Inadequate empiric antimicrobial therapy and mortality in geriatric patients with bloodstream infection: A target for antimicrobial stewardship

• Published in: Journal of infection and chemotherapy : official journal of the Japan Society of Chemotherapy 2018-10, Vol.24 (10), p.807-811
• Main Authors: Honda, Hitoshi, Higuchi, Naofumi, Shintani, Kensuke, Higuchi, Masanori, Warren, David K.
• Format: Article
• Language: English
• Subjects: Aged; Anti-Infective Agents - therapeutic use; Antimicrobial Stewardship; Bacteremia; Bacteremia - drug therapy; Bacteremia - epidemiology; Bacteremia - mortality; Bacterial Infections - blood; Bacterial Infections - drug therapy; Bacterial Infections - epidemiology; Bacterial Infections - mortality; Cohort Studies; Confidence Intervals; Female; Geriatric population; Health Services for the Aged; Hospital Mortality; Humans; Initial antimicrobial therapy; Japan - epidemiology; Male; Odds Ratio; Retrospective Studies; Tertiary Care Centers
• ISSN: 1341-321X; 1437-7780
• DOI: 10.1016/j.jiac.2018.06.009

Bloodstream infections are responsible for a large proportion of deaths among geriatric patients. Japan is a rapidly aging society; however, little is known about the epidemiology of bloodstream infections in geriatric patients in Japanese institutions. We conducted a retrospective cohort study of patients aged ≥65 years old with a bloodstream infection in a Japanese tertiary care hospital in 2013. We defined inadequate empiric antimicrobial therapy as either antimicrobial treatment that was ineffective against subsequently isolated organisms or treatment initiated after notification of a positive culture. Predictors of inadequate antimicrobial therapy and 30-day mortality among geriatric patients with bloodstream infections were evaluated. We identified 275 patients with a bloodstream infection, of which 42.2% of cases (116/275) were healthcare-associated, hospital-onset. The most common source of bloodstream infection was hepatobiliary (28.0%). Inadequate empiric antimicrobial therapy occurred in 29.8% of the patients. Factors associated with inadequate empiric therapy included a history of surgery prior to bloodstream infection during index hospitalization (adjusted odds ratio [aOR] 3.27; 95% confidence interval [CI] 1.18–9.12). In 275 patients, 38 (13.8%) died within 30 days after the first positive blood culture. Predictors of 30-day mortality was Pitt bacteremia score >6 (aOR 9.80; 95% CI 4.72–20.36). Inadequate empiric antimicrobial therapy occurred in approximately one-third of episodes of bloodstream infection in geriatric patients. Severity at the time of bloodstream infection was likely to have contributed to mortality. The initiation of adequate empiric antimicrobial therapy may have important implications for antimicrobial stewardship even in the elderly population.