Keywords

Introduction

Since the 1950s and following the introduction of fracture fixation techniques by the AO group in Switzerland, there has been a revolution of implant designs to allow fixation/reconstruction of fractures of all different anatomical areas of the human body [1]. Both internal and external fixation implants with or without specific anatomical profiles are currently being used in the clinical setting [1].

The objective is that the implant selected to stabilise the injured limb will provide adequate fracture stability to obtain bony union, and restore the affected limb axis, rotation, length and joint congruence [2]. It is anticipated that the implant will provide the appropriate biomechanical environment to allow fracture healing and then no longer be needed for physiologic loading. While implants have been divided to load sharing (Intramedullary nailing) and load bearing (plating systems; locking and non-locking) devices, both are at risk of failure prior to the fracture uniting.

The aetiology of metal work failure is multifactorial including selection of wrong implant, sub-optimal fixation technique, non-compliant patient, fragile bone, non-union and infection amongst others [3,4,5].

Although metal work failure post fracture fixation is infrequent, the overall incidence of this phenomenon is not well reported in the literature. Herein, we report the incidence of fixation failure prior to fracture union in different anatomical sites of the human body.

Proximal Humerus

Proximal humeral fractures are the third most common non-axial osteoporotic fracture, affecting 63/100,000 persons [6]. They most commonly affect elderly females sustaining these injuries from low-energy falls [7]. The majority of humeral fractures are low energy with low rates of non-union and can be managed non-operatively [8]. When operative treatment is planned, this can be either in the form of fixation or arthroplasty.

The use of locking plates has expanded the role of fixation of proximal humerus fractures, gaining better purchase and fixation in osteoporotic bone. Despite this the failure of these devices continues to be reported in between 7 and 14% of cases [9,10,11,12,13]. Factors associated with the loss of reduction when using locking plates include increasing patient age, presence of osteoporosis, initial varus displacement, degree of reduction achieved, residual varus following fixation and medial comminution [10, 11]. The reported rate for fixation failure in a recent systematic review examining the role of intramedullary nails in the management of proximal humerus fractures suggests a failure rate of up to 24%, with risk factors for failure including the use of this device in three and four-part fractures in addition to the aforementioned risk factors [14].

Humeral Shaft

Humeral shaft factors account for between 1 and 5% of all fractures, with an incidence between 13 and 20/100,000 patients [15]. They have a bimodal distribution with an initial peak in young men between the age of 21 and 30 years, often as a result of high-energy trauma; and a second peak in elderly females between 61 and 80 years, more commonly in the setting of low-energy injuries [15]. Operative management can consist of either plate fixation or fixation with an intramedullary nail, and is utilised in up to 60% of cases [16].

Failure of plate fixation is rarely reported, with small series reporting fixation failure in 4–6% of cases, most commonly associated with osteoporotic bone, short plate span and an early return to weight-bearing activities [17,18,19]. Similarly low rates of fixation failure are quoted for intramedullary nailing [19].

Distal Humerus

Distal humeral fractures represent one-third of all humeral fractures with an incidence of 6/100,000 patients [20]. As with humeral shaft fractures they have a bimodal distribution with young men sustaining high-energy fractures, and older women sustaining low-energy injuries [20]. Operative treatment is associated with good clinical outcomes, and therefore the role of non-operative management is reducing, generally restricted to undisplaced fractures or those who are not medically fit enough to undergo anaesthesia [21].

When fixation of distal humeral fractures is selected over arthroplasty options, dual plate fixation, either in a parallel or a perpendicular configuration, is generally undertaken. Fixation failure is reported to occur in between 0 and 27% of these cases [22]. Osteoporosis represents a significant risk factor for failure of fixation, and in its presence consideration should be given to the use of arthroplasty [23]. Other risk factors for failure include the use of perpendicular plating, metaphyseal comminution, inadequate volume of screws in the distal segment, usage of short screws in the distal segment [22, 24, 25].

Olecranon

Olecranon fractures are common injuries sustained in the elderly population, with an incidence of 15/100,000 patients [26]. As intra-articular fractures, an operative approach is generally recommended unless the fracture is undisplaced. In those <65 years, an operative approach is taken in 79% of cases, with this tactic reducing in the over 65 s at 65% [27]. Popular techniques for fixation of these fractures include the use of plate fixation, and tension band wiring [27].

Tension band wiring of olecranon fractures is appropriate with simple fracture patterns in the absence of comminution [28,29,30]. Failure of this technique is reported in between 4 and 16% of cases. Factors that appear to be most associated with failure include the placement of intramedullary wires as opposed to bicortical hold, the use of single knot constructs as opposed to dual knot techniques and failure to adequately secure the proximal end of the K-wire [31,32,33].

Plate fixation is often utilised in the context of increasing patient age, and increasing complexity of fracture pattern. When utilising plates, current failure rates are quoted to be between 3% and 17% [29, 34, 35]. Prior to locking plate technology the majority of plate fixation would be with the limited contact dynamic compression plate (LCDCP), with failure occurring through screw pull-out [36]. The advent of locking systems specifically for the olecranon has reduced this occurrence, though these constructs may still fail in severely osteoporotic bone, and in highly comminuted fractures [37].

Radial Head

Radial head fractures affect 11/100,000 persons, most commonly females in their 60s [38]. Trends towards operative treatment of these fractures are increasing from 69% in 2007, to 85% in 2016 [38]. This is most commonly performed using screw fixation, although plate fixation and radial head replacement remain options for more comminuted fractures.

Screw fixation is rarely associated with failure, reported in between 0 and 15% of cases [39,40,41,42]. Reported risk factors for fixation failure include the presence of osteoporosis, development of non-union, multifragmentary fractures and the use of convergent screw orientations [41, 43]. Plate fixation is less commonly utilised when compared to screw fixation, and as a result there are no clear data available reporting the rates of fixation failure in this cohort.

Forearm

Whilst the highest rates of forearm fractures occur in children, there is a significant increase in these injuries in women aged over 45, and men aged over 70 [44]. The true incidence is poorly defined, but thought to be between 1 and 10/100,000 persons [45]. An operative approach to management is generally advocated due to the risk of non-union, mal-union and subsequent difficulties with forearm rotation [46]. This is most commonly achieved with plate fixation in the adult population.

Failure of fixation is rare in this cohort, reported in just 2–4% of cases [47, 48]. As with many fracture types, the presence of comminution poses a risk of fixation failure. Additional risk factors include failure to provide compression to the fracture, and the use of short plates which has been demonstrated to be of a higher importance than the number of screws utilised in each segment [48].

Distal Radius

Distal radius fractures represent the most commonly sustained fracture seen by orthopaedic surgeons with an incidence of up to 195/100,000 persons in the United Kingdom [6]. They are increasingly frequently seen in female patients over the age of 60 as a result of a fall from standing height [49]. Extraarticular distal radius fractures that maintain an acceptable alignment can be reliable managed non-operatively; however, displaced fractures or those that extend into the joint surface require fixation. Currently between 14 and 16% of distal radius fractures are managed operatively, most commonly by plate fixation (62%) followed by K-wire fixation (30%) [50, 51].

Modern distal radial plate designs have expanded the scope of fixation including more reliable use in osteoporotic bone and distally based fractures. Within the current literature, the failure rates are noted to be between 1 and 13% [52,53,54]. Failure rates are reported to be higher in the setting of early return to weight-bearing, close proximity of the fracture to the volar rim with little plate coverage of the unstable fragment, multifragmentary volar rim fractures (AO23-B3), smaller width of the lunate fragment piece, greater ulnar variance on the pre-operative imaging and failure to achieve adequate articular reduction (Fig. 1.1) [54,55,56,57].

Fig. 1.1
8 radiographs of the hand of a patient with a fracture in the ulna bone in different views. Scans 1 and 2, exhibit the fracture. The internal fixator with K-wires is prominent in scans 3 and 4. The internal fixator fixed with screws is prominent in scans 5 to 8.

A 42-year-old lady was involved in a rollover RTC sustaining an isolated closed distal radius and ulna fracture. She was taken to theatre on the morning following admission where following bridge plating of the ulna, her swelling did not allow for a second approach to the radius which was, therefore, managed with K-wires with a good intraoperative result. Unfortunately, she did not attend her early follow-up and returned at 6 weeks with a pin site infection, and a significant loss of reduction of the intermediate column of her wrist, resulting in incongruency of both her radiolunate and distal radioulnar joints. Given the concerns surrounding infection, she was treated with 4 weeks of antibiotics in order to suppress the infection until radiological union was achieved. She subsequently underwent removal of metalwork from the ulna accompanied by wrist denervation; however, she subsequently never returned for her planned ulna shortening and corrective radial osteotomy

Full size image

Distal Ulna

Distal ulna fractures frequently occur in conjunction with distal radius fractures, with an incidence of 3.8/100,000 persons [58]. The majority of distal ulna fractures can be managed non-operatively, particularly when screened to be stable following the fixation of a distal radius; however, when fixation is pursued, this is most commonly in the form of a plate [59].

Outcome of distal ulna fixation is significantly less frequently reported when compared to the distal radius. In those small series, assessing the outcome of fixation of the distal ulna the reported failure rate is 0%. These studies frequently don’t examine the ulna in isolation, having been fixed in conjunction with fixation of the distal radius [60,61,62,63]. Whilst clinical data do not currently exist, finite element analysis would suggest that the fixation is under the lowest stress when placed on the dorsal surface of the ulna, with three points of distal fixation [64].

Pelvic Ring

Pelvic ring fractures have an incidence of 23/100,000 persons, with a bimodal distribution affecting young males with high-energy mechanisms, and elderly females with low-energy falls [65]. Operative fixation of pelvic ring injuries is infrequently performed, selected in just over 8% of cases [66]. When operative management is selected, this is frequently a combination of percutaneous screw fixation with open reduction and internal fixation with plates, or use of anterior external fixation [66].

Failure of plate fixation is commonly reported, although frequently asymptomatic. Rates of failure are reported in between 5 and 46% of patients; however, less than 10% of these are symptomatic and require reoperation [67,68,69,70,71]. Risk factors for failure of anterior plate fixation include the use of the technique in osteoporotic bone, the use of a single implant as opposed to dual implant and the use of fewer than 3 holes per segment when spanning the symphysis (Fig. 1.2) [66, 68].

Fig. 1.2
A pelvic radiograph reveals a failure of the matta plate that spans the symphysis. A long screw-like structure extends from the ilium on both sides and connects at the center.

Anteroposterior (AP) pelvic radiograph demonstrating a broken 3.5 mm matta plate. As a plate that spans the symphysis, this construct is continually exposed to bending stresses that lead to plate failure by fatigue (in this case the residual pubis diastasis that developed after failure of the plate did not require any further intervention as the patient was asymptomatic)

Full size image

Similarly high rates of fixation failure when employing the technique of anterior external fixation are also reported, in between 23 and 57% of cases [72, 73]. Risk factors for failure of this technique include initial fracture displacement, inadequate reduction particularly in the setting of vertical shear injuries, fixator loosening and the use of this technique in lateral compression type injuries [72, 73].

Fixation of the posterior pelvic ring, typically achieved with percutaneous sacro-iliac (SI) screws, has much lower reported failure rates, occurring in between 4 and 16% of cases [74, 75]. Risk factors for failure of this technique include non-union, intraoperative malpositioning due to either surgeon error or inadequate fluoroscopy, use of a single screw as opposed to two SI screws and patient non-compliance with post-operative weight-bearing instructions [74, 75].

Acetabulum

Acetabular fractures are less commonly seen when compared to pelvic ring fractures, with an incidence of only 3/100,000 [76]. In contrast to pelvic ring injuries, they are more frequently observed in males, often as a result of a high-energy injury [77]. As an articular injury, an operative approach is more readily pursued when compared to the pelvic ring, across both the elderly and the non-elderly population [78]. Where fixation is performed, this is most commonly a combination of screw and plate fixation [78].

Failure of fixation is variably reported in the literature with many studies not directly commenting of fixation failure and instead reporting on rates of conversion to total hip arthroplasty (THA). Within the literature, the reported failure rate varies from 10 to 57% [79,80,81,82]. Risk factors for fixation failure in this population include increasing age, development of non-union, fracture comminution, initial articular displacement, inability to attain an anatomic articular reduction, fracture classification as an associated type particularly T-type with posterior wall involvement, obesity and surgeon error in siting the fixation device [83,84,85].

Proximal Femur

Proximal femoral fractures represent the second most commonly sustained osteoporotic fracture with an incidence of 129/100,000 persons [6]. The majority of these fractures affect the intertrochanteric region (60%), with 32% affecting the femoral neck, and 8% affecting the subtrochanteric region [86]. Management is almost exclusively operative unless the patient is unable to undergo an anaesthetic. Fixation is dependent on the location of the fracture and the degree of comminution, however, frequently involves the use of cannulated screws, a sliding hip screw, or a cephalomedullary nail [87].

Failure of fixation should generally be divided between those implant systems utilised in the management of intracapsular and extracapsular fractures. With regard to intracapsular fractures, the three most commonly utilised systems include the femoral neck system, cannulated screws and the dynamic hip screw with a derotation screw. The failure rates of the femoral neck system is currently reported in between 4 and 6% of cases; however, there is little literature examining this relatively novel implant [88, 89]. Failure rates of cannulated screw fixation are reported in between 13 and 39% cases, compared to failure rates between 0 and 20% when using a dynamic hip screw [90,91,92,93,94,95,96]. Risk factors for failure when managing intracapsular neck of femur fractures include increasing age, initial displacement, technical error in siting the implant, inadequate reduction, inferior cannulated screw distance >3 mm from the calcar, cannulated screw configuration (inverted triangle reduces in lowest failure rate) and a delay to fixation of greater than 24 h [88, 90, 97].

When considering extracapsular neck of femur fractures, the most commonly utilised fixation systems include the dynamic hip screw, and cephalomedullary nails. The rate of fixation failure utilising the dynamic hip screw is reported in between 4 and 28% of cases, whilst the rates of failure with an intramedullary nail are reported in between 0 and 13% of cases [98,99,100,101,102,103,104,105]. Risk factors for failure of fixation in extracapsular neck of femur fractures include increasing age, initial displacement, comminution, inadequate reduction, surgeon error, unstable fracture patterns (A2 or A3 compared with A1), comminution of the lateral cortex, calcar tip apex distance, notching of the screw aperture and reduction in a varus alignment (Fig. 1.3) [98,99,100,101, 106, 107].

Fig. 1.3
6 radiographs of the pelvis. Scan 1 exhibits a fracture in the proximal femoral. The cephalomedullary nail fixed over the fractured site is prominent in scans 2 and 3. Scans 4 and 5 reveal the failure of the nail, while scan 6 indicates the proximal femoral replacement on the right leg.

Initial AP pelvic radiograph demonstrating a subtrochanteric proximal femoral fracture in a 74-year-old male that was managed with a cephalomedullary nail. As can be seen, the reduction was not anatomic with residual translation in the sagittal plane, and a degree of malalignment of the medial calcar. The patient represented at 2 months with varus collapse and failure of the nail through the lag screw aperture. This was successfully managed with a proximal femoral replacement to facilitate early patient mobilisation and rehabilitation

Full size image

Femoral Shaft

The worldwide incidence of femoral shaft fractures ranges between 10 and 21 per 100,000 per year [108, 109]. They have a bimodal distribution affecting young males with high-energy mechanisms, and elderly females with low-energy falls [108]. These fractures are almost exclusively managed operatively. Operative fixation with intramedullary nailing is the gold standard of treatment; however, in transverse fracture patterns use of plate fixation is also observed [110].

The incidence of nail failure is low, reported in between 0.5 and 10% of cases [111, 112]. This is lower than those failure rates seen with plate fixation, which is reported in 1 and 14% of cases [113,114,115]. Risk factors for failure of femoral shaft fixation include undersising of the nail diameter, failure to lock nail, malreduction, comminution, degree of initial displacement, soft tissue stripping, development of delayed union, sagittal plane malalignment and the use of a short fixation working length when utilising a plate (Fig. 1.4) [114, 116].

Fig. 1.4
10 scanned images of the right femoral. Scans 1 and 2 reveal the midshaft femoral fracture. Scans 3 to 5 exhibit that the fractured site is fixed with a compression plate. Scan 6 depicts a failure of the compression plate. Scans 7 to 10 exhibit that the fractured site is fixed with a long rod-like device.

Initial AP and lateral radiograph demonstrating a transverse midshaft femoral fracture in a 12-year-old boy. This was managed with compression plating performed via a lateral approach, as can be seen from the operative films the plate was not pre-contoured, and whilst a good reduction was achieved, there is still some residual gapping on the medial surface. The patient represented at 6 months post-operative with increased pain and swelling of the mid-thigh. Radiographs taken at the time demonstrated evidence of a hypertrophic non-union and breakage of the plate through fatigue. The fixation was removed and an antegrade trochanteric entry nail performed which went onto uneventful union

Full size image

Distal Femur

Fractures of the distal femur are rare with a reported prevalence of 0.5% of all fractures; they have been slowly increasing in incidence over the past decade with most reported incidence of 8.7/100,000 person per annum [117]. These have been reported traditionally as fragility fractures and the increasing incidence is likely due to a shift towards an aging population worldwide. Distal femur fractures have a bimodal distribution, with patients either being young adults involved in high-energy trauma or elderly osteoporotic individuals who experience a fall from standing.

The most common fracture types are the 33-A1 or 33-A2. Type 33-C (complex articular fracture) is less common. Management is dependent on stability of the fracture pattern, involvement of the knee joint as well as patient-related factors. Where operations are deemed necessary, fixation is dependent on the location of the fracture and the degree of comminution. This normally involves the use of plate fixation (fixed angle blade plate vs. buttress plate vs. locking plate) or intramedullary nailing (antegrade vs. retrograde) [118,119,120].

The use of locking plates expanded the role of fixation within the distal femur, gaining better purchase and fixation in osteoporotic bone. Despite this, the failure of these devices has been reported in between 6 and 20% of cases [121, 122]. Factors associated with the loss of reduction when using locking plates include increasing patient age, presence of osteoporosis, initial varus displacement, poor initial reduction achieved, residual varus following fixation and medial or posteromedial comminution [121].

Proximal Tibia

Tibial plateau fractures account for 1% of all fractures and are typically sustained with high-energy mechanisms. The incidence of tibial plateau fractures is 10.3 per 100,000 people annually [123]. They have a bimodal distribution with an initial peak in men younger than 50, often as a result of high-energy trauma; and a second peak in elderly females between years, more commonly in the setting of low-energy injuries leading to tibial plateau insufficiency fractures [123]. In intra-articular fractures, an operative approach is generally recommended unless the fracture is undisplaced. This can be either through the use of plates and screws, external fixator devices or alternatively arthroplasty [124, 125].

Failure of plate fixation has been reported, with small series reporting fixation failure in 30% of cases, most commonly associated with osteoporotic bone, fracture fragmentation and an early return to weight-bearing activities [126]. Failure of fixation elements when utilising a circular fixator is reported in 14% of cases [124].

Tibial Shaft

Tibial shaft fractures are common long bone injuries accounting for 2% of all adult fractures [127]. They have an incidence of 2/100,000 population with a bimodal distribution of peaks at ages 20 and 50 [128]. These injuries may be managed non-operatively if minimally displaced, alternatively they can be treated with Intramedullary nail fixation, external fixator devices or plate osteosynthesis [129]. A cross-sectional survey performed showed that 80% of surgeons treat these Injury patterns with operative intervention [130].

Intramedullary nail fixation failure has been listed as approximately 7.3% [131]. These patients have a higher percentage of open injuries with a higher degree of comminution and had been treated with smaller diameter nails when compared with the group of patients, who had no implant failure. Failure occurred most frequently at the transverse proximal locking screw when a single screw was used [131]. Failure of circular frames is infrequently reported, with most ‘failures’ constituting broken wires which do not necessarily require intervention in 0–5% of cases [132,133,134].

Distal Tibia

The incidence of distal tibia fractures is estimated to be 9.1/100,000 persons per annum [135]. Women appear to have an increasing incidence of distal tibia fractures when stratified by age whilst males have a fairly constant incidence [135]. Distal tibia fracture can be treated with a variety of operative treatment methods including external fixators, intramedullary nailing and internal plate fixation [136,137,138]. Of these fractures there is a reported incidence of 6.9/100,000 distal tibia fractures which are subsequently operated on [139].

Pilon fractures often pose challenging fracture configurations to adequately reduce. There is limited literature assessing failures of differing treatment modalities. Studies suggest a rates of fixation failure between 2 and 10% when utilising plate fixation, and 3% when utilising a circular frame [136, 140,141,142,143]. Most commonly cited issues include malreduction of the fracture site and there has been reported to be an association between the use of anteromedial plates and non-unions [140]. Further risk factors include the presence of comminution and periosteal stripping, often seen in open injuries (Fig. 1.5) [141, 142].

Fig. 1.5
11 radiographs of a right leg with fractures in the pelvic, femur, and tibia. 1 to 4. They exhibit the fractured site. 5 and 6, depict that the fractured site is fixed with an internal fixator, while 7 and 8 reveal their failure. 9, depicts a leg with an external fixator. 10 and 11 exhibit a leg with no fractures.

A 34-year-old pregnant lady presented having been crushed between a van and a car. Her injuries included a lateral compression pelvic fracture, a left distal femur fracture and a right open tibial fracture. Following resuscitation, she was taken to theatre for caesarean section, pelvic fixation and debridement of her open tibial fracture with application of an ankle spanning external fixator. Two days following admission, she was returned to theatre for anterolateral plating of her distal tibial fracture and insertion of an antibiotics impregnated cement spacer with plans to reconstruct her bone defect via the Masquelet technique. She had her second stage Masquelet treatment at 6 weeks post first stage. Unfortunately, her graft failed to fully incorporate resulting in a distal tibial non-union and her plate failed via fatigue at 8 months post second stage. This was successfully managed with bone transport

Full size image

Ankle

Ankle fractures, accounting for 3.9–10.2% of adult fractures, are the most common type of fracture of the lower extremity [144]. They have an incidence rate of 100/100,000 people per year, with the majority occurring secondary to low-energy falls (55%) [6, 145]. Operative management is dependent on the fracture configuration as well as patient-related factors. It may consist of either plate fixation or fixation with an intramedullary nail (Fibular nails/Hind foot nails).

The use of locking plates has significantly expanded the role of fixation within the ankle, gaining better purchase and fixation in osteoporotic bone, leading to a change in treatment paradigm in geriatric ankle fractures with few fixation failures reported. Surgical re-intervention has been reported to range between 1 and 2% [146]. The most common indication for surgical reintervention was syndesmotic malreduction (59%) in a cases series published. This is often secondary to fibula shortening leading to lateral translation with a potential rotational malalignment of the syndesmosis [146]. Furthermore, the importance in reduction of the posterior malleolus has also been shown in biomechanical studies to affect the syndesmosis. Other risk factors for failure fixation include obesity, inability to follow post-operative weight-bearing instructions and the presence of open fractures (Fig. 1.6) [147].

Fig. 1.6
8 radiographs of the right ankle of a patient. Scans 1 and 2 reveal a trimalleolar ankle fracture in different views. Scans 3 to 6 exhibit that the fractured site is fixed with a fibula locking and tubular plates. Scans 7 and 8 depict that the fractured site is fixed with a hindfoot nail.

AP and lateral radiograph of a trimalleolar ankle fracture in a frail 53-year-old female that was managed with open reduction and internal fixation with a fibula locking plate, and fragment specific fixation using 1/3 tubular plates for the posteromedial malleolus and the medial malleolar shear fragment. Due to frailty, the patient was not able to comply with post-operative instructions to non-weight-bearing and represented with increased pain and swelling 1 month post-operatively with repeated radiographs demonstrating proximal translation of the medial malleolus, loss of reduction and of joint congruence. Due to the patient’s frailty, it was elected to revise this construct to a hindfoot nail which allowed the patient to weight bear without restrictions

Full size image

Calcaneus

Calcaneal fractures are the most commonly fractured tarsal bone. The annual incidence of calcaneal fractures are 11.5/100,000 people, with a male to female ratio of 2.4:1, most common sustained following falls from height (70%) [148]. The fractures can be broadly classified into extra-articular injuries (25%) often secondary to Achilles avulsion type injuries or intra-articular fractures (75%) [149]. Operative fixation is often recommended when significant disruption to the ‘angle of Gissane’ or ‘Bohlers angle’ is present. This can be achieved through percutaneous screw fixation, plate fixation, primary subtalar arthrodesis or C-nails [150,151,152,153,154].

Failure of plate fixation has been documented to be between 0 and40% and has been most commonly associated with osteoporotic bone [151, 154, 155]. The increasing use of locking plates has attempted to overcome this. There is paucity in literature detailing rates of fixation failures and the rationale behind this. One case series showed that screw fixation had a 24% probability of failure, plates showed a 36% failure and the most unstable seem to be the C-nails with 42% probability of failure. The authors do suggest fixation failure is often linked to patient factors such as smoking status and non-compliance with post-operative weight-bearing status.

Lisfranc

Lisfranc fractures have an incidence of 16/100,000 persons per year [156]. However, there actual incidence may well be higher due to up to 24% of these injuries being missed on their original radiographs [157]. These injuries are more common in males (4 males: 1 female) and most commonly occur in the third decade of life [158].

If true disruption of the ligamentous Lisfranc complex is present, then surgical management is often recommended. Operative intervention can consist of either open reduction internal fixation (ORIF) or primary arthrodesis [158]. The fixation method has been contentious with some surgeons advocating arthrodesis given the decreased need to return at a later date for removal of metalwork and subsequent fusion. Failure of fixation associated with ORIF can often be linked to over compression during the fixation, malreduction of the fracture site when the plates are applied or plantar trajectory of the ‘home run screw’ [159]. With respect to primary arthrodesis underprepared joints prior to fusion have been implicated with fixation failure, as has an early return to weight-bearing due to poor compliance [159].

Whilst failure of fixation is nor frequently reported, unplanned re-operation rates are similar between ORIF and primary arthrodesis (29.5 vs. 29.6%), most commonly due to post-traumatic arthritis in patients treated with ORIF and non-union in those treated with primary arthrodesis [160].

Discussion

Metal work failure remains a rare complication of fracture fixation, though the overall incidence is poorly defined within the literature. A summary of the current reported rates of fixation failure defined by anatomic site is summarised in Table 1.1.

Table 1.1 Incidence and rates of fixation failure alongside risk factors for fixation failure separated by body site
Full size table

Rates are currently extrapolated from small retrospective series and secondary outcomes of larger trials, varying from 0 to 57% depending on the location of the fracture and the technical application of the technique. Fixation failure is significantly higher in the lower limb where issues with ambulation introduce the risk of early weight-bearing and increased forces to which the fixation construct is exposed to.

Failure was reportedly highest when utilising techniques to stabilise the anterior pelvic ring, be that in the form of an external fixator or a plate. Fixation fails here at a much higher rate as the implant is spanning the symphysis, a joint that whilst stiff will never produce the same strain environment as a healed bone segment. Whilst pelvic ‘fixation failure’ is commonly reported, severe clinical symptoms are infrequently encountered nor is the requirement for removal of symptomatic hardware [67, 69].

Failure was similarly high in areas where high force transmission and poor vascularity predispose to slow healing, such as the femoral neck; in poor quality cancellous bone where fixation constructs struggle to gain adequate hold, such as the calcaneus; and in the pelvis where cancellous bone combined with an inability to prevent high stress due to its core position place significant stress in the implants utilised in the management of fractures here.

Reports regarding fixation failure are sparse, and often reported as secondary outcomes within larger studies. Whilst an extensive database search was conducted to examine its frequency, this report may still miss some studies which were not identifiable on a standard search. Similarly, the definition of fixation failure is not standardised across all studies, with some reporting on all cases where the integrity of the fixation construct was lost, and others simply reporting when a re-operation was required.

Reporting all cases of fixation failure will often identify metalwork complications that have no bearing on the clinical picture, such as the asymptomatic breakage of syndesmosis screws or loss of tension of an olive wire in a healing fracture segment [161]. Nonetheless reporting only those complications that require revision fixation will miss a number of patients that are symptomatic from their metalwork failure, who may need to alter their post-operative course through adjustment of weight-bearing or splintage, but do not require further operative management to achieve union in an acceptable alignment.

Conclusion

The overall incidence of fixation failure is poorly defined within the literature. Moving forward the true incidence of fixation failure does need to be more accurately defined, ideally via larger cohort studies, with a stricter definition that identifies those patients whose clinical course and outcome are altered by the construct failure.