ObjectiveTo investigate whether pulmonary compliance and gas exchange will be sustained during ``total'' perfluorocarbon liquid ventilation followed by ``partial'' perfluorocarbon liquid ventilation when compared with gas ventilation in the setting of the acute respiratory distress syndrome (ARDS).Study DesignA prospective, controlled, laboratory study.SettingA university research laboratory.SubjectsTen sheep, weighing 12.7 to 25.0 kg.InterventionsLung injury was induced in ten young sheep, utilizing a right atrial injection of 0.07 mL/kg of oleic acid followed by saline pulmonary lavage. Bijugular venovenous extracorporeal life support access, a pulmonary artery catheter, and a carotid artery catheter were placed. When the alveolar-arterial O2gradient was more than equals 600 torr and PaO2less than equals 50 torr (less than equals 6.7 kPa) with an FIO2of 1.0, extracorporeal life support was instituted. For the first 30 mins on extracorporeal life support, all animals were ventilated with gas. Animals were then ventilated with equal tidal volumes of 15 mL/kg during gas ventilation (n equals 5) over the ensuing 2.5 hrs, or with total liquid ventilation for 1 hr, followed by partial liquid ventilation for 1.5 hrs (total/partial liquid ventilation, n equals 5).Measurements and Main ResultsAn increase in physiologic shunt (gas ventilation equals 69 plus minus 11%, total/partial liquid ventilation equals 71 plus minus 3%) and a decrease in static total pulmonary compliance measured at 20 mL/kg inflation volume (gas ventilation equals 0.48 plus minus 0.03 mL/cm H2O/kg, total/partial liquid ventilation equals 0.50 plus minus 0.17 mL/cm H2O/kg) were observed in both groups with induction of lung injury. Physiologic shunt was significantly reduced during total and partial liquid ventilation when compared with physiologic shunt observed in the gas ventilation animals (gas ventilation equals 93 plus minus 8%, total liquid ventilation equals 45 plus minus 11%, p less than .001; gas ventilation equals 95 plus minus 3%, partial liquid ventilation equals 61 plus minus 12%, p less than .001), while static compliance was significantly increased in the total, but not the partial liquid ventilated animals when compared with the gas ventilated group (gas ventilation equals 0.43 plus minus 0.03 mL/cm H2O/kg, total liquid ventilation equals 1.13 plus minus 0.18 mL/cm H2O/kg, p less than .001; gas ventilation equals 0.41 plus minus 0.02 mL/cm H2O/kg, partial liquid ventilation equals 0.47 plus minus 0.08, p equals .151). In addition, the extracorporeal life support flow rate required to maintain adequate oxygenation was significantly lower in the total/partial liquid ventilation group when compared with that of the gas ventilation group (gas ventilation equals 89 plus minus 7 mL/kg/min, total liquid ventilation equals 22 plus minus 10 mL/kg/min, p less than .001; gas ventilation equals 91 plus minus 12 mL/kg/min, partial liquid ventilation equals 41 plus minus 11 mL/kg/min, p less than .001). Lung biopsy light microscopy demonstrated a marked reduction in alveolar hemorrhage, lung fluid accumulation, and inflammatory infiltration in the total/partial liquid ventilation animals when compared with the gas ventilation animals.ConclusionsIn a model of severe ARDS, pulmonary gas exchange is improved during total followed by partial liquid ventilation. Pulmonary compliance is improved during total, but not during partial liquid ventilation. Total followed by partial liquid ventilation was associated with a reduction in alveolar hemorrhage, pulmonary edema, and lung inflammatory infiltration.(Crit Care Med 1996; 24:1001-1008)