Osteoporosis is a disease characterized by low bone mass and a deterioration of bone micro-architecture leading to an increased fracture risk. Today's standard treatment aims at reduction of further bone loss using anti-resorptive therapy.1 However, increasing bone mass by anabolic treatment might be more optimal for reducing fracture risk. To date anabolic treatment is limited to the use of recombinant parathyroid hormone or its analog. This treatment requires daily subcutaneous administration, accompanies potential dangerous side effects like hypocalcaemia, and is expensive. Therefore this therapy is limited to patients that do not respond to or have contraindications for bisphosphonates.1
As an alternative to pharmacologic treatment biophysical stimuli have been suggested, but mechanical vibration, pulsed electromagnetic fields and ultrasound have so far not been proven to be beneficial in osteoporosis.2–5 Previously we showed that unfocused extracorporeal shock waves (UESW) with an energy flux density (EFD) of 0.3 mJ/mm2 can also induce anabolic effects in healthy rat bone.6 Pronounced increased bone formation was observed in the cortical bone and the bone marrow 7 days after UESW were applied to the hind leg of healthy male rats. This resulted in increased trabecular and cortical bone mass and improved biomechanical properties. Furthermore, we have shown that UESW can beneficially affect bone micro-architecture in a rat osteoporosis model.7 These experiments were however done with a lower EFD (0.16 mJ/mm2) and we did not find an anabolic bone response.
So far, most research included focused extracorporeal shock waves. These shock waves are used in a variety of musculoskeletal disorders like non-unions and delayed unions, diaphyseal fractures, stress fractures, osteonecrosis of the femoral head, Achilles tendinopathy, and fasciitis plantaris.8–12 Shock waves are acoustical pulses with a high amplitude (±100 bar) and a short rise time (≤10 ns) in the frequency spectrum between 16 Hz and 12 MHz.13 In focused shock wave therapy the waves converge in a focal point. In contrast, unfocused shock waves are produced as a parallel bundle, enabling a homogenous treatment of larger regions.
In experimental studies examining the effects of focused shock waves on bone it has been shown that a single treatment led to an increased differentiation of bone marrow stem cells towards osteoprogenitor cells.14, 15 Furthermore, it has been shown that several growth factors that are important for bone regeneration, including VEGF, TGF-beta 1, and several BMPs, are upregulated after extracorporeal shock wave treatment.14, 16–18
To explore the potential use of UESW to increase bone mass and subsequently reduce fracture risk in osteoporotic patients, we examined the effects of UESW on the bone micro-architecture and biomechanical properties in a rat model for osteoporosis. To investigate the additional clinical value of UESW in the presence of an anti-resorptive treatment with bisphosphonates, conditions with and without ALN (Merck, Whitehouse Station, NJ) were examined.
Twenty-seven female Wistar rats, age 20 weeks, were obtained (Harlan Laboratories, Horst, the Netherlands). The animals were housed in threesomes at the animal facility of the Erasmus MC, with a 12 h light–dark regimen, at 21°C and received standard food pellets and water ad libitum. The research protocol (116-08-02/EUR 1455) was approved by the local committee for Animal Experiments and is in accordance with Dutch law.
Three groups of six animals were used for evaluation of bone changes during a 10-week follow-up period with in vivo micro-CT scanning and these rats were used for biomechanical testing. The remaining nine animals were used for histological and ex vivo micro-CT evaluation only and were not included in the statistical analysis. Of these nine animals, three animals were euthanized 24 h after UESW were applied (no ALN) and six animals were euthanized at 1 week after UESW were applied (three received saline and three received ALN).
All rats received a bilateral ovariectomy (OVX) to simulate osteoporosis. This procedure was performed under sterile conditions using gas anesthesia (oxygen with 2% isoflurane; Rhodia Organique Fine Ltd., Bristol, UK). Buprenorfine (Schering-Plough, Kenilworth, NJ), 0.05 mg/kg/12 h, was given for 2 days. Three treatment groups of 6 animals were made. The first group received subcutaneous injections with saline three times per week, starting the first day after OVX. The second group received subcutaneous injections with ALN, 2.4 µg/kg, 3x/week. This is comparable to a human dose of 70 mg/week.19 Treatment was also started the first day post-operatively. In these two groups UESW were applied 2 weeks after OVX. The third group received subcutaneous saline injections similar to the first group, but UESW treatment was applied 2 days after OVX was performed. Because UESW were applied a short time after OVX the bone volume at time of UESW treatment is comparable to the ALN treated group, in which little bone loss is observed.
Unfocused Shock Wave Therapy
One hind leg was treated with UESW, the contra-lateral hind leg was not treated and served as control. Under gas anesthesia with oxygen and 2% isoflurane both hind legs were shaved and an ultrasonic gel that served as coupling media was applied. The shock wave applicator was placed at the antero-lateral side of the hind leg. Since this is an unfocussed shock wave applicator the shock wave energy is about evenly distributed over a fairly large region, that contains the entire tibia of the rat. A total of one thousand electro-hydraulically generated UESW were applied at 3 Hz with an EFD of 0.3 mJ/mm2 (Dermagold, Tissue Regeneration Technologies, Woodstock, GA, manufactured by MTS, Konstanz, Germany).
In Vivo Micro-CT Scanning
At 0, 2, 4, and 10 weeks after UESW treatment micro-CT scans were made under gas anesthesia (isoflurane/oxygen). The proximal 17 mm of the tibia was scanned (60 kV, 167 µA, using a 0.5 mm aluminium filter, over 196° with a rotation step of 1°), which takes 8 min (Skyscan 1076 scanner, Kontich, Belgium). This resulted in datasets with an isotropic voxel size of 18 µm. Three-dimensional (3D) reconstructions of two regions of interest were made (NRecon software, Skyscan). The first is of the proximal metaphysis of which trabecular bone parameters were calculated (starting just distal of the epiphysial growthplate and continuing until 2.7 mm more distal). The second reconstruction was from the diaphysis. On these reconstructions cortical bone parameters were determined (starting 9 mm distal from the epiphysis and continuing until 3.6 mm more distal). Bony and non-bony structures were separated using a local threshold algorithm (software freely available)20 resulting in binary datasets.21 Cortical and trabecular bone were automatically separated using in-house software. Trabecular architecture in the proximal metaphysis was characterized by determining trabecular bone volume fraction (BV/TV), connectivity density (Conn/TV), structural model index (SMI), and 3D trabecular thickness (TbTh). Cortical architecture in the diaphysis was characterized by cortical volume (CtV) and cortical thickness (CtTh).
Three-Point Bending Tests
Mechanical properties were determined using a three-point bending test. At 10 weeks after UESW, animals were euthanized using an overdose of pentobarbital. The tibiae were dissected, soft tissue removed, and stored in PBS at 4°C. Testing was performed with the bones equilibrated to room temperature. Each tibia was stably positioned on its medial antero-lateral site. The distance between the loading posts was 2.2 cm, in this way the posts were just distal from the condyles and just proximal from the maleoli. To ensure a stable position of the bone five pre-conditioning runs with a displacement rate of 0.01 mm/s and a maximal force of 4 N were done (Single Column Lloyd LRX Testing System, Lloyd Instruments, Hampshire, UK). Hereafter a displacement rate of 0.01 mm/s was applied until fracture of the tibia. Displacement (mm) and force (N) were registered at a sample rate of 20 Hz. Force displacement graphs were made and load to failure and stiffness (tangent of the linear portion of the force displacement curve) were determined.
The tibiae were fixed in 4% paraformaldehyde. The tibiae hind legs used for mechanical testing at 10 weeks after UESW treatment were directly fixed after testing. After 21 days of fixation the proximal half of the tibiae were dehydrated and block embedded in methyl methacrylate (MMA). Sagittal sections (6-µm thick) were made. Histological assessment (e.g., new bone formation, marrow status, and cell characterization) was evaluated using hematoxilin–eosin and thionine staining (0.05% thionine in 0.01 M aqueous sodium phosphate, pH 5.8 for 5 min).
All parameters were statistically analyzed for each time point using paired t-tests, in which the UESW treated and untreated control legs served as a pair (GraphPad Software, San Diego, CA). Changes in BV/TV and CtV at each time point relative to baseline were statistically tested in treated and untreated legs separately, for each treatment group, using Mixed Model regression (SPSS 16 Software IB07, Armouh, NY). Differences of biomechanical parameters of treated legs between the three treatment groups were statistically analyzed with one-way ANOVA with a Tukey's multiple comparison test.
In Vivo Micro-CT Analysis
In saline treated rats receiving UESW 2 weeks after OVX, BV/TV was 22.2% (SD 3.3) at baseline (t = 0). During 2 weeks follow-up trabecular bone loss as a consequence of OVX was much lower in UESW treated legs than in non-treated control legs (Fig. 1a). At 2 weeks BV/TV was 20.2% (SD 3.5) in UESW treated legs compared to 13.8% (SD 3.0) in untreated control legs (P = 0.003). This demonstrates that in UESW treated legs the cancellous bone volume was 46% compared to untreated controls. However, at 4 and 10 weeks no difference could be found. Morphometric analysis of the cancellous bone showed more plate-like trabeculae (lower SMI), thicker trabeculae, and a higher trabecular connectivity in UESW treated legs compared to untreated control legs at 2 weeks after UESW (Table 1). At 4 and 10 weeks no difference between treated and untreated legs were found.
Table 1. Bone parameters determined with in vivo micro-CT scanning, mean with standard deviation
trab.th., mean trabecular thickness; conn.dens., connectivity density; SMI, structural, model index; cort.thick., cortical thickness.
P < 0.05 in UESW treated compared to untreated controls at the same time point.
Cortical volume was significantly higher in UESW treated legs than in control legs during follow-up at all time-points (Fig. 1b). The cortical volume in UESW treated legs increased with 8.0% (SD 2.9), 13.4% (SD 5.5), and 13.0% (SD 6.2) relative to baseline at 2, 4, and 10 weeks respectively. This was less in untreated control: 1.3% (SD 2.0), 3.7% (SD 1.5), and 3.2% (SD 3.3) at 2, 4, and 10 weeks respectively. Cortical thickness was also higher in UESW treated legs compared to control legs at 2, 4, and 10 weeks (Table 1).
In rats receiving ALN, BV/TV was 28.2% (SD 1.6) at baseline. UESW resulted in an increase in BV/TV during follow-up, which resulted in a significantly higher BV/TV compared to untreated contra-lateral legs at 2, 4, and 10 weeks after UESW (2 weeks: 32.8% (SD 2.6) vs 27.4% (SD 1.6) (P = 0.0002), 4 weeks: 33.0% (SD 2.8) vs 27.0% (SD 1.0) (P = 0.0003), and 10 weeks: 27.5% (SD 3.3) vs 21.9% (SD 2.2) (P = 0.0002; Fig. 1a). In UESW treated legs the cancellous bone volume was 20%, 22%, and 26% higher at 2, 4, and 10 weeks respectively. After UESW treatment trabeculae were more plate-like (lower SMI), thicker and had a higher connectivity at all time-points (Table 1).
Cortical volume was significantly higher in UESW treated legs compared to control legs during follow-up (Fig. 1b). Cortical volume increased with 12.2% (2.7), 18.7% (4.5), and 21.5% (3.2) from baseline at 2, 4, and 10 weeks, respectively. In control legs this was 2.8% (SD 1.7), 5.9% (SD 2.0), and 7.7% (SD 2.8) at 2, 4, and 10 weeks respectively. Cortical thickness was also higher in UESW treated legs at all time points during follow-up (Table 1).
In the third group, in which UESW were given 2 days after OVX, BV/TV was 28.4% (SD 3.7), similar to the ALN treated group. UESW did again result in reduced trabecular bone loss after UESW treatment with a BV/TV of 25.3% (SD 3.6) in UESW treated legs compared to 19.5% (SD 3.9) in untreated control legs (P = 0.003) at 2 weeks follow-up (Fig. 1a). This demonstrates a 30% higher cancellous bone volume in the treated legs compared to untreated controls at 2 weeks. At 4 and 10 weeks no difference in BV/TV was found. At 2 weeks, but not at 4 and 10 weeks SMI was lower and connectivity density and mean trabecular thickness were higher in UESW treated legs compared to controls (data not shown). UESW again resulted in higher cortical volume and cortical thickness in UESW treated legs compared to untreated controls (Fig. 1b).
In both the saline and the ALN group UESW treated legs showed significant higher maximal force at failure and higher stiffness (Fig. 2). In saline treated rats that received UESW 2 weeks after OVX the maximal force and stiffness were respectively 14% and 10% higher compared to untreated controls. In ALN treated rats maximal force and stiffness were respectively 15% and 18% higher and in saline treated rats that received UESW 2 days after OVX they were respectively 10% and 19% higher. Furthermore, maximum force and stiffness were significantly higher in ALN treated animals compared to saline treated animals (P = 0.02 for maximum force at failure, and P = 0.01 for stiffness).
Ex Vivo Micro-CT Analysis and Histology
On micro-CT reconstructions at 1 week after UESW treatment, abundant areas of new bone formation were found (Fig. 3a). These areas were seen along the periosteal cortex and in the bone marrow. The latter was seen distally to where trabeculae normally extend and appears to be de novo bone formation. With histology it is demonstrated that in both these areas very active intramembranous new bone formation exists (Fig. 3b). At the cortex many active osteoblasts, osteoid, and immature bone are found (Fig. 3c). Areas with hypertrophic chondrocytes and mineralized callus were observed only sparsely. In the cortex of UESW treated tibiae osteocytes with defragmented or pyknotic nuclei and empty lacunae were found. This seemed more pronounced at the endosteal site. No periosteal detachment or subperiosteal hemorrhage were observed. In the bone marrow active osteoblasts, osteoid, and immature bone were seen abundantly around fibrotic tissue with fibroblasts (Fig. 3d).
Further, on micro-CT reconstructions cortical cracks were identified in three of the six animals (Fig. 4). The undisplaced fractures were seen in the diaphysis only, running perpendicular to one of the three cortical columns (6–10 mm in length). Which column was affected varied. We found no more than one crack per animal. New bone formation was not directly related to these cracks since this was seen at all cortices (Fig. 5).
In this study we show that a single treatment with UESW leads to an increase in trabecular and cortical bone volume, which leads to significantly improved biomechanical properties and thus to a theoretical reduction of the fracture risk. Interestingly, these effects were more pronounced in osteoporotic rats receiving ALN, a bisphosphonate that is widely used among osteoporotic patients.
Today's standard treatment of osteoporosis is an anti-resorptive treatment with bisphosphonates. ALN is widely used in that perspective. To investigate the potential role of UESW for osteoporotic patients, the effect of UESW was investigated in situations with and without treatment with ALN. We found that UESW lead to an increase in trabecular bone volume in both saline and ALN treated animals. The effect of UESW on trabecular bone in saline treated animals was only transient, while in ALN treated animals this increase was preserved during follow-up. Furthermore, the effect of UESW on cortical bone mass was more pronounced in ALN treated rats. ALN has a direct effect on the activity of osteoclasts, which subsequently leads to inhibition of bone resorption.22 Although bone remodeling is mostly a coupled system ESW did induce strong anabolic effects, which suggest that bone modeling was stimulated rather than bone remodeling.
To be assured that these favorable effects were due to the anti-resorptive effects of the bisphosphonate and not due to the higher bone volume at time of UESW treatment, we examined a third group. In this saline treated group UESW treatment was applied 2 days after ovariectomy when BV/TV was comparable to BV/TV in the group that received ALN for 2 weeks after OVX. Again UESW resulted in a decline of trabecular bone loss. The relative loss in BV/TV in saline treated rats that received UESW 2 weeks after ovariectomy and rats that received UESW 2 days after ovariectomy was both around 10% of baseline. But again this effect was not maintained at longer follow-up. This indeed suggests that the anti-resorptive treatment itself and not the amount of trabecular bone at time of UESW treatment was responsible for the increase and preservation of trabecular bone mass after UESW treatment.
Biomechanical testing showed that UESW induced bone changes result in improved mechanical properties which were most pronounced in ALN treated animals. This suggests that a treatment with UESW in the presence of an anti-resorptive treatment might further reduce fracture risk.
Abundant areas of new bone formation are already seen 1 week after UESW are applied. It is remarkable to see that the new bone formation at the periosteal site and in the bone marrow was predominantly intramembranous. In line with this is a report in which focused ESW to rat tibiae resulted in cambium cell proliferation and subsequent intramembranous osteogenesis.23 Furthermore, Takahashi et al. demonstrated that treatment with focused ESW lead to an increased expression of genes that are suggested to relate to intramembranous bone formation.24 In a previous report in which we applied UESW with an EFD of 0.3 mJ/mm2 to the hind leg of non-osteoporotic male rats, we found that UESW resulted in lysis of hematopoetic cells, destruction of adipocytes, and disruption of micro-vessels.6 We did not observe these features when osteoporotic rats were treated with UESW with an EFD of 0.16 mJ/mm2.7 We believe that in the current study the effects of UESW on the bone marrow led to the formation of fibrotic tissue. Subsequently, extensive areas of intramembranous bone formation are formed. These features show great similarity with bone marrow ablation models. These models are used to examine bone regeneration after the bone marrow is mechanically removed.25, 26 Regeneration is followed through specific stages including an inflammatory stage with clot formation, a repair phase with neovascularization and cell migration (including mesenchymal stem cells), and finally a remodeling phase to re-establish the hematopoietic and fat tissue in the bone marrow. Histological images of bone regeneration in bone marrow ablation show great resemblance to our findings. Interestingly, bone regeneration in bone marrow ablation is solely by intramembranous bone formation.27 This might suggest that treatment with UESW, which also results in intramembranous bone formation, induce similar biological responses. This needs however to be determined in future studies.
We do not have any support that UESW therapy induces a systemic effect that affects bone mass at other skeletal regions. If such a systemic effect would exist we expect to have found an effect on bone micro-architecture in the non-treated contra-lateral site during the 10 weeks follow-up. Also, the BV/TV of saline treated rats (2 days after OVX) at 2 weeks follow-up (25.3% SD 3.6) is not statistically different BV/TV of saline treated rats (2 weeks after OVX) at baseline (22.2% (SD 3.3). Furthermore, we have used whole body multi-pinhole SPECT scanning in a previous study to analyze the effect of UESW.6 In that study we also only found a local effect at the treated region.
In the current study we found cortical fractures after UESW treatment. This is in contrast to earlier experiments we have performed both at an EFD of 0.16 and 0.3 mJ/mm2.2, 3 Because these cracks were only sporadically induced and varied in location, they might be the result from an indirect effect of the shockwaves, known as cavitation.13, 28 This indirect effect, can occur when gas bubbles in soft tissue or liquid tissue, in this case blood in the Haversian system, grow by the positive pressure of the shock wave and eventually collapse. During the collapse, energy is released and high mechanical forces occur, which may cause a cortical crack as was also demonstrated in focused shock wave treatment.29, 30 There might be a correlation between the appearance of fractures and the EFD of UESW, but this has not been studied before. Since new bone formation was seen in the diaphysis of all cortices, we believe the biological responses are independent of cortical cracks. Whether these effects can also occur if the same shock waves are applied in larger species is yet unknown.
A disadvantage of (U)ESW is that the application can be very painful, especially when they are produced electro-hydraulically. Although it has been described that treatments with an EFD up to 0.17 mJ/mm2 can be applied without additional analgesia,31 we believe that especially in the older, fragile patient additional analgesia is indicated. In that perspective it has been shown that intravenous supply of Paracetamol, NSAIDs or Tramadol has good pain relieving results in ESW therapy for kidney stones.32 Alternatively, UESW treatment can be applied when the patient receives anesthesia for other indications, for instance when surgical treatment for osteoporotic fractures is indicated, UESW can be applied to the contra-lateral site and other skeletal regions to prevent other fractures.
The current study shows that UESW can potentially be used in the local treatment of osteoporosis. Especially when combined with an anti-resorptive treatment with bisphosphonates, UESW led to increased bone mass and improved biomechanical properties. Since anti-resorptive treatment with bisphosphonates is the standard treatment for osteoporosis, UESW treatment might have important implications for osteoporotic patients. Further reduction of the fracture risk might be achieved by treating those sites that are specifically vulnerable to fracture in osteoporosis. Since several clinical studies used UESW for other modalities, concerns of adverse side effects seem limited. All together these results suggest the need for a clinical study to examine the effects of UESW on bone density in humans. Most relevant skeletal sites to start with include the forearm and proximal femur. Furthermore, more research is needed to investigate the biological causes that lead to the anabolic bone response after UESW are applied.
In conclusion, this study shows that treatment with UESW leads to increased bone mass and improved biomechanical properties especially when treatment is combined with bisphosphonates. These improved properties suggests that UESW can potentially be of benefit for osteoporotic patients.
W. Schaden is medical advisor and shareholder at Tissue Regeneration Technologies, Woodstock, GA.