Liquid ventilation: Gas exchange, perfluorochemical uptake, and biodistribution in an acute lung injury.

Abstract

OBJECTIVE: Compare the physiologic, histologic, and biochemical findings of tidal and partial liquid ventilation (PLV) with gas ventilated lambs with an acute lung injury. DESIGN: Experimental, prospective randomized controlled study. SETTING: School of medicine, department of physiology. SUBJECTS: Eighteen newborn lambs (</=1 wk old). INTERVENTIONS: Injury was established by using HCl saline lavages. Seven lambs underwent tidal liquid ventilation (TLV), five underwent PLV, and six underwent gas ventilation (GV) for 4 hrs. Measurements: Sequential arterial blood chemistries were performed. Ventilation efficiency index, arterial-alveolar Po(2), and physiologic shunt were calculated. Blood and tissue were analyzed for perfluorochemical fluid. Histologic examinations of lungs were performed. MAIN RESULTS: TLV oxygenation was significantly better (p <.001) than PLV and GV. Paco(2) was similar in all three groups. Ventilation efficiency index was significantly better (p <.01) in the TLV group as compared with the PLV and GV groups. Physiologic shunt was significantly less in the TLV injury group (p <.01) than the PLV and GV groups. Perfluorochemical fluid blood level of 2.3 +/- 0.32 &mgr;g/mL in the PLV group was significantly lower (p <.01) than TLV of 7.8 +/- 0.71 &mgr;g/mL; there was a difference (p <.01) as function of time in the TLV and no difference in the PLV injury group. There were no differences in tissue perfluorochemical fluid levels as a function of ventilation ([mean +/- sem] TLV, 219 +/- 26 &mgr;g/g; PLV injury, 184 +/- 26 &mgr;g/g). There was a significant difference in perfluorochemical fluid levels as a function of tissue (p <.001). CONCLUSION: In severe lung injury, this study demonstrates that physiologic gas exchange can be maintained with TLV or PLV. Physiologic shunt was less in the TLV group as compared with PLV or GV. Additionally, perfluorochemical fluid in the blood and tissue is low during PLV and TLV relative to that associated with intravenous administration of perfluorochemical fluid emulsion.