Periprosthetic hip fractures include injuries along the proximal femur and surrounding the acetabulum. They may range from relatively simple fractures with a stable implant to major injuries with components that require revision. Management of such fractures requires a surgeon to borrow and blend principles for fracture treatment with concepts and “rules” of hip arthroplasty. These two disciplines often have conflicting ideas for treating a postoperative periprosthetic fracture, but it is truly an art to blend concepts from both subspecialities to provide optimal outcomes for our patients.
With an increasing number of people undergoing hip arthroplasty, a longer projected lifespan and more active/demanding patients, the number of periprosthetic fractures continues to rise. At our own institution, we have seen an increase in the number of revisions for periprosthetic fracture from 6.2% to 7.2% over the last 10 years. Some studies have projected the prevalence of periprosthetic fractures to increase an average of 4.6% over the next 30 years (which is similar to the rate seen at the authors’ home institution). , This is in part attributed to the aforementioned factors coupled with the growing worldwide trend toward cementless total hip arthroplasty. In the last decade, there has been substantial research regarding the significant morbidity and mortality of these injuries, spurring an influx of new research looking at predictive measures based on patient anatomy and implant design using clinical data and finite element analysis (FEA). The ultimate goal is to potentially identify particular implants that will have lower risks based on anatomic variations that exist. Modern technologies such as machine learning, edge detection software, and FEA hold promise to shed some light on predictive measures with associated solutions for lowering the rates of postoperative periprosthetic fracture. Preventive measures remain paramount, as the financial burden of periprosthetic fractures could result in a tidal wave of costs in the upcoming years. ,
The remainder of the chapter will focus on the epidemiology, diagnosis, classification, management, and outcomes of these fractures for both the femur and acetabulum. It is important to understand the following concepts not only for the treatment of these fractures but to also recognize at-risk cases that may be managed prior to a fracture occurring. By remaining vigilant in the aspects of diagnosis and evaluation, we may be able to achieve an “ounce of prevention.” Similarly, when necessary, working in a timely manner, optimizing modifiable factors and meticulous surgical technique can ensure the best outcomes in these difficult cases.
Periprosthetic Femur Fractures
For the reasons mentioned previously, the incidence of postoperative periprosthetic fractures (PPF) is expected to follow suit, particularly in the femur. A true prevalence has been difficult to establish, yet numerous global joint registries approximate the incidence of PPFs of the femur to range between 0.1% and 2.6%. Frenzel et al. estimated about a 20% increase in the number of PPF cases over a 5-year span. The complications associated with femoral PPF pose a significant financial burden on health care systems worldwide. This is due to the greater costs attributed to increased surgical time, length of stay, and manifestations of perioperative complications. Further, timely management of these injuries is paramount, as Scott et al. recently found an increased risk of medical complications, surgical complications, and 90-day risk of prosthetic joint infection if revision for a PPF is delayed for more than 48 hours.
Preoperative Risk Factors
It is essential to recognize which patients are at higher risk for femoral PPF in order to intervene early and possibly prevent these complications from occurring. In general, risk factors for a postoperative PPF can be broken down into patient-related factors (age, body mass index [BMI], physiologic age, and so on) and surgical factors (planning, surgical technique, implants utilized, and the like). Furthermore, it is important to consider the differences associated with an early versus late PPF of the femur. There are numerous unique relationships and findings that are more related to late periprosthetic fractures, particularly those occurring years after the index procedure. A summary is shown in Table 23.1 .
|Late PPF Risk Factors|
While the focus of this chapter centers around postoperative fractures, it is important to recognize that many early PPFs are thought to be missed intraoperative injuries. One explanation for a “missed” intraoperative fracture is the advent of minimally invasive or “new” surgical approaches in THA. With such procedural changes, the surgical field continues to diminish in size, giving limited line of site during the operation. Further, an approach-specific learning curve must be mastered, and early signs of fracture are often not readily detected on routine imaging. For these reasons, it is crucial to remain hyper-vigilant when inserting a femoral stem through any surgical approach.
For femoral PPFs, female patients and those over the age of 65 years are known to be at higher risk than the general population for these injuries. Additionally, those with systemic metabolic bone disease, neuromuscular deficits, inflammatory arthritis, osteoporosis, and proximal femoral deformities are at higher risk. , , Concerns for long-term bisphosphonate use and atypical PPFs of the femur have been realized, described in the literature and must be a consideration after a total hip arthroplasty.
Intraoperative Risk Factors
Surgical technique and planning can also influence whether a patient is at risk for sustaining a perioperative prosthetic femur fracture. When preparing the femoral canal during a primary or revision THA, it is imperative to minimize eccentric reaming (make sure that you are lateral prior to initiating reaming) and avoid perforation of the cortex (particularly anteriorly or laterally), as both events have been identified as significant contributors to an early PPF. Surgical approach, implant type, and means for fixation (cemented vs. cementless) have been implicated as surgical factors associated with potential fracture. The likelihood of a PPF within the first 30 days after surgery is up to 10 times greater in uncemented versus cemented THA. , Thinner distal cortices and a decreased meta-diaphyseal taper are anatomical roadblocks that must be managed to prevent a PPF after a cementless primary THA. Berend et al. suggest that anatomic characteristics also give rise to a higher chance of a PPF when using the anterolateral approach.
Compared with primary THA, there is an increased risk of up to 4% of PPF in revision THA cases. , Younger patients with a more active lifestyle are at risk of suffering a PPF after revision THA. More specific to revision surgery, it is crucial to be mindful when removing hardware or retained cement, as this could result in a stress riser from thin cortices or a frank perforation of the femur. Such areas of concern should be accounted for during preoperative planning so that you are prepared during surgery to bypass such defects and minimize the potential contribution to a postoperative PPF. The typical rule of thumb is to utilize a longer femoral stem, twice the bone diameter, to prevent a PPF. , , , Additionally, proximal femoral remodeling (occurs in ∼20%–40% of revision cases) should be assessed during the planning stage, as this can be associated with an occult fracture, as well as fracture when trying to remove the in situ femoral component.
Postoperative Risk Factors for Late Periprosthetic Fractures
While the elderly female patients and those with osteoporosis are also likely to suffer from a late PPF, there are several additional aspects that distinguish these cases from early fractures. Patients with long-term indwelling prostheses face a greater threat of incurring a PPF over time due to bone adaptations, greater risk for injury, and wear-related complications. Lindahl et al. analyzed 1049 patients from the Swedish National Hip Arthroplasty Register over a 3-year span. They found that late PPFs were the third most common cause for revision surgery, with most patients demonstrating signs and symptoms of loosening prior to the injury.
Osteolysis and mechanical loosening are commonly recognized as two of the biggest risk factors associated with late PPF. The relationship between the two was previously thought to be in large part due to failures in cement fixation. However, further studies have shown that peri-implant osteolysis may result in aseptic loosening with subsequent PPF. Trunnion and metallic debris (from wear or component impingement) can also spark the initiation of an adverse local tissue reaction in both cemented and uncemented arthroplasty. , ,
Diagnosis and Evaluation
Diagnosis of a PPF of the femur is often easy to make. However, it is important to complete a thorough evaluation that includes a detailed history, physical examination, and assessment of radiographs (of adjacent and contralateral joints, when possible). Often, the initial presentation occurs in relation to a low-energy trauma, such as a fall from level height. Beals and Tower found in their study that only 8% of PPFs occurred as a result of a spontaneous fracture. In up to 50% of cases, patients may report antecedent upper thigh pain before the actual traumatic incident.
Surveying serial radiographic images (if available) is paramount in evaluating a suspected PPF complication. At a minimum, there should be a standard series of radiographs obtained to include anteroposterior (AP), frog-lateral, shoot-through lateral, and AP pelvis views of the affected extremity. Imaging studies should be carefully analyzed for eccentric polyethylene wear, cement mantle fracture, metallic shedding, component loosening, areas of osteolysis, and bony defects. If any of these factors are present, close monitoring is warranted to prevent a future PPF. In addition, this knowledge may direct the management of a current injury. Appropriate counseling on activity levels, management of osteoporosis, and prompt treatment of loose components can help prevent a PPF. A long-term monitoring program set up to track patients over time is imperative in order to accurately assess and avoid PPF complications. Haddad et al. have advocated for routine follow-up with operating surgeons, as they are more familiar with the subtleties of each individual case.
In the setting of an obvious PPF, it is rare to need higher-level imaging studies. However, when the fracture is subtle or occult, metal artifact reduction sequence (MARS) magnetic resonance imaging (MRI) or a technetium bone scan may be warranted to make the diagnosis. Computed tomography (CT) scans can be used to assess peri-implant bone defects but are typically not required to make the diagnosis of a PPF. A detailed review of femoral bone loss classifications are out of the scope of this chapter. However, the authors would be remiss if this was not acknowledged as part of the assessment of a PPF. Routinely, the Paprosky classification for femoral bone defects is followed to help guide management of the revision portion of treating a PPF.
In all instances of a painful THA, whether for fracture or not, it is important to make sure there is not an underlying periprosthetic joint infection (PJI). This can be ruled out by obtaining the appropriate tests as outlined by the Musculoskeletal Infection Society (MSIS) criteria. The presence of a PPF, like any acute fracture, may alter inflammatory marker levels; careful assessment must be followed with a low threshold for performing a joint aspiration. Furthermore, the specific criteria for ruling out infection may vary with an underlying PPF. Van den Kieboom et al. investigated different serum marker thresholds in the setting of a PPF. They found that a white blood cell (WBC) count of 4552 cells/μL, polymorphonuclear (PMN) percentage of 79.5%, erythrocyte sedimentation rate (ESR) of 45.5 mm/h, and C-reactive protein (CRP) of 16.7 mg/L were the optimal threshold levels for detecting a PJI in the setting of a PPF. If revision surgery is required for a concurrent PJI and PPF, then utilizing a two-stage technique with a spacer and fracture fixation, followed by reimplantation when the infection is cleared, may be prudent. Further laboratory testing (metal levels, allergy testing, and so on) is often not necessary in the setting of an acute postoperative periprosthetic fracture, as it will not likely impact the decision-making for treatment and may delay prompt management of the injury.
Once the diagnosis has been made and appropriate evaluation completed, it is important to classify the fracture, as this will help direct the ultimate treatment options available. There have been numerous attempts over the years to describe PPFs in a manner that effectively guides treatment options and proffers a possible prognosis for the patient. In the 1970s and 1980s, fractures were described in relation to stem location, but this system did not achieve these goals. In 1995, Duncan and Masri developed the Vancouver Classification System (VCS), which is today the most widely used system for describing PPFs. It is important to note that there is a separate VCS for intraoperative PPFs. The VCS takes into account three key factors: location of the fracture, stability of the implant, and associated bone quality or bone loss. A more detailed description of the classification system is outlined in Table 23.2 . , ,
|Type A |
Fracture within the intertrochanteric space with a stable implant
Type A fracture involving greater trochanter
Type A fracture involving lesser trochanter
|Type B |
Fractures around the femoral stem or just past the distal end of the stem
Type B fracture with stable implant and good bone quality
Type B fracture with unstable implant and good bone quality
Type B fracture with unstable implant and poor bone quality
|Type C |
Fractures well past the implant
|Type D |
Fractures between implants
|Type E |
Fractures to both bones supporting a joint replacement
a In 2014, this was expanded to include Types D and E fractures and the whole classification was adjusted for periprosthetic fractures of any bone.
There are several studies that have shown both validity and reliability of the VCS for postoperative periprosthetic femur fractures. , , However, these investigations did not distinguish between fractures around cemented versus cementless femoral components. There is recent evidence to suggest that the VCS is less reliable in cementless hip replacements, which may pose a future problem considering the increasing popularity of cementless stems. Moreover, the classification system does not speak to fractures of the acetabulum or between (interprosthetic fracture) THA and total knee arthroplasty (TKA) components. In response to these limitations, Duncan and Haddad expanded the classification, now known as the Unified Classification System (UCS), for all joint replacement PPFs. In addition to the previous three types of PPFs, Class D fractures are assigned to interprosthetic fractures and Class E to fractures involving both the acetabulum and femur (both bones supporting a joint replacement). Despite some added complexity, the UCS has been shown to be reliable, with De Meo et al. finding moderate agreement for the VCS and good agreement for the UCS between junior and senior surgeons.
The primary objectives when it comes to treating PPFs are to maintain or reestablish stability of the implant and general fracture union. Before surgery, preinjury function should be assessed in order to appropriately manage expectations and goals for the operation. Radiographic images are assessed in order to determine fracture location, bone defect/quality present, and to assist in developing a preoperative plan. Identifying implant stability is crucial to the success of PPF fixation. This is where comparing serial images can help reveal subtle changes suggestive of loosening. A general rule when it comes to treating PPFs is that any indication of component loosening requires revision surgery. It is thought that one in five B1 fracture failures is due to misdiagnosis of component fixation. Therefore, intraoperative testing of implant stability is recommended. This further reinforces the need for careful attention to both the fracture pattern and the arthroplasty components before treating PPFs. The remainder of this section will attempt to put together a framework for treatment strategies for both early and late PPFs of the femur.
Early Postoperative Periprosthetic Fractures
The approach to managing early PPFs closely resembles the options for intraoperative prosthetic femur fractures. Regional location and component stability will influence surgical options. Bone quality typically does not influence management, as not enough time has passed for stress-shielding, osteolysis, or loosening before an early fracture occurs. Because they are often longitudinal splits with no disruption to implant stability (possibly unrecognized intraoperative fracture), Type A intertrochanteric fractures are typically managed nonoperatively with modified weight-bearing and/or activity restrictions ( Fig. 23.1 ). It is critical to understand the implant utilized for the primary surgery, as the fixation level and component coating can predict stability and potential for subsequent osseointegration ( Fig. 23.2 ). Once implant stability is compromised, revision to a distally engaging long-stem implant should be considered with supplemental fixation via cables, wires, and/or plating ( Fig. 23.3 ). Similarly, diaphyseal fractures around the stem can also be treated without surgery due to the high rates of union. While historically treated with a spica cast, nondisplaced distal fractures at the tip of a femoral stem should be managed with activity modification or surgical stabilization if the fracture is displaced or starts to propagate distally.
Early PPF of the femur with component loosening requires revision to a long-stem prosthesis that bypasses the fracture by at least two cortical diameters with additional fixation using cables, wires, plates, and/or cortical strut allografts. , Bypassing the fracture by two cortical diameters is critical, as this construct has been demonstrated to improve femoral strength to over 80% of that of the contralateral side. The degree to which the femoral stem is coated and the implant’s shape influences outcomes in early PPFs. Proximally porous coated stems with limited distal fixation have shown little evidence of success for revising a PPF. , In contrast, fully coated cylindrical or splined, tapered devices have been observed to obtain better distal fixation, establish torsional stability, and even improve bending stability with or without the addition of a cortical strut allograft. , , , In order to maximize long-term success, it is advisable to achieve at least 4 to 6 cm of intimate isthmal fixation for extensively coated cylindrical stems and at least 1 to 2 cm of fixation for tapered, splined stems. Springer et al. found a 0% nonunion rate when an uncemented, extensively porous-coated long stem was used in revision surgeries for a Type B3 fracture. Therefore, in early PPF cases with concurrent implant loosening, we recommend revision to an extensively coated implant that gains distal purchase as well as fixation of the fracture.
Late Postoperative Periprosthetic Fracture
The VCS is helpful in determining optimal management strategies for late PPFs. As mentioned previously, Type A fractures occur in the proximal femur within the trochanteric region. They can be further broken down into AG or AL depending on whether the fracture involves the greater or lesser trochanter, respectively. AL fractures are rare and can often be treated nonoperatively. However, an isolated AL fracture should raise the concern for pathologic process—further workup may be warranted or, at the very least, considered. For AG fractures, displacement less than 2 cm can be treated with protected weight bearing, limited abduction for 6 weeks, and continued monitoring with serial radiographs until fracture union (see Fig. 23.1 ). , Injuries with greater than 2 cm displacement should be reduced and secured via an open reduction internal fixation (ORIF) using cable plates, cable grips/claws, or cerclage wiring. Despite the potential for a significant limp and/or hip instability, if the bone quality of the trochanter is severely compromised, then nonoperative management may be the best option. If the fracture is associated with polyethylene wear–related osteolysis, then a head-liner exchange may be indicated, with or without fracture fixation. It is also important to note that AG fractures are often the result of an avulsion fracture due to contraction of the abductor muscles (and the associated sequelae of a Trendelenburg gait and higher risk for dislocation), typically through an area of osteolysis and/or stress-shielding. An isolated AG fracture can occur from a direct blow to the side of the hip from a fall or injury. For osteolytic-induced AG fractures, one successful strategy is to use monofilament stainless steel cerclage wires in a figure-of-eight fashion. Due to limited bone stock, poor bone quality and the significant strength of the abductor muscles, AG fractures can be difficult to successfully manage. Despite these issues, nonoperative management can be successful, with the patient using a cane for ambulation.
While some Type A fractures can be treated without surgical intervention, nonoperative management for B1 fractures leads to higher rates of malunion, nonunion, and even implant failure. Although there is widespread agreement that B1 fractures require ORIF, there is no gold standard when it comes to establishing the ideal construct. Initially, traditional dynamic compression plates were used for fixation, with at least two bicortical screws distal to the fracture. Because proximal fixation was challenging to achieve, some have introduced the combination of both distal bicortical screws and proximal cerclage fixation as a viable construct.
Today, fixation options have expanded to include locked femoral plating, cable plate constructs, strut allografts, or a combination of these techniques. Dennis et al. investigated the biomechanical properties of several plate-plus-screw/cable combinations. The constructs with greatest axial, lateral, and torsional stability were plates fixed proximally with either cables, unicortical screws, or both and distally with bicortical screws. Furthermore, Demos et al. concluded that one can use either standard or locking screws to achieve adequate fixation. Another strategy for fixing Type B1 fractures is to utilize a cortical strut allograft alone or in combination with plate fixation. It has been shown to improve the rate of fracture union, achieve better alignment, and restore femoral bone stock. However, cortical strut allografts are expensive, require host soft-tissue stripping, and can potentially serve as a conduit for disease transmission.
For patients with osteoporotic bone, locking plates are useful for achieving fixation. However, failure rates with locking plates in isolation can reach as high as 43% and are associated with a higher burden of complications when utilized around cemented implants. Strategies to augment chances for success include supplementation with a strut allograft and the use of polyaxial locking plates. , One theory for the difficulty in managing B1 fractures with locking plates focuses on the overall stripping and disruption of the soft tissue and periosteal blood supply during the procedure. It is imperative in these cases to maintain sound fracture management principles in regard to fracture fragment blood supply, the stability of the fixation, and the desired type of bone healing that is being utilized for fracture union. Despite the poor outcomes associated with locked plating, many surveyed surgeons preferred its use over cable-plate constructs. This highlights the need for better studies in order to shed some light on the optimal treatment algorithm for B1 PPFs of the femur.
In B2/B3 fractures, the implant is unstable; our recommendation is to conduct revision surgery with or without ORIF. B3 fractures are further distinguished based on the degree of femoral bone loss, which can be summarized using the Paprosky classification system ( Table 23.3 ). Modern options for revision of B2/B3 fractures include extensively coated cylindrical femoral stems, modular splined, tapered stems, allograft-prosthetic composites, and megaprostheses. , During femoral component revision procedures, it is unlikely to achieve adequate fixation with proximal fixation stems as they are associated with a greater chance for early loosening. In the setting of Paprosky IIIA, IIIB, and IV femoral defects, the bone quality diminishes, as does the area to obtain suitable fixation for a new implant. For type IIIA cases that can achieve 4 cm of “scratch-fit,” a cylindrical diaphyseal engaging implant (between 13 and 18 mm in diameter) is a safe option to consider. However, many surgeons are moving toward regularly using a tapered, splined stem in these cases, either monoblock or modular in nature ( Fig. 23.4 ). Furthermore, modular and monoblock splined, tapered stems are preferred in Type IIIB defects, as the metaphysis provides little support and there is less than 4 cm of good isthmal bone remining. For PPFs in Type IV femurs, the preferred implant options include modular tapered stems, an allograft prosthetic composite, or megaprosthesis ( Fig. 23.5 ). Sufficient preoperative planning is mandatory to ensure that the correct stem length is chosen as well as to assess prior femoral remodeling and areas of bony defects. This will ensure that the appropriate implants and grafts are available as well as those to cover a backup plan if needed.
|I||Metaphyseal bone loss (mild)|
|II||Significant metaphyseal bone loss with intact diaphysis|
|IIIA||Significant metaphyseal bone loss with at least 4 cm of intact diaphyseal cortical bone|
|IIIB||Significant metaphyseal bone loss with less than 4 cm of intact diaphyseal cortical bone|
|IV||Significant metaphyseal bone loss with insufficient diaphyseal support|
Extended trochanteric osteotomy (ETO) is a highly useful technique when revising loose cemented femoral implants, correcting varus remodeling, removing broken implants, or even for just obtaining better exposure. In a case series evaluating 14 periprosthetic fractures in which ETOs were implemented during the revision procedure, all osteotomies healed and all femoral components remained stable over a 2-year span. ETO provides access for more easily inserting femoral components, removing well-fixed cement mantles or pedestals, and correcting deformities without compromising fracture healing. Compared with cemented stems, uncemented stems with cerclage wires/cables are associated with better stability and overall results when treating Type B2/B3 fractures. , Once an ETO is implemented, the overall bone defect classification elevates to a Type IIIA or IIIB and requires appropriate management of the injury.
For PPF patients with severely poor bone stock, vaiable options are to treat with either a megaprosthesis (see Fig. 23.5 ), also known as a proximal femoral replacement (PFR), or allograft prosthetic composite (APC). In a series of 18 cases treated with either a PFR or APC, 7 patients (39%) suffered loosening complications—6 were aseptic and 1 septic. Alternatively, Klein et al. investigated 21 patients treated with a modular PFR and found that 95% were able to ambulate with minimal or no pain at 3-year follow-up. Complications reported from this cohort included 2 dislocations, 1 refracture, and 2 irrigation and debridement procedures for persistent wound drainage. In another study, 13 out of 15 patients treated with an APC reconstruction for a Type B3 fracture returned to preoperative functional levels at 5-year follow-up.
Under the VCS, fractures that occur well below the femoral stem are known as Type C fractures. The consensus strategy for these injuries is to treat the fracture with ORIF via lateral, locked femoral plating. Alternative options include ORIF with standard femoral plates, cable plates, retrograde intramedullary nails, or even allograft strut plating. To avoid future stress risers, it is crucial to overlap fixation with the distal end of the femoral implant. Overall, the literature supports the claim that ORIF management for Type C fractures results in desirable outcomes with few associated complications. , , ,
Periprosthetic femur fractures between an ipsilateral THA and TKA are also known as interprosthetic femur fractures (IFFs), or UCS Type D fractures. The same consideration of fracture location, implant stability, and quality of bone stock still apply when handling IFF complications. The major challenge is gaining adequate fixation on both sides of the fracture with the implants in the way. Historically, these injuries were treated with a Mennen plate plus bone graft from the iliac crest and restricted weight bearing for 3 months. However, further investigation showed poor results with this technique, as patients often required either a knee amputation or hip disarticulation at interval follow-up. Today, standard fixation options for IFFs include locked plating with screws and supplemental cerclage and/or bone grafting. Rozell et al. recommend the use of intramedullary nails if possible, interprosthetic sleeves, revision components, and supplemental grafting for cases of unstable or loose implants. In cases of severe bone stock reduction, they also acknowledged the benefit of PFR or total femur replacement (TFR). Modern docking systems are currently being evaluated but are not approved by the US Food and Drug Administration (FDA) for routine use. These systems clamp over the end of the stems from the THA and TKA to link the constructs and provide stability (essentially serving as an intercalary segment). IFF is a difficult problem to manage, often requiring some level of creativity and improvisation during the surgical procedure.
PPFs are an uncommon adverse event but remain highly complex with serious risk of further complications, morbidity, or mortality. A selected summary of published outcomes for femoral PPFs by treatment method is provided in Table 23.4 . , , , In patients with a PPF, the compromise in bone quality associated with the prior THA also influences the potential for fracture healing. It has been reported that 41% of complications from PPFs occur from the fracture itself compared with 33% being attributed to the arthroplasty. One year after PPF surgery, patients may have up to a 12% chance of requiring a reoperation. Many studies have investigated morbidity and mortality associated with PPFs. It has been shown that in the short term, a PPF has a similar mortality rate compared with a standard hip fracture. However, over time and after controlling for age and comorbidities, the risk of death is lower for a PPF compared with a hip fracture. Haughom et al. determined that the risk of complications in the acute setting is higher after a PPF compared with a native hip fracture. The authors further recommended that a multidisciplinary approach to managing PPFs should be considered, similar to how it is utilized for native hip fractures.