Pietro Melloni, MD, PhD

The elbow is a complex joint. It links the shoulder and the hand, enhances the flexibility of hand motion, and transmits generated forces. The most common elbow injuries are related to chronic overuse, although the joint is vulnerable to acute trauma as well.

 It may also be affected by disease or infection.

MRI has become the secondary procedure of choice, after plain-film x-ray, for evaluating elbow abnormalities. The modality can visualize ligament and tendon injuries, compressive or entrapment neuropathy, bone injuries, inflammatory and synovial conditions, and soft-tissue masses.

Congenital abnormalities in the elbow are rare. The most common is radial head dislocation, which may affect one or both elbows. Another congenital anomaly is synostosis, or osseous fusion of tubular bones between the proximal portions of the radius and ulna. Fusion may be complete or fibrous, and it may limit forearm supination. Synostosis is often bilateral and may be accompanied by other anomalies.

 MRI is rarely used in such cases, but it may be applied to evaluate the relation between soft tissues and osseous structures.

Infections to the elbow are most often due to Staphylococcus aureus. Osteomyelitis, for example, results from hematogenous spread of bacteria or bacterial spread from surrounding tissues. The thinness of the soft-tissue layer covering the elbow means that osteomyelitis is commonly caused by direct inoculation from an open fracture or surgery.

 Other causes of elbow infection include septic arthritis, septic olecranon bursitis, and pyomyositis. Infection may appear as areas of cortical interruption or periostitis. Its extension into surrounding soft tissues may result in cellulitis, abscess, phlegmon, and sinus tract formation.

MRI is useful for diagnosing or excluding osteomyelitis and evaluating its extent. It can also help determine the location and size of abscesses and identify areas of phlegmonous or necrotic tissue that may need surgical management.

Bone tumors are not common in the elbow, and those that do occur are more often benign than malignant.

 It is extremely important to obtain plain-film x-rays prior to MRI so that suspected tumors can be diagnosed and characterized. MRI is indicated whenever the tumor appears aggressive, has a soft-tissue component, is not clearly seen on x-ray, or requires further anatomic delineation. Intravenous contrast is often helpful but not always necessary. Some benign osseous tumors (e.g., osteoid osteoma, osteochondroma, and enchondroma) have a characteristic appearance on MRI. Other benign tumors found relatively frequently in the elbow are chondroblastoma, fibrous dysplasia (Figure 1), giant cell tumor, aneurysmal bone cyst, and unicameral bone cyst.

Primary malignant osseous tumors are very rare in the elbow. Plain-film x-ray should always be obtained first. MRI is particularly useful for defining tumor extent and also helps in tumor staging, which can affect prognosis and therapy. Many malignant elbow tumors are metastases, though this is not a common site for them. Primary cancers that frequently metastasize to bone include melanoma and lung, breast, and kidney cancers.

MRI is not generally indicated for evaluating a known elbow fracture. Unsuspected fracture or contusion, however, may be demonstrated in patients with elbow pain following trauma. Osteochondritis dissecans of the elbow usually affects the capitellum. This condition, which often occurs in adolescents and may be asymptomatic in its early stages, can be related to repetitive trauma.

 Osteochondritis dissecans primarily affects subchondral bone. The underlying articular cartilage surface is initially preserved, but necrosis and softening of subchondral bone secondary to fracture may cause the articular cartilage to separate. This injury eventually leads to osteoarthritis in more than 50% of patients.

Osteochondritis dissecans can be mistaken on MRI for a pseudodefect of the capitellum that is related to the normal anatomy of radiocapitellar articulation. It should also be distinguished from the osteochondrosis known as Panner's disease, which occurs in boys aged four to 10 years old. Imaging findings of Panner's disease are similar to those of avascular necrosis of the femoral head.

 Osteochondral lesions appear as well-defined focal areas of abnormal marrow signal intensity in the subchondral bone, particularly in the capitellum.

Osteonecrosis is relatively rare in the elbow, but it can occur, most often in the capitellum. The condition is caused by repetitive stress and/or microtrauma in children. Additional triggers in adult patients include immunocompromise and use of corticosteroids, alcohol, and tobacco.

 Adults can acquire stress fractures of the olecranon, which are usually associated with throwing activities such as pitching. MRI is highly accurate in diagnosing stress fractures that are occult or equivocal on x-ray.

The elbow is stabilized by the medial (ulnar) and lateral (radial) collateral ligaments. The medial collateral ligament, extending from the medial humeral epicondyle to the coronoid process of the ulna, is the more important of the two. It has three portions, the largest and most important being the anterior band. The lateral collateral ligament is a less robust structure that extends from the lateral epicondyle to the annular ligament.

Clinical evaluation of collateral ligament injuries involving pain and subjective instability can be difficult. Although radiography does not show ligaments directly, x-rays should be obtained to rule out soft-tissue and bone abnormalities in the joint space. Partial or complete ligament tears can be demonstrated easily on MRI. Collateral ligament injuries will show ligament thickening or thinning, high-signal edema and/or hemorrhage around the ligament, ligament waviness or laxity, and ligament discontinuity.

Medial collateral ligament lesions often occur with valgus stress in throwing sports. Acute tears to this ligament can occur together with elbow dislocations due to joint hyperextension. Lateral collateral ligament injuries are rare, but they may occur in association with elbow dislocations or fractures. MRI can visualize associated capsular and annular ligament tears (Figure 2).

The major nerves (ulnar, radial, and median) are each subject to various entrapment and compression neuropathies specific to their particular anatomic locations and dynamic forces at the elbow.

 Prior to MRI, physicians relied principally on clinical findings and electromyography to assess neurologic disease. Now MRI can determine the exact location of compression, visualize loose bodies, and exclude soft-tissue masses such as ganglion, lipoma, hematoma, inflammatory pannus, osteophytes, ectopic calcification, and ossification.

The ulnar nerve is injured more frequently than the other two, due to its position within the fibro-osseous tunnel posterior to the medial humeral epicondyle. MRI will show an entrapped ulnar nerve to be thickened within and above the tunnel, tapering more distally.

 An absent arcuate ligament allows subluxation of the ulnar nerve in some cases. Resulting friction against the medial epicondyle can lead to neuritis. The instability, which usually occurs during arm flexion, may also be asymptomatic.

The radial nerve and its branches are at risk from direct trauma to the elbow or pressure from a space-occupying mass such as a ganglion or lipoma. The median nerve is also at risk from external or internal compression (Figure 3), most often from a soft-tissue mass.

T2-weighted images are useful for assessing edema and signal intensity changes within and around the nerves and for accentuating soft-tissue masses that have displaced them.

Changes to muscles affected by neuropathy can be seen on MRI as well. Muscular edema is the most significant sign of neuropathy, though it is not specific. Edema has also been associated with intense muscular exertion or intramuscular hemorrhage, tumors, polymyositis, acute inflammation or infection, effects of ionizing radiation, and rhabdomyolysis. Muscle atrophy can be observed as muscular edema in the acute/subacute stage of neuropathy. A decrease in muscle diameter is seen from day 21 onwards following denervation. Irreversible fatty changes are seen in patients with chronic neuropathy that has lasted more than six months.

 MRI can additionally depict changes to both the common flexor and common extensor tendons. The spectrum of pathologic changes is the same regardless of whether the medial or lateral side is involved, the latter (tennis elbow) being more common. The appearances of lateral and medial epicondylitis on MRI are similar.

Tendinosis is a degenerative process. Its least severe manifestation is demonstrated as normal to increased tendon thickness with intermediate signal intensity on T1-weighted MRI. Ligaments are seen with either intermediate to low signal intensity, resulting from fibroplastic proliferation, or high signal intensity, owing to fibrovascular proliferation and mucoid degeneration.

 Partial tears are seen on T2-weighted MRI as tendon thinning with high signal intensity. A complete tear is usually caused by a tendon failing within a region of mucoid degeneration. This is seen as tendon disruption and discontinuity, and it is best appreciated on T2-weighted images.

Injuries to the elbow's muscles are common, and these can be detected and characterized on MRI.

Distal biceps tendon has been reported with increased frequency since the introduction of MRI. Injury is typically due to an acute traumatic event with forced elbow hyperextension in midflexion. This results in complete avulsion of the tendon from its insertion in the radial tuberosity (Figure 4). Partial tendon tears and myotendinous junction tears are much less common.

MRI can help determine the location and degree of tendon injury as well as the amount of tendon retraction. Sagittal images are optimal for determining the size of the gap. An axial image should be obtained from the myotendinous junction through the radial tuberosity. Radial bicipital bursitis sometimes appears on T2-weighted MRI as a high-signal fluid collection interposed between the distal biceps tendon and radial tuberosity. Rupture and partial tear of the triceps tendon are rare.

Synovial proliferation may accompany a variety of disease processes, including rheumatoid arthritis, septic arthritis, neuropathic osteoarthropathy, crystal deposition disorders, pigmented villonodular synovitis, and idiopathic synovial osteochondromatosis (Figure 5). Effusion is common with all of these diseases. It is recognized easily on T2-weighted spin-echo MRI as fluid with high signal intensity.

Rheumatoid arthritis and other synovial inflammatory disorders lead to proliferation of the synovial membrane and fluid accumulation. Capsular distention or rupture and synovial cyst formation may also be seen. Altered dynamics at the proximal radioulnar joint and elbow may contribute to the production of cysts, and these may be an early sign of disease. Signal intensity characteristics of the abnormal synovium and fluid are similar on standard MRI sequences. Inflammatory tissue enhances if gadolinium contrast is administered.

Idiopathic synovial osteochondromatosis, related to metaplasia of the synovial lining, is accompanied by synovial proliferation and formation of intrasynovial cartilage nodules. These may calcify, ossify, or become free within the joint cavity and later embed in a distant synovial site.

 Ganglion cysts and bursitis (olecranon and biceps tendon bursitis) are common around the elbow.

Intra-articular osteocartilaginous loose bodies are usually found in association with osteochondritis dissecans, osteochondral fractures, and synovial osteochondromatosis. Trauma is the most common reason for the elbow's articular surface to disintegrate. Free bodies tend to migrate to the olecranon and coronoid fossa.

Soft-tissue tumors are not common in the elbow. Those that do occur can affect a large number of the joint's significant structures.

 Soft-tissue masses are difficult to evaluate on plain-film x-ray, making MRI an essential tool for assessing these pathologies. Multiplanar imaging enables unequaled soft-tissue discrimination and precise definition of local tumor extent.

Although MRI is often nonspecific for tumors, the modality can characterize some soft-tissue lesions. These include common benign lipoma (Figure 6), and vascular hemangioma. The value of MRI in differentiating benign and malignant soft-tissue tumors nonetheless remains controversial.

The prognosis and treatment of malignant tumors is influenced by their location and extent as well as their histological grade. These criteria form the basis of musculoskeletal tumor staging. MRI should be performed prior to biopsy, as postbiopsy images may be affected by anatomic distortion and heterogeneous signal intensity caused by hemorrhage, inflammation, and edema.

 MRI is used after surgery, radiation therapy, and chemotherapy, however, to check for residual tumor, disease recurrence, and healing. Differentiation between recurrent tumor and postoperative changes can be difficult, especially because obliteration of soft-tissue planes, scarring, hemorrhage, or infection may occur postoperatively.

In addition to its use in evaluating injury, disease, and infection of the elbow, MRI can help clarify diagnostic doubts raised by x-ray and CT. This could include evaluation of an implant site and/or checking for osteointegration in patients with nonmetallic prostheses (Figure 7) or checking whether a resorbable screw remained in the elbow.

DR. MELLONI is senior radiologist, and DR. VALLS is head of musculoskeletal radiology, both at Unitat de Diagnostic d'Alta Tecnologia, Corporacio Sanitaria Parc Tauli, Universitat Autonoma de Barcelona, in Sabadell, Spain.

1. Schenk M, Dalinka MK. Imaging of the elbow. An update. Orthop Clin North Am 1997;28(4):517-535.

2. Green WT, Mita MA. Congenital radio-ulnar synostosis: surgical treatment. J Bone Joint Surg Am 1979;61(5):738-743.

3. Butters KP, Morrey BF. Septic arthritis. In: Morrey BF, ed. The elbow and its disorders, 3rd ed. Philadelphia: WB Saunders, 2000:Part VIII.

4. Fitzgerald RH, Kelly PJ. Infections of the skeletal system. In: Howard RJ, Simmons RL, eds. Surgical infectious diseases, 3rd ed. London: Appleton & Lange, 1995:1207-1236.

5. Beltran J, Noto AM, McGhee RB, et al. Infections of the musculoskeletal system: high-field-strength MR imaging. Radiology 1987;164(2):449-454.

6. Dahlin DC, Unni KK. Bone tumors: general aspects and data on 8,542 cases, 4th ed. Springfield, IL: Thomas, 1986.

7. Pappas AM. Osteochondrosis dissecans. Clin Orthop 1981;158:59-69.

8. Shaughnessy WJ, Bianco AJ. Osteochondritis dissecans. In: Morrey BF, ed. The elbow and its disorders, 3rd ed. Philadelphia: WB Saunders, 2000:Part IV.

9. Le TB, Mont MA, Jones LC, et al. Atraumatic osteonecrosis of the adult elbow. Clin Orthop 2000;373:141-145.

10. Sonin AH, Tutton SM, Fitzgerald SW, Peduto AJ. MR imaging of the adult elbow. Radiographics 1996;16(6):1323-1336.

11. Jobe FW, Fanton GS. Nerve injuries. In: Morrey BF, ed. The elbow and its disorders, 3rd ed. Philadelphia: WB Saunders, 2000:Part VIII.

12. Sallomi D, Janzen DL, Munk PL, et al. Muscle denervation patterns in upper limb nerve injuries: MR imaging findings and anatomic basis. AJR 1998;171(3):779-784.

13. Fritz RC, Steinbach LS, Tirman PFJ, et al. MR imaging of the elbow. An update. Radiol Clin Am 1997;35(1):117-144.

Zionts LE, Vachon LA. Demonstration of avulsion of the triceps tendon in an adolescent by magnetic resonance imaging. Am J Orthop 1997;26(7):489-490.

15. Resnick D, Kang HS. Internal derangements of joints. Emphasis on MR imaging. Philadelphia: WB Saunders, 1997;334-386.

16. Schweitzer M, Morrison WB. Arthropathies and inflammatory conditions of the elbow. Magn Reson Imag Clin North Am 1997;5(3):603-617.

Articles

The elbow is a complex joint. It links the shoulder and the hand, enhances the flexibility of hand motion, and transmits generated forces. The most common elbow injuries are related to chronic overuse, although the joint is vulnerable to acute trauma as well.1 It may also be affected by disease or infection.

MR imaging illuminates elbow joint disorders

Hip replacement procedures are becoming in-creasingly common as the population ages. Degen-erative problems remain the main indication for this procedure, although patients with femoral head avascular necrosis, rheumatoid arthritis, complications of fracture, and hip dysplasia may also require hip replacement surgery.

The first hip replacements were performed about 50 years ago. As surgical techniques subsequently evolved, radiological criteria for evaluating prosthesis components and their complications adapted accordingly.

Our hospital has performed more than 200 hip replacements annually over the past 14 years. Surgeons have used a variety of materials to replace the hip joint, including metal, polyethylene, and ceramics. All patients undergo imaging examinations immediately after surgery, at six months, and then yearly or when necessary. This follow-up is most frequently done using plain-film x-rays, occasionally with ultrasound or CT, and, rarely, with MRI.

A retrospective review of 1854 patients (36 to 91 years old; male/female ratio 1:3) who underwent hip arthroplasty at our hospital in an eight-year period revealed that 5.2% experienced acute complications, and 9.2% developed complications later.

The acute complications included infection, cement extrusion, periprosthetic fracture, and immediate dislocation. Late complications, some of which occurred as long as 15 years after surgery, have included aseptic loosening, aggressive granulomatosis, stress shielding, heterotopic ossification, luxation, bone fractures, and prosthetic fracture.

Hip replacements can be cemented at the acetabular and/or femoral stem. Polymethylmethacrylate cement, used for filling the space between the bone and the prosthesis, provides the necessary fixation and force distribution. Cement-free prostheses are coated to provide a porous surface that enables fixation by bone ingrowth. Acetabular fixation may be completed with spikes or screws.

Our standard follow-up imaging evaluation of hip arthroplasty begins with routine plain-film radiography. It is important that the entire prosthesis is included on two orthogonal x-rays. Leg length, acetabular inclination, and varus or valgus stem position should be evaluated and compared with the immediately postoperative study.

Radiography of cemented prostheses will generally show a thin lucent line, less than 2 mm wide, along the cement-bone interface. This is a normal finding that represents fibrous tissue and is stable over time. The line is delimited from adjacent bone by a thin line of sclerosis, the demarcation line.

Lucency of 2 mm or more at the bone-ingrowth surface with cement-free arthroplasty is considered abnormal. Lucency that is less than 2 mm in width should be followed on consecutive plain films for progression. Well-defined areas of lucency around the prosthesis usually indicate bone resorption destruction (histiocytic response), but aseptic loosening and infection can demonstrate a similar appearance.

Infection, which occurs in 1% of artificial joint replacements, can be a serious early complication, or it may appear months or even years after surgery. Radiological findings in patients with indolent infection can be unremarkable or show minimal changes,

 and a normal-appearing radiograph does not exclude infection.

Sinus tract formation or bone destruction may occur with more aggressive infectious agents. Infection may also be present when postoperative imaging shows lytic lesions, progressive interfacial widening, periosteal reaction, or smooth endosteal scalloping with cement lucency, as is typical in aseptic loosening or aggressive granulomatosis disease.

Extensive bone destruction, air in the soft tissue and/or joint, extensive or aggressive periosteal reaction, and a wide or irregular lucent zone are the most suspicious radiological signs of infection. CT or ultrasound-guided joint aspiration or synovial biopsy can detect about 75% to 95% of prosthetic infections.

Polymethylmetacrylate cement ex-trusion is another acute complication. It occurs most commonly through a defect in the medial acetabular wall. It is visible on initial radiographs. Intra-pelvic cement is usually asymptomatic.

Intraoperative or acute periprosthetic fracture is a rare complication in primary cemented prostheses, although it is relatively common in the follow-up of failed hip arthroplasty. It is usually found at the tip of the prosthetic stem; the pelvis is a less common location. This type of fracture is usually not displaced and is easily overlooked.

An artificial hip can generally be expected to last 12 to 15 years. In some cases, however, the hip prosthesis will loosen earlier, causing pain and/or instability. Mechanical loosening is the most typical indication for repeat surgery. Although precise definitions of loosening vary, a hip joint is probably loose if imaging shows a radiolucent zone greater than 2 mm wide.

 This zone may involve the prosthesis/bone interface, the cement/bone interface, or the cement/prosthesis interface, depending on the type of arthroplasty.

Definite loosening is diagnosed when a cemented component migrates (intra- or extramedullar) (Figure 1), changes alignment (varus or valgus; especially increasing varus orientation), or shows progressive widening of the radiolucent zone. Comparison between current and previous images is the best way to detect loosening.

Aggressive granulomatosis, also known as particle or cement disease (Figure 2), is caused by macrophage reaction on either the acetabular or femoral side of the arthroplasty. Lesions of aggressive granulomatosis present as extensive areas of radiolucency around the prosthesis, sometimes in association with smooth endosteal scalloping. Infection, which often has more aggressive features, is the main differential diagnosis.

 The prosthesis may remain stable even when large areas of osteolysis are present.

The presence of a prosthesis alters stress loading on a bone.

 Prostheses can induce increased bone density at points of contact (called stress shielding) and reduced bone mass and osteoporosis in areas of decreased loading. Localized bone loss typically occurs in the greater trochanter. This phenomenon has not been associated with a predisposition to loosening or prosthesis failure.

Heterotopic calcification is relatively common, found in 15% to 50% of cases, and is usually asymptomatic. It can appear a few weeks after surgery and may take two years to mature.

 The supratrochanteral area between the pelvis and femur is the most common site of heterotopic calcifications. Predisposing factors include infection, ankylosing spondylitis, and previous surgery. Low-dose radiotherapy or indomethacin can prevent its occurrence.

Dislocation or luxation of hip arthroplasties may be caused by a patient's poor muscle tone, subsequent trauma, positional factors such as the patient placing the hip beyond the expected range of motion, or component malpositioning. These complications are more common in revision arthroplasties (patients who have received a replacement implant).

A dislocating hip should be evaluated for acetabular component malpositioning (retroversion, abnormal lateral opening, or abnormal horizontal center of rotation), and ab-normal leg length due to a misplaced implant. Again, comparison with previous radiographs is helpful.

Bone fractures usually occur in patients with osteoporosis. The most common location is the tip of the stem. Patients may also present with insufficiency fractures due to increased physical activity. Prosthetic component fracture is rare but can occur in cases of severe trauma or metal fatigue. The acetabular component almost never fractures. The tab used to lock the acetabular component may be seen, but this does not represent a significant component failure. Fracture of the femoral component, also rare, is usually located in the distal portion of the stem.

Imaging evaluation of painful total joint arthroplasty has traditionally been performed using plain-film radiography, arthrography, and nuclear scintigraphy.

 All three diagnostic options provide clinical information on the status of the prosthesis, adjacent bone, and surrounding soft tissues, but all are hampered by poor specificity. Plain-film x-rays remain the cornerstone of diagnosis of hip ar-throplasty complications, but ultrasound, CT, and MRI can play a complementary role in imaging specific problems that might arise.

An ultrasound examination performed shortly after surgery will show fluid collection corresponding to a hematoma and its delayed resorption. Hematoma of the joint is usually organized within three weeks after the operation. It can be visualized in all patients as a homogeneous region with a slightly echogenic appearance. It is not clearly distinguishable from surrounding tissue and is not compressible. Areas of low echogenicity are rarely seen. It is not possible to distinguish hematoma from infection on ultrasound; percutaneous aspiration is necessary if clinical suspicion of infection is high (Figure 3).

CT is usually used to confirm prosthesis osteointegration (Figure 4A). We use different CT slices of the total hip prosthesis to analyze prosthesis osteointegration in the femur. CT can also help confirm radiologic suspicions of intraosseous lesions that are difficult to appreciate on plain-film x-ray (Figure 4B).

 Artifacts from metallic prostheses can be significantly minimized by using a multislice scanner and evaluating surrounding soft-tissue abnormalities immediately adjacent to the prosthesis. Plain-film x-ray nonetheless remains the primary method of evaluating the prosthesis and adjacent osseous structures.

MRI with optimized pulse se-quences to reduce artifacts generated by the arthroplasty can provide clinically relevant information on the prosthesis and surrounding bone, soft tissue, and neurovascular structures, particularly around the femoral prosthetic stem (Figure 5).

Diagnostic-quality MRI of hip arthroplasty complications requires only simple modifications to standard MR sequences.

 MR image quality may improve with the advent of new materials for prostheses and alternative pulse sequences that avoid artifacts.

Careful attention to initial prosthesis placement and comparison of follow-up x-ray images enables detection of subtle abnormalities in patients with painful hip arthroplasties. Ultrasound, CT, and MRI each furnish clinical information on the status of the prosthesis, adjacent bone, and surrounding soft tissue to complement plain-film radiographs.

Ultrasound may have a special role in the evaluation of soft-tissue ab-scesses or fluid collections, and it is useful for guiding percutaneous aspiration of soft-tissue collections. The main advantage of ultrasound over CT and MRI is the lack of artifacts caused by metallic prostheses. But the modality is unable to evaluate deep soft-tissue structures farther away from prostheses. CT can best assess the osteointegration of hip prostheses, while MRI may help identify hip arthroplasty complications.

1. Weissman BN. Imaging of total hip replacement. Radiology 1997;202(3):611-623.

2. Manaster BJ. From the RSNA refresher courses. Total hip arthroplasty: radiographic evaluation. Radiographics 1996;16(3): 645-660.

3. Gibbon WW, Long G, Barron DA, O'Connor PJ. Complications of orthopedic implants: sonographic evaluation. J Clin Ultrasound 2002;30(5):288-299.

4. Eisler T, Svensson O, Engstrom CF, et al. Ultrasound for diagnosis of infection in revision total hip arthroplasty. J Arthroplasty 2001;16(8):1010-1017.

5. Tigges S, Stiles RG, Roberson JR. Appearance of septic hip prostheses on plain radiographs. AJR 1994;163(2):377-380.

6. Cyteval C, Hamm V, Sarrabere MP, et al. Painful infection at the site of hip prosthesis: CT imaging. Radiology 2002;224(2): 477-483.

7. Manaster BJ. Total knee arthroplasty: postoperative radiologic findings. AJR 1995;165:899-904.

8. Tigges S, Stiles RG, Roberson JR. Complications of hip arthroplasty causing periprosthetic radiolucency on plain radiographs. AJR 1994;162(6):1387-1391.

9. Joshi MG, Advani SG, Miller F, Santare MH. Analysis of a femoral hip prosthesis designed to reduce stress shielding. J Biomech 2000;33(12):1655-1662.

10. Temmerman OP, Raijmakers PG, David EF, et al. A comparison of radiographic and scintigraphic techniques to assess aseptic loosening of the acetabular component in a total hip replacement. J Bone Joint Surg Am 2004;86-A(11):2456-2463.

11. Park JS, Ryu KN, Hong HP, et al. Focal osteolysis in total hip replacement: CT findings. Skel Radiol 2004;33(11):632-640.

12. White LM, Kim JK, Mehta M, et al. Complications of total hip arthroplasty: MR imaging-initial experience. Radiology 2000;215(1):254-262.

13. 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 pathologic findings. Radiology 2003;229(3):718-723.

14. Sofka CM, Potter HG. MR imaging of joint arthroplasty. Semin Musculoskelet Radiol 2002;6(1):79-85. 

Pietro Melloni, MD, PhD

Articles

MR imaging illuminates elbow joint disorders

X-ray plays vital role in hip replacement surgery