Surveillance of infections acquired in intensive care: usefulness in clinical practice

Nosocomial infection surveillance is common in the USA and in some European countries but in Italy few hospitals use it. In order to evaluate its usefulness in clinical practice we performed a one year prospective epidemiological study that included 178 patients, admitted to an intensive care unit (ICU) for more than 48 h. Median ICU stay was 16 days. Trauma and neurological diseases accounted for 65% of admissions. The selected population had high severity scores and required a large number of invasive procedures for diagnosis and therapy. The most common infections were: pneumonia 46/1000 ventilator-days; urinary tract infections 17/1000 catheter-days; central venous catheter infections 14.5/1000 catheter-days with 1.7/1000 CVC-related sepsis; bacteraemic sepsis 12/1000 ICU-days. The most frequent pathogens were Staphylococcus aureus, Pseudomonas aeruginosa, other Gram-negative aerobes and Candida spp. Antimicrobial resistance was substantial, with 68% methicillin-resistance in S. aureus and 76% of P. aeruginosa displaying antibiotic resistance. Severe sepsis or septic shock occured in 30 patients (8/1000 ICU-days), and three patients died from septic shock of unknown origin (10% case fatality rate). There were no case fatalities for pneumonia and bacteraemic sepsis. Overall, ICU-acquired infections were not associated with an increased risk of death. (C) 2002 The Hospital Infection Society.