Royal Society of Tropical Medicine and Hygiene Meeting at Manson House, London, 12 December 1996. HIV and pneumococcal infection in Africa. Microbiological aspects.

By the 1930s several studies had shown that Streptococcus pneumoniae was an important pathogen in Nairobi (Kenya) and various risk factors for infection were recognized, including seasonally cold conditions, overcrowding and recent arrival in the city. Research into pneumococcal disease declined with the arrival of penicillin but recently interest has been rekindled by recognition of the pneumococcus as a human immunodeficiency virus (HIV)-associated pathogen and by the emergence of antibiotic resistance. The pneumococcus and its association with HIV were studied during the course of the Wellcome Trust/Kenya Medical Research Institute HIV Programme in Nairobi (1988-1993). There were generally high rates of pneumococcal disease. The pneumococcus (with tuberculosis and salmonellosis) was a major HIV-related pathogen. One study showed HIV seropositivity to confer a relative risk of 17.8 for pneumococcal infection. Recurrent infection accounted for a large proportion (25%) of disease episodes in a longitudinally studied cohort of HIV patients. There were higher pneumococcal carriage rates in HIV-positive than in HIV-negative patients (28% vs. 16%, P = 0.003). High rates of resistance were found to penicillin (25%). Molecular characterization of penicillin-resistant strains identified 11 separate clones, showing great genetic diversity in a small sample of isolates, and there was evidence of horizontal spread of penicillin-binding protein genes between separate lineages. Molecular characterization of isolates from patients with recurrent disease suggested that both relapse and reinfection might occur. There was molecular evidence of transfer of capsular genes between clones (serotype switching). The overall spectrum of serotypes resembled those reported elsewhere, most serotypes being included in the 23-valent vaccine. Higher numbered serotypes were associated with respiratory tract source and antibiotic resistance. Various methods were used to show 82% concordance between pernasal and blood isolates in pneumonia cases. HIV-seropositive patients were more prone to infection with penicillin- and tetracycline-resistant organisms than seronegative patients (penicillin, 27% vs. 7%; tetracycline, 40% vs. 17%), a difference reflected by different serotype profiles in the 2 groups. These studies highlight the importance of the pneumococcus as an HIV-related pathogen in one part of Africa. The high rates of antibiotic resistance are a cause of concern. There should be continued monitoring of resistance patterns, and assessments of the significance of pneumococcal disease made elsewhere in Africa are to be encouraged.