- Email: [email protected]
Imaging of hip arthroplasty
European Journal of Radiology 81 (2012) 3802–3812
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Imaging of hip arthroplasty Theodore T. Miller ∗ Department of Radiology and Imaging, Hospital for Special Surgery, 535 E. 70th Street, New York, NY 10021, United States
a r t i c l e
i n f o
Article history: Received 16 February 2011 Accepted 22 March 2011 Keywords: Hip Arthroplasty Resurfacing Complications Imaging CT MRI
a b s t r a c t The imaging evaluation of the prosthetic hip begins with radiography, but arthrography, aspiration, scintigraphy, sonography, CT and MR imaging all have roles in the evaluation of the painful prosthesis. This article will review the appearance of normal hip arthroplasty including hemiarthroplasty, total arthroplasty, and hip resurfacing, as well as the appearances of potential complications such as aseptic loosening and osteolysis, dislocation, infection, periprosthetic fracture, hardware failure, and soft tissue abnormalities. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The modern era hip arthroplasty began in the 1960s when a stainless steel metal-on-polyethylene prosthesis was pioneered by Charnley [1]. Many different variations and designs have since been introduced, but most follow his principle of a metal femoral head articulating against a polyethylene socket. Hip arthroplasty has become so successful, with some designs having a 25 year survivorship of almost 80% that the hip is the most commonly replaced joint. This article will review the appearance of normal hip arthroplasty, as well as the appearances of potential complications.
2. Normal The hip joint can be artificially restored in several ways. Hemi-arthroplasty replaces only the femoral side of the joint, usually in cases of femoral neck fracture or avascular necrosis in which the acetabular side of the joint is normal. In a unipolar hemi-arthroplasty the prosthetic femoral head articulates directly against the acetabular cartilage, whereas in a bipolar hemiarthroplasty the prosthetic femoral head articulates against a prosthetic cup which is placed into, but not fixed to, the native, unprepared, acetabulum and which articulate against the acetabular cartilage. The acetabular articular cartilage will eventually wear away in both uni- and bipolar hemiarthroplasties, leading to painful degenera-
∗ Tel.: +1 212 606 1127; fax: +1212 734 7475. E-mail address: [email protected] 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.03.103
tive arthritis of the acetabulum, requiring revision to a total hip arthroplasty (Fig. 1). Total hip arthroplasty, in which both the femoral head and the acetabulum are replaced by fixed prosthetic devices, is most often performed for disease processes which have affected both sides of the native joint, such as degenerative and inflammatory arthritis. The acetabular and femoral components may be cemented or noncemented (either press-fit or in-growth), with the most common combination being a cemented femoral stem and non-cemented acetabular cup (Fig. 2). The femoral component should be placed in either mild valgus position, such that the tip of the stem is located in the medial aspect of the medullary canal, or in neutral position with the tip located centrally in the medullary canal. A “cement restrictor”, a plug made of plastic, cement, metal, or bioabsorbable material, is sometimes present in the femoral shaft, just distal to the tip of the stem; this plug keeps the cement contained in the medullary canal during cement injection and stem insertion in order to maintain injection pressure and assure good bonding of the cement with the endosteal trabeculae. The cement mantle should be at least 2 mm thick all the way around the stem [2] to minimize the risk of subsequent cement fracture. A cemented stem may have a surrounding thin radiolucency at the cement–bone interface, representing a fibrous pseudocapsule that forms as a result of trabecular necrosis, due to either the marked exothermic curing of the cement or the surgical reaming of the medullary canal, with an adjacent thin sclerotic line along its outer margin representing reactive bone [3]. An ingrowth stem usually has tiny beads or wires scintered onto its proximal surface to increase its surface area for bonding with bone, and the scintered surface gives the prosthesis a fuzzy
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 1. Hemiarthroplasties. (a) AP radiograph of a unipolar prosthesis shows the prosthetic femoral head articulating directly against the native acetabular cartilage, causing cartilage narrowing (arrow). (b) AP radiograph of a bipolar hemiarthroplasty shows the acetabular cup, which is placed into the native socket but is not fixed to it. The acetabular cartilage is narrowed by the cup (straight arrow). The inferior aspect of the prosthetic femoral head (round tail arrow) is visible.
edge radiographically. The non-cemented component may also be coated with hydroxyapatite in order to induce bone formation onto the component’s surface. Radiographically, the well-fixed ingrowth component shows bony sclerosis extending onto the prosthesis (Fig. 3). There may normally be some cortical or endosteal sclerosis and thickening of the femoral shaft at the level of the tip of the femoral component due to normal transfer of load stresses along the femoral stem [4]. Thin linear lucencies, less than 2 mm wide, representing fibrous ingrowth rather than bony ingrowth may also be seen, most often along the proximal aspect of the stem medially and laterally [4]. The greater trochanter may have decreased radiodensity due to “stress shielding” because the implant shifts physiologic load away from the greater trochanter. Similarly, resorption of the calcar of the femur due to stress shielding may occur with both cemented and
3803
non-cemented stems [3,4], but the extent of resorption depends on the particular model of stem. The acetabular component is usually placed in approximately 40◦ of vertical tilt (“inclination”) from the horizontal on an AP radiograph, depending on the model of the prosthesis and preoperative condition of the acetabulum, and is usually anteverted (i.e., open facing anteriorly) 10–20◦ . Thin radiolucency at the metal–bone interface of a non-cemented acetabular component represents fibrous ingrowth. Most arthroplasty designs utilize a high-density polyethylene (PE) liner in the acetabular cup, articulating with a metal prosthetic femoral head. The prosthetic head should be symmetrically seated within the cup, appearing on a frontal radiograph equidistant from the superior and inferior margins of the cup. The head may be slightly inferiorly located in the cup if the surgeon has used an “off-set” liner which has a thicker superior rim, but a head located superiorly in the cup, even mildly so, is never normal and indicates PE wear. Most femoral heads are metallic, and are thus radiographically dense; ceramic heads, used in an attempt to decrease PE wear, are radiolucent (Fig. 2). The rim of the PE liner usually is flush with the rim of the metal cup, but sometimes additional mechanical constraint is needed, such as for patients with recurrent dislocation of the femoral head. In this group of patients, a constrained liner extends beyond the cup itself, to deepen the articulation and limit the range of motion (Fig. 4). In younger, active patients with disease on both sides of the joint, such as patients with femoroacetabular impingement and secondary degenerative arthritis, hip resurfacing (HR) might be performed instead of total arthroplasty. In both HR and THA the acetabulum is replaced, but in HR a pegged metal cap is cemented over the femoral head, thus preserving femoral neck bone stock for the eventual revision to a THR that the patients may require as they age. HR designs use large diameter metal-on-metal surfaces in order to minimize surface wear [5]. The femoral component should be in 5–10◦ valgus relative to the femoral neck [5]; varus positioning or exaggerated valgus positioning can lead to notching of the femoral neck with resultant fracture or loosening. The femoral peg is usually not cemented, as cementation can cause stress shielding which may lead to loosening [5]. The acetabular component should have inclination and anteversion similar to that of a THA [6] (Fig. 5). Large series of long term (>10 years) survivorship data for HR is lacking since the current generation of HR models was just coming into use about 10 years ago, but medium term (4–6 year) survivorship of HR ranges from 94% to 99%, similar to that of THA [5]. A recent meta-analysis of the literature comparing various outcomes of HR and THA [7] showed that HR patients tend to have higher postoperative activity levels than THA patients, even when controlling for preoperative activity levels [8], and that gait is more natural after HR. However, HR cannot correct for limb length discrepancies greater than 2 cm and has limited ability to correct femoral head–neck offset [5]. Other contraindications to HR include: inflammatory or erosive arthritis (because of the risk of loosening or fracture), osteoporosis or steroid use (because of the risk of fracture), short femoral neck, acetabular dysplasia, poor femoral head bone stock due to large cysts, hypersensitivity to metal, and poor renal function (because of impaired ability to clear metal ions) [5].
3. Complications Complications of hip arthroplasty include aseptic loosening and osteolysis, dislocation, infection, periprosthetic fracture, hardware failure, and soft tissue abnormalities.
3804
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 2. AP radiographs of various combinations of total hip arthroplasty. Notice that the acetabular cups of total arthroplasties are smaller than the cup of a bipolar arthroplasty in order to avoid neck–cup impingement. (a) Metal head against polyethylene liner; cemented femoral stem. (b) Ceramic head against polyethylene liner; cemented femoral stem. The ceramic head is lucent compared to metal. (c) Metal head against polyethylene liner; non-cemented, ingrowth femoral stem. The ingrowth is supposed to occur around the proximal aspect of the stem (arrows). (d) Metal head against metal liner; non-cemented, ingrowth femoral stem. The large head distributes load, decreasing metal wear.
3.1. Aseptic loosening and osteolysis Aseptic loosening of the prosthesis is the most common reason for revision surgery [2,9]. Radiographic appearances of loosening of a cemented femoral prosthesis are: lucency at the
cement–bone interface of >2 mm surrounding the component, progressive widening of the lucency at the cement–bone interface, lucency at the metal–cement interface, and fracture of the cement mantle (Fig. 6). Radiographic appearances of loosening of a non-cemented femoral prosthesis are: lucency at the metal–bone
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3805
Fig. 4. Constrained liner. AP radiograph of a total hip replacement shows the metal ring (arrow) of the inferior edge of the polyethylene liner.
interface >2 mm surrounding the component, development or widening of the lucency at the metal–bone interface, and subsidence of more than 1 cm and/or which continues to progress more than 1 year after placement [3]. Shedding of surface beads can be seen with both loose and stable implants [10]. The radiographic appearance of loosening of cemented and non-cemented acetabular components is lucency >2 mm at the cement–bone or metal–bone interface around its entire circumference [11]. Additional radiographic features of loosening, regardless of whether the components are femoral or acetabular, cemented or non-cemented, are: migration of the component or change of position of the component, fracture of the component, and component motion with stress views. The orthopedic classification for modes of stem failure is listed in Table 1 [12]. The description of the location of the lucencies around a total hip arthroplasty should follow the standard orthopedic descriptions of
Fig. 3. In-growth femoral component. (a) Initial post-operative radiograph. (b) AP radiograph 1.5 years later shows early in-growth of bone (arrow) onto the stem.
Fig. 5. AP radiograph of normal hip resurfacing. The large metal resurfacing head (arrow) is cemented to the femur, but the cement is hidden beneath the head. The stem is not cemented, and is in 5–10◦ of valgus relative to the neck. The acetabular component (round tail arrow) is the same as in a metal-on-metal total hip arthroplasty.
3806
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812 Table 2 Mechanisms of wear [16,17]. 1. Abrasion – the scratching of a surface by some other, usually harder, surface 2. Adhesion – transient bonding of the bearing surfaces to each other, usually due to poor lubrication, which pulls particles from the weaker surface 3. Fatigue – stress of the material beyond its normal mechanical limits, with resultant release of particles
Table 3 Classification of wear [16]. Type 1. Normal articulation between two bearing surfaces Type 2. Articulation between a bearing surface and non-bearing surface (e.g., prosthetic head penetrating through the polyethylene liner to articulate against metal backing) Type 3. Third-body abrasion, caused by a fragment of material interposed between normally articulating surfaces Type 4. Motion between two non-bearing surfaces (e.g., between the backside of the polyethylene liner and the metal acetabular backing, or between the modular head and its connecting taper)
Fig. 6. Femoral stem loosening. AP radiograph shows a cemented stem with irregular and focally wide radiolucency around the entire cement mantle, but particularly medially (arrows) with marked thinning of the cortex. The cement column is broken (round tail arrow) and shifted in position.
femoral and acetabular “zones”: there are seven femoral zones on the AP radiograph, beginning at the proximal lateral aspect and progressing clockwise with zone 4 at the tip of the femoral stem [12]. There are an additional seven zones on the lateral radiograph, numbered 8–14, beginning at the anteroproximal aspect of the femoral stem. The region around the acetabular component is divided into three equal zones, 1–3, from superolateral to inferomedial [13]. The acetabular zones for HR are the same, but unlike the femoral arthroplasty stem the femoral HR peg has only six zones: superior surface of peg (zone 1), tip of peg (zone 2), inferior surface of peg (zone 3) on the AP radiograph, and anterior (zone 4) to posterior (zone 6) on the lateral view [14]. The most common cause of periprosthetic lucency is mechanical loosening, but osteolysis due to either “particle disease” or infection can look similar radiographically. Any of the components of a hip replacement, such as the metal, polyethylene liner, or cement, can become fragmented due to wear, shedding microscopic particles of material which can induce a histiocytic inflammatory reaction [15]. Acetabular PE is the most common source because it is constantly worn by the metal femoral head, an example of abrasive
wear (Tables 2 and 3, and Fig. 7). The average PE wear rate of MOP designs is approximately 100–200 m/year [16]. The particles are engulfed by macrophages, which then release various factors and cytokines, such as interleukins, prostaglandins, and tumor necrosis factor, which cause up-regulation of osteoclasts, leading to osteolysis [17]. Osteolysis is dependent on the number of shed particles (reflected in the volumetric and linear wear rates of the bearing surfaces) and the histiocytic response to those particles. While there is no absolute threshold for the number of shed PE fragments necessary to incite the histiocytic response, the fewer the number of shed particles the less likely an inflammatory response; a linear wear rate of less than 100 m/year has been suggested as a threshold level [18]. Moreover, the type of particle itself is important, since PE particles have a high inflammatory profile, while metal and ceramic particles do not [19,20]. Therefore, in an attempt to minimize the histiocytic response, different materials and combinations of articulations have been investigated. Ceramic-on-ceramic (COC) designs have the lowest coefficient of friction and the lowest wear rate [16,20], but disadvantages are their high cost, dependence on precise positioning, and fracture (.001–.002% incidence) [16]. Metal-on-metal (MOM) designs,
Table 1 Modes of failure of cemented stems [12]. Mode 1. Pistoning (up-and-down motion) 1A. Pistoning of the stem within the cement mantle. 1B. Pistoning of the stem and cement mantle within the medullary canal. Mode 2. Medial midstem pivot (medial migration of proximal stem and lateral migration of stem tip) Mode 3. Calcar pivot (medial–lateral toggling of the stem tip, analogous to the “windshield wiper” phenomenon of uncemented stems) Mode 4. Bending cantilever fatigue (medial migration of the proximal stem with distal fixation of the stem). This is often the result of initial varus positioning of the stem
Fig. 7. AP radiograph shows superior eccentric seating of the femoral head, indicating polyethylene wear.
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3807
in which a cobalt–chromium–molybdenum femoral head articulates against a similar acetabular surface, also has a low wear rate (approximately 3–5 m/year) after an initial “wear-in” period. Metal particles are smaller than PE particles and produce a lower grade histiocytic response [19], with resultant less osteolysis. However, patients with MOM designs have higher serum and urine levels of chromium and cobalt than control subjects [6], raising concern about carcinogenesis. Metal ions can also cross the placenta, and although no teratogenic or mutagenic effects have been documented, it is recommended that MOM implants not be performed in women of child-bearing age [5]. Additional risk factors for elevated metal ions in MOM patients include an acetabular inclination angle >55◦ (because of edge loading), female gender (because of gait), and small femoral head size (because of increased contact stress and resultant wear) [6]. Lastly, highly cross-linked PE is more resistant to wear than standard ultra high molecular weight PE, with a wear rate similar to MOM components [21]. The development of highly cross-linked PE allows the use of MOP designs, which are familiar to surgeons, more forgiving in their positioning, and more versatile than MOM and COC for making off-set liners and constrained liners. However, while highly cross-linked PE is more resistant to wear against smooth bearing surfaces, it is not more resistant to third-body abrasion [21]. Osteolysis due to particle disease is suggested radiographically by focal well-defined radiolucencies around either the acetabular or femoral components (Fig. 8). Osteolysis may occur at sites away from the actual articulating surfaces of the arthroplasty by insinuation of joint fluid and its inflammatory debris around both the femoral and the acetabular components due to the hydrostatic pressure generated by joint movement, a concept called the “effective joint space” [22]; thus, for example, screw holes in an acetabular component provide an avenue for particulate debris to reach pelvic bone (Fig. 9). Radiographs are not sensitive for the depiction of osteolysis and underestimate the extent of involvement [23]. Both CT and MRI are more sensitive for detecting osteolysis and demonstrating its extent than radiographs because of their tomographic nature. In an experimental model of osteolysis, radiographs had 52% sensitivity, CT had 75% sensitivity and MRI had 95% sensitivity [24]. On MR imaging, osteolysis is usually low signal intensity on T1WI, and intermediate to high signal intensity on T2WI [25] and can occasionally be expansile (Fig. 10). Often there is accompanying bulky synovitis with debris. Linear high signal intensity surrounding the femoral stem on STIR images may indicate loosening or a histiocytic response [26]. CT is particularly helpful for evaluation of lysis of the medial wall of the acetabulum [27], and it can be useful for evaluating femoral and acetabular bone stock in preparation for revision surgery. 3.2. Dislocation Dislocation is the second most common reason for revision surgery [9,28] and is multifactorial including the age and gender of the patient, the surgical approach, size of the components and position of the components [28]. Dislocation within the first three months after surgery is usually due to laxity of the immature pseudocapsule of the joint and surrounding soft tissues. Atraumatic dislocation occurring between three months and 5 years after surgery is usually due to component malposition, such as acetabular inclination >60◦ , acetabular anteversion >20◦ , or acetabular retroversion (Fig. 11). Inclination can be assessed on frontal radiographs, and acetabular version is well assessed with CT scan [28]. Dislocation occurring more than 5 years after placement is usually due to gradual stretching of the pseudocapsule and surrounding soft tissue laxity, and women are at greater risk than men [28].
Fig. 8. Osteolysis. (a) AP radiograph of a cemented femoral stem shows small faint lucent areas (arrows). (b) Two years later the radiolucenies have enlarged and new ones have developed.
3808
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 9. Coronally reformatted CT image shows osteolysis (straight white arrows) around the acetabular screw (round tail arrow) and a screw hole (black arrow). Notice the superior eccentric position of the femoral head due to polyethylene wear.
Surgical options for treating recurrent dislocation include correction of acetabular position, placement of a constrained liner which provides stability by deepening the articulating socket thus limiting the femoral range of motion, and revision of the femoral head to a larger size. 3.3. Infection Infection is the third most common reason for revision arthroplasty [9], occurring in 1–5% of hip replacements [17]. Radiographic findings suggestive of infection include a wide irregular radiolucency around the cement–bone or metal–bone interfaces and frank bone destruction. However, a distinction between infectious osteolysis and aseptic osteolysis due to mechanical loosening or particle disease often cannot be made on a single radiograph. Usually, previous radiographs are necessary for comparison, with mechanical
Fig. 10. Axial proton density image through the acetabulum of a patient with a total hip replacement shows dephasing artifact due to the acetabular cup, but a large surrounding area of osteolysis is visible (arrows) with intermediate to high signal intensity and mild expansion of the medial wall.
Fig. 11. AP radiograph of a person with recurrent hip dislocations shows an excessively inclinated acetabular component.
loosening and histiocytic osteolysis usually taking a slowly progressive course, whereas infection usually occurs with a rapid time course and an aggressive appearance. While the appearance of osteolysis cannot distinguish infectious from non-infectious loosening, the presence of femoral periosteal reaction or an adjacent soft tissue collection are highly predictive [29,30]. A fluid collection over the greater trochanter or along the iliopsoas tendon may be a normal communicating bursa or an expected post-surgical space due to disruption of normal soft tissue planes, but a communicating non-bursal cavity with an irregular lining is suspicious for infection [31]. These cavities can be assessed with arthrography, MR imaging, and sonography (Figs. 12 and 13). Various nuclear medicine examinations can also be used to evaluate clinically suspected infection in hip arthroplasty. Three-phase bone scan suffers from poor specificity since a cemented femoral component can normally have increased uptake around the prosthesis for several years after placement, and because a normal non-cemented prosthesis also demonstrates increased radiotracer uptake at the site of bony ingrowth onto the prosthesis. However, the bone scan has excellent negative predictive value and can thus be used as an initial screening test since no uptake reliably excludes infection [32]. Adding a gallium scan to the standard technetium bone scan can improve the diagnostic accuracy for infection to 70–80%: infection is excluded if the gallium scan is normal or has less intense uptake than the corresponding bone scan, and infection is diagnosed when there is uptake of gallium without corresponding Tc uptake or the gallium uptake is more intense than corresponding Tc uptake [32]. The combination of technetium or indium labeled white cells and technetium labeled sulfur colloid has excellent results, with accuracy of over 90%, and is currently the scintigraphic method of
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3809
Fig. 12. Infected hip arthroplasty. (a) Coronal T2-weighted inversion recovery image shows a peritrochanteric collection (horizontal arrow), with communication to the skin (vertical arrow). (b) Coronal T1-weighted image after intravenous contrast shows the peritrochanteric abscess (horizontal arrows) and the superficial draining abscess (vertical arrow). (c) Axial T1-weighted image after intravenous contrast shows the peritrochanateric abscess (round tail arrow) communicating with the posterior aspect of the prosthetic joint (straight arrows). (d) Longitudinal sonographic image during needle aspiration shows the needle (straight arrows) in the abscess (round tail arrow).
choice for evaluating suspected infection. The imaging feature of infection is spatial incongruence, in which there is uptake of the labeled white cells, regardless of intensity, without uptake of the sulfur colloid [32]. Fluoro-deoxy-glucose (FDG)-PET (positron emission tomography) scanning has variable performance for evaluation of prosthetic infection. While Chacko et al. reported 92% sensitivity and 97% specificity [33], Mayer-Wagner et al. reported 67% sensitivity and 83% specificity [34], and Stumpe et al. reported sensitivity of only 22–33%, with an overall accuracy of 69%, which was the same as radiographs and worse than three-phase bone scan [35]. A meta analysis of the literature showed slightly poorer accuracy of PET compared to white blood cell imaging (89% versus 91%), but PET imaging can be completed within a few hours of injection compared to the 24 h delay of white cell imaging [36]. Aseptic loosening also demonstrates FDG uptake and standardized uptake values (SUV) are not helpful for distinguishing infectious from non-infectious loosening because of overlapping values [36]. Chacko et al. suggested that the location of the uptake is more important than the intensity; uptake around the prosthetic head/neck or at the tip of the femoral stem can be normal, while uptake around the shaft of the stem suggests infection [33]. The gold standard for the evaluation of a clinically suspected infected joint is aspiration, with gram stain and culture of the joint
fluid. Although considered the definitive diagnostic test, there is a wide range of reported sensitivities. The sensitivity is better in patients with high clinical suspicion of infection or with periosteal reaction [37].
3.4. Periprosthetic fracture Periprosthetic fractures are rare with total hip replacements, and occur more often around the femoral than the acetabular component. Fracture of the femur may occur during placement of the femoral stem, usually as either focal cortical penetration or longitudinal splitting of the bone, and happens more often with uncemented components than cemented ones (5% versus 0.3%) because of the tight press-fit needed with uncemented stems [38]. The incidence is even higher in revision THA because of diminished bone stock resulting from osteoporosis or prior osteolysis, with an incidence of 6.3% for cemented THAs and almost 18% in non-cemented cases [39]. Fracture of the femur may also occur any time after hip replacement, typically at the level of the tip of the femoral stem because of “stress risers” at this level caused by the difference in stiffness between the metal stem and bony shaft. These fractures also are more common in revision hips (4%) than primary arthroplasties (1.1%), because of deficient bone stock [38].
3810
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 13. Infected hip arthroplasty. (a) Longitudinal sonographic image of the prosthetic hip joint shows a large joint effusion (arrows) around the prosthetic head (H) and neck (N). A is the acetabulum. (b) Oblique sonographic image shows the effusion (straight arrow) communicating with a large peritrochanteric collection (round tail arrow), which in turn has a tract (dashed arrow) toward the skin. (c) AP radiograph after intraarticular injection of contrast shows the irregularly shaped peritrochanteric collection with the thin communicating tract (arrow) to the skin.
Fractures of the femoral neck are the most common complication of hip resurfacings, with reported rates ranging from 0% to 12% [5]. Risk factors for fracture include scalloping of the femoral neck, osteonecrosis of the remaining femoral head, or poor cement integration. 3.5. Hardware failure Hardware failure can affect either the femoral or the acetabular components. The stem of the femoral component can break, representing a metal fatigue stress fracture, because the metal stem is more stiff and less yielding than the surrounding femoral bone, although the incidence of fracture depends on the geometry and metal composition of the stem (Fig. 16). The scintered beads of an in-growth stem may shear-off, indicating either micromotion or frank loosening of the stem, and may act as a cause of third-party abrasion [10]. The superior aspect of the polyethylene liner can become gradually worn down, appearing radiographically as asymmetric superior location of the femoral head within the acetabular cup, constituting type I wear (Table 3 and Fig. 7). In addition to gradual full-thickness wear, the polyethylene liner can frankly break and dissociate from the metal acetabular shell. In this latter scenario, the femoral head will be superiorly seated against the acetabular shell, and displaced broken metal tines may be seen on radiographs [40]. The broken pieces of the liner can be visualized arthrographically as lucent filling defects within the contrast pool. The resulting abnormal articulation of the metal head against the metal acetabular shell can lead to marked shedding of metal debris that distends the hip
Fig. 14. Early metallosis. AP radiograph shows marked superior eccentric seating of the metal femoral head articulating against the metal acetabular cup, an example of Type 2 wear (Table 3). The distended joint capsule (white arrows) is dense due to metal debris, the metal tines (short black arrow) of the polyethylene’s metal backing are becoming visible as they unlock, and a piece of broken polyethylene (long black arrow) is faintly visible within the dense effusion.
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3811
Fig. 16. Coronal proton density MR image shows atrophy of the gluteus medius muscle (straight arrow). There are also areas of high signal intensity acetabular and femoral osteolysis (round tail arrows) and a resultant fracture of the greater trochanter (dashed arrow).
and involving all three AP zones; 4. narrowing of the femoral neck – this may be due to remodeling if stable and less than 10% of the neck diameter, but continued narrowing after 3 years or narrowing of more than 10% may reflect osteolysis or impingement (Fig. 15); 5. scalloping at the neck–rim junction – could be due to impingement or remodeling; 6. osteolysis around the peg. Sclerosis around the tip of the peg, the “pedestal sign”, is of uncertain clinical significance [5,14] 3.6. Soft tissue complications Fig. 15. Osteolysis around a hip resurfacing. (a) AP radiograph shows osteolysis in acetabular zone 3 (black arrow), and narrowing of the femoral neck (white arrows) due to osteolysis. Compare to the appearance of the normal femoral neck in Fig. 5. (b) Axial proton density MR image shows mixed signal intensity osteolysis (straight arrows) adjacent to the femoral neck stem (round tail arrow).
capsule, called “metallosis”, and which appears radiographically as a dense “bubble” around the joint [41] (Fig. 14). The incidence of fracture with current generation ceramic components is approximately 1 in 1000 [16], and is usually caused by poor component placement, leading to edge loading. Fracture is a devastating injury, because of potential damage to the femoral taper on which the head implanted, and because any remaining ceramic shards can cause severe third body abrasion, especially of a softer revision material such as a metal head. The radiographic evaluation of the acetabular component of a hip resurfacing is similar to that of a THR, looking for change in inclination and periacetabular osteolysis. For the radiographic evaluation of the femoral component of a hip resurfacing, Shimmin et al. [5] suggest that the following signs may indicate early failure: 1. change in the peg–neck angle – this may mean femoral head necrosis or collapse, neck fracture, or component loosening; 2. decreasing distance of the tip of the peg to the lateral femoral cortex – this may indicate subsidence, loosening, or head collapse; 3. radiolucent line around the peg – may indicate loosening if greater than 2 mm wide
MR imaging and sonography are useful for the evaluation of the painful or clinically abnormal but radiographically normal hip arthroplasty. Greater trochanteric bursitis and other periprosthetic collections, dehiscence of the posterior pseudocapsule and short external rotators, and tendon tears or tendinosis can be demonstrated [42] (Fig. 16). Sonography can also provide guidance for needle aspiration of the joint or periprosthetic collections (Fig. 12) and for steroid/anesthetic injections such as of the greater trochanteric bursa and the iliopsoas tendon (Fig. 17). A particular soft tissue complication of MOM prostheses is the “adverse local tissue reaction” (ALTR). Patients present with pain, most commonly in the groin but also laterally and in the buttocks, and women are more commonly affected than men [43,44]. The adverse reaction may either be a high-wear macrophagemediated event or a low-wear lymphocyte-mediated event [44,45]. The low-wear lymphocyte-mediated event has been termed “aseptic lymphocytic vasculitis-associated lesions” (ALVAL), and is believed to be a type of hypersensitivity reaction [45]. MR imaging of ALTR’s can demonstrate capsular dehiscence, periprosthetic fluid collections, osteolysis, and focal inflammatory pseudotumors that may be either hypo- or hyperintense on T2-weighted images [43,45,46]. Histologically, the ALVAL lesions are characterized by a lymphocyte-predominant infiltrate and tissue necrosis, whereas the high-wear events are characterized by a macrophage-
3812
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 17. Transverse sonographic image at the level of the pelvic brim (black round tail arrow) during a cortisone injection shows a spinal needle (white straight arrows) positioned with its tip deep to the iliopsoas tendon (white round tail arrow). The iliopsoas bursa (B) is distended by the injectate. The metal acetabular component is demarcated by the straight black arrow.
predominant infiltrate and metal debris without tissue necrosis [45]. 4. Conclusion Radiography should always be the first step in the imaging evaluation of hip arthroplasty, as most abnormalities can be diagnosed radiographically. However, scintigraphy, sonography, CT, and MR imaging can be useful in the evaluation of osteolysis, infection, and periprosthetic soft tissues. In particular, MR imaging may be useful in the evaluation of the radiographically normal hip with pain. References [1] Charnley J. The long-term results of low-friction arthroplasty of the hip performed as a primary intervention. J Bone Joint Surg 1972;54B:61–6. [2] Barrack RL. Early failure of modern cemented stems. J Arthroplasty 2000;15:1036–50. [3] Manaster B. Total hip arthroplasty: radiographic evaluation. RadioGraphics 1996;16:645–60. [4] Kaplan PA, Montesi SA, Jardon OM, Gregory PR. Bone-ingrowth hip prostheses in asymptomatic patients: radiographic features. Radiology 1988;169:221–7. [5] Shimmin A, Beaulé PE, Campbell P. Metal-on-metal hip resurfacing arthroplasty. J Bone Joint Surg Am 2008;90:637–54. [6] Langton DJ, Jameson SS, Joyce TJ, Webb J, Nargol AV. The effect of component size and orientation on the concentrations of metal ions after resurfacing arthroplasty of the hip. J Bone Joint Surg Br 2008;90:1143–51. [7] Marker DR, Strimbu K, McGrath MS, Zywiel MG, Mont MA. Resurfacing versus conventional total hip arthroplasty – review of comparative clinical and basic science studies. Bull NYU Hosp Jt Dis 2009;67:120–7. [8] Zywiel MG, Marker DR, McGrath MS, Delanois RE, Mont MA. Resurfacing matched to standard total hip arthroplasty by preoperative activity levels – a comparison of postoperative outcomes. Bull NYU Hosp Jt Dis 2009;67:116–9. [9] Clohisy JC, Calvert G, Tull F, McDonald D, Maloney WJ. Reasons for revision hip surgery: a retrospective review. Clin Orthop 2004;429:188–92. [10] von Knoch M, Sychterz CJ, Engh Jr CA, Engh Sr CA. Incidence of late bead shedding from uncemented porous coated cups. A radiographic evaluation. Clin Orthop Relat Res 1997;342:99–105. [11] Della Valle AG, Zoppi A, Peterson MGE, Salvati EA. Clinical and radiographic results associated with a modern, cementless modular cup design in total hip arthroplasty. J Bone Joint Surg 2004;86A:1998–2004. [12] Gruen TA, McNeice GM, Amstutz HC. “Modes of Failure” of cemented stemtype femoral components: a radiographic analysis of loosening. Clin Orthop 1979;11:17–27. [13] DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976;121:20–32. [14] Pollard TC, Baker RP, Eastaugh-Waring SJ, Bannister GC. Treatment of the young active patient with osteoarthritis of the hip. A five- to seven-year comparison of hybrid total hip arthroplasty and metal-on-metal resurfacing. J Bone Joint Surg Br 2006;88:592–600.
[15] Konttinen YT, Zhao D, Beklen A, et al. The microenvironment around total hip replacement prostheses. Clin Orthop Relat Res 2005;430:28–38. [16] McKellop HA. Bearing surfaces in total hip replacements: state of the art and future developments. Instr Course Lect 2001;50:165–79. [17] Bauer TW, Schils J. The pathology of total joint arthroplasty II: mechanisms of implant failure. Skeletal Radiol 1999;28:483–97. [18] Dumbleton JH, Manley MT, Edidin AA. A literature review of the association between wear rate and osteolysis in total hip arthroplasty. J Arthroplasty 2002;17:649–61. [19] Silva M, Heisel C, Schmalzried TP. Metal-on-metal total hip replacement. Clin Orthop Relat Res 2005;430:53–61. [20] Hannouche D, Hamadouche M, Nizard R, Bizot P, Meunier A, Sedel L. Ceramics in total hip replacement. Clin Orthop Relat Res 2005;430:62–71. [21] McCalden RW, MacDonald SJ, Rorabeck CH, Bourne RB, Chess DG, Charron KD. Wear rate of highly cross-linked polyethylene in total hip arthroplasty. A randomized controlled trial. J Bone Joint Surg Am 2009;91:773–82. [22] Schmalzried TP, Jasty M, Harris WH. Periprosthetic bone loss in total hip arthroplasty. J Bone Joint Surg 1992;74-A:849–63. [23] Claus AM, Engh Jr CA, Sychterz CJ, Xenos JS, Orishimo KF, Engh Sr CA. Radiographic definition of pelvic osteolysis following total hip arthroplasty. J Bone Joint Surg 2003;85-A:1519–26. [24] Walde TA, Weiland DE, Leung SB, et al. Comparison of CT, MRI, and radiographs in assessing pelvic osteolysis: a cadaveric study. Clin Orthop Relat Res 2005;437:138–44. [25] Potter HG, Foo LF. Magnetic resonance imaging of joint arthroplasty. Orthop Clin North Am 2006;37:361–73. [26] Sugimoto H, Hirose I, Miyaoka E, et al. Low-field strength MR imaging of failed hip arthroplasty: association of femoral periprosthetic signal intensity with radiographic, surgical, and pathological findings. Radiology 2003;229: 718–23. [27] Garcia-Cimbrelo E, Tapia M, Martin-Hervas C. Multislice computed tomography for evaluating acetabular defects in revision THA. Clin Orthop Relat Res 2007;463:138–43. [28] Hamilton WG, McCauley JP. Evaluation of the unstable total hip arthroplasty. Instr Course Lect 2004;53:87–92. [29] Cyteval C, Hamm V, Sarrabere MP, Lopez FM, Maury P, Taourel P. Painful infection at the site of hip prosthesis: CT imaging. Radiology 2002;224:477–83. [30] van Holsbeeck MT, Eyler WR, Sherman LS, et al. Detection of infection in loosened hip prostheses: efficacy of sonography. AJR 1994;163:381–4. [31] Steinbach LS, Schneider R, Goldman AB, Kazam E, Ranawat CS, Ghelman B. Bursae and abscess cavities communicating with the hip. Diagnosis using arthrography and CT. Radiology 1985;156:303–7. [32] Palestro CJ, Love C. Radionuclide imaging of musculoskeletal infection: conventional agents. Semin Musculoskelet Radiol 2007;11:335–52. [33] Chacko TK, Zhuang H, Stevenson K, Moussavian B, Alavi A. The importance of the location of fluorodeoxyglucose uptake in periprosthetic infection in painful hip prostheses. Nucl Med Commun 2002;23:851–5. [34] Mayer-Wagner S, Mayer W, Maegerlein S, Linke R, Jansson V, Müller PE. Use of (18)F-FDG-PET in the diagnosis of endoprosthetic loosening of knee and hip implants. Arch Orthop Trauma Surg 2009;(November) [Epub ahead of print]. [35] Stumpe K, Notzli HP, Zannetti M, et al. FDG PET for differentiation of infection and aseptic loosening in total hip replacements: comparison with conventional radiography and three-phase bone scintigraphy. Radiology 2004;231: 333–41. [36] Reinartz P. FDG-PET in patients with painful hip and knee arthroplasty: technical breakthrough or just more of the same. Q J Nucl Med Mol Imaging 2009;53:41–50. [37] Lachiewicz PF, Rogers GD, Thomason HC. Aspiration of the hip joint before revision total hip arthroplasty. Clinical and laboratory factors influencing attainment of a positive culture. J Bone Joint Surg 1996;78-A:749–54. [38] Berry DJ. Epidemiology: hip and knee. Orthop Clin North Am 1999;30:183–90. [39] Tsidiris E, Haddad FS, Gie GA. The management of periprosthetic femoral fractures around hip replacements. Injury 2003;34:95–105. [40] Werle J, Goodman S, Schurman D, Lannin J. Polyethylene liner dissociation in Harris–Galante acetabular components. J Arthroplasty 2002;17:78–81. [41] Cipriano CA, Issack PS, Beksac B, Della Valle AG, Sculco TP, Salvati EA. Metallosis after metal-on-polyethylene total hip arthroplasty. Am J Orthop 2008;37:E18–25. [42] Cooper HJ, Ranawat AS, Potter HG, Foo LF, Jawetz ST, Ranawat CS. Magnetic resonance imaging in the diagnosis and management of hip pain after total hip arthroplasty. J Arthroplasty 2009;24:661–7. [43] Pandit H, Glyn-Jones S, McLardy-Smith P, et al. Pseudotumours associated with metal-on-metal hip resurfacings. J Bone Joint Surg Br 2008;90:847–51. [44] Schmalzried TP. Metal–metal bearing surfaces in hip arthroplasty. Orthopedics 2009;32(9), pii: orthosupersite.com/view.asp?rID=42831. [45] Campbell P, Ebramzadeh E, Nelson S, Takamura K, De Smet K, Amstutz HC. Histological features of pseudotumor-like tissues from metal-on-metal hips. Clin Orthop Relat Res 2010;468:2321–7. [46] Anderson H, Toms AP, Cahir JG, Goodwin RW, Wimhurst J, Nolan JF. Grading the severity of soft tissue changes associated with metal-on-metal hip replacements: reliability of an MR grading system. Skeletal Radiol 2010;(July) [Epub ahead of print].
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Imaging of hip arthroplasty Theodore T. Miller ∗ Department of Radiology and Imaging, Hospital for Special Surgery, 535 E. 70th Street, New York, NY 10021, United States
a r t i c l e
i n f o
Article history: Received 16 February 2011 Accepted 22 March 2011 Keywords: Hip Arthroplasty Resurfacing Complications Imaging CT MRI
a b s t r a c t The imaging evaluation of the prosthetic hip begins with radiography, but arthrography, aspiration, scintigraphy, sonography, CT and MR imaging all have roles in the evaluation of the painful prosthesis. This article will review the appearance of normal hip arthroplasty including hemiarthroplasty, total arthroplasty, and hip resurfacing, as well as the appearances of potential complications such as aseptic loosening and osteolysis, dislocation, infection, periprosthetic fracture, hardware failure, and soft tissue abnormalities. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The modern era hip arthroplasty began in the 1960s when a stainless steel metal-on-polyethylene prosthesis was pioneered by Charnley [1]. Many different variations and designs have since been introduced, but most follow his principle of a metal femoral head articulating against a polyethylene socket. Hip arthroplasty has become so successful, with some designs having a 25 year survivorship of almost 80% that the hip is the most commonly replaced joint. This article will review the appearance of normal hip arthroplasty, as well as the appearances of potential complications.
2. Normal The hip joint can be artificially restored in several ways. Hemi-arthroplasty replaces only the femoral side of the joint, usually in cases of femoral neck fracture or avascular necrosis in which the acetabular side of the joint is normal. In a unipolar hemi-arthroplasty the prosthetic femoral head articulates directly against the acetabular cartilage, whereas in a bipolar hemiarthroplasty the prosthetic femoral head articulates against a prosthetic cup which is placed into, but not fixed to, the native, unprepared, acetabulum and which articulate against the acetabular cartilage. The acetabular articular cartilage will eventually wear away in both uni- and bipolar hemiarthroplasties, leading to painful degenera-
∗ Tel.: +1 212 606 1127; fax: +1212 734 7475. E-mail address: [email protected] 0720-048X/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2011.03.103
tive arthritis of the acetabulum, requiring revision to a total hip arthroplasty (Fig. 1). Total hip arthroplasty, in which both the femoral head and the acetabulum are replaced by fixed prosthetic devices, is most often performed for disease processes which have affected both sides of the native joint, such as degenerative and inflammatory arthritis. The acetabular and femoral components may be cemented or noncemented (either press-fit or in-growth), with the most common combination being a cemented femoral stem and non-cemented acetabular cup (Fig. 2). The femoral component should be placed in either mild valgus position, such that the tip of the stem is located in the medial aspect of the medullary canal, or in neutral position with the tip located centrally in the medullary canal. A “cement restrictor”, a plug made of plastic, cement, metal, or bioabsorbable material, is sometimes present in the femoral shaft, just distal to the tip of the stem; this plug keeps the cement contained in the medullary canal during cement injection and stem insertion in order to maintain injection pressure and assure good bonding of the cement with the endosteal trabeculae. The cement mantle should be at least 2 mm thick all the way around the stem [2] to minimize the risk of subsequent cement fracture. A cemented stem may have a surrounding thin radiolucency at the cement–bone interface, representing a fibrous pseudocapsule that forms as a result of trabecular necrosis, due to either the marked exothermic curing of the cement or the surgical reaming of the medullary canal, with an adjacent thin sclerotic line along its outer margin representing reactive bone [3]. An ingrowth stem usually has tiny beads or wires scintered onto its proximal surface to increase its surface area for bonding with bone, and the scintered surface gives the prosthesis a fuzzy
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 1. Hemiarthroplasties. (a) AP radiograph of a unipolar prosthesis shows the prosthetic femoral head articulating directly against the native acetabular cartilage, causing cartilage narrowing (arrow). (b) AP radiograph of a bipolar hemiarthroplasty shows the acetabular cup, which is placed into the native socket but is not fixed to it. The acetabular cartilage is narrowed by the cup (straight arrow). The inferior aspect of the prosthetic femoral head (round tail arrow) is visible.
edge radiographically. The non-cemented component may also be coated with hydroxyapatite in order to induce bone formation onto the component’s surface. Radiographically, the well-fixed ingrowth component shows bony sclerosis extending onto the prosthesis (Fig. 3). There may normally be some cortical or endosteal sclerosis and thickening of the femoral shaft at the level of the tip of the femoral component due to normal transfer of load stresses along the femoral stem [4]. Thin linear lucencies, less than 2 mm wide, representing fibrous ingrowth rather than bony ingrowth may also be seen, most often along the proximal aspect of the stem medially and laterally [4]. The greater trochanter may have decreased radiodensity due to “stress shielding” because the implant shifts physiologic load away from the greater trochanter. Similarly, resorption of the calcar of the femur due to stress shielding may occur with both cemented and
3803
non-cemented stems [3,4], but the extent of resorption depends on the particular model of stem. The acetabular component is usually placed in approximately 40◦ of vertical tilt (“inclination”) from the horizontal on an AP radiograph, depending on the model of the prosthesis and preoperative condition of the acetabulum, and is usually anteverted (i.e., open facing anteriorly) 10–20◦ . Thin radiolucency at the metal–bone interface of a non-cemented acetabular component represents fibrous ingrowth. Most arthroplasty designs utilize a high-density polyethylene (PE) liner in the acetabular cup, articulating with a metal prosthetic femoral head. The prosthetic head should be symmetrically seated within the cup, appearing on a frontal radiograph equidistant from the superior and inferior margins of the cup. The head may be slightly inferiorly located in the cup if the surgeon has used an “off-set” liner which has a thicker superior rim, but a head located superiorly in the cup, even mildly so, is never normal and indicates PE wear. Most femoral heads are metallic, and are thus radiographically dense; ceramic heads, used in an attempt to decrease PE wear, are radiolucent (Fig. 2). The rim of the PE liner usually is flush with the rim of the metal cup, but sometimes additional mechanical constraint is needed, such as for patients with recurrent dislocation of the femoral head. In this group of patients, a constrained liner extends beyond the cup itself, to deepen the articulation and limit the range of motion (Fig. 4). In younger, active patients with disease on both sides of the joint, such as patients with femoroacetabular impingement and secondary degenerative arthritis, hip resurfacing (HR) might be performed instead of total arthroplasty. In both HR and THA the acetabulum is replaced, but in HR a pegged metal cap is cemented over the femoral head, thus preserving femoral neck bone stock for the eventual revision to a THR that the patients may require as they age. HR designs use large diameter metal-on-metal surfaces in order to minimize surface wear [5]. The femoral component should be in 5–10◦ valgus relative to the femoral neck [5]; varus positioning or exaggerated valgus positioning can lead to notching of the femoral neck with resultant fracture or loosening. The femoral peg is usually not cemented, as cementation can cause stress shielding which may lead to loosening [5]. The acetabular component should have inclination and anteversion similar to that of a THA [6] (Fig. 5). Large series of long term (>10 years) survivorship data for HR is lacking since the current generation of HR models was just coming into use about 10 years ago, but medium term (4–6 year) survivorship of HR ranges from 94% to 99%, similar to that of THA [5]. A recent meta-analysis of the literature comparing various outcomes of HR and THA [7] showed that HR patients tend to have higher postoperative activity levels than THA patients, even when controlling for preoperative activity levels [8], and that gait is more natural after HR. However, HR cannot correct for limb length discrepancies greater than 2 cm and has limited ability to correct femoral head–neck offset [5]. Other contraindications to HR include: inflammatory or erosive arthritis (because of the risk of loosening or fracture), osteoporosis or steroid use (because of the risk of fracture), short femoral neck, acetabular dysplasia, poor femoral head bone stock due to large cysts, hypersensitivity to metal, and poor renal function (because of impaired ability to clear metal ions) [5].
3. Complications Complications of hip arthroplasty include aseptic loosening and osteolysis, dislocation, infection, periprosthetic fracture, hardware failure, and soft tissue abnormalities.
3804
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 2. AP radiographs of various combinations of total hip arthroplasty. Notice that the acetabular cups of total arthroplasties are smaller than the cup of a bipolar arthroplasty in order to avoid neck–cup impingement. (a) Metal head against polyethylene liner; cemented femoral stem. (b) Ceramic head against polyethylene liner; cemented femoral stem. The ceramic head is lucent compared to metal. (c) Metal head against polyethylene liner; non-cemented, ingrowth femoral stem. The ingrowth is supposed to occur around the proximal aspect of the stem (arrows). (d) Metal head against metal liner; non-cemented, ingrowth femoral stem. The large head distributes load, decreasing metal wear.
3.1. Aseptic loosening and osteolysis Aseptic loosening of the prosthesis is the most common reason for revision surgery [2,9]. Radiographic appearances of loosening of a cemented femoral prosthesis are: lucency at the
cement–bone interface of >2 mm surrounding the component, progressive widening of the lucency at the cement–bone interface, lucency at the metal–cement interface, and fracture of the cement mantle (Fig. 6). Radiographic appearances of loosening of a non-cemented femoral prosthesis are: lucency at the metal–bone
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3805
Fig. 4. Constrained liner. AP radiograph of a total hip replacement shows the metal ring (arrow) of the inferior edge of the polyethylene liner.
interface >2 mm surrounding the component, development or widening of the lucency at the metal–bone interface, and subsidence of more than 1 cm and/or which continues to progress more than 1 year after placement [3]. Shedding of surface beads can be seen with both loose and stable implants [10]. The radiographic appearance of loosening of cemented and non-cemented acetabular components is lucency >2 mm at the cement–bone or metal–bone interface around its entire circumference [11]. Additional radiographic features of loosening, regardless of whether the components are femoral or acetabular, cemented or non-cemented, are: migration of the component or change of position of the component, fracture of the component, and component motion with stress views. The orthopedic classification for modes of stem failure is listed in Table 1 [12]. The description of the location of the lucencies around a total hip arthroplasty should follow the standard orthopedic descriptions of
Fig. 3. In-growth femoral component. (a) Initial post-operative radiograph. (b) AP radiograph 1.5 years later shows early in-growth of bone (arrow) onto the stem.
Fig. 5. AP radiograph of normal hip resurfacing. The large metal resurfacing head (arrow) is cemented to the femur, but the cement is hidden beneath the head. The stem is not cemented, and is in 5–10◦ of valgus relative to the neck. The acetabular component (round tail arrow) is the same as in a metal-on-metal total hip arthroplasty.
3806
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812 Table 2 Mechanisms of wear [16,17]. 1. Abrasion – the scratching of a surface by some other, usually harder, surface 2. Adhesion – transient bonding of the bearing surfaces to each other, usually due to poor lubrication, which pulls particles from the weaker surface 3. Fatigue – stress of the material beyond its normal mechanical limits, with resultant release of particles
Table 3 Classification of wear [16]. Type 1. Normal articulation between two bearing surfaces Type 2. Articulation between a bearing surface and non-bearing surface (e.g., prosthetic head penetrating through the polyethylene liner to articulate against metal backing) Type 3. Third-body abrasion, caused by a fragment of material interposed between normally articulating surfaces Type 4. Motion between two non-bearing surfaces (e.g., between the backside of the polyethylene liner and the metal acetabular backing, or between the modular head and its connecting taper)
Fig. 6. Femoral stem loosening. AP radiograph shows a cemented stem with irregular and focally wide radiolucency around the entire cement mantle, but particularly medially (arrows) with marked thinning of the cortex. The cement column is broken (round tail arrow) and shifted in position.
femoral and acetabular “zones”: there are seven femoral zones on the AP radiograph, beginning at the proximal lateral aspect and progressing clockwise with zone 4 at the tip of the femoral stem [12]. There are an additional seven zones on the lateral radiograph, numbered 8–14, beginning at the anteroproximal aspect of the femoral stem. The region around the acetabular component is divided into three equal zones, 1–3, from superolateral to inferomedial [13]. The acetabular zones for HR are the same, but unlike the femoral arthroplasty stem the femoral HR peg has only six zones: superior surface of peg (zone 1), tip of peg (zone 2), inferior surface of peg (zone 3) on the AP radiograph, and anterior (zone 4) to posterior (zone 6) on the lateral view [14]. The most common cause of periprosthetic lucency is mechanical loosening, but osteolysis due to either “particle disease” or infection can look similar radiographically. Any of the components of a hip replacement, such as the metal, polyethylene liner, or cement, can become fragmented due to wear, shedding microscopic particles of material which can induce a histiocytic inflammatory reaction [15]. Acetabular PE is the most common source because it is constantly worn by the metal femoral head, an example of abrasive
wear (Tables 2 and 3, and Fig. 7). The average PE wear rate of MOP designs is approximately 100–200 m/year [16]. The particles are engulfed by macrophages, which then release various factors and cytokines, such as interleukins, prostaglandins, and tumor necrosis factor, which cause up-regulation of osteoclasts, leading to osteolysis [17]. Osteolysis is dependent on the number of shed particles (reflected in the volumetric and linear wear rates of the bearing surfaces) and the histiocytic response to those particles. While there is no absolute threshold for the number of shed PE fragments necessary to incite the histiocytic response, the fewer the number of shed particles the less likely an inflammatory response; a linear wear rate of less than 100 m/year has been suggested as a threshold level [18]. Moreover, the type of particle itself is important, since PE particles have a high inflammatory profile, while metal and ceramic particles do not [19,20]. Therefore, in an attempt to minimize the histiocytic response, different materials and combinations of articulations have been investigated. Ceramic-on-ceramic (COC) designs have the lowest coefficient of friction and the lowest wear rate [16,20], but disadvantages are their high cost, dependence on precise positioning, and fracture (.001–.002% incidence) [16]. Metal-on-metal (MOM) designs,
Table 1 Modes of failure of cemented stems [12]. Mode 1. Pistoning (up-and-down motion) 1A. Pistoning of the stem within the cement mantle. 1B. Pistoning of the stem and cement mantle within the medullary canal. Mode 2. Medial midstem pivot (medial migration of proximal stem and lateral migration of stem tip) Mode 3. Calcar pivot (medial–lateral toggling of the stem tip, analogous to the “windshield wiper” phenomenon of uncemented stems) Mode 4. Bending cantilever fatigue (medial migration of the proximal stem with distal fixation of the stem). This is often the result of initial varus positioning of the stem
Fig. 7. AP radiograph shows superior eccentric seating of the femoral head, indicating polyethylene wear.
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3807
in which a cobalt–chromium–molybdenum femoral head articulates against a similar acetabular surface, also has a low wear rate (approximately 3–5 m/year) after an initial “wear-in” period. Metal particles are smaller than PE particles and produce a lower grade histiocytic response [19], with resultant less osteolysis. However, patients with MOM designs have higher serum and urine levels of chromium and cobalt than control subjects [6], raising concern about carcinogenesis. Metal ions can also cross the placenta, and although no teratogenic or mutagenic effects have been documented, it is recommended that MOM implants not be performed in women of child-bearing age [5]. Additional risk factors for elevated metal ions in MOM patients include an acetabular inclination angle >55◦ (because of edge loading), female gender (because of gait), and small femoral head size (because of increased contact stress and resultant wear) [6]. Lastly, highly cross-linked PE is more resistant to wear than standard ultra high molecular weight PE, with a wear rate similar to MOM components [21]. The development of highly cross-linked PE allows the use of MOP designs, which are familiar to surgeons, more forgiving in their positioning, and more versatile than MOM and COC for making off-set liners and constrained liners. However, while highly cross-linked PE is more resistant to wear against smooth bearing surfaces, it is not more resistant to third-body abrasion [21]. Osteolysis due to particle disease is suggested radiographically by focal well-defined radiolucencies around either the acetabular or femoral components (Fig. 8). Osteolysis may occur at sites away from the actual articulating surfaces of the arthroplasty by insinuation of joint fluid and its inflammatory debris around both the femoral and the acetabular components due to the hydrostatic pressure generated by joint movement, a concept called the “effective joint space” [22]; thus, for example, screw holes in an acetabular component provide an avenue for particulate debris to reach pelvic bone (Fig. 9). Radiographs are not sensitive for the depiction of osteolysis and underestimate the extent of involvement [23]. Both CT and MRI are more sensitive for detecting osteolysis and demonstrating its extent than radiographs because of their tomographic nature. In an experimental model of osteolysis, radiographs had 52% sensitivity, CT had 75% sensitivity and MRI had 95% sensitivity [24]. On MR imaging, osteolysis is usually low signal intensity on T1WI, and intermediate to high signal intensity on T2WI [25] and can occasionally be expansile (Fig. 10). Often there is accompanying bulky synovitis with debris. Linear high signal intensity surrounding the femoral stem on STIR images may indicate loosening or a histiocytic response [26]. CT is particularly helpful for evaluation of lysis of the medial wall of the acetabulum [27], and it can be useful for evaluating femoral and acetabular bone stock in preparation for revision surgery. 3.2. Dislocation Dislocation is the second most common reason for revision surgery [9,28] and is multifactorial including the age and gender of the patient, the surgical approach, size of the components and position of the components [28]. Dislocation within the first three months after surgery is usually due to laxity of the immature pseudocapsule of the joint and surrounding soft tissues. Atraumatic dislocation occurring between three months and 5 years after surgery is usually due to component malposition, such as acetabular inclination >60◦ , acetabular anteversion >20◦ , or acetabular retroversion (Fig. 11). Inclination can be assessed on frontal radiographs, and acetabular version is well assessed with CT scan [28]. Dislocation occurring more than 5 years after placement is usually due to gradual stretching of the pseudocapsule and surrounding soft tissue laxity, and women are at greater risk than men [28].
Fig. 8. Osteolysis. (a) AP radiograph of a cemented femoral stem shows small faint lucent areas (arrows). (b) Two years later the radiolucenies have enlarged and new ones have developed.
3808
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 9. Coronally reformatted CT image shows osteolysis (straight white arrows) around the acetabular screw (round tail arrow) and a screw hole (black arrow). Notice the superior eccentric position of the femoral head due to polyethylene wear.
Surgical options for treating recurrent dislocation include correction of acetabular position, placement of a constrained liner which provides stability by deepening the articulating socket thus limiting the femoral range of motion, and revision of the femoral head to a larger size. 3.3. Infection Infection is the third most common reason for revision arthroplasty [9], occurring in 1–5% of hip replacements [17]. Radiographic findings suggestive of infection include a wide irregular radiolucency around the cement–bone or metal–bone interfaces and frank bone destruction. However, a distinction between infectious osteolysis and aseptic osteolysis due to mechanical loosening or particle disease often cannot be made on a single radiograph. Usually, previous radiographs are necessary for comparison, with mechanical
Fig. 10. Axial proton density image through the acetabulum of a patient with a total hip replacement shows dephasing artifact due to the acetabular cup, but a large surrounding area of osteolysis is visible (arrows) with intermediate to high signal intensity and mild expansion of the medial wall.
Fig. 11. AP radiograph of a person with recurrent hip dislocations shows an excessively inclinated acetabular component.
loosening and histiocytic osteolysis usually taking a slowly progressive course, whereas infection usually occurs with a rapid time course and an aggressive appearance. While the appearance of osteolysis cannot distinguish infectious from non-infectious loosening, the presence of femoral periosteal reaction or an adjacent soft tissue collection are highly predictive [29,30]. A fluid collection over the greater trochanter or along the iliopsoas tendon may be a normal communicating bursa or an expected post-surgical space due to disruption of normal soft tissue planes, but a communicating non-bursal cavity with an irregular lining is suspicious for infection [31]. These cavities can be assessed with arthrography, MR imaging, and sonography (Figs. 12 and 13). Various nuclear medicine examinations can also be used to evaluate clinically suspected infection in hip arthroplasty. Three-phase bone scan suffers from poor specificity since a cemented femoral component can normally have increased uptake around the prosthesis for several years after placement, and because a normal non-cemented prosthesis also demonstrates increased radiotracer uptake at the site of bony ingrowth onto the prosthesis. However, the bone scan has excellent negative predictive value and can thus be used as an initial screening test since no uptake reliably excludes infection [32]. Adding a gallium scan to the standard technetium bone scan can improve the diagnostic accuracy for infection to 70–80%: infection is excluded if the gallium scan is normal or has less intense uptake than the corresponding bone scan, and infection is diagnosed when there is uptake of gallium without corresponding Tc uptake or the gallium uptake is more intense than corresponding Tc uptake [32]. The combination of technetium or indium labeled white cells and technetium labeled sulfur colloid has excellent results, with accuracy of over 90%, and is currently the scintigraphic method of
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3809
Fig. 12. Infected hip arthroplasty. (a) Coronal T2-weighted inversion recovery image shows a peritrochanteric collection (horizontal arrow), with communication to the skin (vertical arrow). (b) Coronal T1-weighted image after intravenous contrast shows the peritrochanteric abscess (horizontal arrows) and the superficial draining abscess (vertical arrow). (c) Axial T1-weighted image after intravenous contrast shows the peritrochanateric abscess (round tail arrow) communicating with the posterior aspect of the prosthetic joint (straight arrows). (d) Longitudinal sonographic image during needle aspiration shows the needle (straight arrows) in the abscess (round tail arrow).
choice for evaluating suspected infection. The imaging feature of infection is spatial incongruence, in which there is uptake of the labeled white cells, regardless of intensity, without uptake of the sulfur colloid [32]. Fluoro-deoxy-glucose (FDG)-PET (positron emission tomography) scanning has variable performance for evaluation of prosthetic infection. While Chacko et al. reported 92% sensitivity and 97% specificity [33], Mayer-Wagner et al. reported 67% sensitivity and 83% specificity [34], and Stumpe et al. reported sensitivity of only 22–33%, with an overall accuracy of 69%, which was the same as radiographs and worse than three-phase bone scan [35]. A meta analysis of the literature showed slightly poorer accuracy of PET compared to white blood cell imaging (89% versus 91%), but PET imaging can be completed within a few hours of injection compared to the 24 h delay of white cell imaging [36]. Aseptic loosening also demonstrates FDG uptake and standardized uptake values (SUV) are not helpful for distinguishing infectious from non-infectious loosening because of overlapping values [36]. Chacko et al. suggested that the location of the uptake is more important than the intensity; uptake around the prosthetic head/neck or at the tip of the femoral stem can be normal, while uptake around the shaft of the stem suggests infection [33]. The gold standard for the evaluation of a clinically suspected infected joint is aspiration, with gram stain and culture of the joint
fluid. Although considered the definitive diagnostic test, there is a wide range of reported sensitivities. The sensitivity is better in patients with high clinical suspicion of infection or with periosteal reaction [37].
3.4. Periprosthetic fracture Periprosthetic fractures are rare with total hip replacements, and occur more often around the femoral than the acetabular component. Fracture of the femur may occur during placement of the femoral stem, usually as either focal cortical penetration or longitudinal splitting of the bone, and happens more often with uncemented components than cemented ones (5% versus 0.3%) because of the tight press-fit needed with uncemented stems [38]. The incidence is even higher in revision THA because of diminished bone stock resulting from osteoporosis or prior osteolysis, with an incidence of 6.3% for cemented THAs and almost 18% in non-cemented cases [39]. Fracture of the femur may also occur any time after hip replacement, typically at the level of the tip of the femoral stem because of “stress risers” at this level caused by the difference in stiffness between the metal stem and bony shaft. These fractures also are more common in revision hips (4%) than primary arthroplasties (1.1%), because of deficient bone stock [38].
3810
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 13. Infected hip arthroplasty. (a) Longitudinal sonographic image of the prosthetic hip joint shows a large joint effusion (arrows) around the prosthetic head (H) and neck (N). A is the acetabulum. (b) Oblique sonographic image shows the effusion (straight arrow) communicating with a large peritrochanteric collection (round tail arrow), which in turn has a tract (dashed arrow) toward the skin. (c) AP radiograph after intraarticular injection of contrast shows the irregularly shaped peritrochanteric collection with the thin communicating tract (arrow) to the skin.
Fractures of the femoral neck are the most common complication of hip resurfacings, with reported rates ranging from 0% to 12% [5]. Risk factors for fracture include scalloping of the femoral neck, osteonecrosis of the remaining femoral head, or poor cement integration. 3.5. Hardware failure Hardware failure can affect either the femoral or the acetabular components. The stem of the femoral component can break, representing a metal fatigue stress fracture, because the metal stem is more stiff and less yielding than the surrounding femoral bone, although the incidence of fracture depends on the geometry and metal composition of the stem (Fig. 16). The scintered beads of an in-growth stem may shear-off, indicating either micromotion or frank loosening of the stem, and may act as a cause of third-party abrasion [10]. The superior aspect of the polyethylene liner can become gradually worn down, appearing radiographically as asymmetric superior location of the femoral head within the acetabular cup, constituting type I wear (Table 3 and Fig. 7). In addition to gradual full-thickness wear, the polyethylene liner can frankly break and dissociate from the metal acetabular shell. In this latter scenario, the femoral head will be superiorly seated against the acetabular shell, and displaced broken metal tines may be seen on radiographs [40]. The broken pieces of the liner can be visualized arthrographically as lucent filling defects within the contrast pool. The resulting abnormal articulation of the metal head against the metal acetabular shell can lead to marked shedding of metal debris that distends the hip
Fig. 14. Early metallosis. AP radiograph shows marked superior eccentric seating of the metal femoral head articulating against the metal acetabular cup, an example of Type 2 wear (Table 3). The distended joint capsule (white arrows) is dense due to metal debris, the metal tines (short black arrow) of the polyethylene’s metal backing are becoming visible as they unlock, and a piece of broken polyethylene (long black arrow) is faintly visible within the dense effusion.
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
3811
Fig. 16. Coronal proton density MR image shows atrophy of the gluteus medius muscle (straight arrow). There are also areas of high signal intensity acetabular and femoral osteolysis (round tail arrows) and a resultant fracture of the greater trochanter (dashed arrow).
and involving all three AP zones; 4. narrowing of the femoral neck – this may be due to remodeling if stable and less than 10% of the neck diameter, but continued narrowing after 3 years or narrowing of more than 10% may reflect osteolysis or impingement (Fig. 15); 5. scalloping at the neck–rim junction – could be due to impingement or remodeling; 6. osteolysis around the peg. Sclerosis around the tip of the peg, the “pedestal sign”, is of uncertain clinical significance [5,14] 3.6. Soft tissue complications Fig. 15. Osteolysis around a hip resurfacing. (a) AP radiograph shows osteolysis in acetabular zone 3 (black arrow), and narrowing of the femoral neck (white arrows) due to osteolysis. Compare to the appearance of the normal femoral neck in Fig. 5. (b) Axial proton density MR image shows mixed signal intensity osteolysis (straight arrows) adjacent to the femoral neck stem (round tail arrow).
capsule, called “metallosis”, and which appears radiographically as a dense “bubble” around the joint [41] (Fig. 14). The incidence of fracture with current generation ceramic components is approximately 1 in 1000 [16], and is usually caused by poor component placement, leading to edge loading. Fracture is a devastating injury, because of potential damage to the femoral taper on which the head implanted, and because any remaining ceramic shards can cause severe third body abrasion, especially of a softer revision material such as a metal head. The radiographic evaluation of the acetabular component of a hip resurfacing is similar to that of a THR, looking for change in inclination and periacetabular osteolysis. For the radiographic evaluation of the femoral component of a hip resurfacing, Shimmin et al. [5] suggest that the following signs may indicate early failure: 1. change in the peg–neck angle – this may mean femoral head necrosis or collapse, neck fracture, or component loosening; 2. decreasing distance of the tip of the peg to the lateral femoral cortex – this may indicate subsidence, loosening, or head collapse; 3. radiolucent line around the peg – may indicate loosening if greater than 2 mm wide
MR imaging and sonography are useful for the evaluation of the painful or clinically abnormal but radiographically normal hip arthroplasty. Greater trochanteric bursitis and other periprosthetic collections, dehiscence of the posterior pseudocapsule and short external rotators, and tendon tears or tendinosis can be demonstrated [42] (Fig. 16). Sonography can also provide guidance for needle aspiration of the joint or periprosthetic collections (Fig. 12) and for steroid/anesthetic injections such as of the greater trochanteric bursa and the iliopsoas tendon (Fig. 17). A particular soft tissue complication of MOM prostheses is the “adverse local tissue reaction” (ALTR). Patients present with pain, most commonly in the groin but also laterally and in the buttocks, and women are more commonly affected than men [43,44]. The adverse reaction may either be a high-wear macrophagemediated event or a low-wear lymphocyte-mediated event [44,45]. The low-wear lymphocyte-mediated event has been termed “aseptic lymphocytic vasculitis-associated lesions” (ALVAL), and is believed to be a type of hypersensitivity reaction [45]. MR imaging of ALTR’s can demonstrate capsular dehiscence, periprosthetic fluid collections, osteolysis, and focal inflammatory pseudotumors that may be either hypo- or hyperintense on T2-weighted images [43,45,46]. Histologically, the ALVAL lesions are characterized by a lymphocyte-predominant infiltrate and tissue necrosis, whereas the high-wear events are characterized by a macrophage-
3812
T.T. Miller / European Journal of Radiology 81 (2012) 3802–3812
Fig. 17. Transverse sonographic image at the level of the pelvic brim (black round tail arrow) during a cortisone injection shows a spinal needle (white straight arrows) positioned with its tip deep to the iliopsoas tendon (white round tail arrow). The iliopsoas bursa (B) is distended by the injectate. The metal acetabular component is demarcated by the straight black arrow.
predominant infiltrate and metal debris without tissue necrosis [45]. 4. Conclusion Radiography should always be the first step in the imaging evaluation of hip arthroplasty, as most abnormalities can be diagnosed radiographically. However, scintigraphy, sonography, CT, and MR imaging can be useful in the evaluation of osteolysis, infection, and periprosthetic soft tissues. In particular, MR imaging may be useful in the evaluation of the radiographically normal hip with pain. References [1] Charnley J. The long-term results of low-friction arthroplasty of the hip performed as a primary intervention. J Bone Joint Surg 1972;54B:61–6. [2] Barrack RL. Early failure of modern cemented stems. J Arthroplasty 2000;15:1036–50. [3] Manaster B. Total hip arthroplasty: radiographic evaluation. RadioGraphics 1996;16:645–60. [4] Kaplan PA, Montesi SA, Jardon OM, Gregory PR. Bone-ingrowth hip prostheses in asymptomatic patients: radiographic features. Radiology 1988;169:221–7. [5] Shimmin A, Beaulé PE, Campbell P. Metal-on-metal hip resurfacing arthroplasty. J Bone Joint Surg Am 2008;90:637–54. [6] Langton DJ, Jameson SS, Joyce TJ, Webb J, Nargol AV. The effect of component size and orientation on the concentrations of metal ions after resurfacing arthroplasty of the hip. J Bone Joint Surg Br 2008;90:1143–51. [7] Marker DR, Strimbu K, McGrath MS, Zywiel MG, Mont MA. Resurfacing versus conventional total hip arthroplasty – review of comparative clinical and basic science studies. Bull NYU Hosp Jt Dis 2009;67:120–7. [8] Zywiel MG, Marker DR, McGrath MS, Delanois RE, Mont MA. Resurfacing matched to standard total hip arthroplasty by preoperative activity levels – a comparison of postoperative outcomes. Bull NYU Hosp Jt Dis 2009;67:116–9. [9] Clohisy JC, Calvert G, Tull F, McDonald D, Maloney WJ. Reasons for revision hip surgery: a retrospective review. Clin Orthop 2004;429:188–92. [10] von Knoch M, Sychterz CJ, Engh Jr CA, Engh Sr CA. Incidence of late bead shedding from uncemented porous coated cups. A radiographic evaluation. Clin Orthop Relat Res 1997;342:99–105. [11] Della Valle AG, Zoppi A, Peterson MGE, Salvati EA. Clinical and radiographic results associated with a modern, cementless modular cup design in total hip arthroplasty. J Bone Joint Surg 2004;86A:1998–2004. [12] Gruen TA, McNeice GM, Amstutz HC. “Modes of Failure” of cemented stemtype femoral components: a radiographic analysis of loosening. Clin Orthop 1979;11:17–27. [13] DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res 1976;121:20–32. [14] Pollard TC, Baker RP, Eastaugh-Waring SJ, Bannister GC. Treatment of the young active patient with osteoarthritis of the hip. A five- to seven-year comparison of hybrid total hip arthroplasty and metal-on-metal resurfacing. J Bone Joint Surg Br 2006;88:592–600.
[15] Konttinen YT, Zhao D, Beklen A, et al. The microenvironment around total hip replacement prostheses. Clin Orthop Relat Res 2005;430:28–38. [16] McKellop HA. Bearing surfaces in total hip replacements: state of the art and future developments. Instr Course Lect 2001;50:165–79. [17] Bauer TW, Schils J. The pathology of total joint arthroplasty II: mechanisms of implant failure. Skeletal Radiol 1999;28:483–97. [18] Dumbleton JH, Manley MT, Edidin AA. A literature review of the association between wear rate and osteolysis in total hip arthroplasty. J Arthroplasty 2002;17:649–61. [19] Silva M, Heisel C, Schmalzried TP. Metal-on-metal total hip replacement. Clin Orthop Relat Res 2005;430:53–61. [20] Hannouche D, Hamadouche M, Nizard R, Bizot P, Meunier A, Sedel L. Ceramics in total hip replacement. Clin Orthop Relat Res 2005;430:62–71. [21] McCalden RW, MacDonald SJ, Rorabeck CH, Bourne RB, Chess DG, Charron KD. Wear rate of highly cross-linked polyethylene in total hip arthroplasty. A randomized controlled trial. J Bone Joint Surg Am 2009;91:773–82. [22] Schmalzried TP, Jasty M, Harris WH. Periprosthetic bone loss in total hip arthroplasty. J Bone Joint Surg 1992;74-A:849–63. [23] Claus AM, Engh Jr CA, Sychterz CJ, Xenos JS, Orishimo KF, Engh Sr CA. Radiographic definition of pelvic osteolysis following total hip arthroplasty. J Bone Joint Surg 2003;85-A:1519–26. [24] Walde TA, Weiland DE, Leung SB, et al. Comparison of CT, MRI, and radiographs in assessing pelvic osteolysis: a cadaveric study. Clin Orthop Relat Res 2005;437:138–44. [25] Potter HG, Foo LF. Magnetic resonance imaging of joint arthroplasty. Orthop Clin North Am 2006;37:361–73. [26] Sugimoto H, Hirose I, Miyaoka E, et al. Low-field strength MR imaging of failed hip arthroplasty: association of femoral periprosthetic signal intensity with radiographic, surgical, and pathological findings. Radiology 2003;229: 718–23. [27] Garcia-Cimbrelo E, Tapia M, Martin-Hervas C. Multislice computed tomography for evaluating acetabular defects in revision THA. Clin Orthop Relat Res 2007;463:138–43. [28] Hamilton WG, McCauley JP. Evaluation of the unstable total hip arthroplasty. Instr Course Lect 2004;53:87–92. [29] Cyteval C, Hamm V, Sarrabere MP, Lopez FM, Maury P, Taourel P. Painful infection at the site of hip prosthesis: CT imaging. Radiology 2002;224:477–83. [30] van Holsbeeck MT, Eyler WR, Sherman LS, et al. Detection of infection in loosened hip prostheses: efficacy of sonography. AJR 1994;163:381–4. [31] Steinbach LS, Schneider R, Goldman AB, Kazam E, Ranawat CS, Ghelman B. Bursae and abscess cavities communicating with the hip. Diagnosis using arthrography and CT. Radiology 1985;156:303–7. [32] Palestro CJ, Love C. Radionuclide imaging of musculoskeletal infection: conventional agents. Semin Musculoskelet Radiol 2007;11:335–52. [33] Chacko TK, Zhuang H, Stevenson K, Moussavian B, Alavi A. The importance of the location of fluorodeoxyglucose uptake in periprosthetic infection in painful hip prostheses. Nucl Med Commun 2002;23:851–5. [34] Mayer-Wagner S, Mayer W, Maegerlein S, Linke R, Jansson V, Müller PE. Use of (18)F-FDG-PET in the diagnosis of endoprosthetic loosening of knee and hip implants. Arch Orthop Trauma Surg 2009;(November) [Epub ahead of print]. [35] Stumpe K, Notzli HP, Zannetti M, et al. FDG PET for differentiation of infection and aseptic loosening in total hip replacements: comparison with conventional radiography and three-phase bone scintigraphy. Radiology 2004;231: 333–41. [36] Reinartz P. FDG-PET in patients with painful hip and knee arthroplasty: technical breakthrough or just more of the same. Q J Nucl Med Mol Imaging 2009;53:41–50. [37] Lachiewicz PF, Rogers GD, Thomason HC. Aspiration of the hip joint before revision total hip arthroplasty. Clinical and laboratory factors influencing attainment of a positive culture. J Bone Joint Surg 1996;78-A:749–54. [38] Berry DJ. Epidemiology: hip and knee. Orthop Clin North Am 1999;30:183–90. [39] Tsidiris E, Haddad FS, Gie GA. The management of periprosthetic femoral fractures around hip replacements. Injury 2003;34:95–105. [40] Werle J, Goodman S, Schurman D, Lannin J. Polyethylene liner dissociation in Harris–Galante acetabular components. J Arthroplasty 2002;17:78–81. [41] Cipriano CA, Issack PS, Beksac B, Della Valle AG, Sculco TP, Salvati EA. Metallosis after metal-on-polyethylene total hip arthroplasty. Am J Orthop 2008;37:E18–25. [42] Cooper HJ, Ranawat AS, Potter HG, Foo LF, Jawetz ST, Ranawat CS. Magnetic resonance imaging in the diagnosis and management of hip pain after total hip arthroplasty. J Arthroplasty 2009;24:661–7. [43] Pandit H, Glyn-Jones S, McLardy-Smith P, et al. Pseudotumours associated with metal-on-metal hip resurfacings. J Bone Joint Surg Br 2008;90:847–51. [44] Schmalzried TP. Metal–metal bearing surfaces in hip arthroplasty. Orthopedics 2009;32(9), pii: orthosupersite.com/view.asp?rID=42831. [45] Campbell P, Ebramzadeh E, Nelson S, Takamura K, De Smet K, Amstutz HC. Histological features of pseudotumor-like tissues from metal-on-metal hips. Clin Orthop Relat Res 2010;468:2321–7. [46] Anderson H, Toms AP, Cahir JG, Goodwin RW, Wimhurst J, Nolan JF. Grading the severity of soft tissue changes associated with metal-on-metal hip replacements: reliability of an MR grading system. Skeletal Radiol 2010;(July) [Epub ahead of print].