Table 1.
The summary of in vivo studies analyzed in this review on the impacts of mechanical loading on fracture healing/regeneration.
| Mechanical loading model and parameters | Main findings | Immediate /delay loading post-trauma |
|---|---|---|
| Fracture healing | ||
| Clinical trial with axial sliding model (82 human subjects). 1st stage: axial displacement (1 mm, 0.5 hz, 20 min/day); 2nd stage: axial movement allowed over the level of 12 kg. | Both clinical healing and mechanical healing were enhanced in the group subjected to axial micromovement, compared to the control group in a fixed mode | Delay (0–7 days) 251 |
| Clinical trial with diaphyseal tibial shaft fractures (3 volunteers, 2 oblique fractures and 1 transverse fracture). 1st stage: 15 kg loading between week 2 and 8; 2nd stage: full body weight loading. | No difference between transverse fractures and oblique fractures. The shear movement induced by 15 kg loading is shown to be compatible with successful healing | Delay (2 weeks) 254 |
| Sheep fix-sliding model with diaphyseal osteotomy (24 subjects). Nonuniform cyclic tensile strains (0.2 and 0.8 mm displacement, 1,5 and 10 Hz for 500 cycles/day) | The external stimulation applied in this study did not significantly enhance the fracture healing process. However, 0.5 mm/10 Hz stimulation induced the highest periosteal callus area. | Delay (7 days) 252 |
| Sheep model with diaphyseal osteotomy (24 subjects). Axial movement: 1.5 mm displacement with preloaded spring with a force of 40 N; Shear movement: allowing 1.5 mm sliding between distal and proximal bone segment, with 2% rotational slackness. | Shear movement considerably delayed the fracture healing, with only 60% bridging of osteotomy fragments in the Shear group, whereas 100% in the Axial group. Peripheral callus formation in the Shear group also reduced to 64% compared to the Axial group. | Immediate 253 |
| Rabbit fix-sliding model with diaphyseal osteotomy (64 subjects). Rabbits in the Axial and Shear group were subjected to self-body-weight induced compression, with different directions restrained by the fix-sliding devices. | The shear movement led to superior healing 4 weeks after fracture but inferior outcomes 2 weeks after fracture compared to axial interfragmentary movement. | Immediate 255 |
| Mouse model with tibia osteotomy followed by intramedullary nailing (80 subjects). Cyclic compression loading (0.5, 1 or 2 N; 1 Hz for 100 cycles/day, 5 days/week for 2 weeks) | Compared to the immediate loading model, the low magnitude (0.5 N, 1 Hz for 100 cycles/day, 5 days/week for 2 weeks) axial cyclic compression with a short delay (4 days delay) significantly improved fracture healing with increased callus strength. Such improvement diminished with increased loading amplitudes (0.5–2 N). | Immediate and delayed (4 days) 258 |
| Mouse model with femur osteotomy (20 subjects). Cyclic strain (8–16 N, 10 Hz, 3000 cycles, 3 times/week for 4 weeks) | Cyclic strain applied on the mouse fracture model led to significantly higher callus formation and mineralization within the remodeling phase, which is associated with Wnt signaling activation and reduced distribution of sclerostin and RANKL in fracture callus. | Delayed (3 weeks) 259 |
| Human tibia bone surrogates model with transverse osteotomy*. Dynamic compression loading (150/200 N−1 Hz for 5 h) | Dynamic loading increased the transport of bone cells (280% for chondrocytes and 180% for osteoblasts) and growth factors (220% for chondrogenic growth factors and 120% for osteogenic growth factors) in the callus compared to the free diffusion. A moderate transport improvement was observed for the MSCs (22%) and fibroblasts (17%). | Simulated loading 260 |
| Sheep metatarsus fracture model*. Cyclic compression loading (0.02 mm displacement of amplitude; 1, 50, and 100 Hz for 15 min) | Mechanical loading with low amplitude and high frequency (0.02 mm displacement, 50 and 100 Hz) significantly improved the osteogenic activity of the callus. Interstitial fluid flow velocity was the only mechanical variable undergoing a significant increase in amplitude and peak value when the frequency of the external stimulus increased. | Simulated loading 261 |
| Sheep fracture model*. HF/LM Vibration (1% displacement, 90 Hz for 5 min, 2 times/day) | HF/LM-mediated improvement of bone formation could attribute to an increase in the interstitial fluid flow velocity, which promotes endochondral ossification, cell proliferation, migration, and ECM synthesis | Simulated loading 273 |
| Defect healing, bone loss regeneration, and implant healing | ||
| Mouse model with critical-sized tibia defect. The defective limb was subjected to daily loading of 5 N peak load, 2 Hz, 60 cycles for 4 consecutive days. | Loading during the inflammatory phase (post-surgery day, PSD 2–5) delayed hematoma clearance and bone matrix deposition and stimulated cellular proliferation and osteoclast activity. Loading during the matrix deposition phase (PSD 5–8) stimulated cellular proliferation and promoted cartilage and bone formation. Finally, loading during the remodeling phase (PSD 10–13) stimulated cellular proliferation and prolonged the remodeling phase. | Delayed (1 day) 256 |
| Rat model (156 subjects) with anterior cruciate ligament reconstruction. Cyclic axial loading (0.05 Hz, 30 g with 2.2% graft elongation tensile). | Delayed application of cyclic axial loading after anterior cruciate ligament reconstruction resulted in improved mechanical and biological parameters of tendon-to-bone healing, manifested by improved bone formation, less ED1+ inflammatory macrophages, more ED2+ resident macrophages, fewer osteoclasts, and reduced tissue vascularity. | Immediate, early delayed (4 days), and late delayed (10 days) 257 |
| Healthy rat ulna model (29 subjects). Static axial compression (8.5 and 17 N, 10 min/day for day 1–5 and 8–12) and axial dynamic compression (17 N, 2 Hz, 1200 cycles/day for day 1–5 and 8–12) | Static loading could not generate an anabolic bone response, whereas it suppresses skeleton growth. Dynamic loading significantly promotes periosteal and endocortical bone formation compared with static loading. | Immediate 266 |
| Healthy rat ulna model (29). Axial dynamic compression (haversine waveform at 17 N peak value, 360 ×1 cycles/day or 90 ×4 cycles/day, 3 days/week for 16 weeks) | The loaded ulnas showed 5.4% (360 ×1) and 8.6% (90 ×4) greater areal bone mineral density than the control. In addition, bone mineral content was enhanced by 6.9% and 11.7% in the 360 × 1 and 90 ×4 loaded ulnas. | Immediate. 269 |
| Mouse model with type I diabetes-induced bone loss (133 subjects). Cyclic axial compression (1.2–2.4 N, 2 Hz, 120 cycles/day for 3 days) | The combination of mechanical loading and PTHrP-derived peptide overcame the bone loss, fragility, and reduced mechanoresponsiveness caused by diabetes. | Immediate 262 |
| OVX Mouse model with undamaged bone (150 subjects). Compression loading (4 N, 10 Hz, 5 min/d for 2 weeks) was applied on the lumbar spine in the dorsal-ventral direction. | The loaded OVX mice showed a significant increase in the number of osteoblasts and a decrease in the number of osteoclasts via Wnt3a signaling. Spinal loading also elevated the volume of microvascular and VEGF levels. | Immediate (2 weeks after OVX surgery) 263 |
| Rat tibia model with titanium implant placement (77 subjects). Compression loading: HF/LM (40 Hz, 0.5 N); HF/HM (40 Hz, 1 N); LF/LM (2 Hz, 10 N); and LF/HM (2 Hz, 20 N). | Bone fraction and bone-implant-contact rate were increased at the cortical level in response to HF/LM and LF/HM loading. However, BIC at the medullar level was positively influenced only in response to HF-LM loading | Immediate 272 |
| Rabbit tibia model with titanium implants (4 implants/rabbit) placement (10 subjects). Movement loading (20 min daily treadmill running from 0 to 6 weeks) | The test group showed more significant vertical bone growth (1.26 ± 0.48vs 0.32 ± 0.47 mm, p < 0.001), higher ISQ values (11.25 ± 6.10vs 5.80 ± 5.97 p = 0.006), higher BIC (25.14 ± 5.24%vs 18.87 ± 4.45%), and higher bone neoformation (280.50 ± 125.40 mm2 vs. 228.00 ± 141.40 mm2, p = 0.121). | Immediate (with 2 weeks progressive adaption phase) 271 |
*
Finite elemental analysis and mathematical modeling.