♦ Revision hip replacement requires careful preoperative planning
♦ Accurate diagnosis is vital: particular attention must be paid to whether infection is present or not
♦ Extensile approaches are preferred
♦ Appropriate equipment is greatly helpful in explantation of the failed components
♦ Imaging, classification, and templating are useful in determining the best reconstruction techniques.
Revision hip surgery is less commonly performed than primary procedures. The National Joint Register records that approaching 10% of operations are revisions or reoperations. However, this equates to 10 000 hip revisions in 2008 in England alone; the numbers are continuing to rise. In the United States, there were 43 000 revisions in 2002, accounting for 17.5% of hip replacements. Revision rates are 0.7% by 1 year and 1.3% at 3 years. Hip revision surgery accounts for disproportionately high costs to the health service: the operations are time consuming, require an extended hospital stay, and implant costs are higher. In addition, the rates of infection, dislocation, and mortality are all higher in revision surgery (Box 7.11.1).
The surgeon aims to answer a series of questions:
♦ Is the pain coming from the hip? Potential alternative diagnoses must be considered and may require further investigation to obtain an accurate diagnosis (Box 7.11.2).
♦ Is the patient as well as possible? Revision hip surgery is a major undertaking and the patient should be considered as a whole. In particular, the cardiovascular and respiratory systems should be assessed and any malfunction referred to a physician in case investigation and intervention are required. Common examples include lung function testing, cardiac angiography, and angioplasty
♦ Is the hip replacement infected? Differentiation between aseptic loosening and a septic hip is vital and discussed further later in this chapter. Decisions as to whether the procedure should be performed in two separate phases are also influenced by this information. If the hip is infected, identification of the causative organism is very useful
♦ What is the degree of bone loss? The magnitude and pattern of bone loss must be well understood for proper planning to take place. Consideration of modes of failure and sites of anatomical defect will determine whether one or other of the implants can be salvaged and determine the best reconstructive options available. The use of preoperative templating is then used to plan reconstruction of the bone loss.
The key to successful revision hip surgery is preoperative planning. It can shorten the learning curve, anticipate intraoperative challenges, identify the correct approach, allow ordering of necessary equipment and identify appropriate investigations (Box 7.11.3).
Investigations of a painful total hip replacement
Accurate investigation allows for proper planning of revision hip surgery (Box 7.11.4). All hips should have anteroposterior and lateral views of the pelvis and proximal two-thirds of the femur.
Judet views are used to assess the integrity of the anterior and posterior acetabular columns in particular. The iliac oblique view demonstrates the anterior wall and posterior column, whilst the obturator oblique view demonstrates the posterior wall and anterior column.
Computed tomography (CT) is useful for analysis of component orientation as well as existing bone stock. Three-dimensional (3D) reconstruction can provide further evidence of bone loss.
If there is intrapelvic cement or the acetabular component has migrated through the medial wall, angiography can be needed to demonstrate the vessels.
Approaches used in revision hip arthroplasty
Selection of the approach for revision surgery is even more critical than for primary surgery. There is no perfect solution through which all hips can be revised. Important considerations include the site of previous scars, the position and type of components to be removed, and the presence and extent of osseous defects. Adequate visualization is essential but must not be at the price of excessive damage to soft tissue or bone. Revision surgery is not always predictable and exposures should allow for flexibility when operative findings necessitate variance from the preoperative plan.
The posterior approach
The posterior approach is extensile and avoids damage to the abductor mechanism of the hip. It can be performed through extension of most previous lateral or posterior approach scars. More distal exposure of the femur can be achieved through release of the gluteus maximus and iliopsoas tendons. The exposure can continue with release of vastus lateralis, allowing exposure of the entire length of the femur if necessary. The acetabulum is also extremely well visualized. Reports vary as to whether this approach is associated with a higher rate of dislocation at revision surgery but more recent work suggests that the rate can be as low as with other approaches.
The anterolateral approach
This extensile exposure allows excellent visualization of the proximal femur and good exposure of the acetabulum. The vastus lateralis can be reflected anteriorly to allow access to the proximal femur. The risk of gluteal nerve injury is higher in revision surgery, with a corresponding increase in Trendelenburg gait pattern after the procedure.
Trochanteric osteotomy can be carried out to improve visualization of the acetabulum where complex reconstruction is anticipated. A variety of methods have evolved. A simple osteotomy can be carried out at the base of the greater trochanter, as described by Charnley. Whilst this provides excellent exposure of both sides of the hip, there can be problems with vascularity of the trochanter, reattachment, and metalwork failure. These can be compounded by proximal bone loss, weakening the greater trochanteric cortices and giving rise to trochanteric escape in 15–20% of patients in most revision series. The use of ‘partial trochanteric osteotomy’ can reduce these problems. The trochanteric flip begins its osteotomy posteriorly and stops to leave a 1-cm bone bridge supporting the incision of the gluteal tendons and the vastus lateralis, thus reducing the damage to the vascularity of the trochanter. Partial anterior trochanteric osteotomy removes a small sliver of anterior bone with the insertions of gluteus medius and vastus lateralis attached but leaves the insertion of gluteus minimus intact, making reattachment easier and subsequent escape less likely. A trochanteric slide also preserves the attachment of gluteus medius and vastus lateralis on one segment of bone, increasing vascularity, improving the biomechanics (the proximal pull of the fragment by gluteus medius is countered by the distal pull of vastus lateralis) of the reattachment, and reducing non-union rates to 4%.
The extended trochanteric osteotomy is useful in removal of femoral components with solid distal fixation, typically fully porous-coated cementless stems, and stems in significant varus. It has a high rate of union, typically less than 2% have non-union or malunion. The osteotomy begins with a 5–10 cm osteotomy posterior and medial to the greater trochanter, running down the linea aspera. A transverse arm passes anteriorly then the anterior osteotomy is completed anterior and medial to the greater trochanter. It involves elevation of vastus lateralis off the linea aspera and can be a cause of muscle necrosis if a previous anterolateral or lateral approach has been used, in which case the vastus lateralis fibres should be left inserting into the osteotomized fragment. Following extended trochanteric osteotomy, a longer femoral revision prosthesis will be required in order to achieve stable fixation.
The transfemoral approach allows excellent visualization of the femoral canal through exposure of the implant and any cement mantle across its entire length. This is achieved at the price of weakening the available bone stock, so it is used in combination with a very long prosthesis achieving distal fixation (such as the Wagner stem).
The explantation phase requires planning and can be made significantly easier by the use of appropriate specialist instruments but only if the surgeon has thought to have them available.
Hemispherical uncemented cups will work well in the majority of cases. They require a minimum of 50% host bone contact in order to osseointegrate (NB host bone is intact pelvis—allograft is not included). Care must be taken to remove any pseudocapsule or membrane which may reduce the percentage of implant to host bone contact. If host bone contact is anticipated to be below 50%, further techniques will be required.
Table 7.11.1 American Academy of Orthopaedic Surgeons (AAOS) classification of acetabular deficiencies
Central—absence of medial wall
Cavitary deficiencies—an expanded and thin-walled socket
Combined deficiencies—I and II coexist
Pelvic discontinuity—separation of the ilium from the ischiopubic portion of the pelvis
Type I defect (Table 7.11.1)
Segmental defects can be present in the anterior wall, posterior wall, medial wall, or superior margin of the acetabulum. Typically, these defects can be filled with augments or allograft. Augments can include the use of oblong cups or porous metal augments with screw fixation to bone. Impaction techniques of morcellized allograft work well in medial wall defects. Superior segmental defects can be treated with structural allograft and screw fixation to the ilium. Structural graft in particular is prone to resorption and collapse; thus one potential advantage of porous metal augments is avoidance of this complication.
These cavitary defects have intact margins and floor of the acetabulum. They are often amenable to jumbo uncemented cup reconstruction, with or without morcellized allograft.
These defects are a combination of segmental and cavitary bone loss. Their treatment, therefore, can require a combination of techniques. The use of antiprotrusio cages, with or without allograft, is more frequent. These gain superior fixation to the ilium via screws through a superior flange; inferior fixation is via an inferior hook in the obturator foramen.
In these cases, there is pelvic discontinuity and mechanical stability must be restored. Before the acetabulum is reconstructed, posterior reconstruction plating is performed to stop movement between the ilium and the ischium. Anterior plating is not necessary. Acetabular reconstruction cages are the mainstay of treatment; CAD/CAM components can be useful.
Femoral reconstruction (Box 7.11.1)
Cemented revision has advantages of immediate fixation, local delivery of antibiotics, and ability to flow into geometrically abnormal proximal femurs. It can, however, be difficult in severe bone loss and can be hard to re-revise. In addition, some series show high rates of early loosening because the internal femur is highly sclerotic, significantly reducing the microinterlock between cement and bone. If cement is used, meticulous cementing technique is vital. The use of impaction grafting is also successful in the femur, particularly in contained, proximal lesions.
Classification of femoral bone loss must be done after component removal to give an accurate evaluation of bone loss, as defects can be significantly larger after the explantation of the original stem.
If there is a completely intact cement mantle within healthy bone, a shorter stem can be cemented into the original cement mantle: a so-called cement-in-cement revision.
Type 1 defect
There is cancellous bone present, with minimal metaphyseal loss and an intact diaphysis. Proximal fixation is achieved with a porous coated, uncemented stem or use of a cemented prosthesis.
Type 2 defect
There is metaphyseal loss but the diaphysis is intact; this is a common finding at revision of cemented stems. An uncemented, proximally loaded, metaphyseal fitting stem can be used in less severe cases. If there is very poor proximal bone stock, an uncemented stem will require distal fixation and a fully porous coated stem is appropriate, with or without impaction bone grafting.
Table 7.11.2 Paprosky classification of femoral abnormalities
Metaphysis and diaphysis intact
Metaphysis deficient, diaphysis intact, calcar non-supportive
Metaphysis and diaphysis non-supportive; 4cm distal fixation near isthmus.
Metaphysis non-supportive, diaphysis not intact due to severe bone loss; fixation available distal to isthmus
Extensive metadiaphyseal damage; cortical fixation at isthmus not reliable
Type 3a defect
The metaphysis is severely damaged and not supportive but there is at least 4cm of intact diaphyseal bone available for distal fixation. A fully porous coated stem can be used to achieve distal fixation, again with or without bone grafting.
Type 3b defect
The metaphysis is damaged and there is less than 4cm of intact diaphyseal bone distally. In this case a modular cementless stem with distal flutes is required in order to gain rotational stability.
Type 4 defect
There is extensive metaphyseal and diaphyseal bone loss, a widened femur and the isthmus is non supportive. A conical, cementless stem will not achieve stable fixation. Options include extensive impaction grafting (if the proximal cortices are intact), use of a prosthesis allograft composite, or proximal femoral replacement.
Dislocation (Box 7.11.8)
The options for revision will depend upon the cause and direction of the dislocation. Success is more likely if the cause of dislocation is accurately identified. For example, posterior dislocation can be prevented by replacing the socket in a more anteverted position or removal of excessive anterior acetabular bone upon which the femoral neck may impinge. It is not always necessary to remove all components. In modular implants, a longer femoral head or lateralized acetabular liner may be an appropriate method of increasing offset. A larger diameter femoral head and bearing or use of a lipped liner will also increase stability. If this is not possible, the use of trochanteric advancement increases tension in the hip abductors. In cases with a significant soft tissue component to the dislocation (typically multidirectional instability with severe shortening or neurological disorder), the use of a constrained acetabular liner can be necessary. This carries the disadvantage of reduced range of motion and higher rates of loosening. Revision to a bipolar femoral prosthesis alone can be used as a last resort in recurrent cases.
Nerve damage (Box 7.11.9)
The rate of nerve injury is 1–4% in primary hip replacement but rises to around 7% in revision surgery. Reconstruction of a dysplastic acetabulum also carries a higher risk of nerve injury.
The most commonly injured nerve is the posterior (peroneal) division of the sciatic nerve. The posterior division has a thinner perineurium than the anterior (tibial) branch and is more tethered at the fibula head, both of which are postulated to contribute to its increased rate of injury. Sciatic nerve injury is more common if the leg is lengthened at operation; lengthening less than 4cm tends to injure the posterior branch and lengthening over 4cm gives rise to complete sciatic palsy. Sciatic nerve injury is also caused by perineural haematoma, careless retractor positioning, and posterior capsular release, particularly in revision surgery. Dislocation can injure the sciatic nerve either perioperatively or postoperatively.
The femoral nerve is less commonly injured. It passes anterior to the hip, overlying the iliopsoas tendon, and can be injured in anterior capsular release or through careless placement of anterior retractors during acetabular preparation.
Injury to the superior gluteal nerve is rarely recognized after hip surgery; its inferior branch in particular is at risk in anterolateral and lateral approaches to the hip and its injury can lead to abductor weakness and Trendelenburg gait pattern.
Obturator nerve injury is rare and may be associated with extrapelvic extrusion of cement or antiprotrusio cages.
Vascular injury is rarer than neurological compromise, occurring in well under 0.5% of hip replacement. Again, the incidence is far higher in revision surgery: if the acetabular component is significantly protruded into the pelvis, preoperative angiography is recommended.
The femoral vessels run parallel to the femoral nerve and are protected in a similar fashion, i.e. careful placement of an anterior acetabular retractor lateral to the iliopsoas muscle.
Medial wall penetration can injure the iliac vessels, though this is rare. Acetabular screw fixation also risks injury to the iliac vessels posteriorly or the obturator vessels anteriorly.
If bleeding is massive and uncontrolled, exposure and temporary clamping of the iliac vessels may be needed, followed by emergency intervention by a vascular surgeon.
Alberton, G., High, W., and Morrey, B. (2002). Dislocation after revision total hip arthroplasty. An analysis of risk factors and treatment options. Journal of Bone and Joint Surgery, 84-A, 1788–94.Find this resource:
Barrack, R. and Burnett, R. (2005). Preoperative planning for revision total hip arthroplasty. Journal of Bone and Joint Surgery, 87-A, 2800–11.Find this resource:
D’Antonio, J., McCarthy, J.C., and Bargar, W.L. et al (1993). Classification of femoral abnormalities in total hip arthroplasty. Clinics in Orthopedics, 296, 113–9.Find this resource:
D’Antonio, J., Capello, W., and Borden, L. (1989). Classification and management of acetabular abnormalities in total hip arthroplasty. Clinical Orthopaedics and Related Research, 243, 126–37.Find this resource:
DeHart, M. and Riley, L. (1999). Nerve injuries in total hip arthroplasty. Journal of the American Academy of Orthopaedic Surgeons, 7, 101–11.Find this resource:
Della Vale, C.J. and Paprosky, W.G. (2003). Classification and an algorithmic approach to the reconstruction of femoral deficiency in revision total hip arthroplasty. Journal of Bone and Joint Surgery, 85-A(Suppl 4), 1–6.Find this resource:
Della Vale, C.J. and Paprosky, W.G. (2003). Classification and an algorithmic approach to the reconstruction of femoral deficiency in revision total hip arthroplasty. JBJS, 85-A(Suppl 4), 1–6.Find this resource:
Dorr, L.D., and Wan, Z. (1998). Causes of and treatment protocol for instability of total hip replacement. Clinical Orthopaedics and Related Research, 355,144–51.Find this resource:
Schmalzried, T.P., Amstutz, H.C., and Dorey., F.J. (1991). Nerve palsy associated with total hip replacement, risk factors and prognosis. Journal of Bone and Joint Surgery (Am), 73,1074–80.Find this resource: