Understand the risk factors leading to periprosthetic fractures.
During the postoperative surveillance of arthroplasty patients, maintain a high index of suspicion for impending periprosthetic fractures.
Understand the Vancouver classifications for postoperative and intraoperative periprosthetic fractures.
Before undertaking surgical treatment, develop a detailed plan, including alternative strategies for managing unexpected findings such as inadequate bone stock or inadequate fixation.
Ensure that an adequate complement of implants, fixation devices, and bone graft is available.
Maintain familiarity with extensile approaches to the hip.
Ensure familiarity with all implants and bone grafting techniques that may be required to manage periprosthetic fractures, including salvage techniques.
Avoid underestimation of the quality and amount of remaining bone stock, especially in elderly patients.
Avoid missing occult fractures intraoperatively; obtain intraoperative radiographs before closure in select cases.
Minimize soft tissue stripping to optimize the local environment for fracture healing.
Rule out infection before undertaking revision surgery.
Total hip arthroplasty (THA) is among the most effective modern-day surgical procedures. One of the significant complications of THA is a periprosthetic fracture, the prevalence of which is increasing. The reported prevalence of periprosthetic fracture has varied between 0.1% and 18%.
Periprosthetic fractures can be broadly differentiated by whether they occurred intraoperatively or postoperatively and by involvement of the acetabulum or femur. As these injuries increase in frequency, the modern reconstructive hip surgeon must be well versed in the current classification and management of these problems.
INTRAOPERATIVE PERIPROSTHETIC FRACTURES
Intraoperative Acetabular Fractures
Epidemiology and Etiology
Intraoperative periprosthetic fractures of the acetabulum are less common compared with femoral fractures, or even postoperative acetabular fractures, which can result from osteolysis and may be associated with pelvic discontinuity. Such intraoperative acetabular fractures are reportedly more common with cementless compared with cemented components.
Intraoperative acetabular fractures tend to occur during insertion of the component, particularly at the time of impaction. This is a result of the need to obtain a press fit of the component in cementless THA, which requires underreaming of the acetabulum. To minimize the risk, it has been suggested that components not be oversized by more than 2 mm. Acetabular strain has been shown to continue to increase even after seating of a component that is 2 mm oversized. Patient factors contributing to the risk of these fractures include osteopenia from any cause, including osteolytic defects, and the presence of metabolic bone disease such as Paget disease.
Maintaining a high index of suspicion for intraoperative fractures is important. When a concern arises regarding a possible fracture, full assessment of the affected area is required both clinically and radiographically. Unlike postoperative femoral fractures, which may be apparent on radiographs, acetabular fractures are more difficult to identify. A review of the classification system for such fractures is presented elsewhere.
Treatment and Results
Regardless of whether an intraoperative acetabular or femoral fracture has occurred, the goal of treatment is to maximize function. This is achieved through fracture stabilization, prevention of propagation of the fracture, maintenance of component alignment and stability, and fracture union.
Della Valle et al have made recommendations regarding treatment of intraoperative acetabular fractures according to their classification system. If the fracture is associated with a well-fixed component and is not displaced, it can be treated nonoperatively. If the component is unstable, the structural integrity of the anterior and posterior columns must be assessed and supplemental screw fixation used if necessary to provide initial stability of the component. Bone graft should be used to augment bone stock. In addition, if significant mobility at the fracture site or pelvic dissociation is present, the posterior column must be plated to restore pelvic continuity. In some cases of severe acetabular bone loss in which plating the posterior column is impossible, the acetabulum may be reconstructed with allograft bone and a reconstruction cage or a combination of a tantalum metal cup with or without augments and a supporting cage to transfix the pelvis and stabilize the pelvic discontinuity.
INTRAOPERATIVE FEMORAL FRACTURES
Epidemiology and Etiology
Burden of Disease
Substantially greater attention has been paid in the literature regarding intraoperative periprosthetic femoral fractures. The risk for this injury has been shown in several studies to be greatest with cementless components at revision THA. Intraoperative femoral fractures are encountered in 1% to 5.4% of primary cases compared with up to 30% of revisions. When cemented stems are used, this risk ranges from 0.3% to 1.2%.
The surgeon must be alert to the possibility of iatrogenic fractures. Intraoperative radiographs are valuable in avoiding missed occult fractures. The stability of the component before closure must be ensured, particularly if an uncemented component is used.
Treatment of both intraoperative and postoperative fractures has been recommended according to the classification of the fracture. Although not proven reliable like the Vancouver classification for postoperative fractures, the Vancouver classification for intraoperative fractures also can be used to guide management ( Fig. 29-1 ). The intraoperative classification considers the location, pattern, and stability of the fracture. Intraoperative fractures are classified as type A if they involve the proximal metaphysis, type B if the fracture is diaphyseal, and type C if the fracture is distal to the stem tip and not amenable to the largest length revision stem. Each type is further subclassified as subtype 1, having only a cortical perforation; subtype 2, having a nondisplaced crack; and subtype 3, having a displaced unstable fracture pattern.
Treatment and Results
Accurate classification of the fracture type provides appropriate management ( Fig. 29-2 ). Type A1 fractures commonly result in a stable implant and can be treated with bone graft alone, usually obtained from the acetabular reamings. Type A2 fractures can be treated with cerclage wire placement before stem insertion to prevent propagation if a proximally coated stem is used, after which the implant usually is stable. If a fully porous coated stem is used, the fracture may be ignored as long as no distal propagation of the fracture is present. Type A3 fractures are treated with a diaphyseal fitting stem unless the fracture involves only the greater trochanter, in which case trochanteric fixation with wires, cables, or claw plates is appropriate.
Type B1 fractures usually occur during cement removal and are treated by bypassing the affected area by two cortical diameters with a longer stem, which has been demonstrated to restore stability. If a sufficiently long stem is not available, allograft cortical struts can be used. Type B2 fractures usually result from increased hoop stress during implantation of either the broach or implant. These injuries may be missed intraoperatively, in which case they usually are diagnosed on postoperative radiographs. If diagnosed intraoperatively, management should consist of cerclage wire placement and bypassing the affected area. Cortical allograft has been shown to increase the strength of the construct and provide adequate stability with good outcomes and may be used if the bone quality is deemed to be poor. If these injuries are diagnosed postoperatively, treatment should consist of protected weight bearing and close observation. Type B3 injuries usually occur during hip dislocation, cement removal, or final stem insertion. Placement of cerclage wire is appropriate for oblique patterns as long as the fracture is bypassed by at least two cortical diameters by either the stem or cortical onlay allografts; transverse fractures should be treated by bypassing the fracture site with cortical allograft or a longer stem if possible.
Type C1 fractures usually occur during cement removal or canal preparation and are treated with bone graft and bypass of the fracture site with cortical onlay allograft if the perforation is large. Treatment of type C2 fractures consists of cerclage wires and augmentation with cortical strut allograft, whereas C3 fractures are treated with open reduction and internal fixation. Of note, no comparative studies provide evidence regarding the ideal treatment of these various fracture types, but this algorithm has served the authors well over the years.
The largest study exclusively of intraoperative femoral fractures at the time of revision THA was reported by Meek et al. A total of 211 consecutive patients had 64 intraoperative femoral fractures (30%) and 147 patients who did not sustain a fracture. Each patient was treated with revision THA with an extensively coated femoral stem designed for diaphyseal press fit. The most common fracture type, according to the Vancouver classification, was B2 in 39 cases (58%) followed by B1 in 11 (16%). A wide variety of adjuncts were used, most commonly cerclage wire fixation in 25 cases (39%) followed by use of a cortical allograft strut graft in 18 (28%). No difference was found in functional outcome, with bone ingrowth of the stem in 97% and stable fibrous union in 3% at 2-year follow-up.
POSTOPERATIVE PERIPROSTHETIC FRACTURES
Postoperative Acetabular Fractures
Epidemiology and Etiology
Berry et al noted a 0.9% prevalence of periacetabular fractures with pelvic discontinuity at revision surgery. Although not uniformly straightforward to diagnose pelvic discontinuity should be considered, particularly in the setting of trauma and osteolysis.
Periprosthetic fractures of the acetabulum can be difficult to diagnose. Historic features that increase the suspicion of acetabular fracture include persistent groin pain after trauma. In the setting of extensive osteolysis or osteopenia, an episode of known trauma may not be elicited during an interview. In the absence of trauma, the differential diagnosis should include stress fracture of either the acetabulum or pubis, especially in elderly patients with significant osteoporosis.
The radiographic assessment of the acetabulum is compromised by the radiopaque component. Although the anteroposterior pelvis view may show a fracture line, indirect features often are required to establish the diagnosis of acetabular fracture. Berry et al described three radiographic hallmarks of pelvic dissociation. These include a visible fracture line through the anterior and posterior columns, medial translation of the inferior aspect of the acetabulum observed as disruption of Kohler’s (ilioischial) line, and subtle rotation of the hemipelvis inferior to the fracture, resulting in an asymmetric appearance to the obturator foramina. Judet views provide radiographs that directly project the posterior and anterior columns for better discrimination of acetabular integrity and identification of fracture lines. Computed tomographic scans can be helpful in selected cases in which the diagnosis remains in doubt despite appropriate plain radiography.
The sensitivity of current imaging technology is imperfect for detecting all acetabular fractures. Preparing for the occult pelvic discontinuity to be diagnosed intraoperatively and ensuring that appropriate implants are available are prudent actions.
The American Academy of Orthopaedic Surgeons (AAOS) classification was modified by Berry et al to direct treatment decisions. The AAOS type IV acetabular defect refers to pelvic discontinuity. Berry et al added Type IVa, defining an association with a cavitary defect (AAOS type II) or a mild to moderate segmental defect (AAOS type I). Type IVb represented moderate to severe segmental defects or combined defects (AAOS type III). Type IVc was reserved for cases with previous irradiation to the pelvis regardless of the type of bony defect present.
Treatment and Results
Early Postoperative Fractures
Fractures detected in the immediate and early postoperative period can be managed according to fracture stability. For type I fractures with minimal displacement a period of toe-touch weight bearing (6 to 8 weeks) may be sufficient to achieve union, though most cases eventually require surgery. Union probably is more likely if cementless acetabular components augmented with screw fixation into the intact superior hemipelvis were used at the index procedure. If significant displacement of the fracture is noted with subsequent instability of the acetabular component, revision surgery is necessary.
Late Postoperative Fractures
The treatment objectives for acetabular fractures that occur late include restoring column integrity, addressing bony defects, and reimplanting a stable acetabular component. Plating of the posterior column corrects the pelvic discontinuity and provides a foundation for subsequent reimplantation of a cementless implant with screw augmentation. Management techniques for bony defects can use structural or nonstructural bone graft; more recently augmentation with metallic components has been used, such as trabecular metal augments (Zimmer, Warsaw, Ind.) or similar devices currently under development by other manufacturers. A detailed description of these techniques is outside the scope of this chapter; the reader is referred to the work of Sporer and Paprosky and Boscainos et al.
POSTOPERATIVE FEMORAL FRACTURES
Epidemiology and Etiology
Burden of Disease
The prevalence of postoperative periprosthetic fractures of the femur in the literature ranges between 0.1% and 2.1%. The Mayo Clinic Joint Registry reported the largest series available, identifying a prevalence rate of 1% (238 of 23,980) in primary hip arthroplasties and 4% (252 of 6349) in revision hip arthroplasties.
Lindahl reviewed the Swedish arthroplasty registry from 1992 to 2002 and calculated the hazard function for postoperative periprosthetic fracture. The 10-year probability was 0.64% overall, ranging from 0.07% among those with low risk to 2.25% for those with high risk. Factors considered to contribute to the degree of risk were described in a previous report and included young age at the time of primary THA, loosening of the stem, and the type of implant used, with the Charnley and Exeter stems being associated with the greatest risk of fracture.
Risk Factors for Fracture
Several risk factors have been associated with the rising incidence of periprosthetic femur fractures. Patient factors include the increasing prevalence of patients undergoing total hip arthroplasty, elderly patients with hip implants at risk for low-energy falls, and young patients with hip implants engaged in activities subject to a higher likelihood of high-energy trauma events. Surgical technique also can have an impact on the potential for periprosthetic femur fractures. Revision arthroplasty techniques that transfer energy to the tip of the implant stem, such as impaction allograft and cementless press-fit implant insertion, also may predispose to periprosthetic fractures.
The majority of periprosthetic femur fractures (up to approximately 90%) result from a low-energy mechanism, usually a fall from standing height, or no significant trauma at all. High-energy mechanisms account for less than 10% of reported cases.
Appropriate classification of periprosthetic femur fractures can direct subsequent management strategies. The Vancouver classification was developed to direct treatment based on the three most important factors: fracture location, implant stability, and quality of the remaining bone stock ( Fig. 29-3 ). It was subsequently shown to be valid and reliable. The Vancouver classification divides the femur into three anatomic zones: A, the trochanteric region; B, the diaphyseal, including or just distal to the tip of the implant; and C, the distal diaphyseal remote to the distal tip of the implant.
Type A is subclassified into groups A G , representing fractures of the greater trochanter, and A L , representing fractures of the lesser trochanter. Type A G fractures tend to be stable and related to osteolysis of the greater trochanter. Type A L fractures can cause instability of the implant if the medial cortical extension of the fragment is significant.
Type B fractures are subclassified according to the stability of the fracture. In subtype 1 fractures the implant is stable and solidly fixed. In both subtypes 2 and 3 the implant is unstable. The feature differentiating subtype 2 from 3 is the available bone stock; B3 fractures have insufficient bone stock (osteopenia, osteolysis, or comminution) to support a standard cemented or porous coated revision stem.
Type C fractures are remote and distal to the implant stem and therefore not associated with stability of the implant. Fixation techniques based on the fracture pattern, being mindful of bridging the distal extent of the implant, should be used.
Treatment and Results
Ruling out Infection
The usual consideration and investigations for ruling out occult infection before undertaking revision arthroplasty procedures is complicated in the setting of a fracture. In the absence of a fracture, alternate potential sources of infection, or systemic illness serological markers such as erythrocyte sedimentation rate and C-reactive protein are sensitive surrogates of occult infection that may prompt further investigation. These markers are not reliable, however, in the setting of a periprosthetic fracture and therefore are of limited value. Joint aspiration under radiologic guidance for fluid culture is therefore routinely recommended in these cases, particularly when the patient had been symptomatic for some time before the fracture. If infection is encountered, appropriate treatment should be instituted, the details of which are beyond the scope of this text.
Principles of Operative Treatment
Except in cases in which the patient is either not a candidate for surgery because of inability to tolerate the procedure or has a stable fracture, all patients are offered surgery to stabilize the fracture. The risks associated with prolonged bed rest, including respiratory complications and thromboembolic disease, mandate that operative fixation be performed on an urgent basis. Cases historically managed nonoperatively have required prolonged inpatient admission, prolonged recumbency, and slowed mobilization and have shown higher rates of nonunion and malunion compared with operative care.
Appropriate preparation for these challenging cases is imperative. Intraoperative findings commonly do not represent the radiographic assessment made preoperatively. Occult extension and severity of the fracture, the extent of bone loss, and poor bone quality may not be fully appreciated until directly visualized. Therefore, in addition to a standard revision arthroplasty system, implants and allograft for unexpected intraoperative findings should be available. Selected cases that present with unreconstructable femurs may require proximal femoral replacement.
Familiarity with extensile exposures to the acetabulum, hip, and femur is necessary for adequate management of these fractures. The fracture itself offers an opportunity to gain access to the canal to facilitate debridement and cement and plug removal. Intraoperative tissue and fluid should be sent for analysis, followed by administration of prophylactic antibiotics covering the common gram-positive bacteria, such as a first-generation cephalosporin. Intraoperative radiographs should be made if any concern regarding fracture stability or fixation remains at the conclusion of the reconstruction.
Postoperative management should be implemented according to the characteristics of the patient and the stability of the reconstruction. Patients should be mobilized as quickly as possible, adhering to the weight-bearing restrictions appropriate for their reconstruction. In most cases, toe-touch weight bearing is recommended until definitive healing has been demonstrated. Appropriate thromboprophylaxis and antibiotic prophylaxis also should be administered.
Many different treatment options have been described for periprosthetic fractures. Historically, treatments have included nonoperative strategies such as protected weight bearing, traction, and casting or bracing. Modern techniques of operative fixation have supplanted nonoperative techniques except for protected weight bearing in highly selected cases.
Internal fixation endeavors to provide optimal fracture reduction, implant stability, a superior local environment for healing and, ultimately, the most rapid mobilization of the patient. Many techniques have been described, including modified cable-plate devices, compression and locking plates, revision arthroplasty supplemented with or without allograft, and salvage procedures such as proximal femoral replacement or allograft-implant composites. The Vancouver classification directs the treatment algorithm to help the surgeon select an appropriate treatment.
Although outcomes appear to be improving, techniques to manage these fractures still result in variable outcomes. With revision as the end point, Lindahl et al reported the 10-year survival of these injuries to be 73.2% after primary