Groups of children (mean age, 31.4 months) with Haemophilus influenzae type b
meningitis,
epiglottitis, or
septic arthritis were tested for the presence and levels of
bacteremia, capsular
polyribophosphate (PRP) antigenemia, and development of specific antibody in serum after the onset of acute illness. Although
bacteremia cleared promptly after
antibiotic therapy, circulating PRP could be detected in serum for relatively long periods, with 51% of the patients still having detectable
antigen after 30 days postinfection. Even in the presence of specific antibody, antigenemia persisted for as long as 47 days after admission. It was observed that there was no statistically significant correlation between the persistence of antigenemia and age (P greater than 0.2), the initial
antigen concentration (P greater than 0.50), or the development of antibody (P greater than 0.20). The presence of a low magnitude of
bacteremia (less than 300 organisms per ml) was associated with a maximum concentration of 10 ng of PRP per ml. On the other hand, bacterial counts in excess of 10(4)/ml were associated with greater than 1,000 ng of PRP per ml (r = 0.98, r2 = 0.96, P less than 0.001). It was observed that the amount of circulating PRP in the acute phase of illness was related to whether a child developed convalescent-phase antibody. Invariably, the younger children, who primarily had
meningitis, had a PRP concentration of greater than 10 ng/ml and failed to develop an antibody response in any isotype, whereas the older patients, who primarily had
infections other than
meningitis, had a PRP concentration of less than 10 ng/ml and a 45.5% success rate in developing an antibody response (P = 0.006). These findings suggest that there is a direct correlation between the magnitudes of
bacteremia and antigenemia, that
antigen may persist for long periods even in the presence of antibody, and that the level of antigenemia in addition to the patient age is significantly related to the nature of the convalescent-phase antibody response.