Periprosthetic Fractures

James R. Berstock, MBChB, MRCS, FRCS (T&O), MD, PGCert Med Ed

Donald S. Garbuz, MD, MHSc, FRCS

Bassam A. Masri, MD, FRCS

INTRODUCTION

The incidence of periprosthetic fractures (PPFs) around a total knee arthroplasty (TKA) is rising and is believed to have quadrupled between 2000 and 2008 in the United States.¹ This is likely due to increasing numbers of TKAs within an aging, more osteoporotic population.² Approximately 1 in 40 patients who undergo primary TKA will sustain a PPF.³ It has been estimated that intraoperative PPFs occur during approximately 4% of TKAs, often involving the medial femoral condyle.⁴ A study of the Scottish National Database revealed a postoperative 5-year incidence of 0.6% after primary TKA and 1.7% after revision TKA.⁵ PPFs around a TKA may occur in the femur, tibia, or patella and have been defined as occurring within 15 cm of the joint surface or within 5 cm of an intramedullary stem.⁶,⁷,⁸ Distal femoral fractures are the most common PPF occurring in 0.3% to 2.5% of TKAs.⁹,¹⁰,¹¹,¹²,¹³ Patellar fractures occurred in 0.68% in a series from the Mayo clinic,¹⁴ and others report fractures in 0.15% to 21% of resurfaced patellae and 0.05% of nonresurfaced patellae after a TKA.¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹ The least commonly observed are tibial PPFs, reported in 0.4% to 1.7% of TKAs.²⁰

Most fractures result from a low-velocity injury.²¹ Risk factors include inflammatory arthropathy, steroid use, age > 70 years, poor bone stock, neurological disorders, and revision arthroplasty.⁵,²² Prosthesis-related factors include loosening and osteolysis secondary to polythene wear.²¹ Notching during the surgical preparation of the femoral component no longer appears to be a significant risk factor for PPF.²³

Patients presenting with knee PPFs are more likely to be medically comorbid, female, and older when compared with patients undergoing revision TKA for other reasons.²⁴ Consequently, the 1-year mortality following a distal femoral PPF has been reported at 20.6%.²⁵ Complications following the treatment of this complex problem are also common. In a recent review of 58 distal femoral PPFs, readmission within 90 days of treatment occurred in over 20% of patients.²⁴ In a prospective study of 37 PPFs around a TKA, only 68% reached their prefracture mobility by 1 year and 22% had undergone surgical revision for various reasons. Additionally, nonoperative complications occurred in 16%.²⁶

The financial implications of PPFs around TKAs have also been studied. The costs of treatment with revision arthroplasty were on average $37,680, with readmissions each costing $16,806 in 2013. Fracture fixation costs $25,539, with readmissions costing an average of $15,269.²⁴

The goal of treatment is to expeditiously restore function while avoiding complications. This involves restoration of limb alignment, length, and rotation while enabling early mobilization. The treatment of knee PPFs may be challenging because of poor bone quality and the presence of components (proximally and distally), which may or may not be loose. Operative management has demonstrated benefits over nonoperative management of distal femoral fractures without a prosthesis,²⁷ and for the same reasons it forms the mainstay of treatment in PPFs after TKA.

DISTAL FEMORAL PERIPROSTHETIC FRACTURES

Classification

Historic classifications of distal femoral PPFs such as those from Neer et al in 1967,²⁸ DiGioia et al in 1991,²⁹ and Chen et al in 1994³⁰ concern themselves with fracture displacement and provide guidance as to the suitability of nonoperative management or surgical fixation. With the evolution of fixation and revision arthroplasty techniques, a modern group of classifications attempts to direct specific surgical treatment for each fracture type.

Rorabeck and Taylor acknowledged the important role of revision arthroplasty and described three types of fracture in 1997 (Table 57-1).³¹,³² Type 1 represents a nondisplaced fracture with a well-fixed femoral component, type 2 describes a displaced fracture with a well-fixed femoral component, and type 3 is reserved for a loose or failed component. This classification promotes the use of revision arthroplasty for the treatment for type 3 fractures and remains highly influential today.

Su et al classified distal femoral fractures according to location, with a view to guiding the use of retrograde or antegrade nail fixation.³³ Kim et al considered bone stock, component fixation, and fracture reducibility in developing their classification system.³⁴ Further, Backstein et al sought to determine the feasibility of retrograde nailing.³⁵ Frenzel et al include consideration of the timing of treatment after injury.³⁶ Fakler and colleagues integrate femoral component design into their 2017 classification to create 16 fracture types.³⁷

TABLE 57-1 Rorabeck Classification

Type	Description	Treatment
Type 1	Nondisplaced; component intact	Fixation or nonoperative
Type 2	Displaced; component intact	Fixation (nail or plate)
Type 3	Displaced; component loose or failing	Revision arthroplasty

The Unified Classification System (UCS), based on the Vancouver classification of PPFs of the proximal femur, provides a pragmatic guide to treatment while considering fracture location, implant loosening, bone stock, and other implants (Table 57-2).³⁸ Within the UCS, the most important categories with reference to the presence of a knee replacement are types B1, B2, and B3. The type D fracture (interprosthetic) requires special consideration because of the presence of a hip stem/implant within the proximal femur. Type E fractures can be managed as individual fractures of each bone, which should be classified individually.

Nonoperative Management

Deforming forces from the gastrocnemius which cause posterior angulation and rotation of the distal femur are compounded by the shortening forces from the quadriceps and are challenging to control with casting and bracing techniques alone. Malunion occurred in all displaced fractures treated nonoperatively in a series by Moran et al.³⁹ Nonoperative management is therefore reserved for completely nondisplaced fractures with inherent fracture stability and for the frailest of patients who are unlikely to survive surgery. Where appropriate, we recommend 12 weeks of hinged knee bracing with non-weight-bearing for a minimum of 6 weeks and regular radiographic assessment.

TABLE 57-2 Unified Classification System

Type		Example	Treatment
A (apophyseal)		Tibial tuberosity Inferior pole fracture of the patella Epicondylar avulsion fracture	Operative or nonoperative depending on displacement
B (bed of implant)	B1: Well fixed	Distal femur fracture	Fixation (nail or plate)
	B2: Implant loose	Distal femur fracture	Revision
	B3: Implant loose and poor bone stock	Distal femur fracture	Complex revision
C (clear of implant bed)		Femoral diaphysis fracture	Fixation (nail or plate)
D (dividing)		Fracture diving the supporting bone between a THA and TKA.	Plating
E (each of two bones supporting an arthroplasty)		Femur and tibial fracture	Manage each on its merit
F (facing an implant)		Fracture of an unresurfaced patella	Manage individual fracture

Fixation (Retrograde Nailing or Plating)

Distal femoral fractures with a well-fixed femoral component (B1 fractures, e.g., Figs. 57-1 and 57-2) may be treated satisfactorily with internal fixation (Fig. 57-3). The 2008 systematic review of 415 distal femoral PPFs by Herrera et al included data from 29 case series of patients treated with either retrograde intramedullary nailing (RIMN) or plate fixation. They identified an overall nonunion rate of 9%, fixation failure of 4%, infection in 3%, and revision surgery in 13%. This review reports an 87% relative risk reduction of nonunion and a 70% relative risk reduction of revision surgery when RIMN is used when compared with traditional nonlocking plating.⁴⁰

However, there was a nonsignificant trend toward benefit from emerging modern locking compression plate (LCP) technology.⁴⁰

A systematic review of the modern era of treatment by Ebraheim et al in 2015 analyzed all treatments for distal femoral PPFs, with displaced B1 fractures being the most common pattern identified.⁴¹ Ebraheim et al concluded that the most successful treatments were LCP (union rate 87%, complication rate 35%) and intramedullary nail (union rate 84%, complication rate 53%).⁴¹ The most common complications in both groups were nonunion, malunion, delayed union, and the need for revision surgery.

The meta-analysis by Shin et al in 2017 included eight randomized controlled trials comparing RIMN with LCP for distal femoral PPFs. Postoperative Knee Society Scores, time to union, nonunion rates, and revision surgery requirements were not significantly different between LCP and RIMN treatment groups.⁴²

RIMN may have theoretical biomechanical advantages over LCP due to it being coaxial with the anatomical axis of the femur and therefore providing greater stiffness under axial loading than a unilateral locking plate. This may be particularly beneficial when there is medial comminution. However, RIMN can only be used in open-box knee designs where the notch limits the nail diameter
and influences the nail entry point which may be more posterior than required, leading to fracture hyperextension.⁴³ Compatibility studies of common TKA and RIMN designs are reported in the literature.⁴⁴,⁴⁵ The presence of a proximal implant may induce a stress riser at the junction between implants. There also needs to be sufficient bone distally for solid fixation with the distal locking screws. As a result of these limitations and depending on the particular case, LCP might be preferable to RIMN. Also, lateral translation of the femoral component may force an incorrect intraarticulations entry point for nail, forcing the fracture to drift into malalignment, usually in too much valgus.

Only gold members can continue reading. Log In or Register to continue