Semin Musculoskelet Radiol 2018; 22(S 01): S1-S5
DOI: 10.1055/s-0038-1639537
Scientific Presentations and Posters
Georg Thieme Verlag KG Stuttgart · New York

MR Imaging with Metal Artifact Reduction to Differentiate between Patients with and without Infected Total Hip Arthroplasty

Benedikt J. Schwaiger
1   Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Alexandra S. Gersing
1   Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Daniela Muenzel
1   Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Julia Dangelmaier
1   Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Peter M. Prodinger
2   Department of Orthopedic Surgery, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Christian Suren
2   Department of Orthopedic Surgery, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Ernst J. Rummeny
1   Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Germany
,
Klaus Woertler
1   Department of Diagnostic and Interventional Radiology, Klinikum rechts der Isar, Technical University of Munich, Germany
› Author Affiliations
Further Information

Publication History

Publication Date:
26 March 2018 (online)

 

Purpose: To evaluate imaging parameters obtained from magnetic resonance imaging (MRI) with metal artifact reduction to identify patients with infected total hip arthroplasty (THA).

Materials and Methods: In 33 patients (66 ± 10 years; 20 women) with THA, 1.5-T MRI was acquired with view angle tilting and slice-encoding metal artifact correction. Parameters describing changes of metal–bone interface, bone, and soft tissue were assessed. At revision surgery, signs of infection and loosening were evaluated using the criteria of the American Academy of Orthopaedic Surgeons as the standard of reference.

Results: At surgery, prosthetic infection was confirmed in 12, aseptic implant loosening in 15, and conditions unrelated to infection/loosening in 6 patients. The following findings were significantly more common in patients with infection compared with those with loosening: contrast enhancement at the metal–bone interface (sensitivity, 100%; specificity, 43%; p = 0.004), periostitis (100%, 43%; p = 0.004), signal alterations (86%, 77%; p < 0.001), and contrast enhancement in the surrounding soft tissue (82%, 71%; p = 0.011), abscess (63%, 100%; p = 0.004), frondlike synovial hypertrophy (67%, 83%; p = 0.013), and both acetabular and femoral components affected (86%, 100%; p < 0.001). Larger areas were affected in prosthetic infection (p = 0.012), and inguinal/pelvic lymph nodes were larger (receiver operating characteristic-area under the curve, 0.905; p = 0.001; optimal cut-off, 17 mm). To differentiate patients with infection/ aseptic loosening from those with other conditions, all assessed findings were highly significant (p < 0.005 for all).

Conclusion: Identification of patients with infected THA may be possible with parameters derived from MR with metal artifact reduction. The most sensitive parameters were contrast enhancement at the metal–bone interface, periostitis, and enlarged lymph nodes. The most specific finding was the presence of an abscess.