THE ETIOLOGY OF SEVERE COMMUNITY-ACQUIRED PNEUMONIA AND ITS IMPACT ONINITIAL, EMPIRIC, ANTIMICROBIAL CHEMOTHERAPY

Of 259 patients admitted to an intensive care unit with severe acute c ommunity-acquired pneumonia, 173 had primary infections and 86 had sec ondary infections. The commonest organism isolated in each group was S treptococcus pneumoniae (51.3 and 36.6% of known isolates in each grou p respectively). Klebsiella pneumoniae was the next most common isolat e (31.9 and 29.3% respectively). A variety of other Gramnegative organ isms and Staphylococcus aureus accounted for most of the remaining pat hogens. Based on retrospective analysis of data, there appeared to be no difference in the alcohol consumption of patients with infection du e to S. pneumoniae and K. pneumoniae. The overall mortality rate for t he primary infections was 47.4%, with 68.4% of these infections due to K. pneumoniae and 33.9% due to the pneumococcus (P<O.002). Among the secondary infections, the overall mortality rate was 40.8% (not signif icantly different to that of primary infections) with 45.5% due to K. pneumoniae and 23.1% due to the pneumococcus (not significantly differ ent on statistical analysis, probably due to low patient numbers). Our investigation confirms that severe community-acquired pneumonia due t o K. pneumoniae is extremely common, even in patients without obvious risk factors for Gram-negative colonization. This organism is contribu ting to the high mortality rate seen in our intensive care unit among patients with pneumonia, and our empiric therapy for such cases routin ely includes a combination of agents active against this organism (e.g . a cephalosporin and an aminoglycoside).