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
Trauma and Orthopaedic Surgery, Department of Surgery and Cancer, Imperial College London, London SW7 2AZ, UKJoint Reconstruction Unit, The Royal National Orthopaedic Hospital, London, Stanmore HA7 4LP, UKKing Edward VII's Hospital, 5-10 Beaumont Street, Marylebone, London W1G 6AA, UK
Joint Reconstruction Unit, The Royal National Orthopaedic Hospital, London, Stanmore HA7 4LP, UKKing Edward VII's Hospital, 5-10 Beaumont Street, Marylebone, London W1G 6AA, UK
Academic Department of Trauma and Orthopaedic Surgery, School of Medicine, University of Leeds, Clarendon Wing, Floor D, Great George Street, Leeds General Infirmary, Leeds LS1 3EX, UKNIHR Leeds Biomedical Research Centre, Chapel Allerton Hospital, Leeds, UK
With a growing number of patients undergoing total knee replacements globally, coupled with an elderly population, the incidence of periprosthetic fractures around total knee replacements is increasing. As such, this is a highly topical subject that is gaining increasing interest within the orthopaedic community. This review provides a narrative synthesis of the most contemporary literature regarding distal femoral periprosthetic fractures. We review the related epidemiology, initial patient evaluation, the evolution and relevance of the classification systems and treatment options, particularly related to endoprosthetics and hybrid fixation constructs. The latest orthopaedic evidence related to this topic has been included.
The demand for primary total knee replacements (TKRs) is expected to surpass 3.48 million arthroplasties per annum by 2030 in the United States of America (USA) and the United Kingdom (UK) has paralleled this trend [
]. However, as this number grows, the number arthroplasty revisions is also set to increase. For example, within the UK, from 2004 to 2019 there was a 472% increase in the number of knee revisions performed [
]. There are several reasons for this including aseptic loosening, periprosthetic joint infections (PJIs) and periprosthetic fractures (PPF). In 2021, 3.84% (n=3358) of all knee revisions were due to PPFs in the UK [
]. This emphasises that there is a large population within the UK at risk of developing PPFs and that number is set to rise. It is estimated that for primary TKRs the incidence of PPFs is 0.3% to 5.5% and for revisions the number is estimated to be as high as 30% [
This review is a narrative synthesis of the most contemporary literature regarding PPFs. We review the population at risk, their evaluation, the evolution and relevance of the classification systems and current evidence-based treatment options. Due to the complexity and breadth of the topic, we will focus on distal femoral PPFs as they have the highest incidence following TKRs.
Risk factors
Over 90% of PPFs occur following low-energy trauma such as a fall from standing height. High-energy mechanisms have been reported to account for under 7% of all PPFs [
]. When performing a comprehensive patient evaluation, it is useful to consider the numerous factors which predispose patients to PPFs. A useful framework to classify these aetiologies includes patient-specific factors, implant-specific factors and surgical factors (Table 1).
Table 1A summary of the risk factors for periprosthetic fractures.
Patient-specific factors
Implant-specific factors
Surgical factors
Age (bimodal distribution < 60 years old or >80 years old represent high risk groups)
Revision procedure versus a primary TKR
Notching of the anterior femur (controversial)
Deyo-Charlson index score for co-morbidities > 3
Posterior stabilised TKR versus a cruciate retaining implant
Malalignment of components and instability, wear and osteolysis
Specific co-morbidities (diabetes mellitus, cardiac disorders, neurological disorders and any disorder which may increase the risk of sustaining a fall)
Loosening, prior surgery to remove implants, previous diagnosis of non-union or infection
Component removal in revision cases, trial reductions in inadequately sized components with excessive tightness, improper gap balancing, excessive retraction, use of excess force when preparing bone for stems or improper bone resections risk a periprosthetic fracture
Advanced age is considered an independent risk factor in its own right as well as being a risk factor for other patient-specific risks such as osteopenia/osteoporosis and recurrent falls. Comparison of survival analyses in Meek et al.’s review of 47,733 arthroplasties demonstrated that patients older than 70 years were at a significantly increased risk of sustaining a PPF [
]. More recently, the Mayo Clinic Total Joint Registry was reviewed by Singh et al. who reported that there was a bimodal relationship between age and PPFs. They reported that patients under 60 years or greater than 80 years old had a more than 40% increased risk of sustaining a PPF compared to those between 61 to 79 years old [
]. It is clear that the behaviour of different patient groups contributes to their fracture risks. Singh et al. attributed this pattern to a more active lifestyle in younger patients who most likely had underlying conditions that resulted in them developing joint degeneration early [
]. The majority of conditions which may increase the risk of a fall occurring unsurprisingly act as risk factors for PPFs and a Deyo-Charlson index score for comorbidities greater than 3 has specifically demonstrated this [
Specific co-morbidities which may increase fracture risk include neurological disorders (e.g. Parkinson's disease, Epilepsy, Poliomyelitis etc), cardiac disorders, inflammatory arthropathies and the use of medication which may affect the quality of bone (e.g. corticosteroids). Diabetes must be given special attention as it poses a falls-risk and can result in impaired healing due to microvascular disease and peripheral neuropathy [
]. When examining patients, clinicians should pay specific attention to the alignment of the knee and the presence of stiffness which can be associated with a mechanical stress-riser [
]. A number of recent studies have also, reported femoral condyle insufficiency fractures occurring in the unloaded condyles of varus or valgus aligned knees [
]. Knowledge of these factors should improve patient evaluation and counselling pre-operatively.
Implant-specific factors
It is well established that prosthesis design can affect the risk of suffering a fracture. Whenever a prosthesis is being implanted into bone clinicians must consider the elastic moduli of the components and the recipient bone-bed and take measures where possible to select materials with optimal properties. The difference between the elastic moduli should ideally be reduced as much as possible so as to minimise the impact of a stress-riser.
Posterior stabilised (PS) implants involve a resection at the intercondylar notch (colloquially referred to as ‘the box cut’) and this could potentially lead to a stress-riser. Furthermore, if this resection is not accurately aligned, it can lead to a unilaterally thin condyle which experiences greater stress than anticipated and this predisposes it to an early PPF. Care during bone preparation and component trialling can avoid this.
The team at Mayo clinic reported 37 PPFs in 8854 PS TKRs versus 7 PPFs in 7950 cruciate retaining (CR) TKRs and concluded that the relative risk was 4.74 [
]. Although the risk of PPF with uncemented implants is clearly documented in hip arthroplasty, the literature has not linked implant fixation technique to PPFs in TKRs [
]. The relationship between distal femoral bone width and prosthesis width has been an area of interest with the suggestion that smaller bone-to-prosthesis width ratios were considered to be risk factors for PPFs. However, a multivariate regression analysis was not able to demonstrate significance clinically [
Robotic-assisted surgery is on the rise amongst clinicians and involves inserting unicortical tracking pins to register anatomical landmarks. These pin sites can act as stress-risers and although rare, PPFs have been reported to propagate through them [
Periprosthetic fractures through tracking pin sites following computer navigated and robotic total and unicompartmental knee arthroplasty: a systematic review.
]. Smith et al. conducted the most recent systematic review of these fractures and reported large pin diameters (>4 mm), multiple placement attempts, transcortical trajectories and the use of non-self-tapping/drilling pins were the commonest risks [
Periprosthetic fractures through tracking pin sites following computer navigated and robotic total and unicompartmental knee arthroplasty: a systematic review.
Revision arthroplasty is considered to be a major, independent risk factor for PPFs with the incidence ranging from 1.7% to as high as 38% in one series [
]. More recent data published from the Mayo Clinic Joint registry paralleled these findings and showed a 1.1% PPF rate post-primary TKR versus 2.5% post-revision [
]. In the revision TKR cohort of this study, the researchers found that a prior diagnosis of infection, non-union or prior surgery to remove implants increased the PPF risk to a much greater extent than those with loosening, wear or osteolysis. Prior removal of implants was associated with twice the PPF risk compared to loosening, wear and osteolysis [
]. This provides reliable insight into the true risk post-revision and one must consider the condition of the bone microarchitecture and how it's been altered to reduce the PPF risk.
Surgical factors
Post-operative PPFs are ten times more common than intra-operative fractures however intra-operative fractures are extremely under-reported in the literature. The focus of this article is on post-operative femoral PPFs. Fractures may occur at any point during a procedure however are more likely to occur during surgical steps that impose considerable stress to the knee (e.g. component removal in revision cases, trial reductions especially in inadequately sized components, improper gap balancing, excessive retraction, use of excess force when preparing bone for stems or improper bone resections). Attention to detail, scrutiny of the host bone-bed and constant attention to the forces being applied through the knee will help prevent error which can lead to PPFs.
In 2006, Abu-Rajab and colleagues investigated peri-prosthetic bone density after both cemented and uncemented TKRs in 40 patients. They identified a 27% loss of bone mineral density posterior to the anterior flange of the femoral component occurring over the first 2 years post-operatively and therefore concluded that TKRs were at increased risk of fracture during these periods secondary to this stress shielding [
]. Indeed, this is not too dissimilar to studies which have looked at time to fracture post-operatively such as, Gondalia et al.’s report which revealed a mean time of 25.5 months [
Periprosthetic supracondylar femoral fractures following total knee arthroplasty: clinical comparison and related complications of the femur plate system and retrograde-inserted supracondylar nail.
]. It is our opinion, that meticulous technique which minimises the chance of occult injuries or the creation of mechanical stress risers can help prevent PPFs.
The issue of malalignment of components has been the subject of debate as although the literature has clearly identified varus malalignment of the tibial component to be a risk for a tibial PPF malalignment has not been clearly confirmed as a cause for femoral PPFs. It does seem reasonable to consider malalignment as a potential cause for a PPF as ultimately, instability of the knee secondary to malalignment could result in a fall and subsequent fracture. Additionally, malalignment could result in increased wear, osteolysis and aseptic loosening which could lead to a greater chance of fracture following minor trauma [
Notching at the anterior femoral cortex is a controversial issue in the orthopaedic community due to conflicting reports regarding its significance. Biomechanical studies have suggested notching as an important cause for PPFs due to observed reductions in torsional and flexural bone strength in knee models [
]. The effect on stress concentration is often considered to be greatest when notches are greater than 3 mm (millimetres) deep, sharply cut and in close proximity to the femoral component [
]. However, clinical studies have been less transparent. Hoffman et al. reported that 25% of the PPFs in their series were associated with radiographically diagnosed notching which was statistically significant. They also identified that the distance between the anterior flange and the fracture was significantly shorter in patients with notching versus without (3.2 mm versus 39 mm) [
]. Contrary to this Ritter et al. reported notching in 29.8% of cases in their series of 1089 knees. Only 2 femoral PPFs were reported, both in patients without notching. They found no association between notching and femoral PPFs [
]. Gujarathi et al. reported similar findings to Ritters’ group and Minarro et al. looked at whether the PPF pattern was related to the presence of a notch and concluded it was not [
]. Although the clinical significance of notching remains unclear and some consider it insignificant, we would exercise caution given the reduction in biomechanical strength noted and we anticipate that larger population studies would likely yield definitive results.
Patient evaluation and classification
For the clinician to be able to complete a thorough evaluation, one must be familiar with the different types of PPF and the underlying problems which may be associated (as detailed in the previous section). As expected, a thorough history and examination should be completed and specifically look for evidence of long-standing pain which may indicate loosening. One must also be suspicious of a periprosthetic joint infection (PJI) which may account for loosening. Details of post-operative wound complications or recent systemic infections should raise suspicion for this, and we recommend clinicians adopt the most recent guidance provided by the Musculoskeletal Infection Society (MSIS) for the diagnosis of a PJI [
]. Although routine radiographs will diagnose a PPF in the majority of cases one must assess the pattern of injury, extent of comminution, the remaining bone stock available, the quality of bone remaining, extent of loosening and the position of the implants. To fully understand the injury a Computed Tomography (CT) scan is recommended when planning surgical intervention.
Due to the complex nature of TKR PPFs and the multiple factors which must be considered when evaluating patients, a large number of classification systems has been published over time. Each system can be thought of as an evolution from its predecessor with a new factor being included and modern surgical techniques being incorporated into the system and so these systems reflect our progress and development.
The first published classification is that by Neer et al. in 1967 [
]. It was essentially descriptive but considered the injury mechanism, direction of force application and whether the extensor mechanism was in-tact. It wasn't until 24 years later that DiGioia and Rubash modified Neer et al.’s system [
]. They specifically defined PPFs as a fracture occurring within 15 centimetres (cm) of the prosthetic joint line or 5 cm from a stem and quantified the amount of displacement, angulation, and comminution [
]. In 1994, Chen et al. simplified the precedent systems by grouping undisplaced fractures (Neer type 1 equivalents) together and displaced fractures together (Neer types 2 and 3). They considered non-operative and operative treatment options and then made recommendations based on these classes [
]. These initial classification systems are generally considered the first generation of TKR PPF classifications as per Rhee et al. who emphasised that they were of limited clinical use as a specific technique of operative intervention could not be established using them [
]. Rorabeck and Taylor felt that the published classification systems were unable to guide management appropriately and that there was too much inter- and intra- observer variability. They produced their system which they purposely made clear and simple to understand in order to ensure its widespread use and they considered the fixation of the prosthesis in their descriptions with recommended treatment options based on their own experience [
]. This system became the basis for later classifications and is still the most widely used system practically. Guided by this classification, Su et al. presented their own system which considered the fracture line in relation to the femoral component and the feasibility of retrograde intramedullary (IM) nailing [
]. They stated that advancements in available implants and techniques obviated the use of non-operative management unless a patient was deemed medically unfit [
]. Further modifications came from Kim et al., Backstein et al. and ultimately Frenzel et al. who added modifiers to the classification systems to consider bone stick, the fixation of the prosthesis and the fracture timing, respectively (Table 2) [
I = Undisplaced/minimally displaced fractures with <5 mm displacement or <5 degrees of angulation) II = >1 cm displacement IIa = above with lateral femoral shaft displacement IIb = above with medial femoral shaft displacement III = Displacement and comminution
I = Extra-articular, undisplaced (<5 mm of displacement or <5 degrees angulation) II = Extra-articular, displaced (>5 mm or >5°) III = Severely displaced (>10° of angulation or loss of cortical contact and may have an intercondylar or t shaped fracture pattern)
I = Undisplaced fracture with a stable implant II = Displaced fracture but with a stable implant III = Undisplaced or Displaced fracture with a loose component
I = fracture line proximal to femoral component and cement II = Fracture line originating at the proximal aspect of the femoral component and extending proximally III = Fracture line is distal to the most proximal aspect of the anterior flange
I = fractures occurring in patients with preserved bone stock, stable and well-positioned implants IA = above and either undisplaced or easily reducible IB = irreducible fractures requiring reduction and internal fixation II = Fractures with loose or mal-positioned implants but preserved bone stock III = Fractures with a loose or mal=positioned implant and poor bone stock
Femoral periprosthetic fractures were defined as occurring within 15cm of the femoral implant. F1 = Distal fracture fragment has sufficient bone for retrograde nail locking screws F2 = Distal fracture fragment does not have sufficient bone for retrograde nail locking screws. Additional qualifiers: S = Stable implant L = Loose implant G = Good bone stock P = Poor bone stock
Considered suitability of retrograde intramedullary nailing
This system uses the AO/OTA codes for bones and the type of implant. Implantation technique P = Polymethylmethacrylate Cement used. U = Uncemented The fracture pattern type follows the AO/OTA classification A = simple fracture (spiral/oblique/transverse) B = wedge (spiral/bending/multi-fragmentary) C = complex fracture (spiral/segmental/irregular) Prosthesis stability S = stable Q = questionable L = loose Time points of fracture 1 = primary/intraoperatively 2 = secondary 3 = beyond 5 years postoperatively Bone structure 0 = healthy I = moderate osteoporosis with resorption width of <2mm II = Severe osteoporosis with resorption>2mm III = Segmental bone defect, ‘egg-shell’ cortex
We anticipate that classifications will continue to evolve in line with our understanding of these complex injuries however our current preference is to use the Unified Classification System (UCS) which was proposed as a practical tool to guide the correct approach and management for all PPFs (Table 3) [
]. This system has been ‘field-tested’ by Van Der Merwe et al. who stated that ‘the UCS has substantial inter-observer reliability and 'near perfect' intra-observer reliability’ when used by both fellowship trained surgical experts and more junior surgeons in the last 2 years of their training [
Field testing the Unified Classification System for periprosthetic fractures of the femur, tibia and patella in association with knee replacement: an international collaboration.
. The bone involved is identified using the AO/OTA code (Arbeitsgemeinschaft für Osteosynthesefragen/ Orthopaedic Trauma Association). The code for the knee roman numeral V. AO/OTA codes for the Femur and Patella are 3,4 and 34, respectively.
Fracture type based on location
Knee (V) Femur (3)
Treatment summary
Type A = Apophyseal or periarticular/extraarticular fracture with no effect on implant stability
A1
Lateral epicondyle
Management is dependent on displacement and importance of soft tissue attachments. Undisplaced and stable fractures can be managed in a hinged knee brace. Displaced fractures may be fixed to prevent joint instability with cancellous lag screws.
A2
Medial epicondyle
As above
Type B = PPF through the Bed of the implant
B1
PPF around a stable femoral component with good bone stock
Fixation
B2
PPF around a femoral component which is loose but with good bone stock
Revision
B3
PPF around a loose femoral component with poor bone stock/ bone defects
Complex Revision (endoprosthesis or use of allograft-prosthesis composite)
Type C = PPF Clear of the femoral component
C
PPF proximal to the femoral component and cement mantle
Fixation
Type D = an inter-prosthetic fracture/a PPF Dividing the bone that is between 2 joint replacements (e.g., femur fracture in a hip and knee replacement)
D
PPF between hip and knee replacements
‘Block-out analysis to determine plan (this involves assessing one component and blocking out the other then assessing the other component such that a rational plan can be devised)
Field testing the Unified Classification System for periprosthetic fractures of the femur, tibia and patella in association with knee replacement: an international collaboration.
Field testing the Unified Classification System for periprosthetic fractures of the femur, tibia and patella in association with knee replacement: an international collaboration.
The objectives of managing distal femur PPFs are to achieve a stable, well-fixed TKR with satisfactory alignment, length and patellofemoral tracking. The UCS has increased in popularity and in our practise is used as a pragmatic guide to the treatment options available. Non-operative management of PPFs in general may be considered for patients with severe comorbidities and significant anaesthetic risk who are non-ambulatory with the potential for severe life-threatening post-operative complications [
]. Techniques include closed reduction and casting, the use of orthoses (e.g. hinged knee braces or extension splints) and traction. Patients managed with non-operative options require frequent imaging to ensure the fracture has not displaced during the course of the treatment. Additionally, regular inspection of the dermis is essential to ensure no loss of skin integrity has occurred (Fig. 1).
Fig. 1Anteroposterior and lateral radiographs demonstrating a lateral epicondyle periprosthetic fracture sustained following a mechanical fall. Axial and Coronal Metal Artefact CT scan slices further demonstrate the fracture pattern (New Unified Classification System Type A2). This was treated non-operatively in a hinged kneed brace successfully.
In patients deemed medically fit for surgery non-operative management may also be an option for Type A fractures. Type A fractures are defined as fractures that involve the apophyseal or periarticular/extraarticular region with no effect on implant fixation. In the context of the distal femur, they can be subdivided into A1 involving the lateral epicondyle and A2 involving the medial epicondyle (Fig. 2). These patients must be followed up regularly to ensure fracture union occurs, the collateral ligaments do not become de-functioned and instability does not develop.
Fig. 2Anteroposterior and lateral radiographs demonstrating a New Unified Classification System Type B3 periprosthetic fracture treated with revision to a distal femoral replacement successfully.
]. Culp et al. reviewed 61 supracondylar fractures and divided them into two groups; Group A and B. Group A consisted of patients treated with open reduction and internal fixation and group B consisted of patients who treated with casting, traction or knee braces. The rate of non-union and malunion was 43% (n=30) in group B compared to 13% (n=30) in group A indicating that more optimal outcomes with operative [
]. In addition, the literature has shown that patients treated using the closed reduction had a higher relative risk for subsequent surgical procedures to address the higher rates of malalignment, non-union, malunion and arthrofibrosis compared to modern operative techniques [
]. It is also worth noting that the prolonged period of immobilisation may itself pose significant medical risk to the patient. However, this method must not be completely discounted as in patients with severe frailty and significant co-morbidities non-operative techniques may be the only option available [
When operative intervention is planned the choice of technique is determined by a number of factors including fracture location, implant loosening and bone stock (Table 3).
In patients with fractures at the implant bed, preserved bone stock and a stable femoral component (B1 fracture); operative fixation is preferred. Conventionally, this involved using a plate-and-screw construct with the aim of reducing non-/mal-union rates as well as allowing early range-of-motion exercises and a reduced period of immobility. Non-locking plates were initially used and consisted of systems involving dynamic condylar screws [
]. Although non-locking plates achieved higher rates of union and lower rates of malunion compared to non-operative management alone, the approaches were extensile and involved significant soft tissue stripping which imposed significant morbidity to patients [
]. Additionally, Healy et al. demonstrated that although 90% of patients in their series achieved union following conventional plating, 75% required the use of a bone graft [
]. Modern locking plates are minimally invasive with variable angle screw fixation options facilitating fragment stabilisation and allowing for optimisation of the strain environment. Thukral et al. demonstrated that the minimally invasive approach had higher knee society scores (KSS) at 6 and 12-months and, on average, achieved radiological union 54 days sooner compared to conventional plating. Furthermore, bone grafting was not required [
]. Hoffman et al. reported similar results and specifically noted that non-union was less likely in their series when minimally invasive techniques were employed [
]. Despite this, there is significant mortality associated with plating. Struebal et al. analysed 48 patients aged 60 years and above with low-energy distal femur PPFs treated with lateral locking plating fixation. They reported an overall mortality rate of 8% at 30 days and 27% at 1 year and found a significant increase in mortality when surgery was delayed by 4 days or more [
IM nailing is another option for B1 fractures and can be subdivided into antegrade and retrograde IM nailing. Many surgeons elect to use retrograde nailing as it considered to be more reliable for distal fixation [
]. Theoretically, the main advantage of IM nailing is that it requires minimal surgical exposure compared to both locking and non-locking plates, and therefore addresses the morbidity associated with more extensile approaches or prolonged surgery. In addition, when appropriately fitted, it provides a load-sharing construct to allow for early mobilization. When determining suitability for IM nailing one must consider whether the box portion of the implant is large enough to receive the nail, the presence of an ipsilateral hip arthroplasty with a long femoral stem or a patella baja all of which can make it impossible to introduce the nail and so, act as contraindications to the technique.
Our preferred approach is to perform an arthrotomy to allow us to position the nail appropriately and protect the current implants from iatrogenic damage. To achieve adequate stability, nails must be long enough to extend into the diaphysis. As a result, greater caution is required when there is an ipsilateral hip replacement as a short interface between the nail and implant will create a stress riser.
In theory, the IM nailing group should have fewer complications compared to the plating group however, there is no clear consensus regarding this in the literature [
Similar outcomes of locking compression plating and retrograde intramedullary nailing for periprosthetic supracondylar femoral fractures following total knee arthroplasty: a meta-analysis.
]. Ebraheim et al. analysed the complications of 448 fractures over the course of 10 years and found that fractures treated with a locking plate had a complication rate of 35% and fractures treated with an IM nail had an overall complication rate of 53% [
When comparing IM nailing to conventional plating techniques, IM nailing had a significant advantage. It was associated with less infection shorter operating duration, less intraoperative blood loss and fewer total complications [
]. Bezwada et al. treated 30 supracondylar knee fractures with conventional plating or IM nailing. The study reported that the average surgical time was 29 min shorter for the IM nail group (n=18) (p<0.05) with an average blood loss of 100 mls compared to the 450 mls seen with a traditional ORIF [
]. Herrera et al. also demonstrated that compared with non-locked plates, IM nailing had a relative risk reduction of 87 % for developing a non-union. Moreover, a 70 % relative risk reduction for requiring revision surgery in the future is reported [
]. Therefore, it can be argued that both locked plating and IM nailing are superior to conventional plating methods with regards to outcome and morbidity.
Compared to locked plating there is minimal difference in fracture union outcome. A large meta-analysis performed by Shin et al. which showed no statistical difference in clinical or radiological outcome between plating and intramedullary nailing [
Similar outcomes of locking compression plating and retrograde intramedullary nailing for periprosthetic supracondylar femoral fractures following total knee arthroplasty: a meta-analysis.
]. This is echoed by Ristevski et al., who analysed 719 fractures and showed that there was no statistical difference between locked plating and intramedullary nailing with regards to rates of non-union [
In recent years, combined nailing and plating has been proposed. The rationale for this technique is that it results in a more rigid construct allowing early return to full weightbearing and avoiding the morbidity associated with prolonged immobility [
]. However, data is limited though preliminary studies are encouraging. Hussain et al. report of a case-series of 9 fractures, treated with a combination of IM nailing and locked plating. They report a 100% (n=9) union rate at 20 weeks with the added benefit of immediate weight bearing following the procedure, a feature not seen in plating or IM nailing alone [
Stable fixation and immediate weight-bearing after combined retrograde intramedullary nailing and open reduction internal fixation of noncomminuted distal interprosthetic femur fractures.
]. Another study by Christ et al. observed union rates in a total of 40 patients with distal peri-prosthetic fractures. They report a 94% (n=32) union rate and found no statistical difference in outcome when compared with patients who received a bone graft and those who did not receive a bone graft in addition to the combined nailing and plating approach [
]. It is worth noting that all three studies demonstrate a change in ambulation status post operatively. Only 22%–50% of patients returned to their pre-operative ambulatory status after the procedure [
Stable fixation and immediate weight-bearing after combined retrograde intramedullary nailing and open reduction internal fixation of noncomminuted distal interprosthetic femur fractures.
]. In addition, Christ et al. highlighted a significant mortality associated with this technique with 20% (n=8) of elderly patients dying postoperatively or being lost to follow-up [
]. As this technique is currently being investigated, it remains to be seen if it is superior to the established operative techniques described above and there is a clear need for a randomised controlled trial to help determine the safest and most optimal technique.
Revision
Revision arthroplasty or distal femoral replacement (DFR) is the recommended treatment for B3 fractures, which is defined as a fracture with a loose femoral component and poor bone stock (Fig. 2). This population of patients has compromised bone quality and therefore union is unlikely to be achieved with fixation alone. Revision arthroplasty is generally reserved for elderly patients where prolonged immobilization is a concern due to underlying co-morbidities and in patients who are unlikely to maintain compliance with weightbearing restrictions.
An extensile midline incision is made followed by a medial parapatellar arthrotomy. The tibia and distal femur are stripped down to the periosteum and towards the area of resection [
]. The femoral component is removed and used to aid in approximating the length and size of the DFR with much of the fractured bone removed. To achieve adequate alignment, the anterior cortex of the femur is aligned with the trochlear grove superior the proposed site of the osteotomy [
]. The tibial component is then removed with the goal of preserving as much bone as possible and a new tibial component is prepared and implanted.
The reported outcomes following revision arthroplasty are varied. Rao et al. reports a case series in 12 patients treated with a distal femur replacement. All 12 patients were mobilising by the third day [
]. Their Western Ontario and McMaster Universities Arthritis Index (WOMAC) scores improved from the pre-injury state with a mean of 49.62 to 72.54 post-surgery (p value ≤0.05) [
]. Mortazavi et al. analysed 22 PPFs managed with distal femoral replacement and found their clinical KSS scores improved from 71.8 pre-operatively to 82.8 post operatively [
An economic review of the service costs does not appear to demonstrate an obvious difference between fixation strategies and revision to a DFR in the literature. Tandon et al. reports of a case control study where patients between 2005 and 2013 were treated for a PPF. They segregated these patients into two groups; group A (DFR) and group B (ORIFs) and analysed their complication rate and the total financial impact of their procedure and their complications. The DFR group had a complication rate of 19% (n=21) and the ORIF group 40% (n=40) [
]. Although the DFR implant was costly at £7500, the shorter length of stay and total complication rate led to an average cost of £9600 whereas the ORIF group, on average, had longer lengths of stay and a higher complication rate resulting in a cost of £9800. Tandon et al. concluded that DFRs are similar in cost effectiveness compared to ORIFs [
]. Moreover, there is a high rate of prosthesis failure. Pour et al. reported that in their case series of 42 patients treated with a distal femoral replacement had a prosthesis survival rate of 79.6% at one year and 68.2% at five years [
]. A study by Toepfer et al. corroborated similar rates of complications of 64% however this can be attributed to the population having multiple co-morbidities [
Although there is a paucity of data, DFRs have demonstrated expedient weight bearing, shorter stays in hospital and similar cost effectiveness to ORIFs. A recent meta-analysis of 1484 patients performed by Wadhwa et al. reported similar reoperation and total complication rates [
Distal femur replacement versus open reduction and internal fixation for treatment of periprosthetic distal femur fractures: a systematic review and meta-analysis.
]. Further research is required to fully establish the risk and overall success of DFRs, however, in the appropriate patient group this procedure can yield encouraging results. We recommend DFRs in patients who have poor bone stock and a loose component.
CME summary points
•
PPFs incidences are increasing. The Scottish arthroplasty project has shown that the number of PPFs has doubled over an 11-year period.
•
Steroid use, infection, osteoporosis, osteopenia, inflammatory arthropathies and female gender increases the risk of PPFs.
•
Neuromuscular disorders, diabetic neuropathies and Parkinson's disease are some of the comorbidities associated with an increased risk of falls and low impact trauma predisposing to PPFs.
•
Classifications have evolved over the last 30 years to take into account displacement, loosening and bone stock.
•
Meta-analyses have demonstrated no significant difference in union rates between IM nailing techniques versus locking plate constructs in B2 fractures (Unified Classification System).
•
IM nail-locking plate hybrid constructs have emerged as an increasingly popular technique to avoid the morbidity associated with prolonged immobilisation post-operatively.
CME Viva Questions
The answers to these questions are within the text of this article.
CME Viva Question 1
A lady in her late 60s presents to the Emergency department having sustained a mechanical fall in the preceding 2 h and plain radiographs confirm a distal femoral periprosthetic fracture. Please describe your initial assessment and management for the patient.
CME Viva Question 2
How do you classify these injuries? Are you aware of any recent publications which include recommendations for treatment?
CME Viva Question 3
What are the principles of managing elderly patients who sustain B3 periprosthetic fractures?
CME Viva Question 4
Are you aware of any evidence for the use of distal femoral replacements in this group of patients?
CME Viva Question 5
How do the outcomes of locking plate versus IM nailing for UCS B2 fractures compare? How do you choose between the two surgical options?
CRediT authorship contribution statement
Talal Al-Jabri: Conceptualization, Writing – original draft, Writing – review & editing. Mohamed Ridha: Writing – original draft, Writing – review & editing. Robert Allan McCulloch: Writing – review & editing. Chethan Jayadev: Writing – review & editing. Babar Kayani: Writing – review & editing. Peter V. Giannoudis: Conceptualization, Writing – review & editing.
Declaration of Competing interest
The authors declare that they do not have any competing interests.
Funding
The authors did not receive any funding for this article.
Ethical Approval
This was not required for this article.
Acknowledgments
Not applicable.
Consent for publication
Not applicable.
References
Kurtz S.
Ong K.
Lau E.
Mowat F.
Halpern M.
Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030.
Periprosthetic fractures through tracking pin sites following computer navigated and robotic total and unicompartmental knee arthroplasty: a systematic review.
Periprosthetic supracondylar femoral fractures following total knee arthroplasty: clinical comparison and related complications of the femur plate system and retrograde-inserted supracondylar nail.
Field testing the Unified Classification System for periprosthetic fractures of the femur, tibia and patella in association with knee replacement: an international collaboration.
Similar outcomes of locking compression plating and retrograde intramedullary nailing for periprosthetic supracondylar femoral fractures following total knee arthroplasty: a meta-analysis.
Stable fixation and immediate weight-bearing after combined retrograde intramedullary nailing and open reduction internal fixation of noncomminuted distal interprosthetic femur fractures.
Distal femur replacement versus open reduction and internal fixation for treatment of periprosthetic distal femur fractures: a systematic review and meta-analysis.
00170-5&id=gr1.jpg){kind=link}
00170-5&id=gr2.jpg){kind=link}