To define the influence of prognostic factors in patients with community-acquired pneumococcal bacteremia, a 2-year prospective study was performed in 5 centers in Canada, the United States, the United Kingdom, Spain, and Sweden. By multivariate analysis, the independent predictors of death among the 460 patients were age >65 years (odds ratio [OR], 2.2), living in a nursing home (OR, 2.8), presence of chronic pulmonary disease (OR, 2.5), high acute physiology score (OR for scores 9-14, 7.6; for scores 15-17, 22; and for scores >17, 41), and need for mechanical ventilation (OR, 4.4). Of patients with meningitis, 26% died. Of patients with pneumonia without meningitis, 19% of those with >/=2 lobes and 7% of those with only 1 lobe involved (P=.0016) died. The case-fatality rate differed significantly among the centers: 20% in the United States and Spain, 13% in the United Kingdom, 8% in Sweden, and 6% in Canada. Differences of disease severity and of frequencies and impact of underlying chronic conditions were factors of probable importance for different outcomes.

No abstract

To investigate the management of patients with community-acquired pneumonia (CAP) treated in hospital in Sweden, a multicentre retrospective cohort study was performed with medical record review of 982 patients (mean age 63 y) at 17 departments of infectious diseases at hospitals in Sweden. Information on antimicrobial therapy, demographic characteristics, comorbid conditions, physical examination findings, and laboratory and microbiological test results were recorded. Outcome measures were in-hospital mortality and length of hospital stay (LOS). Cultures were obtained from blood in 80% and from sputum in 22% of the patients. A microbiological aetiology was determined for 23% of the patients, with Streptococcus pneumoniae as the dominating agent (9%). The initial antibiotic treatment was mostly given intravenously (78%). Penicillin (50%) or a cephalosporin (30%) was the most common choice. Both of these drugs were usually given as a single agent. The overall mortality was 3.5% and the mean LOS was 6.4 d. Thus, the outcome was favourable despite the empirical antibiotic treatment having a narrow spectrum compared with the broader approach recommended in most recent guidelines on the management of CAP. These findings suggest that a majority of patients who are hospitalized with moderately severe pneumonia can be treated initially with penicillin alone.

Sensitivity patterns of Streptococcus pneumoniae, Haemophilus influenzae and Streptococcus pyogenes were studied prospectively in an outpatient population seeking medical advice for respiratory tract infections (RTI) in the Southern parts of Stockholm. In total, 3,214 nasopharyngeal and 1,907 throat swabs were cultured during January-February 1996. 32% of the patients had received antibiotics during the previous year. Reduced penicillin sensitivity in S. pneumoniae was rare (1.3%) and only seen in patients treated with antibiotics during the previous 4 months. Beta-lactamase production in H. influenzae was found in 13.4% of patients who had been treated with antibiotics during the last 4 months and in 7.9% of the others. No resistance (< 1%) to erythromycin was seen in S. pyogenes. In this population-based surveillance, the levels of resistance in common respiratory tract pathogens were thus low and correlated to previous antibiotic treatment. Strict indications for antibiotic treatment in uncomplicated RTI are advocated to maintain a low resistance rate. Penicillin is still the drug of choice in patients without frequent recurrences of RTI in a setting similar to the one studied.

No abstract