Community-onset Pseudomonas aeruginosa urinary sepsis in elderly people: Predictive factors, adequacy of empirical therapy and outcomes.

Purpose To know the predictive factors for Pseudomonas aeruginosa (PA) urinary sepsis in hospitalised elderly patients coming from community, the adequacy of empirical antibiotic therapy and its outcomes. Methods Retrospective case-control study conducted between 2007 and 2017. Patients aged 65 years or older with PA urinary sepsis were included as cases. Three age-matched controls, with urinary sepsis caused by other microorganisms, were selected for each case. Predictors for PA urinary sepsis were determined by logistic regression analysis. Adequacy of empirical antibiotic therapy and outcomes were compared between both groups. Results A total of 332 patients, from which 83 were cases, were included. Predictive factors for PA urinary sepsis were as follows: male sex (OR 3.16, 95% CI 1.61-6.10; P < .001), urinary catheter (OR 3.25, 95% CI 1.73-6.11; P < .001) and healthcare-associated infection (OR 5.52, 95% CI 1.76 -17.29; P = .003). Inadequate empirical antimicrobial therapy (IEAT) and length of hospital stay were higher in PA group (42% vs 24%, P = .002; 7.45 ± 4.36 days vs 6.43 ± 3.82 days, P = .045, respectively), but mortality was not different (7.2% vs 8.8%, P = .648). Conclusions Pseudomonas aeruginosa urinary sepsis in elderly people was associated with male sex, urinary catheter and healthcare-associated infection. These infections had a higher rate of IEAT and a longer hospital stay than urinary sepsis caused by other microorganisms.