PURPOSE: To determine whether PEEP and lung injury interact to impact on the incidence of ventilator associated pneumonia (VAP).
METHODS: Prospective multicenter international observational study enrolling all patients on mechanical ventilation for one month. All patients were followed until death, extubation, or day 28 of intubation. Three analysis were done, using different defintions of VAP: 1) Chest X-ray (CXR) infiltrate with at least 2 of three criteria (fever, leukocytosis, or purulent sputum), 2) CXR with all three criteria present, or 3) Clinical pulmonary infection score (CPIS) greater than or equal to 5. Cox proportional hazards model with time varying covariates was used.
RESULTS: A total of 2,894 patients were identified who were on mechanical ventilation for 2 days or longer. There were significant interactions between PEEP and lung injury. In the absence of lung injury, higher levels of PEEP (> 9 cm H2O) were associated with increased risk of VAP. In the presence of lung injury, low levels of PEEP (< 5 cm H2O) were associated with increased risk of VAP. In multivariate analysis, PEEP < 5 cm H2O, and high tidal volume with concurrent lung injury were associated with increased risk of VAP. The presence of lung injury further increased the hazard associated with low PEEP. In the absence of lung injury when PEEP was low, higher tidal volumes were protective.
CONCLUSION: Risk of VAP is related to the interaction of tidal volume, PEEP, and lung injury. In the non-injured lung, higher tidal volumes may be protective, but once lung injury occurs, high tidal volumes increase the risk. PEEP < 5 cm H2O increase the risk of VAP.
CLINICAL IMPLICATIONS: PEEP lower than 5 cm H2O should be avoided to minimize the risk of VAP, especially in the presence of lung injury. High tidal volumes in the presence of lung injury also increase VAP risk.
DISCLOSURE: David Ost, None.