Search

This study was undertaken to determine the effect of chronicity of infection on the sensitivity of In-111-labeled leukocyte scanning. A total of 332 scans on 290 patients were reviewed retrospectively. The diagnosis of infection was based on culture results and other laboratory data, autopsy findings, radiographic studies, and clinical course. Duration of infection at the time of scanning was determined by the date of onset of fever and symptoms, elevation of white cell count, positive cultures, and abnormal findings on radiographic studies. Sensitivity of leukocyte scanning was 90% for 69 patients who had infections for 0-14 days and 86% for 86 who had infections for 15 days or longer. This difference is not statistically significant.

To determine the frequency and clinical significance of indium-111 labeled leukocyte activity in the gastrointestinal (GI) tract of patients with fever of unknown origin, we reviewed 312 leukocyte studies involving 271 patients. Radionuclide activity was noted in the bowel in 59 cases. Of these, only 27 were due to the infection or inflammatory disease that caused the patient's fever. The 32 false-positive results were due primarily to swallowed leukocytes or bleeding. In two cases, no explanation was found for the activity in the GI tract. We conclude that bowel activity on In-111 labeled leukocyte scans in patients with fever of unknown origin often does not correlate with the true cause of the patient's fever.

Imaging with leukocytes labeled with indium-111 oxine is a sensitive technique for detecting sites of occult infection. Traditionally, imaging is performed 24 hr after injection. We undertook a prospective study of 35 patients (40 studies) with possible occult infection to see whether a 24-hr delay in imaging is really necessary. Patients were imaged at 1-4 hr and again at 24 hr after injection. The early images had a sensitivity of only 33%, compared with 95% for the 24-hr images. Of the seven studies that were positive on both early and delayed images, 71% had more intense uptake at 24 hr. There were no false-positive early images. We conclude that imaging 1-4 hr after injection with In-111 oxine-labeled leukocytes has a low sensitivity for detecting occult infection. However, a positive early image is specific for a site of infection.

Although 111In-labeled leukocytes have been shown to be a useful technique for detecting infection, it has been postulated that antibiotic therapy may reduce the sensitivity of the leukocyte scan. Many patients with suspected bacterial infections are placed on antibiotics before a definite site of infection has been identified. Three hundred twelve leukocyte scans on 271 patients were retrospectively reviewed and classified as positive or negative, and as to whether or not they were being treated with antibiotics at the time the leukocyte scan was performed. The overall sensitivity, considering all 312 studies, was 90%. One hundred sixty-nine patient studies were on patients receiving antibiotics; 143 studies were on patients not on antibiotics. The sensitivity of the leukocyte scan was 88.7% in patients on antibiotic therapy; it was 92.1% in those who were not receiving antibiotics. The differences in sensitivity between the two groups were not significantly different (p less than 0.05). We conclude that antibiotic therapy does not affect the sensitivity of the 111In-labeled leukocyte scan.

Synthetic vascular graft infection is characterized by late diagnosis due to indolent and nonspecific symptoms. Indium-111-labeled leukocyte imaging holds promise as a diagnostic tool to identify vascular graft infection, but reported data on its accuracy are somewhat sparse and conflicting. In this study, 13 mongrel dogs received Dacron aortic interposition grafts. Seven grafts were contaminated at the time of surgery by topical ATCC Staphylococcus aureus concentrated at 10(8) organisms/ml. Six control animals received no graft contamination. All infected animals were sacrificed on the second postoperative day after 111In leukocyte scanning. The results showed a sensitivity of 71%, specificity of 100%, and accuracy of 85% for the 111In leukocyte study in detecting early graft infections; false-positive leukocyte scans in the early postoperative period were not a problem as has been reported by others. These data indicate that leukocyte scanning for graft infection detection is likely to be clinically valuable.

Indium-111-labeled leukocyte scanning is an excellent technique for detecting abdominal abscesses. However, since labeled leukocytes are normally taken up by the liver and spleen, diagnosing upper abdominal abscesses can be difficult. A prospective study was undertaken to see if having a technetium-99m sulfur colloid liver-spleen scan to compare with the 111In leukocyte scan would improve the reader's ability to diagnose or exclude upper abdominal abscesses. Ninety patients with suspected upper abdominal abscesses were examined with 111In leukocyte scans followed immediately by conventional 99mTc sulfur colloid liver-spleen scans in the same projections. In 57% of patients with upper abdominal abscesses and 26% of all patients studied, the 99mTc sulfur colloid scan was essential to diagnose or exclude an upper abdominal abscess. The liver-spleen scan was considered helpful in another 13%. The addition of a 99mTc sulfur colloid liver-spleen scan to the 111In leukocyte study is useful when evaluating patients with suspected upper abdominal abscesses.

A retrospective review was undertaken to evaluate the frequency and significance of pulmonary activity noted on 306 indium-111 leukocyte studies involving 232 patients with suspected occult infections. Forty-eight studies showed pulmonary activity in one of two patterns of uptake, focal or diffuse. Fourteen of 27 studies (52%) with focal uptake and two of 21 studies (10%) with diffuse uptake were associated with infectious processes. Lung uptake of indium-111-labeled leukocytes was a poor predictor of pulmonary infection in patients studied for occult infection, although the focal pattern was more likely than the diffuse pattern to be associated with infection.

We have noted that the presence of left ventricular anterior/apical aneurysm on contrast ventriculography or echocardiography correlates well with the finding of failure of convergence of the left ventricular walls toward the apex on single photon emission computed tomography (SPECT) thallium-201 images. To assess this observation, we analyzed the SPECT thallium scans of 74 sequential patients who had SPECT thallium scintigraphy and contrast ventriculography and/or echocardiography for evaluation of coronary artery disease. Immediate post-stress and 4-hr redistribution thallium-201 SPECT images, obtained following i.v. administration of 2 mCi of thallium-201, were reconstructed with no attenuation correction using three-dimensional linear and nonlinear filters and displayed in short, vertical-long, and horizontal-long axes. Of the 74 patients, contrast ventriculography and/or echocardiography showed anterior/apical aneurysms in 15 and a posterolateral aneurysm in one. SPECT thallium scans detected 14 of the 16 aneurysms, based on the criterion of failure of convergence of the ventricular walls toward the apex. There were two false-positives. Thus SPECT thallium-201 scintigraphy for the detection of left-ventricular aneurysm in this series had a sensitivity of 94%, a specificity of 97%, and an accuracy of 96%.

The 111In-labeled leukocyte scan is a sensitive and specific technique for the detection and localization of abscesses. However, in reviewing an unselected series of 249 scans, six (2.3%) were false-positive cases in which leukocyte uptake by tumor mimicked an abscess. This represented 12% (6/51) with known tumor at the time of imaging. Five of the cases were primary or metastatic tumors to the soft tissues; the sixth was a skeletal metastasis. The intensity of tumor activity has been characterized as mild in the few cases of leukocyte uptake reported in the literature, suggesting that the degree of uptake is helpful in distinguishing tumor from abscess. In this study, however, variable degrees of tumor-associated activity were seen ranging from mild to very intense. These findings indicate that tumor accumulates labeled leukocytes more frequently than has been previously appreciated, that primary and metastatic neoplasms involving both the soft tissues and skeleton are involved, and that the intensity of uptake is not a reliable criterion to distinguish tumor from abscess.

Bone marrow aspiration and biopsy are excellent techniques for evaluating bone marrow, but this evaluation is limited to a small part of the total blood-forming organ. With the introduction of radionuclide bone marrow imaging, a simple technique became available that overcomes marrow sampling errors by giving a total body view of functioning marrow. Furthermore, the procedure is noninvasive and provides an atraumatic method for evaluating a number of clinical problems including a discrepancy between bone marrow histology and clinical status (possible marrow sampling error), the determination of amount of active marrow after radiation and chemotherapy when further therapy is being considered, detection of sites of extramedullary hematopoiesis, location of the optimal sites for bone marrow biopsy, the diagnosis and staging of diffuse hematologic disorders, detection of metastases, the diagnosis of bone marrow infarcts in hemolytic anemias, and detecting avascular necrosis of the femoral heads. There are two major classes of bone marrow agents: (1) those that are incorporated into the erythroid precursors such as radioiron and (2) colloids that are taken up by the reticuloendothelial system (RES). Indium-111 chloride was originally considered to be an erythropoietic agent but appears to share some properties of RES labels. The best label to use is dependent on the disease being evaluated.

On December 2, 1982, a permanent total artificial heart was implanted into the chest of a 61-year-old man with a progressive and irreversible cardiomyopathy. During the ensuing four-month hospitalization, a number of nuclear medicine procedures were obtained to assist in patient management. These procedures included gated cardiac radionuclide ventriculography, an I-123 iodoamphetamine scan for cerebral perfusion, and In-111 labeled leukocyte imaging. These radionuclide studies demonstrate the advantages of being able to monitor physiologic changes noninvasively and illustrate a potential role for nuclear cardiology in managing patients with a total artificial heart.

There is considerable disagreement as to whether oxine or tropolone is the best labeling agent for indium leukocytes. We have previously looked at the sensitivity of oxine-labeled 111In leukocyte scans for occult infections and now present a similar group of patients imaged with tropolone-labeled 111In leukocytes. Thirty-four patients (38 studies) with possible occult infection were prospectively studied. Patients were imaged 1-4 hr after injection and again at 24 hr postinjection. The early tropolone images had a sensitivity of 53% while the delayed images at 24 hr had a sensitivity of 93%. Based on a previous study, oxine-labeled leukocyte scans have an early sensitivity of 33% and a delayed sensitivity (at 24 hr) of 95%. The differences in sensitivity between oxine and tropolone when imaged early and at 24 hr were not statistically significant. We conclude that there is no significant difference in the ability to detect infection between oxine- and tropolone-labeled leukocytes, both early at 1-4 hr, and on delayed imaging 24 hr after injection.

Radiolabeled blood cells are widely used for both clinical and research studies. In vitro and in vivo studies have shown the tagging process does not significantly affect function. The labeling techniques and clinical uses of labeled leukocytes and platelets are reviewed.