Josephine N. Rini MD, Presenter: Nothing to Disclose

Kuldeep K Bhargava PhD, Abstract Co-Author: Nothing to Disclose

Gene G Tronco MD, Abstract Co-Author: Nothing to Disclose

Hugh L Richardson DPM, Abstract Co-Author: Nothing to Disclose

Russell Caprioli DPM, Abstract Co-Author: Nothing to Disclose

Christopher J. Palestro MD, Abstract Co-Author: Nothing to Disclose

This investigation compares PET imaging of F-18 FDG in vitro labeled leukocytes (FDG WBC) to planar imaging of In-111 labeled leukocytes (In WBC) in patients with suspected orthopedic infections who had surgical/histopathologic confirmation of their final diagnoses.

Thirteen patients, with possible pedal/digital osteomyelitis (8), prosthetic joint infection (2), orthopedic hardware infection (2), and septic arthritis (1), were included. FDG WBC and In WBC were performed within 24 hours in 12 patients and within 4 days in 1 patient. In vitro labeling of autologous human leukocytes with F-18 FDG and In-111 oxine was performed according to published methods. For both FDG WBC and In WBC, tagging efficiencies and cell viability using tryphan blue exclusion technique, were determined. Imaging was performed 3.5-4.5 hours after injection of 5.3-8.4 mCi FDG WBC and approximately 24 hours after injection of 505-593 uCi In WBC. A single reader prospectively interpreted all studies as positive or negative for infection. Imaging results were correlated with surgical/histopathologic results.

Tagging efficiencies were 55-85% (mean 71%) for FDG WBC and 88-94% (mean 90%) for In WBC. Cell viability was 96%-99% (mean 98%) for both FDG WBC and In WBC. FDG WBC results were correct in 10/13 patients. In WBC results were correct in 8/13 patients. FDG WBC and In WBC scans were concordant and correct in 7 patients: 5 true positive (4 pedal osteomyelitis; 1 infected orthopedic hardware) and 2 true negative. Studies were concordant and false negative in 1 patient with septic arthritis. Studies were concordant and false positive in 1 patient with a bone biopsy that was negative for pedal osteomyelitis. Studies were discordant in 4 patients: 1 false positive FDG WBC (fracture); 2 false negative In WBC (pedal and digital osteomyelitis); 1 false positive In WBC (granulating wound). Sensitivity, specificity, and accuracy were: FDG WBC 0.88, 0.60, and 0.77, respectively; and In WBC 0.63, 0.60, and 0.62, respectively.

These data suggest that FDG WBC may be superior to In WBC for the rapid detection of orthopedic infections. These results, however, must be confirmed in a larger series.

Rini, J, Bhargava, K, Tronco, G, Richardson, H, Caprioli, R, Palestro, C, F-18 FDG in Vitro Labeled Autologous Hyman Leukocytes versus In-111 Labeled Leukocytes for Evaluation of Suspected Orthopedic Infections.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4416750.html