If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The burden of a long bone nonunion is substantial with pain, loss of function and psychological distress commonly encountered.
•
Nonunion in clinical practice is most commonly associated with long bone fractures of the forearm, humerus, tibia, clavicle and femur.
•
The majority of nonunions can be effectively managed with conventional surgical techniques.
Abstract
Nonunion following a long bone fracture causes considerable morbidity when it occurs. Risk factors depend on specific fractures but there is a complex interplay of injury severity, comorbidities, patient medication and infection. The majority of nonunions occur after long bone fractures with the tibia, femur, forearm, humerus and clavicle predominating. Despite interest in the biological augmentation of fracture healing, the majority of nonunions can be effectively managed with conventional surgical techniques. In this review we present a review of risk factors for nonunion and the outcome following surgical management.
]. The financial implications can often be extensive for patients due to loss of potential earnings and with an estimated medical expense of up to £79,000 per case being reported [
A precise definition of nonunion is difficult. According to the US FDA definition, a fracture ununited 9 months after injury or one in which there is a failure of progression towards union over the previous three months, can be classified as a nonunion [
]. However, 9 months is a long period of time for patients to wait for defining a nonunion. A specific duration of time to define nonunion must be related to the location of the fracture and the severity of the original injury, two factors which will have a profound influence on time to union. In general, lower energy closed fractures can be expected to progress to union in much shorter time than high energy open long bone fractures. Rather than relying on a specific time frame to define a nonunion it has been proposed that a more practical definition of nonunion is a fracture that will not unite without further intervention.
Traditional classification and treatment of nonunion patterns was based on plain radiographs. A hypertrophic nonunion is characterised by abundant callus formation and is usually considered to have excessive motion but good biological potential for healing at the fracture site. An oligotrophic nonunion has poor callus formation and is considered to have viable fracture fragments but has a combination of excess motion but also impaired biological potential for healing at the fracture site. An atrophic nonunion has no callus formation with no biological capacity to heal without some biological stimulus as part of the treatment, most commonly in the form of a bone graft [
]. These categories are no longer considered valid for a variety of reasons. Firstly, many fractures treated nonoperatively in the past are now treated by internal fixation and the classic radiographic patterns of hypertrophic and atrophic nonunions are less commonly seen. Secondly, histological studies of stiff atrophic and hypertrophic nonunions indicate no difference in the vascularity of the fracture site between the two types [
Current literature supports the view that the majority of long bone nonunions can be addressed with operative fixation to reduce the mechanical strain environment to promote union without the need for biological augmentation [
]. However, in certain nonunion cases careful consideration needs to be given to the biological factors affecting healing and risk factors in the host which may influence fracture healing. In this article we review the current evidence relating to the causes and results of conventional surgical management of long bone fracture nonunion.
A structured review of the relevant literature indexed in PubMed was undertaken with keywords “nonunion” or “delayed union” combined with “diagnosis”, “definition”, “risk factors”, “clavicle”, “humerus”, “forearm”, “femur” and “tibia”. For the tibia and femur these were also combined with “exchange nailing” and “circular frame”. Focus was placed on the contemporary surgical management of long bone nonunion in order to aid clinical decision making. The results are presented with the focus on risk factors and surgical treatment of specific fractures.
Risk factors for nonunion
Local factors
Establishing key risk factors to predict nonunion has been the subject of much interest in orthopaedic literature. Despite an abundance of research on the topic, there is controversy in relation to the relative importance of specific risk factors. This reflects widespread variation in study design, definition of nonunion, heterogeneity of fracture types and patient populations [
Probably the most important local factor affecting bone healing is the energy of the injury causing the fracture and the associated soft tissue injury. Open injuries with associated soft tissue stripping have an intrinsic risk of poor healing as reported for clavicle [
]. Bone loss as a result of open fracture is another risk factor associated with nonunion but this is not common and in one study was reported in 0.4% of all fractures requiring admission to a large tertiary trauma hospital [
]. Spontaneous healing can usually be expected with conventional fixation techniques if the defect is less than 2cm or involving no more than 50% circumference of the bone [
]. Although increasing displacement and the presence of comminution are often considered as risk factors for most long bone injuries, the evidence of this assumption rests largely on retrospective studies which are prone to bias [
]. The association of multiple fractures with polytrauma is thought to be an independent predictor for poor bone healing, although not completely understood, a persistent systemic inflammatory response and the association with high energy injuries is likely to explain this finding [
Individual risk factors for deep infection and compromised fracture healing after intramedullary nailing of tibial shaft fractures: a single centre experience of 480 patients.
Is There an impact of concomitant injuries and timing of fixation of major fractures on fracture healing? A focused review of clinical and experimental evidence..
J Orthop Trauma.2016; 30 (Lippincott Williams and Wilkins): 104-112
While the energy of the injury causing the fracture is important, it is clearly not the only element. Most clavicle fractures occur as the result of what would be considered low energy trauma, but in completely displaced fractures the nonunion risk is approximately 15% following non-operative treatment [
]. Similarly, many humeral diaphyseal fractures will occur as a consequence of low energy trauma but recent series report nonunion rates of up to 20% following non-operative treatment [
]. Perren's strain theory would suggest excessive motion interferes with fracture healing and it may reflect the difficulty of immobilisation of clavicle and certain humeral shaft fractures.
In any fracture complicated by nonunion following fixation infection should always be considered. Diagnosis is not straightforward as there may be few or no clinical signs to confirm infection and inflammatory markers such as C reactive protein may be normal. Clinical suspicion should be raised in cases where there is a history of initial superficial wound infections, there is persistent pain or swelling of the affected limb. Radiographs may also demonstrate findings consistent with infection such as progressive osteolysis and early failure of fixation. It has been estimated that 5% of nonunions are associated with an occult infection. At the time of any revision surgery samples should be sent both for microbiological and histological examination to look for infection in all nonunion cases [
Mills L., Tsang J., Hopper G., Keenan G., Simpson A.H.R.W.. The multifactorial aetiology of fracture nonunion and the importance of searching for latent infection. Bone Jt Res [Internet]. 2016 [cited 2020 Apr 23];5 Oct 1 (10):512–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27784669
Surgeons need to consider modifiable patient risk factors for nonunion as optimising these may improve the prospects of success with surgical intervention.
Most orthopaedic surgeons would regard age as an independent predictor of nonunion (82% from a recent survey) [
]. The evidence for this is poor and observational data from large epidemiological databases have found that nonunion incidence peaks around the age of 35 to 44 and decreases thereafter [
]. This may reflect the fracture epidemiology in certain age groups. For example, elderly patients have a high incidence of low energy fractures in metaphyseal bone such as distal radial fractures which are recognised as having a low risk of nonunion. By contrast, younger patients will have a higher incidence of high energy fractures.
Smoking is generally considered to have a negative effect on bone healing and this viewpoint is supported by most, but not all, studies which have evaluated this as a risk factor [
]. The adverse effects of smoking on union rates have been found in the clavicle, humerus, tibia, and femoral shaft. Studies that did not show an association with smoking and nonunion were characterised by small patient cohorts. Large systematic reviews and studies based on large databases clearly demonstrate that smoking is a risk factor for nonunion. In their systematic review which included over 6000 patients Scolaro et al reported an odds ratio for bone nonunion at 2.32 for current smokers against non-smokers [
]. These include a toxic effect of hyperglycaemia on osteoblasts, an osmotic diuresis of calcium associated with glucosuria, functional hypoparathyroidism and in some patients renal osteodystrophy [
]. As might be expected the risk is probably higher in patients with poor diabetic control and this has been shown in one study where raised haemoglobin A1c levels, were associated with complications including wound breakdown, loss of fixation and an increased risk of nonunion [
]. Better diabetic control may reduce the risk but in their large database study Zura et al found diabetes to be an independent risk factor irrespective of whether it was controlled or uncontrolled [
] it was not associated with an increased risk of nonunion. In a large database study of over 309,000 fracture patients, obesity was a risk factor for nonunion with an odds ratio of 1.19, so the effect would seem to be modest [
]. The increased risk associated with obesity may be explained, to some extent by the difficulty in achieving an appropriate strain environment for fracture healing with non-operative treatment in obese patients where adequate immobilisation of the fracture is difficult to achieve. However, there is increasing recognition of metabolic changes associated with obesity which may affect fracture healing. These include decreasing osteoblast differentiation and bone formation, an increase in tissue pro-inflammatory cytokines promoting osteoclast activity and bone resorption, and a reduction in intestinal calcium absorption, thereby decreasing calcium availability for bone formation [
]. Additionally, large body habitus is thought independently to influence fracture mobility and risk of nonunion with humeral shaft non-operative management [
]. It is possible that obesity effects the normal homeostasis of bone metabolism and fracture healing but this relationship is complex and remains incompletely understood [
Numerous medications have been associated with increasing the risk of nonunion. Steroids although often considered to negatively influence bone healing have no evidence of a significant effect from a recent review [
]. Non-steroidal analgesic consumption (NSAIDS) twelve months prior to a fracture have been associated with nonunion with an estimated odds ratio of 2.6 [
]. Specifically for long bone fractures, NSAIDS following post intramedullary nailing of femoral shaft fractures was found to be strongly associated with nonunion and delayed healing [
]. In an randomised controlled trial (RCT) designed to determine if 6 weeks of indomethacin was effective at reducing heterotopic ossification after acetabular fractures, a significant increase in nonunion was observed in the long bone fractures that these patients had also sustained [
]. However, a preclinical study in a rat model suggested that if NSAIDs were used for just for 48 hours post fracture, there was no detrimental effect [
] NSAIDS and opioid medications were both strongly associated with an increased risk of nonunion and weaker associations were noted with anti-convulsant medication and anti-coagulants. It is important to note that these associations may not be causal. In the case of NSAIDS and anti-convulsant therapy there are known effects on the cyclo-oxygenase pathways and bone metabolism respectively that might indicate a causal association. The association with opioid medication may simply reflect the greater likelihood that patients with nonunions will have a more prolonged treatment course possibly with multiple surgical interventions and will have a higher analgesic requirement than patients who heal uneventfully within expected timeframes. Comorbidities and chronic disease states including osteoporosis have been associated with poor bone healing although not universal to all long bone fractures with some contradictory findings in the literature [
The development of nonunion appears to be multifactorial with a complex interplay between the fracture, patient comorbidities, medications and social behaviours [
Mills L., Tsang J., Hopper G., Keenan G., Simpson A.H.R.W.. The multifactorial aetiology of fracture nonunion and the importance of searching for latent infection. Bone Jt Res [Internet]. 2016 [cited 2020 Apr 23];5 Oct 1 (10):512–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27784669
]. Despite the abundance of literature, accurate prediction of patients at increased risk of nonunion based on factors present from the time of injury remains challenging for all fractures. In the next section we consider the current literature on outcomes of conventional surgical management of aseptic diaphyseal nonunion.
Current evidence of the management for long bone nonunion
Clavicle
Older studies by Neer and Rowe reported very low rates of clavicular nonunion with rates less than 1% [
]. The majority tend to be symptomatic requiring operative intervention, but a fibrous union can be tolerated in the elderly or low demand patients.
Over the past decade there have been over a dozen randomized controlled trials that compared operative fixation to non-operative management for displaced mid-shaft fractures with no cortical contact. The nonunion risk with plate fixation was 1.4% vs 14% with non-operative management [
]. Targeted fixation of displaced fractures destined for nonunion would be advantageous given routine plate fixation of all displaced fractures is not cost-effective and potentially results in over treatment [
]. However these risk factors lack specificity and accurate prediction of patients who will go on to develop a nonunion is currently an unsolved problem [
] evaluating operative versus non-operative treatment of mid-shaft clavicle fractures, 14 studies with 1469 participants were included. The authors pointed out that that almost all of these studies had design features that carry a high risk of bias, which limited the strength of their findings. They were able to confirm that surgery did not appear to be associated with any functional benefit at one year and that the main source of treatment failure was mechanical failure (3.4%) in the surgery group and symptomatic nonunion (11.6%) in the conservative-treatment group.
There is increasing evidence that delayed primary fixation in the first three months post-injury is a valid alternative without an apparent increase in complications [
Open reduction and internal fixation of clavicular fractures after a delay of three months is associated with an increased risk of complications and revision surgery.
]. Delayed recovery following non-operative management may predict union and can be accurately judged on a patient reported QuickDASH score and other objective clinical features at six weeks post-injury [
Smoking status and the disabilities of the arm shoulder and hand score are early predictors of symptomatic nonunion of displaced midshaft fractures of the clavicle.
Displaced midshaft clavicle fracture union can be accurately predicted with a delayed assessment at 6 weeks following injury: a prospective cohort study.
J Bone Joint Surg Am [Internet].2020; (cited 2020 Feb 5];10.2106/JBJS.19.00955. Available from)
For established nonunion after non-operative treatment plate fixation is considered the treatment of choice. Historical case series with plate fixation and bone grafting suggested a low rate of complications with union achieved in the majority of cases [
Open reduction and internal fixation of clavicular fractures after a delay of three months is associated with an increased risk of complications and revision surgery.
]. Although bone grafting was traditionally used in clavicular fracture nonunion, more recent series suggest it is not required for the majority of cases [
Open reduction and internal fixation of clavicular fractures after a delay of three months is associated with an increased risk of complications and revision surgery.
The classic Sarmiento technique of functional bracing of humeral shaft fractures suggested nonunion is extremely rare occurring in only 2% of closed fractures and 6% in open injuries [
]. As with clavicle fractures more recent series have reported nonunion rates of approximately 20% for humeral shaft fractures managed non-operatively [
]. Purely transverse fractures have also been reported to have a higher incidence of nonunion, potentially because the fracture surface available for healing is small. Recent evidence suggests that fracture mobility at six weeks post non-operative management has good accuracy to predict nonunion with 82% sensitivity and 99% specificity when present [
Driesman A.S., Fisher N., Karia R., Konda S., Egol K.A.. Fracture site mobility at 6 weeks after humeral shaft fracture predicts nonunion without surgery. 2017 [cited 2020 Feb 9]; Available from: www.jorthotrauma.com
]. Trials comparing non-operative treatment of humeral shaft fractures with plating have reported lower nonunion rates with plate fixation but in longer term follow-up there does not appear to be any benefit in the functional outcome [
The most popular treatment of humeral nonunion has been compression plating, with or without bone grafting. The majority of humeral shaft nonunions are considered to be atrophic in pattern and therefore plating with additional bone grafting has been the most popular treatment. Although generally considered to be a successful procedure, post-operative complications can be encountered with a persistent nonunion rate of 16% [
] in certain case series. A meta-analysis of humeral nonunions identified 36 studies of 672 patients treated with plate fixation and bone grafting with a reported union rate of 98% [
]. This was superior to the results of intramedullary nailing with bone grafting (88% union rate, 164 cases) or without bone grafting (66% union rate, 71 cases). Only two studies reported the results of plating without bone grafting with a 95% union rate, but these studies combined only had 19 cases. For aseptic diaphyseal humeral nonunion, plating with bone grafting remains the treatment of choice based on current evidence.
Forearm
Skeletally mature patients with displaced diaphyseal forearm fractures will usually undergo plate fixation to minimise the risk of malunion associated with non-operative management. Open injuries with segmental defects are thought to be at increased risk of poor bone healing [
]. Successful union following plate fixation is thought to occur in the majority of cases with a reported union rates between 95 – 100% in most series with conventional plating techniques [
] reporting the results of plating for forearm nonunion there were 101 cases and only 3 persistent nonunions. These studies included 21 isolated radial nonunions, 42 ulnar nonunions and 38 nonunions involving both bones. Standard techniques using 3.5mm dynamic compression plating were used in most cases, generally augmented by bone grafting for cases judged to have oligotrophic or atrophic nonunion. Ring et al reported plating and bone grafting of 35 cases of forearm nonunions with segmental bone defects [
]. All fractures united within 6 months. In their series, they reported rather variable functional outcomes reflecting a cohort of patients with more complex fractures which had originally been open or had been complicated by infection.
In conclusion, forearm nonunion can be successfully treated by plate fixation in the majority of cases, irrespective of whether the nonunion involves radius, ulnar or both bones. Based on the published literature, additional bone grafting is advisable in any patients with bone loss or an atrophic pattern of nonunion.
Tibia
Nonunion following a tibial diaphyseal fracture is found in approximately 9% of cases when considering the spectrum of both closed and open injuries [
]. As expected open fractures have a higher incidence of nonunion and the risk correlates to the severity of the injury as classified by Gustilo and Anderson grading [
In modern orthopaedic practice, intramedullary nailing is the most popular initial management of unstable tibial diaphyseal fractures. When aseptic nonunion occurs, exchange nailing is a straightforward treatment with a high success rate. Although static versus dynamic screw configuration is not thought to influence success, a combined procedure with a fibular osteotomy has been found to reduce time to union [
] (Fig. 2). The need for exchange nailing is generally rare and in a large series of over 1500 tibia fractures, it was required in only 5% of all cases [
]. The indications and timing for exchange nailing must take into account the severity of the original injury. Most closed tibial diaphyseal fractures will heal between 12 and 16 weeks following injury. Absence of radiographic evidence of progression to union by 16 weeks is therefore a relative indication for exchange nailing. Open fractures, particularly type IIIA and IIIB, can be expected to unite in longer time frames, often taking longer than 20 weeks [
For uncomplicated nonunion without infection or bone loss in closed diaphyseal fractures the success rate of exchange nailing is high. Court-Brown et al reported on 33 cases of exchange nails for tibial nonunion with 100% success achieving union in closed, type 1, 2, and 3A open fractures [
]. The largest cohort in the literature of aseptic tibial nonunion was recently published by Hierholzer, they prospectively followed 188 patients and found 88% united after a single exchange nailing [
]. It is also thought that exchange nailing is superior to dynamization in the presence of comminution or persistent fracture gap from the initial procedure [
Tsang reported more variable outcomes however and found a 63% success rate of achieving union with a single tibial nail exchange procedure in 96 patients [
]. Their series included 31 infected nonunions and type III open tibial shaft fractures. For aseptic nonunions the success rate of an exchange nailing was 70%. Fracture union was ultimately achieved in 89 of 96 patients (92.7%) with one or more exchange nailing procedures. Risk factors for failure of exchange nailing included presence of infection, bone gap of more than 5mm, and an atrophic pattern.
The authors therefore advise that patients should be made aware that their nonunion is unlikely to unite with a single exchange nailing procedure, in the context of known infection or a sclerotic fracture [
]. If none of the 4 cortices have periosteal new bone formation near to the fracture site, it appears that repeated exchange nailing is unlikely to be successful and a different technique should be considered. Alternative strategies for an established septic nonunion include fine wire circular external fixation frame application combined with open debridement of the fracture, with or without acute shortening and ipsilateral corticotomy. This technique has been reported to have high union rates but has the morbidity associated with a prolonged frame time for patients [
]. Monolateral external fixator rather than the traditional circular frame may also be a reasonable alternative to undertake distraction osteogenesis [
Prospective randomized comparison of quality of regenerate in distraction osteogenesis of ring versus monolateral fixator in patients with infected nonunion of the tibia using digital radiographs and CT.
Bone Jt J [Internet].2019; 101-B (cited 2020 Oct 26Available from): 1416-1422
For uncomplicated aseptic tibial diaphyseal nonunion exchange nailing is the treatment of choice. For cases associated with bone loss and infection, more complex reconstructive techniques are often required. In the presence of known infection with a periosteal reaction adjacent to the fracture, repeated exchange nailing with antibiotic cover is still a feasible treatment option. If there is densely atrophic bone with no periosteal reaction adjacent to the fracture, then excision of the dead bone and limb reconstructive techniques with fine wire circular external fixation is the preferred option.
Femur
Reamed intramedullary nailing for femoral diaphyseal fractures has resulted in a low nonunion rate of 2% [
]. With regards to the technical aspect of the principal procedure, diaphyseal fractures with a nail diameter of less than 70% of the isthmus width has been correlated with need for exchange nailing [
]. Tsang showed in a series of 40 patients that underwent exchange nailing, with or without bone grafting, successful union was achieved in 96.2% of aseptic cases and 81.8% of infected nonunions [
]. However, in those with infection approximately, half required an additional surgical procedure. Other independent centres have reported a 96%-82% union rates following a single exchange nail for aseptic nonunion [
]. Additionally, an atrophic pattern, open fracture, associated ipsilateral upper limb injury and lack of immediate weight-bearing after index procedure have also been associated [
Dynamization is occasionally attempted as a less invasive method to achieve union in femoral diaphyseal fractures with an axial stable fracture configuration. A recent review suggested this has significantly inferior success to achieve union when compared to exchange nailing [
Uniting the un-united: should established non-unions of femoral shaft fractures initially treated with IM nails be treated by plate augmentation instead of exchange IM nailing? A systematic review..
Strateg Trauma Limb Reconstr.2018; 13 (Springer-Verlag Italia s.r.l.): 119-128
Uniting the un-united: should established non-unions of femoral shaft fractures initially treated with IM nails be treated by plate augmentation instead of exchange IM nailing? A systematic review..
Strateg Trauma Limb Reconstr.2018; 13 (Springer-Verlag Italia s.r.l.): 119-128
). Combination of nail exchange with an additional compression plate has also been found to be a useful solution in complex revision cases for failed constructs or nonunion [
Similar to that of the tibia, in the presence of established infected nonunion exchange nailing alone requires caution. A staged approach with removal of hard ware, debridement of the fracture site, local and systemic antibiotics followed by delayed definitive fixation has been proposed but can cause prolonged patient morbidity [
]. Exchange nailing with local and systemic antibiotic cover is a reasonable option in certain cases but patients should be warned they may need more than one exchange nailing procedure [
Biological augmentation of nonunion needs to be considered with a mobile atrophic pattern of nonunion or where there is bone loss. Autologous bone grafting remains the preferred option as it has osteoconductive, osteoinductive and osteogenic properties. It is still the most commonly used method of biological augmentation. The main drawback is the well-documented morbidity associated with harvesting bone graft from the iliac crest. As an alternative bone graft can be obtained from the medullary canal of long bones using the RIA (reamer/irrigator/aspirator) which enables bone graft to be collected during reaming of a long bone [
]. More recently in a pre-clinical model, similar results to bone grafting have been reported with autologous bone marrow aspirate suggesting this could be a less morbid adjunct to healing [
There has also been much interest in synthetic alternatives. Urist identified osteoinductive properties of demineralised bone matrix and discovered this property was due to the presence of bone morphogenetic proteins [
MicroRnA-186 improves fracture healing through activating the bone morphogenetic protein signalling pathway by inhibiting sMAD6 in a mouse model of femoral fracture.
]. A number of these proteins, notably BMP-2 and BMP-7, were subsequently marketed as alternatives to autologous bone grafting but these materials have fallen in popularity. A recent review found the current evidence of BMP use for long bone fractures, mainly in open tibias, had conflicting results of benefit and currently is not routinely used in clinical practice in the U.K. [
]. Several other pharmacological bone inducing agents exists but currently the evidence of their cost-effectiveness and clinical role is yet to be clearly established [
Nonunion remains a challenge to predict and common risk factors need to be evaluated critically for a specific fracture. Risk factors such as smoking, alcohol intake, non-steroidal use and diabetic control are amenable to modification and this may improve the prognosis for treatment. Infection as a contributory cause should be considered particularly after open fracture or in fractures initially treated surgically which fail to heal. When nonunion is encountered there is good evidence available to guide decision making and successful union can usually be achieved with conventional trauma fixation techniques in the majority of cases. Biological augmentation of fracture healing is not necessary in most cases but needs to be considered in higher risk cases with a mobile atrophic pattern.
Declaration of Competing Interest
There are no conflicts of interest to declare for any of the authors involved in this study.
Individual risk factors for deep infection and compromised fracture healing after intramedullary nailing of tibial shaft fractures: a single centre experience of 480 patients.
Is There an impact of concomitant injuries and timing of fixation of major fractures on fracture healing? A focused review of clinical and experimental evidence..
J Orthop Trauma.2016; 30 (Lippincott Williams and Wilkins): 104-112
Mills L., Tsang J., Hopper G., Keenan G., Simpson A.H.R.W.. The multifactorial aetiology of fracture nonunion and the importance of searching for latent infection. Bone Jt Res [Internet]. 2016 [cited 2020 Apr 23];5 Oct 1 (10):512–9. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27784669
Open reduction and internal fixation of clavicular fractures after a delay of three months is associated with an increased risk of complications and revision surgery.
Smoking status and the disabilities of the arm shoulder and hand score are early predictors of symptomatic nonunion of displaced midshaft fractures of the clavicle.
Displaced midshaft clavicle fracture union can be accurately predicted with a delayed assessment at 6 weeks following injury: a prospective cohort study.
J Bone Joint Surg Am [Internet].2020; (cited 2020 Feb 5];10.2106/JBJS.19.00955. Available from)
Driesman A.S., Fisher N., Karia R., Konda S., Egol K.A.. Fracture site mobility at 6 weeks after humeral shaft fracture predicts nonunion without surgery. 2017 [cited 2020 Feb 9]; Available from: www.jorthotrauma.com
Prospective randomized comparison of quality of regenerate in distraction osteogenesis of ring versus monolateral fixator in patients with infected nonunion of the tibia using digital radiographs and CT.
Bone Jt J [Internet].2019; 101-B (cited 2020 Oct 26Available from): 1416-1422
Uniting the un-united: should established non-unions of femoral shaft fractures initially treated with IM nails be treated by plate augmentation instead of exchange IM nailing? A systematic review..
Strateg Trauma Limb Reconstr.2018; 13 (Springer-Verlag Italia s.r.l.): 119-128
MicroRnA-186 improves fracture healing through activating the bone morphogenetic protein signalling pathway by inhibiting sMAD6 in a mouse model of femoral fracture.
30955-4&id=gr1.jpg){kind=link}
30955-4&id=gr2.jpg){kind=link}
30955-4&id=gr3.jpg){kind=link}