[Mortality and prognostic factors in patients admitted with community-acquired bacteremia]

Community-acquired bacteremia is a frequently clinical problem and is associated with a high mortality. The Pitt bacteremia score that is calculated using body temperature, mental status, and some respiratory and circulatory features is an useful tool in order to know the prognosis of patients with bacteremia. Objectives were to know microbial etiology of bacteremia in our community, percentage of patients that received an inadequate empiric antibiotic treatment and independent prognostic factors associated with mortality, including Pitt bacteremia score.

Observational study of a cohort of patients over 18 years admitted at a tertiary hospital due to an infection with a community-acquired bacteremia. Twenty-eight cases were rejected because possible blood culture contamination.

Seventy-five patients were included, the mean age was 68.6 +/- 18.2 years, 38 women (51%). Mortality rate was 23% [IC95% 13.8-33.8]. More frequent bacteria isolated were Escherichia coli, Staphylococcus aureus, coagulase negative staphylococci, Pseudomonas aeruginosa, Streptococcus viridans, enterococci and Klebsiella spp. Mortality in patients treated with an inadequate antibiotic therapy (18%, 3 patients) was not lower than in adequately treated patients (24%, 14 patients, p = 0.42). Independent prognostic factors related to mortality were serum albumin concentration, OR = 5.17 (IC 95% 1.45-16.7) for every downing step of 1 g/dl, the Pitt bacteremia score OR = 1.50 (IC 95% 1.01-2.24) for every unit increase, and a high score at McCabe and Jackson classification OR = 5.08 ( IC 95% 1.43-16.7).

An inadequate empiric antibiotic therapy was not associated with a worse vital prognosis. Independent prognostic factors related to mortality were serum albumin concentration, the Pitt bacteremia score, and the McCabe and Jackson classification.