A rapid antigen assay that could detect pneumococcal disease in febrile children would allow for timely treatment of occult bacteremia, pneumonia, and meningitis. A preliminary study of a rapid urine pneumococcal assay showed high rates of sensitivity and specificity. In this study, Neuman and Harper evaluated the performance of this assay in children at low and high risk of developing occult pneumococcal infection.

The authors enrolled five groups of children, three months to five years of age, who presented to the emergency department. The groups involved in the study included children with pneumococcal bacteremia, febrile children with focal consolidation on chest radiography, febrile nonbacteremic children with leukocytosis, febrile nonbacteremic children with a normal white blood cell count, and afebrile children with no evidence of current or recent bacterial infection. Children who met the above criteria by blood culture, chest radiography, and white blood cell count of 10,000 mm³ (10 × 10⁹ per L) or less or 20,000 mm³ (20 × 10⁹ per L) or greater, as well as afebrile children, provided urine specimens as a routine part of the emergency department evaluation or after categorization. The main outcome of this study was the diagnostic test characteristics of the pneumococcal antigen assay in the respective groups.

A total of 346 children were enrolled in the study and divided into their respective groups. Children with bacteremia and lobar pneumonia were the most likely to have positive antigen test results (96 and 76 percent, respectively). The control populations had positive antigen test results of 15 percent in the febrile groups and 8 percent in the afebrile groups. No statistically significant difference was noted among patients in the febrile, nonbacteremic, normal white blood cell groups and those in the febrile, nonbacteremic, leukocytosis groups.

The urine antigen assay in this study was highly sensitive in identifying children with pneumococcal bacteremia, as well as specific among children without fever. These results compare favorably with the measurement of white blood cell counts in detecting patients at risk for bacteremia. Although the false-positive rate was low in afebrile children, the usefulness of the test was limited by the finding that 15 percent of febrile children also tested positive using this assay. The reason for this rate is uncertain but might include disruption of mucosal barriers in patients with nasal colonization. Of note, this study was conducted before the widespread use of the conjugate pneumococcal vaccine. With a resulting overall decrease in pneumococcal disease, more false-positive assay results are to be expected.