Kuzyk, Paul R.T. MD, MASc, FRCSC; Watts, Evan MD, MSc, FRCSC; Backstein, David MD, MEd, FRCSC

Total knee arthroplasty (TKA) is one of the most common reconstructive procedures, with >700,000 such procedures performed in the United States each year.¹ Over the last decade, the rate of primary TKA in the United States has more than doubled, and a similar trend in the number of periprosthetic fractures around the knee has been observed. As a result, orthopaedic surgeons are increasingly seeing patients who have experienced this complication.

Multiple strategies have been described to manage periprosthetic fractures around the knee, all with the aim of achieving a functional, well-aligned, and stable knee. The mainstay of treatment has been osteosynthesis with either retrograde intramedullary nailing or fixed-angle plate fixation. The use of revision TKA is considered for periprosthetic fractures associated with loose or malaligned implants, with or without severe bone loss²,^3-7 (Figure 1).

Figure 1

Clinical outcomes of revision TKA are highly variable, and controversy exists regarding the optimal management of periprosthetic fractures. Complications after internal fixation include loss of fixation, nonunion, malalignment, malrotation, prosthetic loosening, and infection.⁸ Known risk factors for fixation failure include poor bone quality, malaligned TKA implants, and very distal femoral fractures. The main benefits of revision arthroplasty in these scenarios are early mobilization and improvement in knee range of motion. Here, we discuss decision making for management of periprosthetic fractures around the knee via revision TKA, with a focus on distal femur replacement and techniques involving the use of porous metal cones and sleeves.

Etiology

In the past decade, large registry databases in the United States and globally have reported the overall incidence of periprosthetic fractures around a primary TKA to be 0.3% to 1.1%.^9-12 Incidence rates increase to 2.5% to 8% in the setting of revision TKA.^9-12 Periprosthetic fractures around a primary TKA occur most frequently in the distal femur, followed by the tibia and patella. Elderly women are more commonly affected than are elderly men. Typically, the mechanism of injury is a low-energy impact in combination with an axial or torsional force. In the United States, the overall mortality rates after distal femoral fractures (native and periprosthetic) have been shown to be 6% at 30 days, 18% at 6 months, and 25% at 1 year.¹³

Risk factors for periprosthetic fracture tend to be grouped into patient factors and surgical factors. Patient factors found to be associated with increased fracture rates include osteoporosis, inflammatory arthritis, neurologic disorders, prolonged corticosteroid use, obesity, and older age. Surgical factors include inadvertent cortical perforation, anterior notching, excessive or eccentric box cuts, implant malposition, forceful impaction of trial implants, undetected intraoperative fracture, and previously placed implants.

Interprosthetic femoral fractures are rare, occurring in 1.25% of patients who have ipsilateral hip and knee arthroplasties.¹⁴ Mortality rates after revision surgery for interprosthetic femoral fractures have been shown to approach 50%.¹⁵ Decreased cortical bone thickness in the femoral shaft is the most important predictor of interprosthetic fracture in biomechanical studies. The interprosthetic distance between the tip of the ipsilateral total hip and total knee prostheses has not been shown to increase the risk of fracture in a biomechanical cadaver model.¹⁶

Classification Systems

Several systems of classifying periprosthetic knee fractures have been published^17-21 (Table 1). A periprosthetic knee fracture is one in which the fracture line is within 15 cm of the joint surface or within 5 cm of the intramedullary stem.²⁰These classification systems categorize fractures based on their location and characteristics. The Unified Classification System covers the management of all periprosthetic fractures, regardless of the affected body site (Table 1). It is a simple assessment and management tool that is based on three principles.²¹ First, the clinician must identify the location of the fracture and determine whether the fracture involves the bone supporting the implant or exists distant to the implant. Second, the clinician must scrutinize the implant fixation and determine whether the bone-implant interface was stable before the fracture and remained so after injury. Finally, the clinician must consider the adequacy of the bone stock and the strength supporting the implant and determine whether fracture fixation or revision can be performed without the need for additional major reconstruction procedures.

Preoperative Workup

After performing a careful assessment for other serious injuries, the attending clinician should obtain a detailed history of the TKA and the events surrounding the injury. Significant medical comorbidities, such as syncope, cardiopulmonary compromise, head injury, or stroke, should be identified and managed by the appropriate specialist before the periprosthetic fracture is addressed.

A history of prefracture pain, swelling, or instability may suggest loosening of a preexisting implant, infection, or polyethylene wear. Previous surgical notes and radiographic images also may provide useful information regarding the surgical approach, the brand and types of TKA implants used, the status of the collateral ligaments, and associated interoperative and/or postoperative complications.

During the physical examination, the clinician should assess previous incisions, skin and soft-tissue quality, the risks of wound complications, and neurovascular status. Assessment of the function of the medial collateral ligament is critical in determining the need for constrained implants. The function of the ligament may be difficult to assess acutely; assessment can be reserved for the intraoperative examination.

Standard radiographs of the knee, including AP and lateral views, should be obtained. Full-length femoral or tibial radiographs should be obtained when stemmed implants are present to identify any other preexisting devices that may have implications for the planned reconstruction. Radiographs should be assessed for evidence of loosening and osteolysis, as well as for implant wear or malposition. Subtle loosening of implants should be considered in the setting of fracture lines that extend under the entire prosthesis, progressive widening of the cement-bone or bone-prosthesis interface, lucencies at the metal-cement interface, cement cracking or fragmentation, and/or progressive shedding of beads from a coated prosthesis. If further information is needed to assess fracture characteristics, regions of bone loss, and implant positioning, the clinician may consider obtaining a metal subtraction CT scan and, possibly, a dual energy scan.

Laboratory studies are used to assess underlying infection and the risk of blood loss. A complete blood count, extended electrolyte levels, coagulation profile, and levels of inflammatory markers (eg, erythrocyte sedimentation rate, C-reactive protein level) are commonly obtained. Inflammatory marker levels may be slightly elevated because of the fracture. If there is concern for an underlying infection (eg, substantially elevated levels of inflammatory markers), preoperative joint aspiration may be performed.

Medical optimization is an important aspect of treatment in patients with periprosthetic fractures. An interdisciplinary approach to treatment has been shown to decrease morbidity and mortality rates in elderly patients with fractures.²²Systemic testing (eg, electrocardiography, echocardiography, chest radiography) should be performed at the discretion of the surgeon and the consulting service. The benefits associated with early surgery, including the benefits associated with early mobilization and pain control, must be measured against the benefits of the patient’s preoperative optimization.

Surgical Indications

The most common indications for revision TKA for periprosthetic fracture are implant instability and insufficient bone stock for internal fixation.^4-8 Grossly unstable implants or fractures associated with implant malalignment or suspected implant loosening are indications, as well. Relative indications include previously failed fixation of periprosthetic knee fractures and fractures associated with periprosthetic infection. Despite the established benefit associated with early mobilization in elderly patients with other lower extremity fracture patterns (eg, hip fractures), the indications for revision TKA for early mobilization and weight bearing, in the context of periprosthetic fractures around the knee, are not well described.

Management

Traditionally, internal fixation with retrograde intramedullary nailing or plate fixation has been the mainstay of treatment for periprosthetic fractures around the knee. Fracture union rates range from 80% to 100%, with no significant difference in outcomes between these modern fixation techniques (eg, locking plates, retrograde intramedullary nailing).²³ To improve union rates, especially in patients with limited bone stock, Kumar et al²⁴ achieved good results by inserting fibular allograft into the medullary canal and applying buttress plate fixation. However, this fixation technique required prolonged periods of protected weight bearing, which is especially problematic in vulnerable patient populations with increased general medical risks associated with immobility. In a meta-analysis of 195 supracondylar periprosthetic fractures, Chen et al²⁵ reported superior results when fractures were managed with revision arthroplasty rather than with open reduction and internal fixation. A more recent study reported similar findings, with arthroplasty providing better results than open reduction and internal fixation for management of periprosthetic femoral fractures, especially in the setting of osteopenic bone.²⁶

One difficulty in managing periprosthetic fractures around the knee is determining the indication for revision arthroplasty; few outcome studies on revision TKA for periprosthetic fractures exist. Those that do exist are mostly small case series on periprosthetic fracture management involving the use of revision TKA for fractures with loose implants or very distal fractures. Deshmukh et al²⁷ reported their results of femoral revision arthroplasty in 16 knees with acute, very distal (Su type III) supracondylar periprosthetic fractures. They showed that fracture union and full return to preinjury activity were achieved in all patients with the use of cemented, constrained revision TKA implants. Hou et al²⁸ reported on the outcomes of four patients with interprosthetic femoral fractures who underwent revision with long-stemmed implants to manage loose prostheses. All patients returned to preoperative functional status. Other studies have reported on expanded indications for revision TKA. Abbas and Morgan-Jones⁵ reported on their experience of revision TKA for the management of failed primary fixation after TKA periprosthetic fracture. In five patients with fracture nonunion and one with malunion after fixation, revision TKA allowed for early mobilization, with an average time to union of 8 to 18 weeks. In the setting of distal femur periprosthetic fractures with limited bone stock, Jassim et al⁶ suggested distal femoral replacement to provide adequate management when other methods of fixation were not possible. This technique was generally used in patients with low activity demands and poor bone stock, and it allowed for immediate weight bearing and early mobilization.

Implant Selection

Despite a growing level of evidence supporting the use of revision TKA for periprosthetic fractures around the knee, specific indications for the optimal type of implant to best manage these fractures are not well defined. Implant options consist largely of standard TKA revision implants, including unconstrained and constrained implants, as well as bone-replacing implants used for oncologic techniques, such as endoprosthetic implants or allograft prosthetic composites (APCs).

Host bone stock is an important factor in selecting the type of implant to be used for revision TKA. The size of the defect and overall quality of the host bone will help to determine the implants necessary to maintain the functionality and stability of the knee postoperatively. Small bone defects traditionally have been managed with polymethyl methacrylate, with or without reinforcing screws. Moderate-sized metaphyseal and diaphyseal defects (20 to 40 mm²) may be addressed with modular augments, including wedges, cones, and metal augments. Derome et al²⁹ evaluated 29 patients who underwent revision TKA with highly porous trabecular metal cones to fill moderate-sized bone defects. At an average follow-up of 33 months, no patient had experienced radiographically evident tantalum cone loosening, and two patients had been treated for periprosthetic fracture. Augmentation of large bone defects is usually managed with APCs or endoprosthetic replacement implants for revision TKA. Traditionally, the decision to choose an APC over an endoprosthesis has been based on patient age and activity level.³⁰ The historic indications for APCs were younger physiologic age (<50 years) and a high level of activity. However, the use of APCs has been mostly abandoned because of the considerable number of complications and revision surgeries related to infection, nonunion, and graft resorption or failure that have occurred. Saidi et al³¹ showed that surgical time and blood loss were significantly reduced (P < 0.05) in patients with standard revision TKA implants (139 minutes and 361 mL, respectively) and distal femur replacement prostheses (131 minutes and 300 mL, respectively) compared with patients who received APCs (217 minutes and 1,025 mL, respectively). In the oncology literature, the 5-year survivorship of such bone replacement implants after tumor resection is 80% to 90%.³² Cannon³³ reported on the use of megaprostheses in 27 patients with comminuted periprosthetic fractures associated with a loose prosthesis. That study showed good to excellent results, with low complication rates, rapid mobilization, and no need for further revision.

The integrity of the medial collateral ligament is an important consideration in managing periprosthetic fractures around the knee because ligamentous integrity will largely direct the selection of a constrained or unconstrained prosthesis.³⁴To our knowledge, no studies have directly compared the outcomes and survival of constrained versus unconstrained prostheses in the management of periprosthetic fractures. The main disadvantage of using constrained implants in revision arthroplasty is their historically lower survival rate compared with standard nonconstrained implants. In a study of 44 patients treated with a rotating hinged knee implant, including 4 patients with periprosthetic fractures, Pour et al³⁵reported a survival rate of 79.6% at 1 year and 68.2% at 5 years. The mean follow-up was 4.2 years (range, 2.0 to 8.0 years). The main cause of failure was mechanical loosening. To improve implant survival, investigators have used slightly less constrained implants. In a long-term outcome study of revision TKA involving the use of constrained condylar knee prostheses, Kim et al³⁶ reported a 16-year implant survival rate of 94.7% with an end point of loosening. In addition to the varying degrees of soft-tissue constraint, the selected prosthesis must also address the bone loss that is usually encountered during revision surgery.

Complications of revision TKA for management of periprosthetic fractures are similar to those encountered during other revision knee surgeries. These complications include deep wound infection, patellar complications, extensor mechanism disruption, prosthetic dislocation, periprosthetic fracture, and implant breakage. However, treatment of patients with periprosthetic fractures is generally more complex because of older age, poor bone stock, and generally, a higher number of comorbidities. These factors can place patients at an increased risk of perioperative complications. Pour et al³⁵ reported that 14% of patients experienced a medical complication postoperatively and that 16% of patients had a surgical complication requiring revision within the first 3 years postoperatively. In particular, women aged >70 years should be warned of a markedly increased risk of periprosthetic fracture after TKA.

Surgical Options

The major considerations for implant selection include fracture characteristics, bone stock, and ligamentous integrity.Figure 2 shows a basic treatment algorithm. In addition to traditional revision TKA, the main surgical options for management of periprosthetic fracture after TKA include techniques for reconstruction of tibial and femoral bone stock (eg, highly porous augments, allograft), and endoprosthesis.

Reconstruction of the Tibia and Femur

Selection of the appropriate reconstruction techniques depends on the fracture characteristics and bone stock after implant removal. Implants are usually removed by using a reciprocating saw and osteotomes to disrupt the implant-cement interface. Typically, both the femoral and tibial implants are revised because this ensures the reestablishment of both knee alignment and balance. Conversion to stemmed implants is also required to provide additional fixation for a more highly constrained implant and to achieve fixation when the metaphyseal bone is deficient. The use of noncemented stems is preferred; however, cemented stems are used in patients with poor diaphyseal bone stock.

To reconstruct the tibia after a fracture, the surgeon recuts the tibial plateau to expose fresh bone by means of an intramedullary tibial cutting guide. Bone defects on the tibial side are noted, and a strategy is formulated for bone reconstitution with augments or porous metal cones or bone replacement, as needed. Segmental medial or lateral uncontained defects of ≤20 mm² that lack a cortical rim are managed with metal augments. Management of bone defects >20 mm² may require a porous metal metaphyseal augment or structural allograft (Figure 3). Circumferential defects (involving both the medial and lateral sides) may be reconstructed in a similar manner with both medial and lateral augments; however, a polyethylene liner of increased thickness (eg >12 mm) may also be used to reconstitute the joint line. Therefore, circumferential defects ≥40 mm may be managed with a combination of augments and a polyethylene liner. Contained defects ≥40 mm with a cortical rim may be managed with metaphyseal sleeves or porous metal cones. For defects >40 mm, a tibial APC or proximal tibial endoprosthesis is needed to provide a stable construct. The tibial baseplate is sized and the rotation is set by aligning the center of the trial implant with the middle to middle one third of the tibial tubercle. A trial tibial baseplate with an appropriately sized stem is then assembled and inserted.

Figure 3

The femur is reconstructed with the knee flexed to 90° so the desired implant will provide a stable flexion gap while avoiding medial-lateral overhang. We use diaphyseal fitting stems whenever possible because they can achieve an excellent fit and assist with limb alignment. In revision surgery, many landmarks, such as the Whiteside line (AP axis) and the posterior condylar line, are absent; thus, femoral rotation is based on the transepicondylar axis whenever possible. The surgeon inserts a trial femoral implant and stem with a trial polyethylene liner of a thickness chosen to provide a stable flexion gap. The surgeon then brings the knee into extension carefully and chooses the proximal-distal position of the femoral implant that will allow full extension and a stable extension gap.

Femoral bone defects are managed in a similar fashion as those on the tibial side. Segmental, uncontained defects of up to approximately 20 mm in size may be reconstructed with augments, depending on the revision system used. Circumferential defects <20 mm can be managed similarly by using augments on both the medial and lateral side. Circumferential defects >20 mm require the use of trabecular metal cones or metaphyseal sleeves (Figure 4).

Figure 4

Distal Femoral Endoprosthesis

In the setting of a distal femoral fracture with bone that is too comminuted or too distal for standard revision TKA, or in which bone stock prevents acceptable fixation techniques, a distal femoral endoprosthesis is typically the best choice (Figure 5). The clinician should consult company-specific indications, instrumentation, compatibility, and surgical technique for distal femoral replacement implants before selecting this technique. Historically, APC grafts were considered to be the standard of care to reconstruct these complex fracture patterns in patients with poor bone stock, as shown in Figure 6. Although this practice may still be considered for younger patients (<50 years) with comminuted distal femoral fractures, its utility in elderly patients with distal femoral periprosthetic fractures has been largely abandoned for endoprostheses.

Figure 5

Figure 6

Revision TKA with a distal femoral replacement prosthesis typically begins with a perpendicular cut in the distal femoral shaft to remove distal bone and the initial prosthesis. Preparation for this cut involves a 360° dissection of tissue off the femoral bone while protecting important posteriorly based neurovascular structures. Once completed, the surgeon can use the distal fragment as a lever to free all tissue so the distal bone and implant can be removed safely. The tibial implant revision is performed at this time.

The femoral shaft is prepared with hand reamers and trialed until a secure fit is achieved. Assembling the trial implant will allow the knee to be assessed via physical examination and will help to determine the length and rotation of the femoral implant. The examination should focus on implant stability during flexion and extension, as well as patellar tracking. Patellar maltracking and knee hyperextension are the most important considerations and should be avoided.

Proximal Tibial Endoprosthesis

A proximal tibial endoprosthesis should be considered in the setting of a proximal tibial periprosthetic fracture that cannot be salvaged with fixation techniques or standard revision implants. This surgical technique requires careful preoperative planning, beginning with an evaluation of the fracture and vascular pattern around the knee. The clinician should be well acquainted with the major branching patterns of the popliteal artery distal to the popliteal fossa, to limit the possibility of iatrogenic injury. The distal extent of the incision should be incorporated into the full-thickness anteriorly based incision. The two most critical components to the success of this procedure are maintenance of the extensor mechanism and preservation of soft-tissue coverage of the implant anteriorly. If the tibial tubercle is involved in the fracture and avulsed with bone, it should be repaired back to the proximal tibia using screws and/or cerclage wires. If the proximal tibial bone stock is severely deficient and must be replaced with an endoprosthesis, the surgeon should select an implant that has a porous-coated proximal tibial component and fixation holes and thus allows for attachment of the patellar tendon (with or without the tibial tubercle) directly to the implant. Active knee extension should be restricted postoperatively for ≥8 weeks to allow for healing of the extensor mechanism reattachment in this scenario. Soft-tissue coverage of the tibial implant may include a rotational flap containing the medial head of the gastrocnemius muscle.

Revision TKA with proximal tibial replacement generally begins with a perpendicular cut in the tibial shaft to remove distal bone and prosthesis. The tibial shaft is prepared with hand or power reamers and trialed until a secure fit is achieved. At this time, the surgeon would perform the femoral implant revision with a stemmed implant. Assembling the trial implant will allow the knee to be assessed by means of a physical examination. The tibial tubercle and joint line are assessed for alignment and rotation based off the ankle, as in standard TKA techniques.

Postoperative Regimen

The goal of revision TKA surgery for periprosthetic knee fractures is rapid mobilization with immediate range-of-motion exercises and rapid progression to full weight bearing. The advantage of distal femoral replacement and most standard revision TKA techniques is the allowance of immediate and full weight bearing.

The patient should receive pharmacologic thromboembolism prophylaxis on the first postoperative day. First-generation cephalosporin prophylactic antibiotics are usually administered intravenously for 24 hours. Dressings and drains are removed at 48 hours postoperatively, and urinary catheters are discontinued within the first 2 days.

Summary

The primary goal of revision arthroplasty for periprosthetic fracture is to restore knee alignment, rotation, and stability, as well as to allow for early mobilization and return of function. Variable results have been achieved using treatment algorithms involving the use of contemporary osteosynthesis techniques to manage complex fractures or failed primary TKA. In these scenarios, revision TKA should be considered as a reconstructive option. Traditional indications for revision TKA include grossly unstable implants, fracture lines immediately adjacent to the implant components, and poor bone stock for internal fixation. The use of constrained implants should be considered in the setting of ligamentous instability. Applying these principles of revision TKA for periprosthetic knee fractures will allow orthopaedic surgeons to manage these injuries more effectively.

References

References printed in bold type are those published within the past 5 years.