Scintigraphy with indium-labeled white blood cells has been reported to be sensitive and specific in the diagnosis of low-grade sepsis of the musculoskeletal system. We reviewed the records of fifty patients who had suspected osteomyelitis or suspected infection about a total joint prosthesis and who underwent scintigraphy with technetium-99m methylene diphosphonate and scintigraphy with indium-111 oxine-labeled white blood cells before an open surgical procedure. Any patient who received preoperative antibiotics was not included in the study. For all of the patients, gram-stain examination of smears, evaluation of a culture of material from the operative site, and histological examination were done. The patients were divided into two groups. Group I was composed of twenty-four patients, each of whom had a prosthesis in place and complained of pain. Group II was composed of twenty-six patients for whom a diagnosis of chronic osteomyelitis had to be considered. With the indium scans alone, there was only one false-negative result (in Group II), but there were eighteen false-positive results (eight patients in Group II and ten patients in Group I). Although scintigraphy with indium-labeled white blood cells is quite sensitive, it is not specific in detecting chronic osteomyelitis; a negative scan should be considered highly suggestive that osteomyelitis is not present. Specificity can be increased by interpreting the indium scan in conjunction with the technetium scan.