453

Introduction: Lung protective ventilation (LPV) and fluid conservative strategy (FCS) in ARDS may not be widely adopted in practice. Effect of hospital-wide ARDS surveillance on LPV and FCS adherence by intensivists has not been described. Hypothesis: ARDS surveillance team will improve adherence to LPV and FCS. Methods: Study was conducted in 979-bed hospital during November 2011-April 2012 (baseline) and May-August 2012 (intervention). Respiratory therapist reviewed daily all mechanically ventilated ICU patients on full assist, screening for ARDS by oxygenation data. Physicians reviewed clinical data and diagnosed ARDS. Every ARDS observation was classified as (1) LPV (FCS) indicated/adherent, (2) LPV (FCS) indicated/not adherent, (3) LPV (FCS) contraindicated, (4) LPV (FCS) no longer applies (weaning started). During intervention, surveillance team gave real-time feedback to ICU teams on presence of ARDS in patients in group (2). ICU teams decided on implementing LPV (FCS). We measured adherence to LPV (FCS) during baseline and intervention. Results: During baseline, there were 1224 daily observations of mechanically ventilated patients. Of these, 369 (30.1%) were classified as ARDS. During intervention, there were 543 observations; of these, 118 (21.7%) were classified as ARDS (p=0.001). Adherence to LPV: during baseline, 48.5% of ARDS observations were adherent, 35.2% not adherent, 6.3% contraindicated, and in 10% LPV was no longer applicable. During intervention, 70.0% adherent, 16.1% not adherent, 9.3% contraindicated, and 7.6% no longer applicable. LPV adherence increased while non-adherence decreased during intervention (both p<0.0001). Adherence to FCS: during baseline, 35.4 % of ARDS observations were adherent, 17.9% not adherent, 44.2% contraindicated, and in 2.5% no longer applicable. During intervention, 30.8% adherent, 29.9% not adherent, 35.9% contraindicated, and 3.4% no longer applicable. There was trend toward decrease in FCS adherence (p=0.051), while FCS nonadherence significantly increased (p<0.0001). Conclusions: ARDS surveillance with feedback was associated with improvement of adherence to LPV but did not improve adherence to FCS. ARDS prevalence decreased during intervention.