Efficacy of low-intensity pulsed ultrasound in the prevention of osteoporosis following spinal cord injury

Bone Vol. 29, No. 5 November 2001:431– 436

Efficacy of Low-intensity Pulsed Ultrasound in the Prevention of Osteoporosis Following Spinal Cord Injury S. J. WARDEN,1 K. L. BENNELL,1 B. MATTHEWS,1 D. J. BROWN,2 J. M. McMEEKEN,2 and J. D. WARK3,4 1

Center for Sports Medicine Research and Education, School of Physiotherapy, University of Melbourne, Parkville, VIC, Australia Victorian Spinal Cord Service, Austin and Repatriation Medical Center, Heidelberg, VIC, Australia 3 Department of Medicine, University of Melbourne, Parkville, VIC, Australia 4 Bone and Mineral Service, Royal Melbourne Hospital, Parkville, VIC, Australia 2

Key Words: Bone; Dual-energy X-ray absorptiometry (DXA); Fracture; Osteoporosis; Quantitative ultrasound (qUS); Ultrasound.

Ultrasound (US), a high-frequency acoustic energy traveling in the form of a mechanical wave, represents a potential site-specific intervention for osteoporosis. Bone is a dynamic tissue that remodels in response to applied mechanical stimuli. As a form of mechanical stimulation, US is anticipated to produce a similar remodeling response. This theory is supported by growing in vitro and in vivo evidence demonstrating an osteogenic effect of pulsed-wave US at low spatialaveraged temporal-averaged intensities. The aim of this study was to investigate whether low-intensity pulsed US could prevent calcaneal osteoporosis in individuals following spinal cord injury (SCI). Fifteen patients with a 1– 6 month history of SCI were recruited. Active US was introduced to one heel for 20 min/day, 5 days/week, over 6 weeks. The contralateral heel was simultaneously treated with inactive US. Patients were blind to which heel was being actively treated. Active US pulsed with a 10 ␮sec burst of 1.0 MHz sine waves repeating at 3.3 kHz. The spatial-averaged temporal-averaged intensity was set at 30 mW/cm2. Bone status was assessed at baseline and following the intervention period by dual-energy X-ray absorptiometry and quantitative US. SCI resulted in significant bone loss. Bone mineral content decreased by 7.5 ⴞ 3.0% in inactive US-treated calcanei (p < 0.001). Broadband US attenuation and speed of sound decreased by 8.5 ⴞ 6.9% (p < 0.001) and 1.5 ⴞ 1.3% (p < 0.001), respectively. There were no differences between active and inactive US-treated calcanei for any skeletal measure (p > 0.05). These findings confirm the negative skeletal impact of SCI, and demonstrate that US at the dose and mode administered was not a beneficial intervention for SCI-induced osteoporosis. This latter finding may primarily relate to the inability of US to effectively penetrate the outer cortex of bone due to its acoustic properties. (Bone 29: 431– 436; 2001) © 2001 by Elsevier Science Inc. All rights reserved.

Introduction Osteoporosis is a common sequelae of spinal cord injury (SCI). Occurring primarily at sublesional cancellous-rich sites, bone mineral content (BMC) can decrease by as much as 70% following SCI.6,13 This bone loss results in increased bone fragility21 and a subsequent increase in the risk for low-trauma fractures.18,36 As these fractures are associated with a high rate of complications,12,16 prevention of the osteoporosis following SCI is important. Ultrasound (US) represents a potential intervention for SCIinduced osteoporosis. US refers to a high-frequency nonaudible acoustic energy that travels in the form of mechanical waves.5 A mechanical wave is one in which energy is transmitted by the movement of particles within the medium through which the wave is traveling.34 As these waves travel as a relatively focused beam (typical effective radiating area ⫽ 5 cm2), US can be directed onto specific regions to exert a local mechanical stimulus. Given the inherent mechanosensitivity of bone,8 it has been hypothesized that US mechanical energy may be an osteogenic stimulus.20,25 This theory is supported by a growing body of evidence demonstrating the skeletal effects of pulsed-wave US at low (⬍100 mW/cm2) spatial-averaged temporal-averaged intensities. In vitro, low-intensity pulsed US stimulates cellular changes associated with mechanotransduction and bone formation in whole bone. In ST2 cells, US induced mRNA changes similar to the response of these cells to mechanical stretching. Changes included the transient expression of the immediate-early response gene c-fos and elevated mRNA levels for insulin-like growth factor-1, osteocalcin, and bone sialoprotein.25 In MC3T3-E1 cells, US induced a threefold increase in prostaglandin E2 production.19 This was shown to result from the upregulation of cyclooxygenase-2, a known mediator involved in a bone-forming response to mechanical loading.10 In addition to these cellular changes, low-intensity pulsed US in isolated bone rudiments has been shown to stimulate endochondral ossification,26 and collagenous and noncollagenous protein synthesis.28 These in vitro findings suggest that low-intensity pulsed US is an osteogenic stimulus.

Address for correspondence and reprints: K. L. Bennell, Ph.D., Center for Sports Medicine Research and Education, School of Physiotherapy, University of Melbourne, Parkville, VIC 3010, Australia. E-mail: [email protected] © 2001 by Elsevier Science Inc. All rights reserved.

431

8756-3282/01/$20.00 PII S8756-3282(01)00599-3

432

S. J. Warden et al. Ultrasound treatment of SCI-induced osteoporosis

Bone Vol. 29, No. 5 November 2001:431– 436

Table 1. Subject data

Subject Gender Age Height Weight number (M/F) (years.month)a (cm)a (kg)a 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Time since Level of Complete (C)/ Lowest level Circumstances injury spinal cord Quadriplegic (Q)/ incomplete (I) ASIA of motor surrounding spinal (days)a lesion paraplegic (P) lesion scoreb functioning cord injury

M M M M M M M M M M M M M M M

19.0 20.4 16.8 24.2 19.9 40.0 30.7 21.5 26.7 28.8 17.7 18.2 19.6 36.3 19.3

197 170 170 181 184 169 187 175 179 158 158 173 175 180 178

69 60 55 62 67 73 70 75 63 56 60 62 71 66 68

139 153 104 138 145 92 153 46 113 73 81 61 123 137 97

Mean ⫽ SD ⫽

23.9 7.3

175.6 10.2

65.1 6.0

110.3 34.5

C-6 C-5 C-5 C-5 C-5 T-5 C-5 T-10 T-8 T-5 T-8 T-6 C-5 C-6 C-6

Q Q Q Q Q P Q P P P P P Q Q Q

C C C I I C C C C I C C C I C

A A A A A A A A A A B A A A A

C-7 C-5 C-5 C-7 C-5 T-5 C-5 T-10 T-8 T-7 T-8 T-6 C-5 C-7 C-6

Footballc MVA Rugby Snow skiing Footballc MVA MBA MBA Fall from height MBA Fall from height MVA Diving accident MVA MVA

KEY: MBA, motor bike accident; MVA, motor vehicle accident. a At entry to study. b American Spinal Injury Association (ASIA) Impairment Scale22: A, no motor or sensory function preserved below the neurological level; B, sensory, but not motor, function is preserved below the neurological level. c Australian rules football.

Further support for a beneficial skeletal effect of low-intensity pulsed US has been provided by in vivo investigations. In vivo, the most frequent application of low-intensity pulsed US is during fracture repair.38,40 In animal fracture models, such US has been shown to facilitate the rate of endochondral bone formation,7,37 increasing the return of bone mineral density (BMD) at the fracture site.31 This results in a 38% acceleration in mechanical strength return.27 In humans, low-intensity pulsed US has been shown to facilitate the return of BMC during limb-lengthening procedures30 and induce a 30%–38% reduction in the time to union in fresh fractures.15,20,24 When applied to nonunited fractures, the same US has been shown to stimulate union in ⬎85% of cases.11,23 Despite this indirect evidence suggesting that low-intensity pulsed US may stimulate osteogenesis, there has been limited research investigating this potential in intact bone. However, preliminary direct evidence for a beneficial effect does exist. With as little as 4 weeks of treatment, Arai et al.2 found low-intensity pulsed US to not only prevent femoral neck bone loss in five osteoporotic bed-ridden patients, but to increase BMD levels by 8.9%, on average, above baseline levels. BMD on the contralateral nontreated side decreased by 4.0% within the same period. This finding suggests an anabolic effect of low-intensity pulsed US on intact bone. From this, it is apparent that further research is necessary to evaluate whether US should be pursued further as a possible clinical modality for the intervention of osteoporosis. To validate the preliminary case-series findings of Arai et al.,2 the current study investigated US effects using a randomized controlled trial experimental design. The aim was to establish whether US had a preventative effect on calcaneal bone loss, to the extent observed by Arai et al.,2 during a 6 week period of acute bone loss following SCI. This was assessed using standard clinical measures— dual-energy X-ray absorptiometry (DXA) and quantitative US (qUS).

Materials and Methods Study Design A within-subject, double-blind, randomized, placebo-controlled trial was performed. Subjects Fifteen patients with a SCI were recruited from the Victorian Spinal Cord Service at the Royal Talbot Rehabilitation Center (Kew, VIC, Australia). Subjects were recruited on the basis of a 1– 6 month (30 –180 day) history of SCI. The majority of bone loss following SCI occurs within the first 6 months postinjury.4,14 A further inclusion criteria was a grade A or B injury on the American Spinal Injury Association Impairment Scale,22 with the absence of motor control below the neurological level of T-12. This ensured complete paralysis of the lower limb musculature, including the hip flexors. This was necessary to eliminate voluntary loading of the calcanei, which could influence skeletal changes. Subjects were excluded if they were ⬎40 years of age and hence at risk of age-related bone changes, not medically stable as determined by their individual doctor, were unable to travel to the assessment site, were to be discharged from inpatient rehabilitation within the duration of the US intervention period (6 weeks), had a pre-existing bone disease, had currently or previously injured either calcaneum, or were taking pharmacological interventions known to affect bone. Subject data are outlined in Table 1. All subjects provided written informed consent and all procedures were undertaken with prior approval from the Austin and Repatriation Medical Center Human Research Ethics Committee (Heidelberg, VIC, Australia). Ultrasound Intervention US was applied to each subject for 20 min/day, 5 days/week (Monday–Friday) over a consecutive 6 week period. Two active

Bone Vol. 29, No. 5 November 2001:431– 436

US-treatment heads were coupled to the medial and lateral surfaces of one calcaneus using US gel pads (Ultraphonic Focus Conforming Gel Pad, Pharmaceutical Innovations, Inc., Newark, NJ). Treatment heads were centered over semipermanent skin markers placed on the skin during qUS assessment (see later) and were held by a positioning jig. Two inactive US-treatment heads were similarly coupled to the contralateral calcaneus. Patients were blinded to which calcaneus was being actively treated. Active US was produced by two custom-built US units. Both units consisted of an arbitary waveform generator (HP 33120A, Hewlett-Packard, Blackburn, VIC, Australia), a custom-built amplifier (BBA-20-5, Rfdi Industries, Sans Souci, NSW, Australia), and an US treatment head (1459.901, Enraf-Nonius, Delft, The Netherlands). These units were configured to produce pulsed-wave US with a 10 ␮sec burst of 1.0 MHz sine waves repeating at 3.3 kHz. The spatial-averaged temporal-averaged intensity was set at 30 mW/cm2. This intensity was based on that shown to influence osteoblasts in vitro,19,25 augment fracture repair in both animals3,27,33 and humans,11,15,20,23,24,30 and increase BMD in osteoporotic patients.2 Output characterization of each US unit was performed at baseline using hydrophonic scanning. Determined according to the Food and Drug Administration’s performance standards for US devices,9 each unit possessed an effective radiating area of 4.72– 4.85 cm2 and beam nonuniformity ratio of 5.59 –5.81. Weekly checks of US unit performance were performed using a digital oscilloscope (TDS 544A, Tektronix, Inc., Beaverton, OR) and US power meter (UPM-DT-1, Ohmic Instruments, Easton, MD). No variation in US output was detected during the time-course of the investigation. The skeletal effects of low-intensity pulsed US were determined by qUS and DXA. These were performed at entry to the research project (initial) and at the completion of the 6 week intervention period (final). Assessments were performed at the Royal Melbourne Hospital (Parkville, VIC, Australia) with the examiner being blinded to the intervention groups. Quantitative Ultrasound A CUBA Clinical qUS unit (McCue, Hampshire, UK) was used to measure the speed of sound (SOS) and broadband US attenuation (BUA) within each calcaneus. Subjects remained in their wheelchair for testing. Before testing, the skin over the test site was cleaned with alcohol. At the initial measure the foot was placed in the testing chamber and the US transducers positioned against the medial and lateral surfaces of the heel. At the center of each transducer a semipermanent mark was placed on the subject’s skin. This demarcated the site for bone assessment and US intervention, and remained in situ for the duration of the intervention period. For qUS measurement, a liberal layer of coupling gel (Lectron II Conductivity Gel, Pharmaceutical Innovations) was applied to the skin. The foot was placed in the testing chamber with the lid acting as a calf support. The transducers were positioned over the previously placed skin markers. Each measurement was performed three times with interim repositioning as per the manufacturer’s recommendations. The mean of the three trials was taken as the final value. Short-term reliability in five healthy individuals assessed five times with interim repositioning showed coefficients of variation of 0.54% for SOS and 3.92% for BUA. Dual-energy X-ray Absorptiometry Calcaneal BMC was measured by DXA (Model QDR-1000W, Hologic, Inc., Waltham, MA). Using a lumbar spine protocol to

S. J. Warden et al. Ultrasound treatment of SCI-induced osteoporosis

433

Figure 1. Subregional analysis of calcaneal DXA scan. Metal wires attached to a Perspex holder arranged to meet at right angles at the semipermanent skin marker on the medial aspect of the heel. Image of metal wires traced to the point where they meet at right angles. Region of interest (ROI) box centered over this point.

perform regional scans of each calcaneus, the scanning point resolution was 1 mm for all scans. For scanning, subjects were positioned side-lying on the side of the calcaneus of interest. The ankle joint was placed in 20° plantarflexion. To facilitate accurate scan analysis, two metal wires mounted on a Perspex holder were positioned to intersect at right angles at the semipermanent skin marker on the medial aspect of the heel. A subregional analysis was performed to analyze the scans. The metal wires visible on the scan image were traced to the point where they met at right angles (Figure 1). A region-ofinterest (ROI) box maintained at a constant 18 ⫻ 18 pixels was centered over this point. With the scanning point resolution, the ROI box dimensions equaled 1.8 ⫻ 1.8 cm. This represented the largest box that could be contained within the area of bone treated with US. As the box analyzed a constant sized area of bone, DXA-derived BMD will not be reported. Short-term reliability for the scanning procedure on our DXA scanner in five healthy individuals scanned five times with interim respositioning showed a mean coefficient of variation (CV) for BMC of 0.86%. Statistical Analysis The skeletal effects of SCI and low-intensity pulsed US were determined using two-way, two-repeated-measures analyses of variance (ANOVAs) with time (initial vs. final) and intervention (active US vs. inactive US) as independent variables. All comparisons were two-tailed and performed with the STATISTICAL PACKAGE FOR SOCIAL SCIENCES (SPSS; Norusis/SPSS, Inc., Chicago, IL). The level of significance was set at 0.05. Results Table 2 and Figure 2 show the effects of SCI and low-intensity pulsed US on DXA and qUS measures of the calcaneum. No significant interactions were found between intervention (active US and inactive US) and time (initial and final) for any skeletal measure. A significant main effect for time was found. Thus, SCI had a significant negative influence on bone during the 6 week intervention period. There was no significant main effect for intervention indicating an inability of low-intensity pulsed US to prevent SCI-induced calcaneal bone changes.

434

S. J. Warden et al. Ultrasound treatment of SCI-induced osteoporosis

Bone Vol. 29, No. 5 November 2001:431– 436

Table 2. Effect of low-intensity pulsed ultrasound (US) on skeletal measures of the calcaneum in individuals following spinal cord injury (mean ⫾ SD) Active-US-treated calcaneum

ANOVA p valuesa

Inactive-US-treated calcaneum

Skeletal measure

Initial

Final

Initial

Final

Interaction

Time main effect

Treatment main effect

BMC (g)b BUA (dB/MHz)c SOS (m/sec)c

1.76 ⫾ 0.26 80.4 ⫾ 23.6 1626.7 ⫾ 39.1

1.62 ⫾ 0.26 73.2 ⫾ 20.2 1605.9 ⫾ 47.6

1.74 ⫾ 0.22 83.5 ⫾ 19.5 1638.2 ⫾ 44.6

1.61 ⫾ 0.21 75.9 ⫾ 17.2 1612.9 ⫾ 41.7

0.617 0.752 0.544

⬍0.001 ⬍0.001 ⬍0.001

0.693 0.315 0.129

KEY: BMC, bone mineral content; BUA, bone ultrasound attenuation; SOS, speed of sound. Two-way, two-repeated measures analysis of variance (ANOVA). b Measured using dual-energy X-ray absorptiometry. c Measured using quantitative ultrasound. a

Discussion The findings of the current investigation demonstrate that lowintensity pulsed US at the dose introduced was unable to protect against SCI-induced calcaneal bone changes, as assessed by standard clinical assessment tools. No significant differences in either DXA or qUS measures were found between active and inactive US-treated calcanei. These nonsignificant findings oppose those of Arai et al.2 who demonstrated an anabolic effect of low-intensity pulsed US on osteoporotic patients. However, their study was an uncontrolled case series investigating a limited number of patients (n ⫽ 5). The findings of the current investigation support those of Spadaro and Albanese32 who found low-intensity pulsed US to have no influence on hindlimb BMD in growing rodents. However, given they were primarily investigating the effect of US on bone growth and did not induce an osteopenic state, their findings are limited in terms of evaluating US as an intervention for osteoporosis. Our findings also support those of Wimsatt et al42 who showed that low-intensity pulsed US had no influence on immobilization-induced pigeon-wing bone loss. Their nonsignificant finding may have resulted from an US treatment protocol of twice-weekly intervention, which appears too infrequent for a meaningful skeletal response. There are a number of possible explanations for the absence of a beneficial US effect in the current investigation. One is the relatively short intervention period. The US intervention period in the current study was 6 weeks. It was impractical to extend this due to patient discharge from rehabilitation and the need for frequent US intervention. Despite the short intervention period, we believe it was of sufficient duration to allow an US protective

Figure 2. Effect of low-intensity pulsed ultrasound on percentage change (mean ⫾ SD) in skeletal measures of the calcaneum in individuals following spinal cord injury.

effect to be detected. Intervention periods of 6 weeks or less have previously been shown to be sufficient for a bone response to low-intensity pulsed US. In the previous study by Arai et al.,2 4 weeks was sufficient for US to not only prevent bone loss during a period of active loss, but to induce an anabolic effect. Similarly, in a recent study by Takikawa et al.,33 6 weeks of US treatment was sufficient to stimulate bone union in a rodent model of established fracture nonunion. Additional support for the length of the intervention period in the current investigation is provided by the highly significant bone loss detected. Over the 6 week intervention period, BMC changes in inactive US-treated calcanei were equivalent to those occurring with 5 years of postmenopausal aging.17 Given the US benefits observed in previous investigations and the extent of bone loss detected in the current study, we believe the intervention period was of sufficient duration to see at least a partial protective US effect. This was not detected, with closer inspection of our data revealing no trends toward bone preservation. As with most intervention studies that have observed a nonsignificant effect, statistical power needs to be questioned. We addressed our research question with sufficient statistical power. Power calculations for DXA measurement of calcaneal BMC indicated that with 15 subjects there would be an 80% probability of detecting a difference of 2.8% between the active and inactive US-treated calcanei. This is a clinically relevant difference in terms of the risk of osteoporotic fracture. Although a longer intervention period may have elucidated a smaller partial protective effect, an effect smaller than could be detected would be of minimal clinical significance with regard to the clinical utility of US in the intervention of osteoporosis. As such, the findings indicate a true absence of a protective effect of US at the dose and mode administrated on SCI-induced bone loss of the calcaneum. The lack of a beneficial US effect on calcaneal changes may relate to the acoustic properties of bone. In comparison to the acoustic properties of the surrounding soft tissues, bone has a high absorption coefficient, a high relative acoustic impedance, and an ability to propagate shear waves.39 These properties restrict US propagation into intact bone. For example, as up to 40% of US energy is reflected at the soft tissue-bone interface and ⬎80% of the remaining energy is attenuated in the first millimeter of propagation in bone,39 only 12% or 3.6 mW/cm2, of incident energy would have remained after 1 mm of propagation into the calcaneum adjacent to each US treatment head. Setting 3.6 mW/cm2 as an optimistic threshold for evoking a biological effect, a maximum of 2% of the average calcaneal bone width was actually treated in the current study. This limited volume of bone effectively treated may have contributed to the absence of an US effect. Our results do not exclude the possibility of a localized

Bone Vol. 29, No. 5 November 2001:431– 436

cortical effect. Given the aforementioned acoustic properties of bone, it is possible that US had a focal effect on the outer cortical shell. Our assessment tools were not empowered to measure such an effect. As we treated no more than 2% of the assessed bone volume, the evoked changes in this volume would need to be 50 times the minimum detectable change in the entire bone volume to be detectable by DXA with adequate statistical power. Consequently, changes of around 140% would have needed to be evoked in the limited volume effectively accessible to the US to be detectable with adequate power. Although there is interest in determining whether US could stimulate a localized cortical effect, this was not the purpose of the current study. Also, the clinical benefits of any isolated cortical effect need to be debated. For US to represent a useful clinical modality for osteoporosis, strength changes need to be induced. The ability of US to stimulate sufficient cortical changes to influence bone strength to a clinically significant level remains questionable. Similarly, the limited ability of currently available tools to detect subtle cortical changes in the clinical setting would make it difficult to assess cortical US effects with precision. US was introduced in the current study on the basis of it being a form of mechanical stimulation. However, the mechanical loading generated by US appears to be inconsistent with that required for bone adaptation. Ignoring attenuation between the US treatment head and bone surface, US at a spatial-averaged temporal-average intensity of 30 mW/cm2 exerts a cortical spatial-averaged temporal-average force of approximately 2 mg/ cm2. With our pulsing regime, the cortical spatial-average temporal-peak force is 60 mg/cm2. This force is substantially below the threshold for bone formation shown in an alternate bone loading model,35 and is not considered by the investigators to be of sufficient magnitude to induce intraosseous strain. The calcaneum proved to be a valid site to investigate the potential clinical utility of US in the prevention of SCI-induced osteoporosis despite not being a site prone to osteoporotic fracture. The calcaneum consists of an inner metabolically active trabecular network and a thin outer cortical shell. This structure is representative of the distal femur and proximal tibia, sites prone to osteoporotic fracture following SCI.16 Similarly, the extent of bone loss in the calcaneum in the acute stages following SCI corresponds to sites prone to SCI-induced osteoporosis. In the current study, the rate of decline in inactive-US-treated calcaneal BMC was 1.25%/week. In the distal femur, Garland et al.14 reported a BMD loss of 1.47%/week in the first 4 months after SCI. The characteristics of bone loss following SCI may have contributed to the absence of a beneficial US effect. In the acute stages following SCI, bone loss is predominantly due to loss of trabecular bone with cortical bone showing a delayed loss.13 In light of the selective cortical effect of US, the delayed response of cortical bone following SCI may have altered its responsiveness to US. Similarly, the pathomechanics of bone loss following SCI may have played a role in the absence of a significant finding. Bone changes in the acute stages following SCI are due to increases in bone resorption with minimal change in bone formation.29 The effects of US on bone resorption have not been established and, as US is considered to have its skeletal benefits by effects on bone formation, the unaltered formation rate following SCI may have influenced the calcaneal response to US. The current study demonstrates that intact bone shows a disparate response to US than fractured bone. One reason for this may relate to the presence of a cortical defect in fractured bone. This fracture gap reduces the cortical attenuation of US, providing an entry point into the injured area. This potentially enables a greater bone volume effect. As US is not considered to induce intraosseous strain, its effects once in the fracture gap may be

S. J. Warden et al. Ultrasound treatment of SCI-induced osteoporosis

435

mediated by phenomena other than mechanical loading. Potential phenomena include stable cavitation and microstreaming. A second possible reason for the disparate finding between intact and fractured bone relates to the tissue-level processes taking place. During fracture repair there are numerous additional stages prior to bone formation. The beneficial effects of US during fracture repair may be due to effects on a range of reparative processes and not solely osteogenesis.3 Although US has been shown to generate a bone-forming response when applied to isolated bone cells in vitro, the findings of these studies need to be interpreted cautiously. In vitro, the aqueous medium in which cells are grown allows room for any contained bubbles to freely expand and contract. This enhanced potential for stable cavitation, combined with facilitated microstreaming and bulk fluid movement, may amplify the response of isolated bone cells to US.1,41 In summary, this investigation has confirmed the negative skeletal effects of SCI and demonstrated the inability of lowintensity pulsed US at one particular dose to prevent SCI-induced bone loss. The absence of a beneficial effect may relate to a number of factors, with the inability of US to effectively penetrate the outer bone cortex due to its acoustic properties being prominent. Being the first controlled study of the effect of US on osteoporosis in humans, the present findings do not completely exclude US as an intervention for osteoporosis. It is possible that US may have benefits on alternate models of osteoporosis. Similarly, it is possible that alternate doses of US may have beneficial effects on intact bone. These effects are likely to be restricted to the outer bone cortex. The clinical benefits of such cortical changes need to be assessed in terms of their ability to be translated into strength changes and, therefore, their ability to alleviate the main consequence of osteoporosis— bone fractures.

Acknowledgments: The authors thank Dr. Adrian Richards and Adam Stirling (Ultrasound Standards, National Measurement Laboratory, Division of Telecommunications and Industrial Physics, CSIRO) for providing assistance with ultrasound design and calibration and experimental methodology, Eliah Andrews (School of Physiotherapy, University of Melbourne) for assisting with ultrasound interventions, the rehabilitation staff at the Victorian Spinal Cord Service for their assistance, and the participating patients at the Royal Talbot Rehabilitation Center for their commitment during this difficult time. This work was funded by the William Buckland Foundation (ANZ Charitable Trusts), Aza Research, and the Physiotherapy Research Foundation.

References 1. AIUM. Mechanical bioeffects from diagnostic ultrasound: AIUM consensus statements. J Ultrasound Med 19:77– 84; 2000. 2. Arai, T., Ohashi, T., Daitoh, Y., and Inoue, S. The effect of ultrasound stimulation on disuse osteoporosis. Trans Bioelectr Repair Growth Soc 13:17; 1993. 3. Azuma, Y., Ito, M., Harada, Y., Takagi, H., Ohta, T., and Jingushi, S. 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– 680; 2001. 4. Biering-Sørensen, F., Bohr, H. H., and Schaadt, O. P. Longitudinal study of bone mineral content in the lumbar spine, the forearm and the lower extremities after spinal cord injury. Eur J Clin Invest 20:330 –335; 1990. 5. Chivers, R. C. Fundamentals of ultrasonic propagation. In: Preston R. C., Ed. Output Measurements for Medical Ultrasound. London: Springer; 1991; 19 –33. 6. Dauty, M., Verbe, B. P., Maugars, Y., Dubois, C., and Mathe, J. F. Supralesional and sublesional bone mineral density in spinal cord-injured patients. Bone 27:305–309; 2000. 7. Duarte, L. R. The stimulation of bone growth by ultrasound. Arch Orthop Trauma Surg 101:153–159; 1983.

436

S. J. Warden et al. Ultrasound treatment of SCI-induced osteoporosis

8. Duncan, R. L. and Turner, C. H. Mechanotransduction and the functional response of bone to mechanical strain. Calcif Tissue Int 57:344 –358; 1995. 9. Food and Drug Administration. WEAC Procedures for Laboratory Testing of Ultrasonic Therapy Devices. Rockville, MD: FDA; 1982. 10. Forwood, M. R. Inducible cyclo-oxygenase (COX-2) mediates the induction of bone formation by mechanical loading in vivo. J Bone Miner Res 11:1688 – 1693; 1996. 11. Frankel, V. H. Results of prescription use of pulse ultrasound therapy in fracture management. In: Szabo´ , Z., Lewis, J. E., Fantini, G. A., and Savalgi, R. S., Eds. Surgical Technology International VII. San Francisco, CA: Universal Medical; 1998; 389 –393. 12. Freehafer, A. A. and Mast, W. A. Lower extremity fractures in patients with spinal-cord injury. J Bone Jt Surg 47A:683– 694; 1965. 13. Frey-Rindova, P., de Bruin, E. D., Stu¨ ssi, E., Dambacher, M. A., and Dietz, V. Bone mineral density in upper and lower extremities during 12 months after spinal cord injury measured by peripheral quantitative computed tomography. Spinal Cord 38:26 –32; 2000. 14. Garland, D. E., Stewart, C. A., Adkins, R. H., Hu, S. S., Rosen, C., Liotta, F. J., and Weinstein, D. A. Osteoporosis after spinal cord injury. J Orthop Res 10:371–378; 1992. 15. Heckman, J. D., Ryaby, J. P., McCabe, J., Frey, J. J., and Kilcoyne, R. F. Acceleration of tibial fracture-healing by non-invasive, low-intensity pulsed ultrasound. J Bone Jt Surg 76A:26 –34; 1994. 16. Ingram, R. R., Suman, R. K. and Freeman, P. A. Lower limb fractures in the chronic spinal cord injured patient. Paraplegia 27:133–139; 1989. 17. Kado, D. M., Browner, W. S., Blackwell, T., Gore, R., and Cummings, S. R. Rate of bone loss is associated with mortality in older women: A prospective study. J Bone Miner Res 15:1974 –1980; 2000. 18. Keating, J. F., Kerr, M., and Delargy, M. Minimal trauma causing fractures in patients with spinal cord injury. Disabil Rehabil 14:108 –109; 1992. 19. Kokubu, T., Matsui, N., Fujioka, H., Tsunoda, M., and Mizuno, K. Low intensity pulsed ultrasound exposure increases prostaglandin E2 production via the induction of cyclooxygenase-2 mRNA in mouse osteoblasts. Biochem Biophys Res Commun 256:284 –287; 1999. 20. Kristiansen, T. K., Ryaby, J. P., McCabe, J., Frey, J. J., and Roe, L. R. Accelerated healing of distal radius fractures with the use of specific, lowintensity ultrasound. J Bone Jt Surg 79A:961–973; 1997. 21. Lee, T. Q., Shapiro, T. A., and Bell, D. M. Biomechanical properties of human tibias in long-term spinal cord injury. J Rehabil Res Dev 34:295–302; 1997. 22. Maynard, F. M., Jr., Bracken, M. B., Creasey, G., Ditunno, J. F., Jr., Donovan, W. H., Ducker, T. B., Garber, S. L., Marino, R. J., Stover, S. L., Tator, C. H., Waters, R. L., and Wilberger, J. E. International standards for neurological and functional classification of spinal cord injury. Spinal Cord 35:266 –274; 1997. 23. Mayr, E., Frankel, V., and Ru¨ ter, A. Ultrasound: An alternative healing method for nonunions? Arch Orthop Trauma Surg 120:1– 8; 2000. 24. Mayr, E., Rutzki, M. M., Rutzki, M., Borchardt, B., Hausser, H., and Ru¨ ter, A. Beschleunigt niedrig intensiver, gepulster Ultraschall die Heilung von Skaphoidfrakturen? Handchir Mikrochir Plast Chir 32:115–122; 2000. 25. Naruse, K., Mikuni-Takagaki, Y., Azuma, Y., Ito, M., Ohta, T., Kameyama, K.-Z., and Itoman, M. Anabolic response of mouse bone-marrow-derived stromal cell clonal ST2 cells to low-intensity pulsed ultrasound. Biochem Biophys Res Commun 268:216 –220; 2000. 26. Nolte, P. A., Klein-Nulend, J., Albers, G. H. R., Marti, R. K., Semeins, C. M., Goei, S. W., and Burger, E. H. Low intensity ultrasound stimulates endochondral ossification in vitro. J Orthop Res 19:301–307; 2001.

Bone Vol. 29, No. 5 November 2001:431– 436 27. Pilla, A. A., Mont, M. A., Nasser, P. R., Khan, A., Figueiredo, M., Kaufman, J. J., and Siffert, R. S. Non-invasive low-intensity pulsed ultrasound accelerates bone healing in the rabbit. J Orthop Trauma 4:246 –253; 1990. 28. Reher, P., Elbeshir, E. I., Harvey, W., Meghji, S., and Harris, M. The stimulation of bone formation in vitro by therapeutic ultrasound. Ultrasound Med Biol 23:1251–1258; 1997. 29. Roberts, D., Lee, W., Cuneo, R. C., Wittmann, J., Ward, G., Flatman, R., McWhinney, B., and Hickman, P. E. Longitudinal study of bone turnover after acute spinal cord injury. J Clin Endocrinol Metab 83:415– 422; 1998. 30. Sato, W., Matsushita, T., and Nakamura, K. Acceleration of increase in bone mineral content by low-intensity ultrasound energy in leg lengthening. J Ultrasound Med 18:699 –702; 1999. 31. Shimazaki, A., Inui, K., Azuma, Y., Nishimura, N., and Yamano, Y. Lowintensity pulsed ultrasound accelerates bone maturation in distraction osteogenesis in rabbits. J Bone Jt Surg 82B:1077–1082; 2000. 32. Spadaro, J. A. and Albanese, S. A. Application of low-intensity ultrasound to growing bone in rats. Ultrasound Med Biol 24:567–573; 1998. 33. Takikawa, S., Matsui, N., Kokubu, T., Tsunoda, M., Fujioka, H., Mizuno, K., and Azuma, Y. Low-intensity pulsed ultrasound initiates bone healing in rat nonunion fracture model. J Ultrasound Med 20:197–206; 2001. 34. ter Haar, G. Basic physics of therapeutic ultrasound. Physiotherapy 73:110 – 113; 1987. 35. Turner, C. H., Forwood, M. R., Rho, J.-Y., and Yoshikawa, T. Mechanical loading thresholds for lamellar and woven bone formation. J Bone Miner Res 9:87–97; 1994. 36. Vestergaard, P., Krogh, K., Rejnmark, L., and Mosekilde, L. Fracture rates and risk factors for fractures in patients with spinal cord injury. Spinal Cord 36:790 –796; 1998. 37. Wang, S.-J., Lewallen, D. G., Bolander, M. E., Chao, E. Y. S., Ilstrup, D. M., and Greenleaf, J. F. Low intensity ultrasound treatment increases strength in a rat femoral fracture model. J Orthop Res 12:40 – 47; 1994. 38. Warden, S. J., Bennell, K. L., McMeeken, J. M., and Wark, J. D. Acceleration of fresh fracture repair using the Sonic Accelerated Fracture Healing System (SAFHS): A review. Calcif Tissue Int 66:157–163; 2000. 39. Warden, S. J., Bennell, K. L., McMeeken, J. M., and Wark, J. D. Can conventional therapeutic ultrasound units be used to accelerate fracture repair? Phys Ther Rev 4:117–126; 1999. 40. Warden, S. J., Wong, W. T., Bennell, K. L., McMeeken, J. M., and Wark, J. D. Facilitation of fracture repair using low-intensity pulsed ultrasound. Vet Comp Orthop Traumatol 13:158 –164; 2000. 41. WFUMB. World Federation for Ultrasound in Medicine and Biology Symposium on safety of ultrasound in medicine. Ultrasound Med Biol 24(Suppl. 1):S1–S55; 1998. 42. Wimsatt, J., Dressen, P., Dennison, C., and Turner, A. S. Ultrasound therapy for the prevention and correction of contractures and bone mineral loss associated with wing bandaging in the domestic pigeon (Columbia livia). J Zoo Wildlife Med 31:190 –195; 2000.

Date Received: December 12, 2000 Date Revised: May 14, 2001 Date Accepted: May 16, 2001