Comparative Study of Legionella pneumophila and Other Nosocomial - Acquired

We studied, in a prospective way, the characteristics of definitively diagnosed nosocomially acquired pneumonias in our hospital over 36 months. Out of 55 cases, 27 were due to Legionella pneumophila and 28 to other, nonLegionella bacteria. The cases oflegionellosis concentrated in July, August, and December. The only risk factors that showed significant differences (p<O.O5) were general anesthesia and surgery and immunosuppressive disease, which were more frequent in the non-Legionella group, as were chronic liver disease and lowering of consciousness level. The absence of severe underlying disease, chronic or not, was uncommon in both groups, but more frequent in the Legionella group. We observed no differences in the clinical features of the two groups. Mean values of gammaglutamyltranspeptidase and total bilirubin were higher (p<O.O5) in the non-Legionella group. The only x-ray data that showed significant difference were pleural effusion, more frequent in the non-Legionella group (p<O.O2). The mortality rate of legionellosis was 14.6 percent compared to 35.7 percent for the non-Legionella group (p<O.O5). We conclude that a sure differential diagnosis based on clinical, roentgenographic and analytical features of both groups is not possible. The relatively low mortality rate of the Legionella group, when compared to other series of nosocomial legionellosis, could be due to the standard use of erythromycin in the therapeutic approach to nosocomialacquired pneumonia in our hospital. (Chest 1991; 99:344-50)