Extravascular lung water and pulmonary vascular permeability index as markers predictive of postoperative acute respiratory distress syndrome: a prospective cohort investigation.

OBJECTIVE: Robust markers of subclinical perioperative lung injury are lacking. Extravascular lung water indexed to predicted body weight and pulmonary vascular permeability index are two promising early markers of lung edema. We aimed to evaluate whether extravascular lung water indexed to predicted body weight and pulmonary vascular permeability index would identify patients at risk for clinically significant postoperative pulmonary edema, particularly resulting from the acute respiratory distress syndrome.
DESIGN: Prospective cohort study.
SETTING: Tertiary care academic medical center. PATIENTS: Adults undergoing high-risk cardiac or aortic vascular surgery (or both) with risk of acute respiratory distress syndrome.
INTERVENTIONS: None.
MEASUREMENTS AND MAIN RESULTS: Extravascular lung water indexed to predicted body weight and pulmonary vascular permeability index measurements were obtained intraoperatively and in the early postoperative period. We assessed the accuracy of peak extravascular lung water indexed to predicted body weight and pulmonary vascular permeability index as predictive markers of clinically significant pulmonary edema (defined as acute respiratory distress syndrome or cardiogenic pulmonary edema) using area under the receiver-operating characteristic curves. Associations between extravascular lung water indexed to predicted body weight and pulmonary vascular permeability patient-important with important outcomes were assessed. Of 150 eligible patients, 132 patients (88%) had extravascular lung water indexed to predicted body weight and pulmonary vascular permeability index measurements. Of these, 13 patients (9.8%) had postoperative acute respiratory distress syndrome and 15 patients (11.4%) had cardiogenic pulmonary edema. Extravascular lung water indexed to predicted body weight effectively predicted development of clinically significant pulmonary edema (area under the receiver-operating characteristic curve, 0.79; 95% CI, 0.70-0.89). Pulmonary vascular permeability index discriminated acute respiratory distress syndrome from cardiogenic pulmonary edema alone or no edema (area under the receiver-operating characteristic curve, 0.77; 95% CI, 0.62-0.93). Extravascular lung water indexed to predicted body weight was associated with the worst postoperative PaO2/FIO2, duration of mechanical ventilation, ICU stay, and hospital stay. Peak values for extravascular lung water indexed to predicted body weight and pulmonary vascular permeability index were obtained within 2 hours of the primary intraoperative insult for the majority of patients (> 80%).
CONCLUSIONS: Perioperative extravascular lung water indexed to predicted body weight is an early marker that predicts risk of clinically significant postoperative pulmonary edema in at-risk surgical patients. Pulmonary vascular permeability index effectively discriminated postoperative acute respiratory distress syndrome from cardiogenic pulmonary edema. These measures will aid in the early detection of subclinical lung injury in at-risk surgical populations.