Technetium-99m-labeled anti-granulocyte antibodies in suspected bone infections

The introduction of 99mTc-labeled anti-granulocyte antibodies seemed to provide advantages in comparison with formerly used in vitro methods to label autologous white blood cells for inflammation imaging. For this reason, we have undertaken a study to evaluate the clinical significance of this method. Thirty unselected patients with suspected bone infections were studied prospectively using the monoclonal 99mTc-labeled anti-granulocyte antibody. Twenty patients were referred with suspected infections of the peripheral bones (Group I), as well as 10 patients with suspected infections of the spine (Group II). Planar whole-body scans were performed 4 hr and 20 to 24 hr after administration of 500 MBq of the labeled antibody. Scans were considered positive for a bacterial (septic) infection when a focally increased antibody accumulation occurred. All scans were evaluated in blinded fashion by two experienced readers. Of the 20 studies from Group I patients, four false-positive scintigraphic findings were observed, and one false-negative, resulting in a specificity of only 64% and a sensitivity of 89%. In Group II (10 studies), five scans were true-negative, and five false-negative. For both groups, the specificity of the scintigraphic method was quite low (75%), and the sensitivity was also relatively low (57%). The results of this study demonstrate that in an unselected patient population in whom the diagnosis is not known, scintigraphy with 99mTc-anti-granulocyte antibodies is not a reliable method for detecting septic inflammatory lesions: In addition, use of this method excludes septic lesions with only a moderate likelihood (83% negative predictive value).