A 48-year-old man presented with increasing left groin pain that was exacerbated by weight-bearing activities. He underwent primary, cementless, left total hip arthroplasty 15 years earlier for posttraumatic arthrosis. He reported that his limb has been getting shorter. Physical examination revealed an antalgic gait related to his left lower extremity, a moderate Trendelenburg lurch, a stiff left hip, a shortened left lower extremity, and groin pain with a straight leg raise. Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) values were normal. Radiographs (see Fig. 39.3 ) demonstrated massive pelvic osteolysis and obvious socket migration.
Although we were prepared to use structural support (i.e., allograft or porous metal) and a cup cage, we were able to treat him with a large hemispherical shell and particulate bone grafting. At 2 years postoperatively, radiographs showed a stable socket and graft incorporation. At the most recent follow-up assessment, the patient was pain free and could ambulate without assistive devices.
A thorough history, physical examination, laboratory investigations, and imaging studies must be performed. All data are properly analyzed to identify the cause of a failed total hip arthroplasty.
Infection should always be considered in the differential diagnosis.
Revision surgery should be goal oriented and meticulously planned.
The surgeon should be prepared for the worst-case scenario.
When in doubt, cases are referred to a subspecialist.
Primary total hip arthroplasty (THA) is one of the most successful procedures in modern medicine, with a clinical success rate exceeding 90% at 10 years. However, early and delayed failures are recurring problems that often require revision procedures.
With an aging population, the demand for primary total joint arthroplasty has increased, and the number of revision hip procedures is expected to increase to 96,700 by 2030 (137% increase compared with 2005). The growing elderly population and broadening of indications, including performing total joint arthroplasties in younger, active patients, will contribute to this increased volume of revisions.
Some authorities have tried to estimate the rate of change. Looking at a 12-year follow-up period, Katz and colleagues projected that the rate of revision THA in the Medicare population would increase by 2% per year in the first 18 months and then 1% per year thereafter. Revision surgery of any kind is often associated with prolonged operative time, additional blood loss, and higher rates of infection, thromboembolism, dislocation, and nerve palsy. Postoperative complications after revision surgery are increasingly prevalent, especially in the elderly population. Given the need to reduce costs and improve value in the current health care system, it is imperative that the indications for revision surgery be well defined and goal oriented.
Because the array of challenges associated with revision THA exceed those of primary THA, preoperative assessment and planning are crucial. This process should include a comprehensive patient history, physical examination, imaging, laboratory tests, and a review of systems to identify medical comorbidities that may influence the treatment plan. Preoperative planning should include a meticulous review of high-quality radiographs, other imaging studies, and templating. This systematic approach enables the formulation of many goal-oriented plans and leads to optimal selection of the implants, instruments, equipment, bone graft, and personnel needed to ensure that the procedure is executed as effectively and efficiently as possible.
In evaluating the need for revision THA, the correct failure mode must be elucidated to provide successful treatment. Knowledge of certain patient comorbidities, such as diabetes, obesity, and rheumatoid arthritis, can direct the surgeon to the likely failure mechanism and assist with risk stratification. Referred sources of pain (e.g., lumbar stenosis) must be ruled out. The surgeon must decide whether the level of disabling pain is sufficient to warrant a major operation. Alternatives to revision THA, especially for elderly patients who are typically less physically active and may have lower demands and expectations, may include activity modification, weight loss, use of external support, and pain management through pharmacologic interventions.
Pain is the most common symptom of patients presenting with a failed primary THA. Pain around the hip and pelvis can be divided into extrinsic and intrinsic causes. Pain extrinsic to the hip joint may be caused by lumbar spinal disease, arterial insufficiency, malignancy, iliotibial band syndrome, or hernia. Intrinsic causes of hip pain include sepsis, aseptic loosening, modulus mismatch, periprosthetic fracture, instability, bursitis, iliopsoas snapping or tendinitis, and adverse local tissue reaction.
A systematic characterization of the pain, including onset, location, radiation, exacerbating factors, and severity, can help the surgeon to identify potential causes. Pain that has persisted since the initial operation may be caused by infection, failure of osseointegration of cementless implants, periprosthetic fracture, or an incorrect initial diagnosis. Discomfort that begins after an initial pain-free period may result from aseptic loosening, fracture, iliopsoas irritation, instability, acute hematogenous infection, or chronic infection. Pain localized to the groin often correlates with intrinsic hip pathology, whereas lateral pain occurs with trochanteric bursitis or pathology intrinsic to the abductor mechanism. Pain that is posterior in the buttock region or radiates down past the knee may have a lumbar origin, whereas pain that occurs at night or while resting can be attributed to infection or neoplasm. Pain due to socket pathology typically hurts in the groin, buttocks, or “deep inside,” whereas localized, nonradiating thigh pain is often attributable to femoral stem pathology.
Failure of the primary THA may have a variety of causes. The most common causes of failure and indications for revision procedures in the United States Medicare population as of 2006 included instability and dislocation (22.5%), mechanical loosening (19.7%), periprosthetic infection (14.8%), osteolysis or aseptic loosening (8%), periprosthetic fractures, and adverse local tissue reaction associated with metal-on-metal bearings.
Indications for Revision Surgery
The reported incidence of instability or dislocation after primary THA ranges from 1% to 15%. Dislocation is more common among patients with impaired cognitive abilities, soft tissue laxity, and underlying hip fractures. The use of small-diameter femoral heads, skirted heads, and acetabular liners with elevated rims and the failure to restore leg lengths and the offset may increase the risk of dislocation.
Instability after THA can be classified as early or late. Early dislocations are usually positional and can often be treated nonoperatively by bracing and activity modification. Careful scrutiny of radiographs is essential to confirm proper implant position and restoration of the offset and limb lengths. The most common reasons for recurrent instability are soft tissue laxity and acetabular malpositioning. Although the surgical technique used to correct recurrent instability depends on the cause of instability, it is not always easily discernable ( Fig. 39.1 ). Surgical options to treat recurrent instability include correction of malpositioned implants, use of a larger articulation, use of dual-mobility components, use of a constrained acetabular liner, and trochanteric advancement to optimize soft tissue tension.
Infections are a serious complication after THA and are a common indication for revision procedures. A detailed patient history should investigate systemic risk factors for infection, previous perioperative hospital course, history of drainage, presence of pain or fever, and any pain-free interval. Infections associated with total joint arthroplasties are common and are categorized according to the time interval to onset of infection and the mechanism of infection. Acute postoperative infections or early infections are symptomatic within a few days to weeks after the THA and are caused by microorganisms that enter the joint during the operation or immediately thereafter through the unhealed operative site or drain site. Treatment at this stage may involve irrigation, débridement, and modular head and liner exchange, although the indications for this procedure may be waning.
Late chronic infections are caused by the proliferation of microorganisms, usually through hematogenous spread ( Fig. 39.2 ). They are usually addressed with a two-stage exchange, although a one-stage exchange can be considered in a competent host with an antibiotic-sensitive organism without the need for bone grafting in a second stage. Laboratory tests to screen for infection include C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR). The most conclusive test for periprosthetic infection involves a positive culture of an aspirate from the joint. Radiographs may also be reviewed for suggestive signs of infection, such as generalized bone resorption without wear, periosteal reaction, and dispersed foci of osteolysis. However, most patients with infection have radiographic findings that are indistinguishable from those of aseptic loosening.
Treatment options for periprosthetic infections include irrigation and débridement of the hip with modular head and liner exchange, one-stage exchange arthroplasty, and resection arthroplasty with or without placement of an antibiotic-laden cement spacer followed by reimplantation (i.e., two-stage exchange). The standard treatment in North America for chronic periprosthetic hip infection is two-stage exchange with resection of implants, thorough irrigation and débridement, and administration of parenteral antibiotics for a minimum of 6 weeks. Surveillance of ESR and CRP values should be combined with continued aspirate cultures, followed by reimplantation when levels of inflammatory markers normalize and aspirate cultures are negative. In cases in which a bone graft is not needed, the host is competent, and the microorganism is sensitive to antibiotics, one-stage revision has been used under strict protocols with 85% success rates in selected European centers.
Osteolysis and Aseptic Loosening
Although wear particle–induced osteolysis and aseptic loosening will become less common in the future, they remained common indications for revision surgery in 2012. As of 2006, wear-related complications were the second most common reason for revision, representing 27.7% of all revisions in the United States Healthcare Cost and Utilization Project Nationwide Inpatient Sample database. Wear particles generate an immune response mediated by inflammatory molecules such as interleukin-1, interleukin-6, and tumor necrosis factor-α and by monocytes and macrophages; all of these responses lead to the activation of osteoclasts and eventually to periprosthetic bone resorption. Loosening of cementless acetabular components often manifests with groin pain, whereas femoral loosening may manifest as thigh pain. Pain that occurs when a patient stands up from a seated position, also known as start-up pain , is often caused by an aseptic implant loosening. 10
After a careful history, evaluation of a patient for aseptic loosening usually includes a workup for infection. Tests for inflammatory markers such as CRP and ESR should be ordered. Elevation of these values should prompt suspicion of a septic rather than aseptic process, and aspiration of the joint should be considered. Radiographic evaluation involves examination of serial radiographs, specifically looking for evidence of osteolysis or radiolucent lines in the acetabulum and femoral components as described by Harris and colleagues and Engh and co-workers. Eccentricity of the femoral head in the socket seen on radiographs is pathognomonic for polyethylene wear. Radiographic evidence of migration, complete radiolucent lines around the porous coating, and pedestal formation associated with radiolucent lines may signify cementless implant loosening. Radiographic signs of loosening of cemented implants include complete radiolucencies at the implant–cement and bone–cement interfaces, cement fracture, implant fracture, and migration.
Pain is often the chief complaint for patients who develop osteolysis around cemented acetabular components. In these patients, the radiographic pattern of osteolysis tends to be linear and progresses to involve the entire bone–cement interface. When the entire bone–cement interface is involved, the cup becomes loose. Timing of the intervention should correlate with the patient’s symptoms, because many patients with loose cemented sockets report only mild pain that does not warrant immediate surgery.
The indications for operative treatment of osteolysis around cementless acetabular components are less well defined. Patients with loose cementless sockets have pain, and the indication for revision is straightforward ( Fig. 39.3 ). Controversy exists regarding the timing of operative intervention when the socket remains well fixed. In these cases, patients often are pain free, and the issue becomes one of operating on asymptomatic patients. A few small areas of osseointegration can keep the cup stable, and patients can remain pain free despite expansile lesions. Most experts agree that osteolytic lesions that progress over a 3- to 6-month period warrant operative intervention.