SNM Annual Meeting Abstracts
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     



J Nucl Med. 2008; 49 (Supplement 1):10P
This Article
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Slotcavage, R.
Right arrow Articles by Alavi, A.
PubMed
Right arrow Articles by Slotcavage, R.
Right arrow Articles by Alavi, A.

General Clinical Specialties: Musculoskeletal

Musculoskeletal

Detection of infected knee arthroplasty using FDG-PET

R. Slotcavage1, M. Vadher1, J. Parvizi2, H. Zhuang1, D. Torigian1 and A. Alavi1

1 Division of Nuclear Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; 2 Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania


Formula

37

Objectives: A remaining challenge in orthopaedics is the non-invasive determination of infection in prosthetic joints. Recently, FDG-PET has shown promise to detect infection. This abstract serves to update our prior paper with new results using FDG-PET to non-invasively identify infection in patients with painful knee arthroplasty.

Methods: FDG-PET was performed on the affected knee(s) in 52 patients with painful prostheses after clinical evaluation and laboratory data were inconclusive for infection. Images were interpreted in blinded fashion, with infection identified based on increased FDG uptake at the bone-prosthesis interface versus adjacent tissue. Images then underwent a second blinded read by a single senior nuclear medicine radiologist to eliminate interobserver variability. Final clinical outcome was available for 31 patients (36 knees) of 52 enrolled, determined via surgical, laboratory, and clinical follow-up.

Results: Of 36 knee prostheses with full follow-up, FDG-PET correctly identified infection in 6 of 7 (sens=85.7%), with false positive (FP) results in 2 of 29 (spec=93.1%). PPV was 75.0% (6/8), NPV was 96.4% (27/28), and accuracy was 91.7% (33/36) when interpreted by staff radiologists. In the second blinded read, sensitivity remained 85.7% (6/7), but specificity increased to 100.0% (29/29), PPV to 100% (6/6), NPV to 96.7% (29/30), and accuracy increased to 97.2% (35/36).

Conclusions: These data suggest that FDG-PET shows further promise in the accurate detection of infected knee arthroplasty. Unlike our hip prosthesis data, interobserver variability appears extremely low when analyzing knee prostheses, allowing greater accessibility of this technique to those with less experience interpreting FDG-PET data.

Research Support: Grant from NIH





This Article
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Slotcavage, R.
Right arrow Articles by Alavi, A.
PubMed
Right arrow Articles by Slotcavage, R.
Right arrow Articles by Alavi, A.