The DNA oligomer 5'-d(TGCGGCCTCTCAGTCCCGCAClTICATCITCC)-3' specifically recognizes Haemophilus influenzae 16S rRNA. We report here the use of this oligonucleotide, with a fluorescein label tagged on its 5' end, as a probe for the in situ detection of nonencapsulated nontypeable H. influenzae in sections of adenoid tissue from 10 children who were clinically infection free but were having their adenoids removed because of nasal obstruction. In some cases, the reticular crypt epithelium was focally infiltrated by H. influenzae. The reservoir for these bacterial colonizations, in all likelihood long standing, seemed to be macrophage-like cells found in the subepithelial layers in all 10 cases. These mononuclear cells contained up to 200 intracellular H. influenzae cells. In the transmission electron microscope, macrophage-like cells with intracellular bacteria with coccoid morphology, at least some of which were dividing, were seen. Adenoid cell suspensions, enriched for macrophages by use of paramagnetic beads coated with monoclonal antibodies against the CD14 marker, yielded up to 1,100 CFU of nontypeable H. influenzae per 105 cells after killing of

That nontypeable H. influenzae (NTHI) can reside intracellularly in human adenoid tissue has been suggested by use of in situ hybridization of a fluorescein labelled 16S rRNA-targeted oligonucleotide probe (FISH). Adenoid tissues from 43 children operated on in a clinically infection-free interval were investigated. FISH revealed H. influenzae in macrophage-like cells, located subepithelially in the crypts in all 43 adenoids. Furthermore, H. influenzae was detected in 22/22 adenoids using immunohistochemistry with the monoclonal antibody MAHI-3 recognizing a conserved H. influenzae LPS inner-core region. FISH and staining with monoclonal antibodies against immunophenotypic markers were performed simultaneously in order to characterize the cellular interrelations in this microenvironment. The findings of widespread presence of H. influenzae in cells of which some strongly expressed the CD14 marker of the monocyte/macrophage lineage may correspond to an important aspect of the colonization mechanisms whereby NTHI persists in the nasopharynx of children.

Homogenized adenoid tissue from 55 children (28-153 months) undergoing adenoidectomy because of nasopharyngeal obstruction was investigated by means of quantitative aerobic bacterial culture. The children were divided into two groups, the hypertrophy alone group--AH (n = 29)--and the hypertrophy with longstanding secretory otitis media group--SOM (n = 26). A nasopharyngeal culture was obtained preoperatively from 38 of the cases. Non-typeable H. influenzae (NTHI) was found in twice as many cases in the AH group as in the SOM group, 21/29 (72%) compared to 11/26 (42%) (p < 0.05) and in a significantly higher mean concentrations, 5.7 x 10(5) CFU/g compared to 1.9 x 10(5) CFU/g (p = 0.02). For the other aerobic potentially pathogenic bacteria no such difference was found. The bulk of the NTHI-positive cases and the cases with the highest concentrations were found in the children below the age of 6 years. In the nasopharyngeal cultures NTHI alone or together with S. pneumoniae and/or B. catarrhalis was found in 29% of the cases in both the AH group and SOM group. NTHI was found in only 50% of the nasopharyngeal cultures corresponding to a positive quantitative culture (10/20). These findings suggest that NTHI is harboured within the adenoid and could thereby chronically stimulate the local immune defense. However, the present study indicates that there is no aerobic bacterial overload in the adenoid tissue in children with SOM compared to children without middle-ear disease.

nasal polyps do not respond with down-regulation of GR mRNA or cytokines following topical corticosteroid treatment. The proposed corticosteroid resistance may be inherent, or induced by a change of local tissue bioavailability.

Using monoclonal antibodies and immunohistochemistry, we compared adenoid tissue from 35 children with or without secretory otitis media. Numerous cells infiltrating the reticular crypt epithelium expressed HLA-DR, as did < 10% of the epithelial cells. Of the antigen-presenting cells, CD1a+ dendritic cells showed intraindividual and interindividual variations; CD68+ macrophages and CD22+ B cells were uniformly distributed. The relative frequencies of CD4+ and CD8+ cells were 6.6 +/- 2.0 versus 2.3 +/- 1.2 (p < .001) in the reticular crypt epithelium and 18 +/- 4.5 versus 1.5 +/- 0.9 (p < .001) in the germinal centers. The IL-2 receptor was expressed on < 0.1% of CD3+ T cells. Over 90% of intraepithelial CD3+ T cells were of the CD45RO+ memory phenotype. The proliferation marker Ki67 was almost exclusively found in the germinal centers. That the analyzed parameters showed a similar pattern in both clinical groups suggests that the presence of secretory otitis media may not correlate to specific alterations in the immune microenvironment of the adenoid.