An 81-year-old woman presented with pain in the right hip and thigh that was associated with progressive difficulty in ambulating and putting weight on the right leg. She had undergone a hybrid total hip arthroplasty (THA) approximately 13 years earlier. Radiographs revealed loosening of the cemented stem with subsidence, cement mantle fracture, varus collapse, osteolysis in all Gruen zones, and Paprosky type IIIB bone loss ( Fig. 48.1 , A ). The uncemented acetabular component was well fixed. Results of the infection workup were negative, confirming aseptic loosening and failure of the femoral component.
The patient was given optimal medical care before surgery. Preoperative planning was performed to ascertain the component’s length, diameter, and offset and to determine the length of the extended trochanteric osteotomy (ETO) needed. A fluted, tapered, modular implant was chosen. Due to peculiar femoral remodeling and greater trochanteric overhang, removal of the femoral component and cement was performed with the use of an ETO. A prophylactic cable was applied distal to the osteotomy, followed by progressive manual reaming to obtain endosteal chatter (i.e., engaging hard endosteal bone to determine component size). After insertion of the distal segment of the implant, an appropriately sized trial proximal body was used to optimize leg length, offset, version, and stability, and it was confirmed with intraoperative radiographs. After the final components were implanted, the undersurface of the ETO fragment was prepared, and it was reattached with multiple cables. A socket revision was performed concomitantly to maximize hip stability. Final radiographs were obtained (see Fig. 48.1 , B to D ).
With the increasing number of primary total hip arthroplasties (THAs) being performed, the number of revision THAs is expected to rise exponentially. In the United States alone, THA revisions may grow by 137% between 2005 and 2030. The revision surgeon may encounter various degrees of femoral bone loss at the time of revision. Depletion of femoral bone stock typically occurs as a result of multiple previous procedures involving insertion and removal of implants, infection, aseptic loosening, osteolysis due to particle debris, stress shielding, or periprosthetic fractures. Femoral revision with bone loss provides significant challenges to achieving implant fixation, stability, and durability.
Numerous options are available for femoral reconstruction in revision THA based on the extent of bone loss and the quality of remaining bone stock. They include cemented fixation, cementless fixation with the use of proximally porous-coated implants, extensively porous-coated cylindrical implants, and modular and nonmodular, fluted, tapered stems. In femurs with advanced bone loss, options include impaction bone grafting and cemented stem fixation, allograft-prosthetic composites, and proximal femoral replacement (i.e., megaprosthesis).
Indications for Revision
Classification of Femoral Bone Loss
Many classifications have been proposed for quantifying bone loss in the revision setting. They include the Mallory, Saleh, and Paprosky, schemas and the American Academy of Orthopaedic Surgeons (AAOS) classification proposed by D’Antonio and colleagues (see Chapter 46 ). The one most widely used is the Paprosky classification, which is based on the quality and quantity of metaphyseal and diaphyseal bone loss, because it provides an algorithm for planning femoral reconstruction.
The Paprosky system divides femoral bone loss into four types and provides reconstructive recommendations. Type I consists of minimal metaphyseal cancellous bone loss and an intact diaphysis. This type of defect is commonly seen after removal of a cementless femoral component without a biologic ingrowth surface. Cemented or cementless, proximal or extensively porous-coated fixation can be used for revision.
In type II, the femur has extensive metaphyseal cancellous bone loss with an intact diaphysis. This type of bone loss is encountered after removal of a cemented femoral component. Because cemented fixation is unreliable in this situation, cementless fixation is recommended with a device for obtaining fixation in the metaphysis or diaphysis, or both.
In type IIIA, the femur has a severely damaged and unsupportive metaphysis along with more than 4 cm of intact diaphyseal bone available for distal fixation. This defect is typified by a femur after removal of a grossly loose femoral component that was inserted with a first-generation cementation technique. This level of bone loss necessitates use of a cementless, diaphyseal-fitting implant such as an extensively porous-coated, cylindrical stem.
In type IIIB, the femur has a severely damaged and unsupportive metaphysis along with less than 4 cm of diaphyseal bone available for distal fixation. This type of femur is often seen after failure of a cemented femoral component that was inserted with a cement restrictor or a cementless femoral component associated with substantial distal osteolysis. Use of a fluted, tapered, modular type of cementless stem is recommended.
In type IV, the femur has extensive metaphyseal and diaphyseal bone loss along with a widened femoral canal and nonsupportive isthmus. The reconstruction options include impaction bone grafting, an allograft-prosthetic composite, and proximal femoral replacement.
Because the results of cemented femoral fixation in the revision setting have been dismal, its use has declined. High rates of femoral loosening and repeat revision have been documented, especially in earlier reports. This may be directly related to a marked reduction in the cement–bone interface shear strength as documented by biomechanical testing. Dohmae and colleagues reported that the shear strength of cement–bone interface in the revision setting was only 20.6% of that in the primary procedure, and Rosenstein and co-workers reported it to be 30% weaker than in the primary procedure. With improved cementation techniques, success rates have been somewhat mixed. The use of cemented stems is restricted to Paprosky type I bone loss.
The in-cement or cement-in-cement technique may be used with good success rates in cases of aseptic loosening of the femoral component with an otherwise intact and stable cement mantle. This technique involves “freshening” the existing cement mantle before recementing. It can also be used when revising a broken femoral component with an intact distal mantle, for removal of a femoral component to improve exposure of the acetabulum, and for recurrent dislocation due to component malposition.
Proximally Porous-Coated Stems
Proximally porous-coated stems can be used in femurs with minimal bone loss. Berry and associates reported that survivorship of this implant worsened with increasing bone loss, mostly because extensive metaphyseal bone loss did not allow initial or long-term biologic fixation. The study authors found a survivorship of only 58% at 8 years with revision for aseptic femoral failure as the end point. Another frequent complication is intraoperative proximal femoral fracture, with an incidence as high as 26% to 46%. Inability to obtain an independent fit and fill into metaphyseal and diaphyseal bone is the greatest limitation of monoblock, proximally porous-coated stems.
Proximally porous-coated, modular stems allow intraoperative ease in obtaining adequate and independent sizing and fit into metaphyseal and diaphyseal portions of the femur and adjustment of version. These stems have provided good results when used in Paprosky type I, II, and IIIA femurs, but there is a diminished survival rate in Paprosky types IIIB and IV bone loss. In a randomized, prospective trial, Iorio and colleagues found no difference in 5-year survivorship and outcome measures when femoral fixation with a third-generation cementing technique was compared with modular, cementless metaphyseal fixation in cases of type I and II bone loss.
Extensively Porous-Coated, Cylindrical Stems
Extensively porous-coated, cylindrical stems have been widely used and studied by several investigators, who have reported 90% to 95% survivorship at 10 years. The principle is to bypass deficient bone and obtain fixation in the diaphysis. In the two series using these stems reported by Weeden and Paprosky and Engh and colleagues, the rate of failure of osseointegration leading to reoperation at long-term follow-up was as low as 4%. In another report, the use of this stem design was successful in approximately 90% of Paprosky type II and IIIA femurs, with a 50% osseointegration failure rate for type IIIB and 100% osseointegration failure rate for type IV cases. As a consequence, the investigators did not recommended its use in type IIIB and IV bone loss. In another study, Sporer and Paprosky noticed a tendency toward mechanical failure of this device in type IIIB femurs when the endosteal canal diameter exceeded 19 mm. There is also concern about proximal stress shielding with the larger stems, especially in osteoporotic women.
Fluted, Tapered Stems
Fluted, tapered stems are used to transmit a load to the femoral diaphysis through mechanical and biologic fixation, bypassing the deficient proximal bone. The tapered, conical geometry provides axial support, and flutes provide rotational stability. These stems are made of titanium alloy and theoretically have less stiffness, which may protect against stress shielding. Initially, they were introduced as a monoblock component in Europe. Several investigators have reported favorable results with the monoblock, tapered, fluted stems (e.g., Wagner SL revision stem), albeit with some concerns regarding subsidence.
Modularity has gained popularity because of the ability to obtain independent fit and fill of metaphyseal and diaphyseal regions and the ease of restoration of leg length, version, and offset. Several surgeons have reported excellent midterm results with various designs of these stems. The incidence of clinically significant subsidence seems to have been reduced. However, stem undersizing is a risk factor for subsidence, and the importance of preoperative planning and surgical technique cannot be overemphasized. The risk of subsidence may also increase in hips with more advanced bone loss (type IIIB or worse), presence of an extended trochanteric osteotomy (ETO), or periprosthetic fracture. In a retrospective study comparing extensively porous-coated, cylindrical, cobalt-chrome stems with tapered, fluted, modular, titanium stems, Richards and co-workers found improved quality of life measures, decreased complication rates, and better preservation of bone stock with the latter. Rodriguez and colleagues reported proximal bony regeneration in 73% of femurs, but stress shielding occurred in 26% of hips in the mid-diaphyseal region. Occasional stem fractures just distal to the modular junction have been reported. Most have occurred with the initial versions of these stems, which featured smaller junctions.
Impaction Bone Grafting and Cemented Fixation
Impaction bone grafting and cemented fixation may be used in femurs with advanced (Paprosky type IV) bone loss if the proximal cortical tube is intact. The technique involves packing cancellous allograft in the femoral canal to create a neomedullary canal into which a highly polished, collarless, double-tapered stem is cemented using contemporary cementing techniques. The graft is gradually vascularized and incorporated, with resultant restoration of bone stock. In femurs with cortical defects, use of a mesh also has been described. Overall, this procedure is technically challenging, time-consuming, and expensive due to the amount of bone graft required. Despite a higher risk of subsidence and iatrogenic femoral fracture, it does have the potential to restore femoral bone stock in a young patient, with long-term implant survivorship between 90% and 99%.
Use of allograft-prosthetic composites involves implanting a long-stem femoral prosthesis distally within the medullary canal of the host femur while the proximal part of the prosthesis is secured within a proximal femoral allograft for initial stability. The allograft also acts as a strut graft for restoration of bone stock, adjustment of leg length, and attachment of soft tissues. In a meta-analysis encompassing nonneoplastic and neoplastic types of femoral bone loss, Rogers and colleagues reported a pooled success rate of 81% at 8 years, with a structural failure rate of 15% and infection rate of 8%. Babis and associates reported 69% survivorship at 10 years, with a higher risk of failure in hips with Paprosky type IV bone loss and three or more previous revisions. Overall, it appears that this is a viable method when performed by experienced surgeons. The disadvantages include nonunion, graft resorption, and risk of disease transmission.
Proximal Femoral Replacement
Proximal femoral replacement is most feasible in elderly patients with low activity levels and massive proximal femoral bone loss. It is less technically challenging compared with the use of allograft-prosthetic composites, and it is less time-consuming and more readily available. Malkani and colleages reported an 8% femoral repeat revision rate at 11 years’ follow-up. However, with any revision used as the end point, survivorship was only 64% at 12 years. There was a 22% dislocation rate in this series, which was similar to the instability rate in the series reported by Parvizi and co-workers. This procedure may provide functional long-term results, but it should be reserved for salvage situations.
Our Preferred Technique
Our algorithm is based on the quality of host bone and on the indication for surgery. For femurs with minimal bone loss (Paprosky type I or II), we have used a fully porous-coated stem ( Fig. 48.2 ) or a monoblock, fluted, tapered stem ( Fig. 48.3 ) with good results. For a Paprosky type IIIA and higher-grade bone loss, a periprosthetic fracture, or an ETO, we prefer a fluted, tapered, modular stem.