Empiric treatment in hospitalized community-acquired pneumonia. Impact on mortality, length of stay and re-admission.

OBJECTIVE: To evaluate adherence to guidelines when choosing an empirical treatment and its impact upon the prognosis of community-acquired pneumonia (CAP).

METHODS: A prospective multicentre study was conducted in 425 CAP patients hospitalized on ward. Initial empirical treatment was classified as adhering or not to Spanish guidelines. Adherent treatment was defined as an initial antimicrobial regimen consisting of beta-lactams plus macrolides, beta-lactam monotherapy and quinolones. Non-adherent treatments included macrolide monotherapy and other regimens. Initial severity was graded according to pneumonia severity index (PSI). The end point variables were mortality, length of stay (LOS) and re-admission at 30 days.

RESULTS: Overall 30-day mortality was 8.2%, the mean LOS was 8+/-5 days, and the global re-admission rate was 7.6%. Adherence to guidelines was 76.5%, and in most cases the empirical treatment consisted of beta-lactam and macrolide in combination (57.4%). Logistic regression analysis showed that other regimens were associated with higher mortality OR=3 (1.2-7.3), after adjusting for PSI and admitting hospital. Beta-lactam monotherapy was an independent risk factor for re-admission. LOS was independently associated with admitting hospital and not with antibiotics.

CONCLUSIONS: A high adherence to CAP treatment guidelines was found, though with considerable variability in the empirical antibiotic treatment among hospitals. Non-adherent other regimens were associated with greater mortality. Beta-lactam monotherapy was associated with an increased re-admission rate.