Surveillance of ICU-associated infections

Introduction

Healthcare-associated infections (HAI) affect around 9% of all patients in hospital [1]. It is generally accepted that the infection rate in the ICU is higher than elsewhere in the hospital and these patients are subject to many invasive procedures that increase their risk of acquiring an HAI. Our aim was to conduct a prospective pilot audit of HAI in ICUs in Scotland using the Hospitals in Europe Link for Infection Control through Surveillance (HELICS) protocol for surveillance of nosocomial infections in ICUs using the current Scottish audit database (Ward Watcher software; Critical Care Audit Ltd, Yorkshire, UK).

Methods

Between 1 May 2005 and 31 August 2005, five of the 25 adult, general ICUs in Scotland participated for 3 months each in a prospective pilot audit of HAI in ICUs. Daily data were recorded on enabling identification of the first episode of infection: HELICS Level 2 surveillance of blood stream infections (BSI), catheter-related infections and ventilatorassociated pneumonia surveillance.

Results

In this 3-month pilot study a total of 386 patients were admitted to the five pilot sites, 52% (199) of whom stayed in the ICU for at least 2 days. For these five ICUs the average occupancy was 78.3%, the first 24-hour APACHE II score 19.8 and the average length of ICU stay 5.7 days. In 32 patients (16%) there were 44 episodes of infection diagnosed using the criteria: BSI = 11.4%, PN = 68.2% and CVC = 20.5%. The overall infection rate was 30.5 infections (95% CI 22.2–40.9) per 1000 patient-days. Twenty-three (72%) patients developed one episode of infection, five (16%) had two episodes, three (9%) developed three episodes and one (3%) patient had four episodes. The majority of pneumonias (60%) were PN4 because quantitative microbiological analysis (PN1 and PN2) is not available routinely.

Conclusion

Of all patients admitted to ICU during the pilot study 16% developed an ICU-associated infection. As expected, the majority of infections diagnosed were pneumonias (68%). This is consistent with the findings of the European Prevalence of Infection in Intensive Care (EPIC) study [2].