Learning from Galileo: Ventilator-associated Pneumonia Surveillance

Ventilator-associated pneumonia (VAP) continues to be a major cause of morbidity and mortality and excess cost in critically ili patients. Traditional VAP surveillance, the key strategy needed to define the frequency of the phenomenon, and to gauge the success of prevention efforts, is complicated and subjective. Several studies have shown that the CDC's standard definition for VAP has a high interobserver variability, low sensitivity, and low specifìcity. Klompas and Klompas and colleagues have proposed a simpler definition of VAP; more-over, Klompas and colleagues have suggested shifting the focus of surveillance from VAP to ventilator-associated complications (VAC). Here we describe the results of the VAP-VAC surveillance among intensive care unit (ICU) patients of a teaching hospital in centrai Italy. Methods The study was conducted in a 900-bed teaching hospital in centrai Italy, in three medical-surgical ICUs from June to December 2011, including ali patients ventilated for more than 48 hours. For each study patient, a surveillance card was filled in with the following information: age, sex, days of hospitalization, and days of mechanical ventilation (MV). VAP and VAC were recorded, by physicians, according to the criteria proposed by Klompas and colleagues. VAC was defined as an increase in the patient's daily minimum positive end-expiratory pressure by 25 cm H20 sustained for at least 2 days or an increase in the daily minimum Fio, by at least 15 points sustained for at least 2 days after a minimum of 2 days of stable or decreasing daily minimum positive end-expiratory pressures and Fi0,s, respectively. The incidence of VAP and VAC was expressed as the number of first episodes per 1,000 ventilator-days. Moreover, the following outcomes were considered: duration of mechanical ventilation, ICU length of stay, and mortality. We compared conttnuous var-iable distributions using Wilcoxon rank-sum test and the categor-ical variable using the Fisher exact test The level of significance was set at P < 0.05. Results One hundred twenty-seven ICU patients with mechanical ventilation were monitored. The VAP rate was 1.32/1,000 MV-days (n = 2), and the VAC rate was 125/1,000 MV-days (n = 19). A signifìcant difference for duration of mechanical ventilation, hospital stay, and mortality was registered between VAC-positive and VAC-negative patients (P < 0.05), whereas VAP was a statistically signifìcant risk factor only for mortality. Dlscussion An incidence rate of 132/1000 MV-days was registered adopting Klompas's definition of VAP. The VAC definition identified a population of patients with longer ICU length stay, more days of MV, and higher mortality compared with patients without VAC These data are in agreement with Klompas's finding where the VAC definition was able to identify patients with increased mortality, duration of mechanical ventilation, and ICU length stay. Therefore, applying the VAC definition, it was possible to identify a population "at risk" whose complications were severe enough to require an increase in ventilator support. Our experience, although it has limitations due to the small number of patients, may highlight the feasibility and the useful-ness of the model, and above ali, the strategie importance of the identification of VAC-positive patients. In fact, it is likely that lowerìng VAC rate in an ICU could improve clinica! outcomes in terms of duration of hospital stay, duration of mechanical ventilation, and mortality. Moreover, VAC can be easily and quickly detected, and all the defining crìteria are objective. In conclusion, we would like to remember Galileo Galilei when he affirmed that is important "to measure what is measurable and make measurable what is not so", because, regarding infection prevention, we know that only what is effectively measurable makes a better outeome