Definite, probable, and possible bacterial aetiologies of community-acquired pneumonia at different CRB-65 scores

According to the recommendations of the Swedish Community-Acquired Pneumonia (CAP) guidelines, the selection of empirical antibiotic therapy should be based on the CRB-65 rule. The guidelines recommend empirical therapy directed predominantly against Streptococcus pneumoniae for patients with low CRB-65 scores and broad-spectrum therapy for patients with high CRB-65 scores. In order to study the utility of the recommendations, we analyzed the data from an aetiological study previously performed on 235 hospitalized adult CAP patients at our medical centre. A definite, probable, or possible bacterial aetiology was noted in 194 cases (83%), including 112 cases (48%) with S. pneumoniae aetiology. The following frequencies of definite-probable aetiologies were noted in the patients with CRB-65 score 0-1 (n=155) and CRB-65 score 2-4 (n=80): S. pneumoniae 30% and 35%, Haemophilus influenzae 6.5% and 14% (p=0.063), Mycoplasma pneumoniae 15% and 5.0% (p=0.019), Chlamydophila species 2.6% and 1.2%, Legionella pneumophila 1.9% and 0%, and Staphylococcus aureus 1.3% and 1.2%, respectively. The high frequency of S. pneumoniae in the study supports the recommendations to predominantly cover this bacterium in the empirical therapy of patients with low CRB-65 scores. In the case of treatment failure in these patients, the study indicates that coverage against M. pneumoniae and H. influenzae should be considered.