm to acquire two-dimensional (2D) images and measure bone volume (BV, mm 3 ) at week before OVX and at 3, 6, and 9 weeks after OVX (Fig. 2). The cross-sectional geometric positioning enabled anatomical regions of interest (ROI) including both trabecular and cortical bone to be scanned simultaneously. The volume of interest (VOI) corresponding to the proximal tibia was selected from a 1.8-mm length of bone, located.54 mm below the growth plate, based on a three-dimensional (3D) reconstruction model of bone rendered from each scanned ROI (2D images) (Fig. 2). 12 Analysis of Effective Structural Modulus From the scanned 2D images and reconstruction models, 3D finite element (FE) models were generated to quantify the effect of LI on stiffness related to bone structure. All lct images were captured using BIONIX (CANTIBio, Suwon, Korea) and converted into 3D

3 244 WOO et al. FIGURE 1. Configuration of the LI system. (a) The right hind limb is stimulated using the ultrasound transducer attached to the wall of the water bath; (b) parameters used for ultrasound stimulation. FIGURE 2. The in vivo lct system and the range of the VOI. BV [mm 3 ] Beginning of Time [Week] FIGURE 3. Changes in BV (mm 3 ) measured in both the and limbs using the direct method available with in vivo lct. There was no significant difference at each time point between the and limbs (p >.5) and the statistical decreases were observed from only between week (beginning of OVX) and 9 weeks (after OVX) in each group (*p <.5). tetrahedral FE models with 18 lm mesh size using a mass-compensated thresholding technique. 24 In this technique, a unique threshold value was selected for each of the images matching the BV (mm 3 ) of the * TABLE 1. Material properties used in the FE models. Young s modulus Poisson s ratio Value 12.5 GPa.3 specimen obtained using in vivo lct (Fig. 3). The BV was measured using a direct method available with in vivo lct. Table 1 shows the values we used for Young s modulus and Poisson s ratio, 1,29 generally referred to mouse models. The trabecular material in this FE analysis was assumed to be isotropic and perfectly elastic. Displacement boundary conditions were applied to the specimens to simulate a uniaxial compression test to quantify the effect of the LI on bone stiffness. For the analysis of lfe models, the effective structural modulus (stress/strain) was measured by applying a compressive displacement of an uniaxial.5% strain in the linear region of the stress strain curve (Fig. 4). 29 All FE analyses were performed using the commercial FE software package ABAQ 6.4 (HKS, Pawtucket, RI, A).

4 Evaluation of the Potential Clinical Application of LI 2441 FIGURE 4. Application of compressive displacement to the FE model to calculate effective structural modulus. Compressive displacement and boundary conditions were applied to the top and bottom bone surfaces, respectively. Analysis of Elastic Modulus The elastic modulus was determined in Hounsfield units, which are related to bone quality, for each element calculated by Eq. (1) suggested by Rho et al. 17 using Mimics 12.3 software (Materialise, Leuven, Belgium). The elastic modulus distributions for the bones in the and limbs were then used to quantify the degree of improvement in bone quality achieved using LI. E ¼ 5:54 q 326 ð1þ ðq ¼ :916 HU þ 114Þ; where q, HU, and E are the density, Hounsfield unit, and elastic modulus, respectively. Statistical Analysis All statistical probabilities were examined using either one-way analysis of variance (ANOVA) with Tukey s post hoc multiple comparisons or a paired t-test. One-way ANOVA was used to analyze the degree of alteration in bone structural modulus (bone quantity) and bone elastic modulus (bone quality) over time in the or limbs. A paired t-test was used to identify differences in bone structural modulus or bone elastic modulus between the and limbs. All comparisons were performed with the Statistical Package for the Social Sciences (SPSS, Chicago, IL, A). The significance level (p) for the statistical analysis was set at.5. RESULTS After analyzing the efficiency of OVX in the mice, the effects of LI on the osteoporotic bones were studied. As expected, the OVX surgery resulted in significant bone changes (Fig. 5). It was identified from the 3D reconstructed models (Fig. 5) that trabecular bone was disappeared after OVX surgery. In addition, the BV (mm 3 ) measured using the direct method with in vivo lct was slightly decreased in the limb compared with that in the limb over time (Fig. 3). However, there was no significant difference at each time point between the and limbs (p >.5) and the statistical decreases were observed from only between week (beginning of OVX) and 9 weeks (after OVX) in each group (p <.5). Figure 6 shows the mechanical effect of LI on bone loss following OVX. Before stimulation, the effective structural modulus in both the and limbs decreased to 81 and 94%, respectively, from week to week 3 with the development of osteoporosis induced by ovarian hormone deficiency. With LI stimulation beginning at week 3, the effective structural modulus in the limbs increased significantly (p <.5) from 4271 ± 975 to 5559 ± 694 MPa at 3 and 9 weeks, respectively, whereas the effective structural modulus in the limbs remained constant (p >.5). The difference in effective structural modulus between the and limbs was also compared at each time point. Significant increase in the effective structural modulus was observed in limbs compared to the limbs at week 9, whereas there was no significant difference between both limbs at weak 3 (the beginning of irradiation) and week 6. The elastic modulus converted from the Hounsfield units, obtained from voxels in lct images, was higher in the limbs compared with the limbs (Fig. 7). The histogram analyses showed that the elastic modulus in the limbs was significantly greater than that in the limbs at 9 weeks (p <.5), and more than 113% and 165% of the highest values (346 and 3723 MPa) were attributed to the effects of LI (Fig. 7a). Figure 7b shows the structural changes of typical FE models, calculated using Mimics 12.3 software, based on in vivo lct measurement. The distribution of high values increased gradually in the limbs over time, whereas those in limbs generally decreased. The structural improvement achieved with LI was verified in the limbs compared to the limbs. DISCSION AND CLIONS Our experiment is the first controlled study to investigate the potential of LI for treating osteoporosis from a biomedical perspective in mouse models. Mouse models, useful for investigating chronic diseases that currently plague public health, 5,6,29 have been used for the study of the correlation between metabolism and osteoporosis. In this study, LI stimulation had positive effects (increased effective structural and elastic modulus)

5 2442 WOO et al. FIGURE 5. Identification of osteoporosis from 2D lct images and 3D reconstructed models. (a) Representative 2D lct image, (b) representative 3D reconstructed model. Trabecular bone disappeared after OVX. Effective structural modulus [MPa] Beginning of * * Time [Week] FIGURE 6. Changes in the effective structural modulus in the and limbs over time (*p <.5). on osteoporotic bone, as show quantitatively from in vivo lct measurement and qualitatively in the 3D reconstructed bone models. LI stimulation may effectively decrease the risk of osteoporotic bone fracture by increasing the mechanical characteristics of osteoporotic bone via improvements in both its effective structural and elastic modulus. The results imply that LI would be very effective clinically in preventing osteoporotic bone fracture. In physical therapy, LI stimulation is used extensively as a therapeutic tool, although a few studies have reported that has no clinical effect on bone loss or bone fragility. 27 In the cases with no clinical effect on bone loss or bone fragility, the mechanical loading produced by may not be conducive to bone adaptation. Studies of have suggested that the adaptation of bone to mechanical stimuli was dependent on both the bone strain magnitude and the loading frequency. To determine the relationship between peak load and peak strain in the rat model, a single rosette strain gauge could be bonded to the hind limbs and strain ranges might be selected from low (76 le) to high peak (7 le). 2

6 Evaluation of the Potential Clinical Application of LI 2443 (a) Volume [mm 3 ] (b) Color * Elastic Modulus [MPa] * Elastic modulus [MPa] Limbs * Limbs week week 3 week 6 Week 9 FIGURE 7. Changes in the elastic modulus in the and limbs. (a) Relationship between the elastic modulus and volume (x-axis, elastic modulus; y-axis, volume) (*p <.5), (b) representative distributions of the elastic modulus in the and limbs over time. Therefore, both of these factors need to be controlled so that the mechanical loading with produces an effective compressive force at the bone surface. Insufficient bone strain or excessive loading with may not stimulate remodeling-driven bone adaptation. 23 Many previous studies investigated the therapeutic potential of LI stimulation on osteoporosis for trabecular bone only. As a number of parameters must be controlled in treatment, the bone strain magnitude and loading frequency of may be neutralized in trabecular bone. However, because we stimulated the cortical bone surface directly with LI, we hypothesize that LI has greater effects on cortical bone than trabecular bone. Although the bone strain magnitude and loading frequency may not have been optimized in terms of the treatment of osteoporosis in our study, these parameters may have a justifiable effect on the cortical and trabecular bone of the proximal tibia, based on the our results of the beneficial effect of LI on bone. Also, as referring to the fundamental rules that govern bone adaptation, 23 a short duration of mechanical loading is necessarily applied in the present study to initiate an adaptive response. Elevated duration of skeletal loading does not yield proportional increases in bone mass. As loading duration is enlarged, the bone formation response tends to saturate. This phenomenon of diminishing returns is best demonstrated in the studies by Rubin and Lanyon 18 and Umemura et al. 25 The improved effective structural and elastic modulus in the proximal tibia would verify our experimental hypothesis. Our findings suggest that LI may reduce risk of osteoporotic bone fracture by improving bone

7 2444 WOO et al. (a) Trabecular number (%) Trabecular thickness (mm) (c) Trabecular number (%) (b) Trabecular number (%) Trabecular thickness (mm) Trabecular thickness (mm) FIGURE 8. Representative changes of trabecular thickness distribution in the and limbs [x-axis, trabecular thickness (mm); y-axis, trabecular numbers (%)]. (a) 3 weeks (beginning of ), (b) 6 weeks, and (c) 9 weeks. stiffness. As shown in Fig. 3, OVX clearly induced osteoporosis in a mature mouse model. There was no significant difference at each time point between the and limbs in BV (%). However, we confirmed the distributed changes of trabecular thickness shifted from the lower values to the higher values in the limbs compared with the limbs at 9 weeks, whereas there was no meaningful difference between the and limbs at 3 and 6 weeks (Fig. 8). The increase of trabecular numbers (%) in the high extent of trabecular thickness distribution (Fig. 8c) could explain the significant difference between the and limbs in the effective structural modulus at 9 weeks (Fig. 6). The LI irradiation reduced the bone porosity, resulting in an increase in the morphological and mechanical characteristics of osteoporotic bones. Additionally, the improved bone mineral density was expected by the benefits of LI. Bone mineral density is currently the accepted indicator of bone strength and fracture risk. In the present study, the trabecular numbers (%) per trabecular thickness might reflect changes of the BMD in the limbs from a biomechanical point of view (Fig. 8), although we did not measure this. We confirmed that structural changes were highly related with an increase in the effective structural modulus in limbs, compared with limbs. Shiraishi et al. 21 also reported that mechanical load values were significantly correlated with the BMD as well as structural changes. Based on these results, we conclude that the trabecular numbers (%) per trabecular thickness closely associated with changes in the bone stiffness might be used as indicators of BMD (Fig. 8). The simulated compression test also showed that the ovarian hormone deficiency during weeks 3 resulted in a pronounced decrease in the mechanical characteristics of the hind limbs of the mature mice (Fig. 6). The subsequent increased effective structural modulus (bone quantity) in the limbs could be related the bone gain due to LI. Furthermore, the improved elastic modulus (bone quality) achieved with LI suggests that new bone formed in the limbs, but not in the limbs. In our models, we observed a temporary increase in effective structural modulus in the limbs (weeks 3 6; Fig. 6). It was impossible to perfectly avoid mechanical wave hitting the limbs. We also postulate that the functional adaptation caused by LI stimulation in the limbs influenced the bone stiffness in the limbs. If the functional adaptation to mechanical loading is regulated neuronally, then our results were associated with neuronal signaling. In fact, Sample et al. 2 concluded that functional adaptation to loading of a single bone in young rapidly growing rats was regulated neuronally during skeletal modeling and involved multiple bones. Using the well-established minimally invasive rat ulna

8 Evaluation of the Potential Clinical Application of LI 2445 end-loading model, they showed that responses to loading of the right ulna were occurred in multiple thoracic limb bones. Their study suggested the existence of a crosstalk mechanism between limbs that facilitates functional adaptation of the skeleton to mechanical loading. Ultimately, the skeletal responses to mechanical stimulation are more complex than previously thought and include adaptive responses in distant bones that were not loaded. In summary, the application of LI to osteoporotic bones preserved the bone microarchitecture of the limbs compared to that in the limbs. LI could be an effective treatment for improving the effective structural and elastic modulus of bone, as we demonstrated quantitatively and qualitatively. In the further study, our findings require confirmation using concepts of nanomechanics and molecular biology to perform the detailed validation. A possible explanation of the determinant factors (such as appropriate bone strain magnitude and loading frequency of LI energy) to bone adaptation needs to be demonstrated in the biophysical mechanism of action of LI. Therefore, we are currently investigating these in an ongoing study, the results of which may increase confidence in the results presented here. It is also important to confirm whether the responses to LI of the hind limbs are regulated neuronally, as neuronal signaling is an important regulator of LI-induced bone formation. Given the predictive ability of OVX mouse models for the preclinical evaluation of interventions aimed at preventing the bone loss and bone fragility associated with osteoporosis in humans, our findings suggest that LI has potential clinical benefits for preventing osteoporotic fractures. ACKNOWLEDGMENTS This study was supported by a grant of the Korea Healthcare technology R&D Project, Ministry for Health, Welfare &Family Affairs, Republic of Korea (A892). REFERENCES 1 Aynaci, O., et al. The effect of ultrasound on the healing of muscle-pediculated bone graft in spinal fusion. Spine 27: , Azuma, Y., et al. Low-intensity pulsed ultrasound accelerates rat femoral fracture healing by acting on the various cellular reactions in the fracture callus. J. Bone Miner. Res. 16:671 68, Briggs, A. M., et al. The effect of osteoporotic vertebral fracture on predicted spinal loads in vivo. Eur. Spine J. 15: , Carvalho, D. C., and A. Cliquet Jr. The action of lowintensity pulsed ultrasound in bones of osteopenic rats. Artif. Org. 28: , Elliott, D. M., and J. J. Saver. Young investigator award winner: validation of the mouse and rat disc as mechanical models of the human lumbar disc. Spine 29: , Gerasimenko, Y., et al. Epidural stimulation: comparison of the spinal circuits that generate and control locomotion in rats, cats and humans. Exp. Neurol. 29(2): , Hantes, M. E., et al. Low-intensity transosseous ultrasound accelerates osteotomy healing in a sheep fracture model. J. Bone Joint Surg. 86: , Heckman, J. D., et al. Acceleration of tibial fracturehealing by non-invasive, low-intensity pulsed ultrasound. J. Bone Joint Surg. 76:26 34, Kaufman, J. J., et al. Ultrasound simulation in bone. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 55: , Kinney, J. H., et al. Three-dimensional morphometry of the L6 vertebra in the ovariectomized rat model of osteoporosis: biomechanical implications. J. Bone Miner. Res. 15: , Kristiansen, T. K., et al. Accelerated healing of distal radial fractures with the use of specific, low-intensity ultrasound. A multicenter, prospective, randomized, double-blind, placebo-controlled study. J. Bone Joint Surg. 79: , Laib, A., et al. 3D micro-computed tomography of trabecular and cortical bone architecture with application to a rat model of immobilisation osteoporosis. Med. Biol. Eng. Comput. 38: , Mosekilde, L. Consequences of the remodelling process for vertebral trabecular bone structure: a scanning electron microscopy study (uncoupling of unloaded structures). Bone Miner. 1:13 35, Neer, R. M., et al. Effect of parathyroid hormone (1 34) on fractures and bone mineral density in postmenopausal women with osteoporosis. N. Engl. J. Med. 344: , Phillips, F. M., et al. In vivo BMP-7 (OP-1) enhancement of osteoporotic vertebral bodies in an ovine model. Spine J. 6:5 56, Pilla, A. A., et al. Non-invasive low-intensity pulsed ultrasound accelerates bone healing in the rabbit. J. Orthop. Trauma 4: , Rho, J. Y., et al. Relations of mechanical properties to density and CT numbers in human bone. Med. Eng. Phys. 17: , Rubin, C. T., and L. E. Lanyon. Regulation of bone formation by applied dynamic loads. J. Bone Joint Surg. 66-A:397 42, Rutten, S., et al. Low-intensity pulsed ultrasound increases bone volume, osteoid thickness and mineral apposition rate in the area of fracture healing in patients with a delayed union of the osteotomized fibula. Bone 43: , Sample, S. J., et al. Functional adaptation to loading of a single bone is neuronally regulated and involves multiple bones. J. Bone Miner. Res. 23: , Shiraishi, A., et al. A comparison of alfacalcidol and menatetrenone for the treatment of bone loss in an

9 2446 WOO et al. ovariectomized rat model of osteoporosis. Calcif. Tissue Int. 71(1):69 79, Siffert, R. S., and J. J. Kaufman. Ultrasonic bone assessment: the time has come. Bone 4:5 8, Turner, C. H. Three rules for bone adaptation to mechanical stimuli. Bone 23:399 47, Ulrich, D., et al. Finite element analysis of trabecular bone structure: a comparison of image-based meshing techniques. J. Biomech. 31: , Umemura, Y., et al. Five jumps per day increase bone mass and breaking force in rats. J. Bone Miner. Res. 12: , Wang, S. J., et al. Low intensity ultrasound treatment increases strength in a rat femoral fracture model. J. Orthop. Res. 12:4 47, Warden, S. J., et al. Skeletal effects of low-intensity pulsed ultrasound on the ovariectomized rodent. Ultrasound Med. Biol. 27: , Wimsatt, J., et al. Ultrasound therapy for the prevention and correction of contractures and bone mineral loss associated with wing bandaging in the domestic pigeon (Columba livia). J. Zoo Wildlife Med. 31:19 195, Woo, D. G., et al. Relationship between nutrition factors and osteopenia: Effects of experimental diets on immature bone quality. J. Biomech. 42: , Yang, K. H., and S. J. Park. Stimulation of fracture healing in a canine ulna full-defect model by low-intensity pulsed ultrasound. Yonsei Med. J. 42:53 58, 21.