Review Article Periprosthetic Fractures Around Loose Femoral Components

Abstract Roshan P. Shah, MD, JD The development of periprosthetic fractures around loose femoral Neil P. Sheth, MD components can be a devastating event for patients who have undergone total hip arthroplasty (THA). As indications for THA expand Chancellor Gray, MD in an aging population and to use in younger patients, these fractures Hassan Alosh, MD are increasing in incidence. This review covers the epidemiology, risk Jonathan P. Garino, MD factors, prevention, and clinical management of periprosthetic femoral fractures. Treatment principles and reconstructive options are discussed, along with outcomes and complications. Femoral revision with a long-stem prosthesis or a modular tapered stem is the mainstay of treatment and has demonstrated good outcomes in the literature. Other reconstruction options are available, depending on bone From Columbia University, New York, quality. Surgeons must have a sound understanding of the diagnosis NY (Dr. Shah), the University of and treatment of periprosthetic femoral fractures. Pennsylvania, Philadelphia, PA (Dr. Sheth, Dr. Gray, and Dr. Alosh), and the Pennsylvania Orthopedic Center, Alvern, PA (Dr. Garino). otal hip arthroplasty (THA) The incidence of periprosthetic hip Dr. Shah or an immediate family Treliably treats hip pain caused by fractures is increasing. Bhattacharyya member has stock or stock options articular cartilage degeneration; et al5 found a 216% increase between held in Pfizer, Merck, GlaxoSmithKline, Alnylam, and Intuitive Surgical. however, the development of peri- 2002 and 2006. Several reasons for Dr. Sheth or an immediate family prosthetic femoral fractures after the increase have been proposed.1,2 member serves as a paid consultant to THA is a devastating complication. First, the total prevalence of patients Zimmer. Dr. Garino or an immediate Two registry studies form the foun- living with THA is increasing. Sec- family member has received royalties from Smith & Nephew; is a member of dation of our understanding of these ond, with time, the number of pa- aspeakers’ bureau or has made paid fractures. Between 1969 and 1999, tients experiencing osteolysis to presentations on behalf of Smith & 1,249 fractures out of 30,329 hip a varying degree, as well as compo- Nephew; serves as a paid consultant to arthroplasty cases were studied in the nent loosening, will increase. Third, Smith & Nephew and DePuy; has 1 received research or institutional Mayo Clinic Joint Registry. Peri- as patients age, there is a greater risk support from Zimmer; and serves as prosthetic fractures in the Swedish for the development of osteoporosis a board member, owner, officer, or National Hip Arthroplasty Registry and periprosthetic fracture caused committee member of the were studied retrospectively between by minor trauma. Fourth, with the Pennsylvania Orthopedic Society. 2 Neither of the following authors nor any 1979 and 1998 and prospectively success of THA and its expanding immediate family member has received between 1999 and 2000.3 The latter indications, more patients are young anything of value from or has stock or study found that femoral peri- and active; consequently, this patient stock options held in a commercial prosthetic fractures were the third population has a greater exposure to company or institution related directly or indirectly to the subject of this article: most frequently reported reason for higher energy trauma and therefore Dr. Gray and Dr. Alosh. reoperation after THA, accounting an increased risk of periprosthetic J Am Acad Orthop Surg 2014;22: for 9.5% of the revisions between fracture. Fifth, the expanding class of 482-490 1999 and 2000.3 Other studies have patients with THA logically leads to reported the prevalence of late peri- a greater number of patients requir- http://dx.doi.org/10.5435/ JAAOS-22-08-482 prosthetic hip fractures to be ing revision THA. between 0.1% and 18%,4 with an Several treatment options have Copyright 2014 by the American Academy of Orthopaedic Surgeons. annual incidence of between 0.045% been proposed based on an accurate and 0.13%.2 classification of the fracture type and

482 Journal of the American Academy of Orthopaedic Surgeons

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Roshan P. Shah, MD, JD, et al

implant stability. Preoperatively dif- technique, and the type of implant department after sustaining spon- ferentiating between a loose and used.9,10 Osteolysis and loosening are taneous or low-energy trauma and a stable femoral implant can be most directly related to Vancouver maybeseeninthetraumabayafter challenging, but certain signs and type B2 and B3 fractures. Routine sustaining high-energy trauma. symptoms can be helpful. Fractures clinical follow-up is necessary to Advanced Trauma Life Support occurring around loose implants identify patients at risk for loosening, guidelines should always be used for require femoral stem revision. and regular radiographic evaluations initial patient evaluation following have been shown to be cost effec- trauma. Conducting an appropriate tive.11 Biomechanical studies have physical examination will help Vancouver Classification demonstrated that loose femoral reveal associated musculoskeletal stems have a nearly 60% reduction in injuries. A thorough history pro- The Vancouver classification is the the torque to failure compared with vides information about the medical most common classification system well-fixed stems.12 In a study by Beals causes of spontaneous or low-energy used to describe periprosthetic frac- and Tower,13 27% of patients with injury, such as syncope, cardiopul- tures. This classification is based on fractures had evidence of loosening monary compromise, or stroke; any fracture location, implant stability, preoperatively. The Swedish registry underlying disorders should then be and integrity of the residual bone showed that 70% of fractures managed by the appropriate spe- stock.6,7 Type B fractures occur along involved loose prostheses, with 23% cialist. As with any femur fracture, the length of the femoral stem. Type B known to be loose and 47% first significant blood loss may occur, fractures are further subdivided by identified as loose at the time of sur- requiring close monitoring of car- stem stability and bone stock. Type gery.2 It is unclear what contribution diopulmonary vital signs and vol- B1 fractures are fractures with a sta- infection has to loosening and sub- ume status. Medical co-management ble femoral component. Type B2 sequent fracture. The inflammatory is the most effective method by fractures are fractures with a loose markers erythrocyte sedimentation which to optimize care of patients femoral component but with sup- rate and C-reactive protein have poor with periprosthetic fractures before portive femoral bone stock. Type B3 specificity in the setting of a frac- surgical intervention. fractures are fractures with a loose ture.14,15 However, intraoperative Previous surgical notes should be femoral component and associated aspiration for cell count and culture obtained, especially if the index pro- poor integrity bone stock, wherein studies provides valuable information cedure was performed at an outside the metaphyseal and diaphyseal bone when the suspicion for infection is institution, to properly identify the stock is deficient and unsupportive. high. We use standard cut-off values currently implanted devices. If prior The incidence and prevalence of each of 3,000 WBC/mLand80%poly- records are unavailable, consultation Vancouver type are not well known, morphonuclear cells for diagnosing with other surgeons and industry rep- for the same reasons that make the true infection. Because of fracture bleed- resentatives can help determine the incidence of all periprosthetic fractures ing, we correct for the contribution of components and the manufacturers. elusive. In the Swedish registry report, serum white blood cells with the Finally, preoperative surgical planning 53% of the fractures were type B2 and following formula: is critical in achieving clinical success. only 4% were type B3.8 When the Arrange to have all extraction devices group examined whether the index 5 WBCcorrected WBCobserved and revision equipment available for surgery was a primary THA or a revi- ÂÀ 2 WBC · RBC = the implanted components. In addi- sion THA, they found that fractures serum fluid ÁÃ tion, a thorough and systematic eval- occurring after primary THA were RBC ;16 serum uation of the acetabular component more commonly type B2, whereas should be conducted intraoperatively fractures occurring after revision THA because an acetabular revision requir- were more commonly type B1.2 where WBC is white bloods cells and RBC is red blood cells. ing more than a liner exchange may be necessary (ie, loose component, poor Risk Factors component position, poor implant track record, shell or locking mecha- General risk factors for the develop- Management nism damage, extensive corrosion). ment of periprosthetic fractures after Radiographic studies consisting of THA include osteolysis and loosen- Preoperative Evaluation an AP pelvis and AP and cross-table ing, trauma, age, gender, osteoporo- Patients with periprosthetic fractures lateral views of the hip should sis, index diagnosis, revision surgery, generally present to the emergency be obtained to assess for fracture

August 2014, Vol 22, No 8 483

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Periprosthetic Fractures Around Loose Femoral Components

morphology, prosthetic stability, radiographs can show signs of loos- splines on the stem provide rotational component malposition, bone stock, ening (eg, subsidence, circumferential stability. Severely ectatic femoral ca- the location of any osteolysis, and the radiolucent lines, evidence of a sup- nals may require proximal bone presence of wear. Lateral frog-leg portive pedestal in a proximally fixed reconstitution with either structural views are also valuable in patients stem, and cement mantle fracture).19 allograft (ie, allograft prosthesis who can tolerate the positioning of the Ultimately,evidenceofloosening composite) or a mega prosthesis (ie, leg. Full-length femur radiographs can may not be revealed until the intra- proximal femoral replacement). reveal bony deformities, such as operative assessment is conducted; The complexity of the fracture and proximal femoral varus remodeling, therefore, all implants and revision the degree of proximal femoral re- total knee arthroplasty, or the pres- instrumentation must be made avail- modeling determines whether the frac- ence of other hardware that may able for all periprosthetic fracture ture is reduced and fixed first or interfere with the planned recon- surgeries. Although others have rec- whether the prosthesis is implanted and struction. In some circumstances, CT ommended routine hip dislocation the proximal bone is wrapped around may be used as an adjunct for assess- and stability testing, the trade-off of the implant. In the presence of little ing the fracture, areas of bone spot- extra soft-tissue dissection is not comminution and few fracture frag- welding, regions of bone loss, and the trivial. When preoperative evaluation ments, it may be preferable to reduce position of the component. suggests that a type B1 fracture is and fix the fracture before implanta- present, we rely on thorough testing of tion. This optimizes compression and the prosthesis-bone interface through bony contact, thus adhering to stan- Distinguishing Vancouver the fracture site to determine implant dard fracture treatment principles. Type B1 From Type B2/B3 stability. In certain situations, the best strategy Historically, outcomes were poor for may require the implantation of a dis- periprosthetic fractures.13,17,18 How- tally fixed revision prosthesis first while Surgical Principles for ever, with the advent of the Van- exploiting the fracture as a window. An couver classification in 1995, results Loose Stems extended trochanteric osteotomy of the have improved, thus reinforcing the The diagnosis of a loose femoral stem proximal fracture fragment may be importance of the system in guiding in the face of a periprosthetic fracture required when the proximal femur diagnosis and treatment. It is not necessitates femoral component demonstrates varus and retroversion uncommon for type B2 fractures to revision. Surgical treatment princi- remodeling.21 Severe proximal femo- be mistaken for type B1 fractures. ples for loose femoral components ral remodeling will not allow for the Determining whether a stem is well- are based on several factors, includ- insertion of a long stem through the fixed or loose is critical; this is likely ing fracture morphology, the host deformity without destroying the the most important step when man- bone quality, patient age, and func- proximal femur. In this scenario, the aging these patients. When fractures tional status.20 When adequate bone diaphyseal preparation is facilitated are misclassified, failure rates are stock remains, the principles of through direct visualization of the higher. In the Swedish registry, a 30% reconstruction require the removal femoral canal and bypassing the reoperation rate was found for B1 of the loose device, cement, and proximal deformity. After the stem is fractures treated with open reduction biomembrane; reconstruction of the implanted, the proximal fragments can and (ORIF), whereas host femoral shaft; and the implan- be debulked and reduced around the the failure rate for B2 fractures treated tation of a long-stemmed prosthesis proximal stem with cables or Luque with revision was only 18.5%.3,8 The to obtain stable distal fixation. In the wires. Proximal femoral remodeling is authors inferred that some of their B1 past, extensively porous-coated de- a more common occurrence with loose fractures were actually B2 fractures vices have been considered the most cemented femoral stems.22 and should have been revised for dependable for the management of optimal treatment. these fractures; however, modular Implant Selection for Relevant history that contributes to stems are becoming more popular. making this determination includes With fractures occurring in the set- Vancouver Type B2 preexisting groin or thigh pain, pain ting of major bone loss (ie, Vancouver Fractures with non–weight-bearing range of type B3), reconstruction typically en- motion, progressive limb shortening, tails device removal and débridement. Extensively Porous-coated and persistent symptoms or signs of This is followed by the use of Stems infection. Radiographic comparison a modular tapered stem that requires The monoblock, extensively porous- with immediate post-arthroplasty ,4 to 6 cm of interference fit; the coated, noncemented prosthesis has

484 Journal of the American Academy of Orthopaedic Surgeons

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Roshan P. Shah, MD, JD, et al

Figure 1 using extensively porous-coated stems showed that 111 of 113 frac- tures (98%) had united by a mean follow-up of 5.6 years, whereas only 6 showed evidence of loosening.24 Complications associated with these implants have been described. Garcia- Rey et al24 reported subsidence of .1 cm in 19 patients (48%) between 6 and 12 weeks. Twelve of 23 treated Vancouver type B2 fractures (52%) and 10 of 17 treated Vancouver type B3 fractures (59%) subsided more than 1 cm. Clinically, six patients (15%) had a limb-length discrepancy of .1 cm, and two patients had a discrepancy of .2 cm. Nine patients (23%) required the use of one crutch for a limp. There were no dislocations, three postoperative hematomas (8%), three intraoperative greater trochanter fractures (8%), two missed fractures (5%) at the distal stem tip (one of whichresultedinadisplacedfrac- ture), one supracondylar femoral Monoblock, extensively coated, noncemented stem for a type B2 fracture. A, fracture (3%), and two reoperations Hip radiograph showing fracture and stem subsidence. B, Postoperative (5%) for removal of cerclage wires radiograph showing satisfactory stem position and fracture reduction; offset is because of superficial infection. overcorrected. Sheth et al10 reported on 21 type B2 and B3 fractures treated with exten- sively coated, diaphyseal implants and been the reliable implant for femoral of the neck independently of the found complications in seven patients reconstruction in type B2 fractures stem may result in hip instability. (33%), including instability, sub- (Figure 1). These devices bypass the Noncemented, extensively porous- sidence, infection, heterotopic ossifi- fracture and achieve distal diaphy- coated long stems have had favorable cation, and osteotomy nonunion. seal fixation, with at least 4 to 6 cm results in the literature. Garcia-Rey of interference fit, which is recom- et al24 reported on 20 type B2 and mended for optimal stability of the 15 type B3 fractures treated with Modular Tapered Stems implant.23 these devices without allograft. All Both monoblock tapered stems (ie, Monoblock stems are available in fractures united, and no patients re- Wagner-type stems) and modular either a straight or a bowed geometry. ported thigh pain at an average of tapered stems allow for diaphyseal Straight stems confer three-point fixa- 8.3 years. Average Merle d’Aubigné fixation, rotational stability through tion in a bowed femur; however, there and Postel scores were 5.8 for pain, splines, reduced risk of anterior perfo- is a risk of anterior distal cortical per- 5.2 for function, and 4.9 for range of ration, and independent control of foration when the stem length exceeds motion. Interestingly, the authors component anteversion. Modular the tolerance of the native femur’s reported an increase in the cortical tapered stems are more commonly of curvature. Bowed stems may index and cortical bone thickness used than monoblock designs and have be necessary when the stems surpass despite the use of diaphyseal fixation become the new dependable implant of a length of 8 inches; however, lim- and concerns for stress shielding. femoral reconstruction.25-27 These itations exist regarding the ability This effect was greater for stems devices comprise a distal component to change the anteversion of the smaller than 16 mm and in patients that engages the diaphysis and prosthesis because of the bow. The with mild or moderate preoperative a modular proximal component that inability to change the anteversion osteoporosis. A review of five studies mates via a Morse taper, allowing

August 2014, Vol 22, No 8 485

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Periprosthetic Fractures Around Loose Femoral Components

27 Figure 2 for4to6weeks.Abdeletal reported on 25 Vancouver type B2 and 19 Vancouver type B3 fractures treated in this manner in a 4.5-year follow-up. Forty-three of 44 patients (98%) reached radiographic union with a Harris Hip score of 83. Munro et al26 reported on 38 Vancouver type B2 and 17 Vancouver type B3 frac- tures with a 4.5-year follow-up. Fifty- four fractures (98%) healed by 24 months, with an average WOMAC (Western Ontario and McMaster Universities Osteoarthritis Index) score of 76 for B2 fractures and 77 for B3 fractures. There were two stem re- visions and eight reoperations. Com- plications reported in both studies included instability, subsidence, and infection.26,27

Long-stem Cemented Prosthesis In rare circumstances, when the surgeon predicts a low likelihood of biologic fixation (eg, postradiation therapy, severe osteoporosis, severe Modular tapered stem for a type B2 periprosthetic fracture. A, Hip radiograph showing fracture and stem subsidence. B, Postoperative radiograph showing femoral bone loss), reconstruction satisfactory position of the modular tapered stem with good fracture reduction. may be best addressed with a ce- mented long-stem prosthesis. The surgeon should achieve anatomic independent control of anteversion, with poor bone stock. A prophylactic fracture reduction, optimally with length, and offset (Figure 2). Modular cableshouldbeplacedaroundthe compression, before reconstruction tapered stems can obtain stability with intact femur distal to the fracture to to minimize cement extrusion through ,4 cm of interference fit within the prevent propagation. The modular, the fracture site, which can impede diaphysis and rely on splines for fluted, tapered stem is then impacted fracture healing. Haidukewych et al28 rotational stability of the implant.26 into the distal femur, and modular recommended fracture reduction and Proper sizing of these implants is components are trialed on the proxi- fixation with cables, gentle cement important for axial stability because mal aspect of the stem until the desired pressurization, and intraoperative subsidence is a frequent mechanical leg length and hip stability are ob- radiographs after implantation to complication; the tendency is to tained. The rotational orientation of search for extruded cement. undersize the stem because of the risk the modular components are marked of creating an intraoperative femoral using a standard reference with fracture.26 electrocautery. Final implants are Cortical Strut Allografts Our preferred technique when using impacted onto the stem, with atten- First described in 1989, cortical strut a modular tapered stem is to recon- tion paid to recreating the desired grafts can restore bone stock and struct the femur first, followed by rotation. The proximal fracture frag- improve the structural mechanics of fracture reduction. Under manual ments are then debulked, using a high- the reconstruction. Historically, good reaming, a conical support is created speed burr, and wrapped around the results were reported for periprosthetic with sequentialtapered reamers, based stem using cables or Luque wires. fractures treated with cortical strut on the template diameter and the Typically, patient rehabilitation allograft and cerclage wires.29,30 In length of the stem. Manual reaming consists of protected weight bearing a study of 19 patients by Chandler reduces the risk of femoral blow-out following surgery and no abduction et al,31 16 patients (84%) had bony

486 Journal of the American Academy of Orthopaedic Surgeons

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Roshan P. Shah, MD, JD, et al

Figure 3

Proximal femoral replacement for a type B3 periprosthetic fracture. A, Hip radiograph showing deficient proximal and lateral bone with multiple fracture lines and stem subsidence. B and C, Postoperative radiographs showing a well-positioned proximal femoral replacement with retention and healing of the deficient proximal bone. (Courtesy of Brett Levine, MD, Rush University, Chicago, IL)

union and good clinical function after augmentation with a cortical strut patterns, augmented with a circumfer- 5 months. In a study by Head et al,32 allograft and/or impaction grafting, ential wire mesh for transverse fracture results of .90% bony union with or replacement with an allograft- patterns. The mesh should span at least cortical struts were reported with prosthetic composite (APC) or mega two cortical diameters above and a follow-up of .10 years. Cortical prosthesis. This scenario is seen in the below the fracture site. The canal is strut grafts are used infrequently and multiply revised hip, with repeated prepared and a suitable prosthesis is only for support of host bone stock, devascularization from canal ream- selected that will extend at least two specifically on the tension surfaces of ing, multiple cable applications, and cortical diameters below the fracture. A the femur (ie, anterior and lateral). prior osteotomies. Proximal replace- cement restrictor is placed in the canal, Park et al20 described an overall ment is also a reasonable choice and corticocancellous bone is impacted union rate of 94% in a series of 18 whenever a distal press-fit stem re- into the canal. Reamers are used to patients. In type B2 fractures, union quires additional proximal support or recreate the canal and further impact times dropped from 52 weeks with in the setting of an oncologic patho- the bone graft, followed by broaches revision alone to 17 weeks if a plate or logic periprosthetic fracture. and special tamps to reconstitute the strut-onlay was used. The use of cor- metaphyseal bone stock. The stem is 33 tical strut grafts remains controversial Impaction Grafting then cemented into place. Lee et al because of concerns over soft-tissue found fracture healing and reconstitu- stripping and its impact on healing. Impaction grafting has been described tion of bone stock in all seven patients for Vancouver type B2 and B3 peri- in their report, with good clinical re- 33 prosthetic fractures and for revision sults in six patients (86%). Vancouver Type B3 THA generally.34 Although rarely Fractures used, this technique may be used suc- cessfully in severely ectatic bone and Allograft Prosthesis Fractures of severely deficient bone simple fracture patterns. Grafting can Composite can cause extensive comminution and address areas of fracture comminution, Allograft prosthesis composite (APC) devascularized bone fragments, re- as well as reconstitute a capacious reconstructions involve cementing sulting in a deficient and unsupportive canal. We recommend fracture reduc- the proximal portion of a long stem proximal femur. This necessitates tion with cables for oblique or spiral into an allograft, and then impacting

August 2014, Vol 22, No 8 487

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Periprosthetic Fractures Around Loose Femoral Components

Table 1 Summary of Reporting on Type B2 Periprosthetic Fractures No. of Subjects Mean Patient Mean Follow-up at Final Fracture Treatment Age in Years in Months Study Follow-Up Type Study Type Method (Range) (Range)

Springer et al39 111 76 B2 and Case series Cemented stem, 65.3 (37 to 91) 64.9 (2 to 185) 35 B3 proximally porous- coated stem, extensively porous- coated stem, allograft- prosthesis composite, or endoprosthesis Bhattacharyya 51 43 B2 and Case-control Long stem revision 79.1 (range NS) 26 (range NS) et al5 8B3 1/2plate or allograft strut, ORIF alone Mukundan 59 42 B2 and Case series Long stem revision 74.2 (57 to 100) 24 (range NS) et al40 17 B3 1/2 plate, distal- locking long stem revision Corten et al41 31 B2 Case series Cemented long stem 82 (56 to 93) 33 (0 to 132) prosthesis with allograft strut Fink et al42 32 22 B2 and Case series Long stem noncemented 67.4 (39 to 90) 32.2 (24 to 60) 10 B3 modular revision Matharu et al43 46 33 B2 and Case series Noncemented, fully 76.7 (57 to 94) 24 (6 to 73) 13 B3 porous-coated long stem revision, or femoral endoprosthesis Marx et al44 29 15 B2 and Case series Noncemented Wagner 67.8 (40 to 82) 74 (range NS) 14 B3 long stem revision Neumann 55 35 B2 and Case series Noncemented modular 74 (45 to 84) 67 (60 to 144) et al45 20 B3 tapered long stem revision Garcia-Rey 35 20 B2 and Case series Noncemented, fully 80 (51 to 86) 100 (36 to 204) et al24 15 B3 porous-coated long stem revision Sheth et al10 21 18 B2, 1 B3, Case-control Noncemented, fully 66.5 (47 to 82) NS 2 B2/A(g) porous-coated long stem revision

NS = not specified, ORIF = open reduction and internal fixation

the distal stem into the host distal literature. In a series of 25 Vancouver femur for an appropriately sized femur. For optimal fixation, the distal type B3 hips treated with APCs, 20 stem. The allograft, once thought to femur is prepared with a step cut, and hips (80%) demonstrated healing incorporate and allow soft-tissue a reciprocal cut is made in the allo- between the allograft and host bone healing, resorbs with time. Ulti- graft. The remaining proximal host at 5 years, and 21 patients (84%) had mately, the allograft conveys a me- bone and soft-tissue attachments can no or only mild pain.35,36 Radio- chanically weaker construct (ie, the be wrapped around the construct as graphic graft resorption was mild in bone can fracture) compared with vascularized autograft and fixed with four hips (16%) and moderate in that of a metal mega prosthesis. cerclage cables. Finally, any residual two hips (8%). host greater trochanter or abductor This technique is time consuming tissue is fixed to this construct.35 and falling out of favor. The allograft Mega Prosthesis APC outcomes have demonstrated canal may require substantial ream- Proximal bone stock can be replaced reasonable clinical outcomes in the ing to match an ectactic distal host with a tumor prosthesis37,38 (Figure 3).

488 Journal of the American Academy of Orthopaedic Surgeons

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Roshan P. Shah, MD, JD, et al

After prosthesis removal and 33% and 12%, respectively. Sub- contents. In this article, references 3, débridement, the femur is cut trans- dividing further, a higher mortality 5, and 22 are level II studies. Refer- versely, as proximally as possible to rate for ORIF than that for revision ences 1, 2, 4, 8, 10, 15, 16, 19, and provide a good circumferential bony arthroplasty approached significance 24 are level III studies. References 6, shelf. The canal preparation and stem for both type B1 and B2 fractures. 7, 13, 17, 18, 20, 21, 23, 26, 27, and placement depends on whether the Part of the benefit may be explained 29-45 are level IV studies. References proximal femoral replacement is ce- by the mobility and full weight 9, 14, 25, and 28 are level V expert mented or press-fit. Anteversion is bearing following revision compared opinions. controlled by the orientation of the with the restricted weight bearing References printed in bold type are implantation, and offset can be varied following ORIF. those published within the past 5 based on the implant modularity. Leg Eighteen percent of patients expe- years. length can be controlled using modu- rienced a postoperative complica- lar body segments in varying sizes. tion, including bleeding (3.4%), 1. Berry DJ: Epidemiology: hip and knee. Orthop Clin North Am 1999;30(2): Proximal femoral replacements dislocation (3.2%), wound infection 183-190. have had mixed clinical results at 4 (2.7%), and deep vein thrombosis 38 38 2 2. Lindahl H, Malchau H, Herberts P, years. McLean et al reported the (0.8%). Early reoperations were Garellick G: Periprosthetic femoral results of 15 proximal femoral common, with 10.4% of patients fractures classification and demographics replacement prostheses in a series of undergoing reoperation within the of 1049 periprosthetic femoral fractures from the Swedish National Hip 8 Vancouver type B3 periprosthetic first 12 months. In total, 23% of Arthroplasty Register. J Arthroplasty 2005; fractures and patients with failed patients underwent reoperation at 20(7):857-865. 8 ORIF of femoral fractures with poor a mean interval of 22 months. 3. Lindahl H, Garellick G, Regnér H, bone stock. They reported Medical Thirteen percent of the reoperations Herberts P, Malchau H: Three hundred and twenty-one periprosthetic femoral Outcomes Study 12-Item Short Form were the result of loosening. fractures. J Bone Joint Surg Am 2006;88(6): scores of approximately 50 for both 1215-1222. the physical and mental components. 4. Sarvilinna R, Huhtala HS, Sovelius RT, Their complications included three Summary Halonen PJ, Nevalainen JK, Pajamäki KJ: dislocations, two failures of seven Factors predisposing to periprosthetic Periprosthetic fractures are increasing fracture after hip arthroplasty: A case one-stage procedures for infection, in incidence as the number of patients (n = 31)-control study. Acta Orthop Scand and one refracture below the proxi- 2004;75(1):16-20. undergoing primary THA increases. mal femur replacement. 5. Bhattacharyya T, Chang D, Meigs JB, The goal of treatment is to obtain sta- Estok DM II, Malchau H: Mortality after ble fracture fixation and distal stem periprosthetic fracture of the femur. J Bone Joint Surg Am 2007;89(12):2658-2662. Outcomes and fixation. A long, fully coated stem was Complications historically the mainstay of treatment, 6. Duncan CP, Masri BA: Fractures of the although modular tapered stems are femur after . Instr Course Lect 1995;44:293-304. Table 1 summarizes the treatment gaining in popularity. Other recon- structive options also exist based on 7. Masri BA, Meek RM, Duncan CP: outcomes for Vancouver B2 and B3 Periprosthetic fractures evaluation and fractures.5,10,24,39-45 Mortality is high the bone quality. Despite significant treatment. Clin Orthop Relat Res 2004; following periprosthetic hip fractures. first-year mortality rates and other 420:80-95. The Swedish registry found that 1.2% complications, patients who sustain a 8. Lindahl H, Malchau H, Odén A, of patients died within the first week periprosthetic fracture can have good Garellick G: Risk factors for failure after outcomes when the appropriate treat- treatment of a periprosthetic fracture of the postoperatively, and 9.4% of patients femur. J Bone Joint Surg Br 2006;88(1): died within 12 months of surgery.2 ment is chosen. It is imperative that the 26-30. treating surgeon have a high level of Bhattacharyya et al5 found a cumula- 9. Franklin J, Malchau H: Risk factors for tive mortality rate of 21% and a suspicion that a stem may be loose and, periprosthetic . Injury 2007;38(6):655-660. 1-year mortality rate of 11%. This if confirmed, appropriate implants and was similar to mortality following hip special equipment must be available to 10. Sheth NP, Brown NM, Moric M, revise the femoral component. Berger RA, Della Valle CJ: Operative fractures generally and significantly treatment of early peri-prosthetic femur higher than the mortality following fractures following primary total hip primary total joint arthroplasty. arthroplasty. J Arthroplasty 2013;28(2): References 286-291. For all type B fractures in the report by Bhattacharyya et al,5 the mortality 11. Lavernia CJ: Cost-effectiveness of early Evidence-based Medicine: Levels of surgical intervention in silent osteolysis. rates after ORIF and revision were evidence are described in the table of J Arthroplasty 1998;13(3):277-279.

August 2014, Vol 22, No 8 489

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited. Periprosthetic Fractures Around Loose Femoral Components

12. Harris B, Owen JR, Wayne JS, Jiranek WA: in femoral revision total hip arthroplasty. 35. Maury AC, Pressman A, Cayen B, Zalzal P, Does femoral component loosening J Arthroplasty 2002;17(4, suppl 1)134-137. Backstein D, Gross A: Proximal femoral predispose to femoral fracture? An in vitro allograft treatment of Vancouver type-B3 comparison of cemented hips. Clin Orthop 24. García-Rey E, García-Cimbrelo E, Cruz- periprosthetic femoral fractures after total Relat Res 2010;468(2):497-503. Pardos A, Madero R: Increase of cortical hip arthroplasty. J Bone Joint Surg Am bone after a cementless long stem in 2006;88(5):953-958. 13. Beals RK, Tower SS: Periprosthetic periprosthetic fractures. Clin Orthop Relat fractures of the femur. An analysis of 93 Res 2013;471(12):3912-3921. 36. Haddad FS, Garbuz DS, Masri BA, fractures. Clin Orthop Relat Res 1996;327: Duncan CP: Structural proximal femoral 238-246. 25. Srinivasan A, Jung E, Levine BR: allografts for failed total hip replacements: Modularity of the femoral component in A minimum review of five years. J Bone 14. Pike J, Davidson D, Garbuz D, Duncan CP, total hip arthroplasty. J Am Acad Orthop Joint Surg Br 2000;82(6):830-836. O’Brien PJ, Masri BA: Principles of Surg 2012;20(4):214-222. treatment for periprosthetic femoral shaft 37. Klein GR, Parvizi J, Rapuri V, et al: fractures around well-fixed total hip 26. Munro JT, Garbuz DS, Masri BA, Proximal femoral replacement for the arthroplasty. J Am Acad Orthop Surg Duncan CP: Tapered fluted titanium treatment of periprosthetic fractures. 2009;17(11):677-688. stems in the management of Vancouver J Bone Joint Surg Am 2005;87(8): B2 and B3 periprosthetic femoral 1777-1781. 15. Chevillotte CJ, Ali MH, Trousdale RT, fractures. Clin Orthop Relat Res 2014; Larson DR, Gullerud RE, Berry DJ: 472(2):590-598. 38. McLean AL, Patton JT, Moran M: Femoral Inflammatory laboratory markers in replacement for salvage of periprosthetic periprosthetic hip fractures. J Arthroplasty 27. Abdel MP, Lewallen DG, Berry DJ: fracture around a total hip replacement. 2009;24(5):722-727. Periprosthetic femur fractures treated with Injury 2012;43(7):1166-1169. modular fluted, tapered stems. Clin Orthop 16. Ghanem E, Houssock C, Pulido L, Han S, Relat Res 2014;472(2):599-603. 39. Springer BD, Berry DJ, Lewallen DG: “ ” Jaberi FM, Parvizi J: Determining true Treatment of periprosthetic femoral leukocytosis in bloody joint aspiration. 28. Haidukewych GJ, Langford JR, fractures following total hip arthroplasty J Arthroplasty 2008;23(2):182-187. Liporace FA: Revision for periprosthetic with femoral component revision. J Bone fractures of the hip and knee. Instr Course 17. Johansson JE, McBroom R, Joint Surg Am 2003;85-A(11):2156-2162. Lect 2013;62:333-340. Barrington TW, Hunter GA: Fracture of the 40. Mukundan C, Rayan F, Kheir E, ipsilateral femur in patients wih total hip 29. Pak JH, Paprosky WG, Jablonsky WS, Macdonald D: Management of late replacement. J Bone Joint Surg Am 1981;63 Lawrence JM: Femoral strut allografts in periprosthetic femur fractures: (9):1435-1442. cementless revision total hip arthroplasty. A retrospective cohort of 72 patients. Clin Orthop Relat Res 1993;295: 18. Lewallen DG, Berry DJ: Periprosthetic Int Orthop 2010;34(4):485-489. 172-178. fracture of the femur after total hip 41. Corten K, Macdonald SJ, McCalden RW, arthroplasty: Treatment and results to date. 30. Brady OH, Garbuz DS, Masri BA, Bourne RB, Naudie DD: Results of Instr Course Lect 1998;47:243-249. Duncan CP: The treatment of periprosthetic cemented femoral revisions for 19. Engh CA, Massin P, Suthers KE: fractures of the femur using cortical onlay periprosthetic femoral fractures in the Roentgenographic assessment of the allograft struts. Orthop Clin North Am elderly. J Arthroplasty 2012;27(2): biologic fixation of porous-surfaced 1999;30(2):249-257. 220-225. femoral components. Clin Orthop Relat 31. Chandler HP, King D, Limbird R, et al: The 42. Fink B, Grossmann A, Singer J: Hip Res 1990;257:107-128. use of cortical allograft struts for fixation of revision arthroplasty in periprosthetic 20. Park SK, Kim YG, Kim SY: Treatment of fractures associated with well-fixed total fractures of vancouver type B2 and B3. periprosthetic femoral fractures in hip joint prostheses. Semin Arthroplasty 1993; J Orthop Trauma 2012;26(4):206-211. arthroplasty. Clin Orthop Surg 2011;3(2): 4(2):99-107. 43. Matharu GS, Pynsent PB, Dunlop DJ, 101-106. 32. Head WC, Malinin TI, Emerson RH Jr, Revell MP: Clinical outcome following 21. Levine BR, Della Valle CJ, Lewis P, Mallory TH: Restoration of bone stock in surgical intervention for periprosthetic hip Berger RA, Sporer SM, Paprosky W: revision surgery of the femur. Int Orthop fractures at a tertiary referral centre. Hip Extended trochanteric osteotomy for the 2000;24(1):9-14. Int 2012;22(5):494-499. treatment of vancouver B2/B3 33. Lee GC, Nelson CL, Virmani S, periprosthetic fractures of the femur. 44. Marx A, Beier A, Jung L, Lohmann CH, Manikonda K, Israelite CL, Garino JP: J Arthroplasty 2008;23(4):527-533. Halder AM: Peri-prosthetic femoral Management of periprosthetic femur fractures treated with the uncemented 22. Foran JR, Brown NM, Della Valle CJ, fractures with severe bone loss using Wagner revision stem. Hip Int 2012;22(3): Levine BR, Sporer SM, Paprosky WG: impaction bone grafting technique. 286-291. Prevalence, risk factors, and management J Arthroplasty 2010;25(3):405-409. of proximal femoral remodeling in revision 45. Neumann D, Thaler C, Dorn U: hip arthroplasty. J Arthroplasty 2013;28 34. Gie GA, Linder L, Ling RS, Simon JP, Management of Vancouver B2 and B3 (5):877-881. Slooff TJ, Timperley AJ: Impacted femoral periprosthetic fractures using cancellous allografts and cement for a modular cementless stem without 23. Weeden SH, Paprosky WG: Minimal 11-year revision total hip arthroplasty. J Bone Joint allografting. Int Orthop 2012;36(5): follow-up of extensively porous-coated stems Surg Br 1993;75(1):14-21. 1045-1050.

490 Journal of the American Academy of Orthopaedic Surgeons

Copyright Ó the American Academy of Orthopaedic Surgeons. Unauthorized reproduction of this article is prohibited.