Does etiological investigation have an impact on the outcomes of community-acquired pneumonia? - A prospective cohort study.

Abstract

There is lack of evidence that etiological investigation influences outcomes in community-acquired pneumonia (CAP). Guidelines recommend diverse approaches to this matter. Our aim was to find if etiological investigation has an impact on CAP management and outcomes. Prospective cohort study, conducted over a two years' period, in a community-based hospital, including all adult patients with CAP. Univariate and multivariate logistic regression modeling were performed to understand the association of etiological identification with CAP outcomes, particularly hospital mortality. A total of 660 cases of CAP were included, with a mean±sd age of 74±15 years and 58.9% of males. Etiology was documented in 33% of cases. Antibiotic (ATB) was modified in 148 patients, in 51 (34%) motivated by microbiological results. There was no significant impact on hospital mortality of microbiological documentation (35.5% vs 31.2%, p=0.352), or the fact that ATB was modified due to microbiological findings (27.0% vs 36.9%, p=0.272). When stratified by 3 subgroups of risk for drug-resistant pathogens (zero, one or two risk factors: being bed-ridden and/or ATB use within 90 days), etiology identification still did not influence mortality. When adjusted for CURB-65, Charlson's index, being bed-ridden, having had ATB or hospitalization within 90 days or coming from long-term care facilities, microbial identification was not associated with lower mortality. Etiological investigation of patients with CAP does not have an association with hospital mortality, irrespective of the risk for drug-resistant pathogens.