Impact of healthcare-associated infection and antimicrobial therapy in intensive care

Abstract

Objectives To characterise antimicrobial use in patients with and without laboratory-confirmed healthcare-associated infections (HAIs), to contribute additionally to a study of HAI impact in two intensive care units (ICUs). Methods From January until June 2008, a prospective observational non-randomised study was conducted in two ICUs. HAI was defined as a positive culture collected 48 h after ICU admission, so only laboratory-confirmed infections were included. Antimicrobial therapy was included for HAI-detection criteria to eliminate cases of contamination or/and colonisation. All data were collected from the hospital online clinical database and analysed using proper statistical software. Key findings Of the 213 patients included, 93 were found to have at least one HAI. The main sites of infection were the respiratory tract, bloodstream and urinary tract. Among the 234 HAIs detected the most frequently isolated microorganisms were Candida albicans, methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa. Piperacillin/tazobactam, meropenem, vancomycin and ceftriaxone were the most prescribed antimicrobials, including in the group of patients without laboratory-confirmed HAIs. The HAI group had a prolonged length of stay and antimicrobial treatment, and higher antimicrobial treatment costs (P < 0.001). Conclusions The prescription pattern of antimicrobial drugs was strongly influenced by the microorganisms causing the HAIs. The occurrence of HAIs had a great clinical and economic impact in these ICUs, and antimicrobial therapy in patients with HAIs cost three times more, associated with a doubled length of stay. Nevertheless, patients without HAIs also showed significant prevention costs.