Bacteremia results in substantial morbidity and mortality among geriatric patients. Japan is a rapidly aging society, however, little is known of the epidemiology of bacteremia in geriatric patients in Japanese institutions.
We performed a retrospective cohort study of patients age >65 years old with bacteremia at a 790-bed, Japanese tertiary care hospital from January to December 2013. We defined treatment mismatch as either empiric antimicrobials not active against subsequently isolated organisms, or antimicrobial agents not administered before notification of positive cultures. Factors associated with a 30-day mortality of bacteremia in geriatric population were evaluated.
We identified 314 episodes of bacteremia in 279 patients; median age was 77 (range 65-94) and 166 (60%) were male. Among 314 episodes of bacteremia, 132 (42%) were community-onset, 55 (18%) healthcare-associated community onset, and 127 (40%) hospital-onset. The most common sources of bacteremia were hepatobiliary (29%) and genitourinary tract (23%). Among the 314 bacteremia episodes, 97 (31%) had treatment mismatch [73 (75%) inappropriate empiric antimicrobial therapy; 24 (25%) no antimicrobials administered]. Treatment mismatch was associated with hospital-onset bacteremia [adjusted odds ratio (aOR) 3.52; 95%CI 1.78-6.94], and candidemia (aOR 8.69; 95% CI 1.87-40.38). Thirty-nine patients (14%) died <30 days after the first positive blood culture. Factors associated with a 30-day mortality were septic shock < 48 hours after first positive blood culture [adjusted hazard ratio (aHR) 5.25; 95%CI 2.16-12.76], Pitt bacteremia score > 3 (aHR 1.28; 95%CI 1.15-1.43), and no administration of antimicrobial agents (aHR 3.08; 95%CI 1.29-7.35)
Treatment mismatch occurred in about one-third of geriatric Japanese patients with bacteremia. Not receiving empiric antimicrobial therapy was independently associated with 30-day mortality.
K. Shintani, None
M. Higuchi, None
Y. Tagashira, None
D. K. Warren, None
See more of: Poster Abstract Session