Abbreviations BMD bone mineral density COX-2 cyclooxygenase-2 FDA Food and Drug Administration LIPUS low-intensity pulsed ultrasound MSC mesenchymal stem cells PGE2 prostaglandin E2 RANKL receptor activator of nuclear factor kappa-B ligand VEGF vascular endothelial growth factor

Post-fracture rehabilitation has been a major clinical burden worldwide, which results in limited mobility and weakened musculoskeletal function of an able body. It is considered as a complex orthopedic challenge, where insufficient and delayed treatment can cause numerous complications such as bone weakening, abnormal healing, and function losing.1 The duration of the patient's recovery from the fracture is determined by certain factors such as the site of injury, strength of impact, types of bone involved, and biological processes.2, 3 A fracture or a broken bone may result in nonunion, a condition of the bone that fails to heal even after 9 months of fracture; whereas delayed unions are defined as the condition of the bone that failed to show radiographic progression between 3 and 9 months after a fracture event.4 It is estimated that 10% of the people are facing problems related to nonunion and delayed unions in the United States alone.5 It was estimated that nearly 4.39 million people experienced fractures due to trauma in China in 2014.6 Even though a fractured bone can undergo self-regeneration, there are unhealed bones, with the cause remaining unknown.7 Many orthopedic surgeons are concerned about the fracture care of the patients.8 Traction is the most commonly used method classically in which alignment of bone is done by stretching certain parts around the broken areas. For some cases, immobilization of fractured bone is needed, by using braces, plaster casts, and splints to stabilize the broken bone. Furthermore, surgical insertion of metal plates, rods, and screws are used to stablilize and fix the bone soon after a severe fracture. Apart from these, treatments, analgesics, and anti-inflammatory medications are commonly used as pain relief and to suppress the inflammation caused by the injury.9-11 However, the choice of the treatment procedures for the fracture healing greatly depends on many factors, including the nature of fracture, location of the bone, type of fracture, and so on. Recent developments in stimulation technologies on bone fracture healing have increased significant interests of the researchers to find an effectual way to improve the healing and to speed up the recovery. Among them, ultrasound stimulation has been found to be a safe approach to accelerate the bone fracture healing process.12, 13 Studies reported that ultrasound stimulation enhances bone formation by accelerating the process at the inflammatory stage14-16 and subsequent phases of fracture healing.17, 18

Therapeutic Ultrasound for Fractured Bone

Ultrasound stimulation for the human body is safe and noninvasive. The therapeutic use of ultrasound started in the early 1930s.19 Initially, a frequency of 800 kHz and an intensity between 4000 and 5000 mW/cm2 were used in the treatment of neuralgia, myalgia, and other diseases. The higher ultrasound intensity caused more heating in biological tissues.19, 20 In the 1940s, ultrasound treatment was limited to treat young bones in humans and dogs.21 It was criticized that ultrasound treatment might induce bone damages until the discovery of Barth,22 who reported that a low dose of ultrasound did not affect the bone or surrounding tissues. Meanwhile, other early clinical studies revealed that ultrasound stimulation with higher intensity ranging from 5000 to 25000 mW/cm2 caused complications such as necrosis, ceased bone healing and formation of fibrous tissue.23, 24 It was also reported that ultrasound had the ability to stimulate osteogenesis.25 Maintz used ultrasound with low intensities ranging from 500 to 2000 mW/cm2 for treatment in limbs of rabbit and noticed the formation of new periosteal bone.26 The first successful formation of new callus in the fracture site was noticed using continuous ultrasound stimulation with an intensity of 1500 mW/cm2.27 In order to minimize the thermal effects on soft tissues, it was proposed to use low-intensity and pulsed ultrasound signals for stimulation, which resulted in bone growth in tibia of rabbits with an intensity of 200 mW/cm2.28 Later in 1983, Xavier and Duarte treated 27 nonunion cases with low-intensity pulsed ultrasound (LIPUS) and generated successful treatment results.29 When the nonunions were treated for 20 minutes per day for 18 days, they showed a success rate of 70%. The efficacy of LIPUS in fracture repair process with 38% of acceleration rate in tibial fractures was further demonstrated by Heckman and colleagues in 1994 with a spatial average-temporal average intensity of 30 mW/cm2.30

The first commercially available LIPUS product is EXOGEN® device, which was approved by the Food and Drug Administration (FDA) for treatment of fresh fractures in 1994. Kristiansen et al reported a shortened time to achieve bone union, acceleration of radiographic stage of healing, and a significant decrease in loss of reduction was achieved with LIPUS treatment.31 In the year 2000, FDA approved the use of LIPUS for nonunion treatment.32 The most commonly prescribed LIPUS bone growth stimulators are Exogen 4000+™, Exogen 3000™, Exogen 2000+™, and Exogen 2000™.33 It was reported that LIPUS with an ultrasound intensity of 30 mW/cm2 and a dosage of 20 minutes/day could accelerate the bone maturation in distraction osteogenesis in rabbits.34 LIPUS treatment healed 86% of cases in nonunions in an average of 22 weeks.32 Heybeli et al used low ultrasound intensity of 11.8 mW/cm2 and demonstrated increased bone density in rat femora as well as radiographic fracture healing.35 A further study reported by Rutten and colleagues showed LIPUS was effective in the treatment of established tibial fracture, for which they achieved 73% of overall success rate with 40% reduction in healing time.36 The studies conducted by Gebeaur et al and Nolte et al in the treatment of nonunion fractures have achieved 85 and 86% of success rate, respectively.32, 37 Because of the positive reports from the above studies, the usage of LIPUS on delayed unions and nonunions was supported by the UK National Institute for Health and Care Excellence (NICE) in 2010.38 The most widely used LIPUS parameters for stimulation consisted of an ultrasound frequency of 1.5 MHz burst waveform with 200 μs on and 800 μs off, signal repetition of 1 kHz, a spatial average-temporal average intensity (ISATA) of 30 mW/cm2, and a dosage of 20 minutes per day.12 Up to now, most of the experiments reported have used EXOGEN® devices approved by FDA, though there are other systems being used, with the intensity of the pulsed wave commonly below 100 mW/cm2.39

Role of LIPUS on Fracture Healing

LIPUS therapy has been widely accepted for enhancing endochondral bone formation.17, 34 It has also been demonstrated to increase the blood flow near the injured area18 and reduce the healing time in the cases of scaphoid fractures, tibial, and distal radius fractures.30, 31, 40 The therapeutic ultrasound used in LIPUS is harmless and does not require any subsequent surgeries,41 and higher efficiency of the treatment can be achieved when it is performed in the initial stage of the fracture impact. In addition, LIPUS treatment can be used along with the metallic fixtures, without causing any adverse side effects to the tissues.42 Apart from these, the clinical application of ultrasound therapy has been extended to healing of maxillofacial bones.43-45 LIPUS has shown positive effects on patients, irrespective of age, smoking, fracture gap, and absence of fibular fracture as well as distal fracture location.46 The acceleration of callus formation in diabetic fractures has also been reported.47 Konno et al further reported that LIPUS increased the acceleration of callus formation in the stimulation side compared with the nonstimulation side.16

Clinical Evidences on Fracture Healing Effects on Fresh Fractures

Fractures less than 1 week are considered as fresh fractures.48 It is clinically proved that LIPUS plays an important role in the fresh fracture healing with an ultrasound intensity of 30 mW/cm2.49 Some of the studies that showed effective results using LIPUS treatment on fresh fractures are listed in Table 1.

Table 1. Details of Studies About the Effects of LIPUS on Fresh Fractures Study Location of Fracture No of Patients (LIPUS Group) No of Patients (Control Group) Start of Treatment After Fracture (Days) Treatment Time per Day (Minutes) Treatment Period (Days) Maximum Healing Period (Days) Surgery or Not Patient Follow-up Rate Healing Time (Days) Major Findings Heckman 199430 Tibial shaft 33 34 4 ± 0.3 20 140 180 No 88%

Active group: 96 ± 4.5

Control group: 154 ± 13.7

Significant decrease in healing time, weight bearing not affecting efficacy of the results, smoking patients healed in 115 ± 11.2 days, (LIPUS accelerates bone formation even in smokers), 38% clinical and radiographic acceleration Kristiansen 199731 Radius 30 31 3 ± 0.4 20 70 128 No 75%

Active group: 61 ± 3

control group: 98 ± 5

Significant decrease in healing time, weight bearing excluded, smoking persons healed in 48 ± 5 days, 34–39% acceleration Emami 199850 Tibial shaft 15 17 4 20 75 361 Yes 100%

Active group: 155 ± 22

Control group: 125 ± 11

Healing time not reduced with intramedullary rods fixed on fractures, healing time reduced in smokers up to 40–50% Mayr 200040 Scaphoid 15 15 2 ± 3.5 20 42 43.2 ± 10.9 No 100%

Active group: 43.2 ± 10.9

Control group: 62 ± 19.2

Decrease in healing time, 31% of acceleration of bone formation, 70% of cases healed Leung 200451 Tibia 16 14 N/A 20 90 84 ± 14 Yes 100%

Active group: 66 ± 15

Control group: 110 ± 21

Disappearance of tenderness of fracture noticed at 6.1 ± 2.1 weeks, full weight bearing at 15 weeks Handolin 200552 Lateral malleolus 11 11 14 20 42 63 Yes 100%

Active group: 84

Control group: 84

Bone healing was assessed using multidetector CT, no significant effect on lateral malleolar fracture healing Lubbert 200853 Clavicle shaft 52 49 5 20 28 27 No 85%

Active group: 26.77

Control group: 27.09

Ability to return to work in 17 days, healing time not reduced Gan 201454 Metatarsals, fibula, tibia 10 13 N/A 20 28 84 No 100% N/A No significant changes in the MRI grading and bone marrow edema size measured. Busse 201655 Tibial shaft 250 251 N/A 20 365 N/A Yes 73% N/A 9% ability to weight bearing, reduction in healing time, fails in improving functional recovery with intramedullary nails LIPUS, low-intensity pulsed ultrasound; N/A, not available or not reported.

These studies demonstrated that LIPUS could effectively reduce the healing time in fresh fractures. One factor that affects healing process is smoking habit.31 LIPUS stimulation has shown acceleration of bone formation and reduction of healing time even in smoking persons.30, 31, 50 The meta-analysis conducted by Lou et al also suggests that LIPUS treatment has positive effects on adult fresh fractures.49 However, some patients were reported to have adverse effects with LIPUS treatment, such as muscle cramps, swelling in the cast, and skin irritation.30, 53 Furthermore, some studies reported that LIPUS did not reduce the healing time and functional recovery with metallic fixations.50, 52, 55 A recent review has suggested that LIPUS holds Grade B recommendation in case of fresh fracture healing.33 Furthermore, LIPUS is more suitable for nonoperative treatments because during the operative treatments there is a higher possibility of osteonecrosis.56 The studies listed in Table 1 show that the functional recovery of the LIPUS treatment has been sparsely studied. Future research should be focused on the treatment outcomes such as pain reduction, weight-bearing ability, and time to return to work.

Effects on Nonunions

Nonunions are the fractures that have failed to heal even after 9 months of the fracture event and with the possibility of healing with or without treatment intervention.4 Treatment of nonunions using internal or external fixations along with bone grafts is considered as the “gold standard.” The surgical treatments for nonunions can achieve success rates between 70 and 90%, depending on the location of the fracture and the type of treatment.41 The success rate for nonunions when treated with LIPUS is determined by three key factors, namely age of the fracture, size of the maximum gap, and stability of the fracture site.57 The detailed outcomes of LIPUS-treated nonunion fractures at various bone fracture locations are shown in Table 2.

Table 2. Details of Studies About the Effects of LIPUS on Nonunions Study Fracture Location No. of Patients Start of Treatment After Surgery (Days) Treatment Time per Day (Minutes) Treatment Period (Days) Max Healing Period (Days) Patient Follow-up Rate Overall Success Rate Major Findings Nolte 200258 Femur, tibia, radius, scaphoid 29 < 90 20 119 154 93% 91% Healing noticed in smokers, ability to bear weight without pain Gebauer 200537 Femur, tibia, radius, humerus 66 120 20 168 180 94% 85% Healing noticed in few smokers; higher success rate seen in ≤1 year of fracture time Rutten 200736 Tibia 71 90 20 160 179 98% 73% Healing rate in smokers 63%, nonsmokers 84%, better healing achieved if the treatment started within 3 months Jingushi 200759 Humerus, radius, femur, tibia, ulna 72 180 20 219 219 N/A 89.7% Higher efficiency noticed if the treatment started within 6 months after surgery Zura 201560 Multiple bones 767 365 20 179.5 179.5 ± 127.9 21% 86.2% Data from patient registry, higher success rate noticed in fractures >1 year Elvey 202061 Hand and wrist bones 26 84 20 104 365 100% 62% 2 nonunion cases had second surgery after 12 months of LIPUS treatment. No significant difference found in bone union rate of LIPUS vs surgery LIPUS, low-intensity pulsed ultrasound; N/A, not available or not reported.

Nolte et al reported that LIPUS treatment for established nonunions had no side effects.58 Ultrasound helps to achieve the bone union similar to surgical means without causing any complications to patients, which is particularly crucial to the elderly patients with low healing ability.37 It has been discovered that the efficiency of bone union using LIPUS treatment was closely related to the time duration between the most recent surgery and ultrasound treatment and it should be less than 3 months.36 Furthermore, it is showed to be more effective in the treatment of postoperative nonunions when LIPUS therapy was started within 6 months after the surgery.59 It was also reported that LIPUS was an effective approach for acute fractures like long bones, especially nonunions.62 It appears that better bony union would be achieved if the treatment is started at the right time after the fracture events. In addition to timing, a recent report on established nonunions demonstrated that LIPUS treatment on bone healing also depended on factors such as fracture type and treatment approach to the injury.63 Meanwhile, it was reported that LIPUS failed to promote the healing of nonunion fractures in the case of the fracture gap size greater than 1 cm, where the average healing time was 5.3 months, with only 20 out of 60 patients reported to have bone formation.64 The systematic review and meta-analysis conducted by R. Leighton et al supported the use of LIPUS treatment for nonunions as the average success rate was greater than 80%.65 In summary, LIPUS treatment appears to be a suitable substitute for the surgical therapy in nonunions for its lower cost and fewer complications. However, there was no report that LIPUS on nonunion cases show any improvement on weight-bearing ability, pain reduction, and time reduction in radiographic healing. Further research in this area needs to be conducted.

Effects on Delayed Unions

Delayed union can be referred to as the fractures that failed to show radiographic progression between 3 and 9 months.66 Similar to nonunions, the treatment is mostly preferred to be started within 6 months after the most recent operation to achieve better results.59 LIPUS effects on delayed unions are summarized in Table 3.

Table 3. Details of Studies About the Effects of LIPUS on Delayed Unions Study Fracture Location No. of Patients Start of Treatment After Fracture (Day) Treatment Time per Day (Minutes) Treatment Period (Days) Maximum Healing Period (Days) Patient Follow-up Rate Overall Success Rate Major Findings Schofer 201067 Tibial shaft 51 ≥120 20 112 112 90% N/A Increased BMD, reduced bone gap area Rutten 201268 Fibula 20 >180 20 150 365 100% 65% LIPUS decreased healing time by 29% Watanabe 201357 Long bones 101 90 20 Until healed 180 100% 74.3% Some bones unhealed due to instability, gap size not >9 mm Farkash 201569 Scaphoid 29 >90 20 67 67 N/A 76% Immobilization to be a major part of the treatment Teoh 201870 Metatarsal 30 101 20 75 88 ± 5.9 100% 90% Three patients found with nonunion even after LIPUS treatment BMD, bone mineral density; LIPUS, low-intensity pulsed ultrasound; N/A, not available or not reported.

LIPUS has been reported to improve bone mineral density (BMD), thus can be an adjuvant therapy after surgical intervention.67 One study has reported that LIPUS could achieve 74.3% bone union without any surgical procedures.