[Changes in the spectrum of the causative agents of community-acquired pneumonias (1982-1992)]. / Wandel im Erregerspektrum ambulant erworbener Pneumonien (1982-1992).

BACKGROUND AND OBJECTIVE: The picture of pneumonia acquired out of hospital is continually changing. Microbial aetiology as well as predisposing factors, course and prognosis of this disease were prospectively analysed over a period of 10 years. PATIENTS AND METHODS: In three prospective studies, undertaken between 1982 and 1992, data from 448 patients with out-of-hospital acquired pneumonia were analysed. Microbiological evidence was obtained through blood cultures, from purulent sputum or bronchial secretions, as well as by determining serum antibodies or antigens in urine. RESULTS: Proof of causative organism was achieved in 282 patients (64.1%), with identification of 337 different ones. The spectrum of causative microbes shifted during the study period from predominantly gram-positive bacteria to largely atypical organisms (from 12.5% during 1982/1983, to 36.7% in 1991/1992). Although pneumococci continued to dominate, Chlamydia pneumoniae appeared as an important cause of pneumonia, at 11.4% the most important atypical organism in 1991/1992. -Most patients (74.3%) had a chronic underlying disease. The death rate remained relatively constant at 12.9% during the 10-year period. CONCLUSION: The observed changes in microbial spectrum should be taken into account when specific antibiotic treatment is given. History as well as clinical, radiological and laboratory findings are of only limited value as pointers to the possible microbial cause.

[Community-acquired pneumonia]. / Ambulant erworbene Pneumonien.

In a prospective study 212 patients were analysed who, between 1. 10. 1982-31.12. 1983 and 1. 10. 1985-31. 12. 1986, had been admitted to hospital because of pneumonia. The causative organism was identified in 127 of the 212 patients (60%). Pneumococcus was the most common organism (n = 64), as demonstrated by culture and immunological techniques of determining antigen or antibody. Next most common was Legionella (n = 15) of various species. Mixed infections were found in 11 patients, in all instances associated with pneumococci. There were 24 deaths (11.3%). It is concluded from these results that (1) determination of pneumococcal antigen in sputum, but not in urine or serum, can improve the identification of the causative organism; (2) Legionella is one of the most common causes of pneumonia acquired outside of hospital; and (3) adequate serological diagnosis of Legionnaire's disease is possible only if a large number of different species are tested for.