Comparison of three latex agglutination kits and counterimmunoelectrophoresis for the detection of bacterial antigens in a pediatric population.

Three commercial latex agglutination kits (Bactigen, Wampole Laboratories, Cranbury, NJ; Directigen, Hynson, Westcott & Dunning, Baltimore, MD; Wellcogen, Wellcome Diagnostics, Dartford, England) used for the detection of bacterial polysaccharide antigens (Haemophilus influenzae type b, Streptococcus pneumoniae, and Neisseria meningitidis) were compared with counterimmunoelectrophoresis and Gram stain for the identification of systemic bacterial disease in children. Urine and (when available) cerebrospinal fluid specimens were saved for all patients. Positive blood or cerebrospinal fluid culture isolates included 36 with H. influenzae type b, 11 with S. pneumoniae, 3 with N. meningitidis and 6 with other organisms. All latex kits performed similarly for the detection of H. influenzae type b antigen with a sensitivity range of 91 to 100%. The four methods performed poorly for the detection of S. pneumoniae (23 to 50%) and N. meningitidis (0%) antigen. Gram stain of cerebrospinal fluid appeared to be equally sensitive to the antigen detection methods for patients with meningitis. The false positive rates for the latex kits and counterimmunoelectrophoresis ranged from 2.8 to 9.2%, with Wellcogen having the lowest rates. The false negative rates ranged from 6.5% to 12% with Directigen having the lowest rate.

Clinically significant cross-reactions with counterimmunoelectrophoresis between pneumococcus type 6 and Haemophilus influenzae type b.

Biologically and clinically significant cross-reactions may occur between the immunochemically similar capsule antigens of Haemophilus influenzae type b and Streptococcus pneumoniae type 6 by using counterimmunoelectrophoresis (CIE). In three cases of culture-proven S. pneumoniae type 6 disease, a false-positive H. influenzae type b CIE result was detected in a body fluid. Two of three cases were also positive when tested by CIE with pneumococcal omnisera. However, in one of the culture-proven cases of pneumococcal type 6 disease, the omniserum was negative and only the burro 132 H. influenzae type b antiserum demonstrated a positive precipitant reaction. In addition to the three case isolates, we evaluated nine blood or cerebrospinal fluid isolates of pneumococcus type 6, using a variety of both species- and type-specific antisera, by CIE. Two burro antisera detected 12 of 12 (100%) of the S. pneumoniae type 6 strains. A rabbit H. influenzae type b antiserum did not detect any (0 of 12; 0%) of the S. pneumoniae type 6 strains evaluated. All (12 of 12; 100%) of the S. pneumoniae type 6 strains were detected with both rabbit omniserum (S. pneumoniae) and rabbit type 6 antiserum. Our data illustrate the importance of being aware of the immunologic cross-reactivity of the H. influenzae type b capsule with the pneumococcus type 6 capsule and the possibility of false-positive results with CIE when patient specimens are interpreted.

Effect of rifampin chemoprophylaxis on carriage eradication and new acquisition of Haemophilus influenzae type b in contacts.

We conducted a multicenter trial designed to assess the efficacy of three different drug regimens on eradication of Haemophilus influenzae type b (HIB) from the nasopharynx of household contacts of patients with invasive type b Haemophilus disease. The drug regimens studied were rifampin, 20 mg/kg, once daily for four days; rifampin, 10 mg/kg, twice a day for four days; and placebo, once daily for four days. Shortly after admission of the index patient to the hospital, 26% of 492 household contacts were found to be colonized with HIB. Both rifampin regimens eradicated carriage significantly better than placebo at 10 and 30 days (P = .001). However, among contacts whose cultures were initially negative, new acquisition of the organism occurred infrequently in this 30-day follow-up period regardless of the drug or placebo regimen prescribed. We also measured the concentration of anticapsular antibody in sera obtained from contacts younger than 6 years of age. Samples were obtained soon after admission of the index patient to the hospital and 30 days later. Several carriers younger than 2 years of age had low concentrations of antibody in both specimens. In contrast, nearly all carriers 2 to 5 years of age had high concentrations of antibody even in the first sample. Children who were not carriers usually had low antibody concentrations which did not increase during the period of observation. Our results suggest that most intrafamilial spread of HIB occurs prior to hospitalization of the index patient and stimulates immunity in contacts older than 2 years of age.(ABSTRACT TRUNCATED AT 250 WORDS)