Healthcare-associated, community-acquired and hospital-acquired bacteraemic urinary tract infections in hospitalized patients: a prospective multicentre cohort study in the era of antimicrobial resistance.

The clinical and microbiological characteristics of community-onset healthcare-associated (HCA) bacteraemia of urinary source are not well defined. We conducted a prospective cohort study at eight tertiary-care hospitals in Spain, from October 2010 to June 2011. All consecutive adult patients hospitalized with bacteraemic urinary tract infection (BUTI) were included. HCA-BUTI episodes were compared with community-acquired (CA) and hospital-acquired (HA) BUTI. A logistic regression analysis was performed to identify 30-day mortality risk factors. We included 667 episodes of BUTI (246 HCA, 279 CA and 142 HA). Differences between HCA-BUTI and CA-BUTI were female gender (40% vs 69%, p <0.001), McCabe score II-III (48% vs 14%, p <0.001), Pitt score ≥2 (40% vs 31%, p 0.03), isolation of extended spectrum β-lactamase-producing Enterobacteriaciae (13% vs 5%, p <0.001), median hospital stay (9 vs 7 days, p 0.03), inappropriate empirical antimicrobial therapy (21% vs 13%, p 0.02) and mortality (11.4% vs 3.9%, p 0.001). Pseudomonas aeruginosa was more frequently isolated in HA-BUTI (16%) than in HCA-BUTI (4%, p <0.001). Independent factors for mortality were age (OR 1.04; 95% CI 1.01-1.07), McCabe score II-III (OR 3.2; 95% CI 1.8-5.5), Pitt score ≥2 (OR 3.2 (1.8-5.5) and HA-BUTI OR 3.4 (1.2-9.0)). Patients with HCA-BUTI are a specific group with significant clinical and microbiological differences from patients with CA-BUTI, and some similarities with patients with HA-BUTI. Mortality was associated with patient condition, the severity of infection and hospital acquisition.