Several chelates are available for leukocyte labeling. Studies indicate that cells labeled with any of the chelates have a sensitivity for infection of 90% to 95% when imaged at 24 hours postinjection. The sensitivity of 111In-labeled leukocytes at earlier imaging times is more controversial. There has been concern about the utility of labeled leukocytes in musculoskeletal infection. Recent leukocyte studies show a high sensitivity for infected prostheses, even though these infections are often walled off and do not cause systemic symptoms. However, leukocytes frequently miss osteomyelitis of the spine for reasons that are not known. Although some investigators do not recommend the use of 111In-labeled leukocytes in chronic infections, we have found a high sensitivity for infections that are 2 or more weeks old. Autopsy studies from the preantibiotic era indicate that bacterial infections with common organisms have high levels of neutrophil infiltration for months. Labeled lymphocytes from mixed-cell preparations also may play a role in detecting these inflammatory sites. Questions have been raised about the effect of antibiotic therapy on leukocyte sensitivity. Antibiotics do not appear to have a significant effect on scan sensitivity. By reducing the number of bacteria at an inflammatory site, antibiotics reduce the amount of chemotactic inhibitors. In addition, some antibiotics have been shown to directly stimulate leukocyte chemotaxis. Other factors that can theoretically reduce leukocyte function, including hemodialysis, hyperalimentation, hyperglycemia, and steroids, do not appear to reduce labeled leukocyte sensitivity for infection. The specificity of leukocyte uptake is reduced in the gastrointestinal tract and lungs. In these sites, uptake correlates with infection or the true cause of the patients' fever in only 10% to 50% of cases.