USEFULNESS OF SCANNING PROCEDURES FOR DIAGNOSIS OF FEVER OF UNKNOWN ORIGIN IN CHILDREN

During a 5-year study period, 109 patients were referred to a large children's hospital for evaluation of prolonged fever of unknown origin, defined as temperature greater-than-or-equal-to 38-degrees-C (100.4-degrees-F) for 3 weeks or longer and negative findings on initial examination. A two-phase protocol of outpatient followed by inpatient diagnostic studies was instituted for most patients. Confirmed diagnoses were achieved in just 36 of these children (33%) in the following disease categories: infectious, 24 (22%); autoimmune, 7 (6%); and neoplastic, 2 (2%). Scanning or special procedures and the number with positive results (in parentheses) were as follows: abdominal ultrasonography, 43 (8); abdominal computed tomography, 14 (3); indium scan 11 (5); gallium scanning, 4 (1), upper gastrointestinal tract series, 13 (2); technetium bone scanning 15 (2); bone marrow examination, 16 (1); and cranial computed tomography, 7 (0). These studies rarely led to an unsuspected diagnosis. It appears most appropriate in evaluating fever of unknown origin in children to obtain only basic laboratory studies such as a complete blood cell count, urinalysis and culture, chest radiograph, tuberculin skin test, and, in the older child, an antinuclear antibody titer. When these test results are negative, almost all children can be observed clinically for progression of illness or a focus that might then direct specific diagnostic procedures.