Recent evidence suggests that bacteria and subcellular bacterial components may contribute to the pathogenesis of chronic otitis media with effusion (OME). Liu et al found that 75% of effusions from children with chronic OME contained bacteria, usually gram-positive cocci in Gram's stained smears. However, only 52% of the effusions yielded bacteria on culture; hence, killed bacteria may be retained in the middle ear space. Endotoxin and pneumococcal capsular polysaccharides have been detected in many sterile middle ear effusions (MEEs) from children with chronic OME. Since Hemophilus influenzae, an endotoxin-containing gram-negative microorganism, and Streptococcus pneumoniae are principal causes of acute otitis media, the presence of subcellular components of these organisms in fluid from chronic OME patients suggests that these products may contribute to the pathological transition from acute to chronic OME. The significance of pneumococcal cell surface components in the pathogenesis of pneumococcal meningitis and pneumonia has recently been explored in rabbit infection models. Pneumococcal cell wall invoked significantly more inflammatory changes in CSF and bronchial lavage fluid than pneumococcal capsular polysaccharide. Pleocytosis and abnormal chemistries were observed in both models. Since pneumococcal cell wall has an inflammatory effect on meningeal and respiratory tissues, we hypothesized that pneumococcal cell wall retained in the middle ear would also invoke inflammatory changes similar to those characteristic of naturally acquired OME that could be measured by cytologic and biochemical analysis of middle ear fluid and by quantitative histopathologic study of the middle ear mucoperiosteum.