Partial liquid ventilation reduces pulmonary neutrophil accumulation in an experimental model of systemic endotoxemia and acute lung injury

Objective: To determine whether pulmonary neutrophil se questration and lung injury are affected by partial liquid ventilation with perfluorocarbon in a model of acute lung injury (ALI). Design: A prospective, controlled, in vivo animal laboratory study. Setting: An animal research facility of a health sciences university. Subjects: Forty one New Zealand White rabbits. Interventions: Mature New Zealand White rabbits were anesthetized and instrumented with a tracheostomy and vascular catheters. Animals were assigned to receive partial liquid ventilation (PLV, n = 15) with perflubron (18 mL/kg via endotracheal tube), conventional mechanical ventilation (CMV, n = 15) or high frequency oscillatory ventilation (HFOV, n = 5). Animals were ventilated, using an Flo(2) of 1.0, and ventilatory settings were required to achieve a normal Paco(2). Animals were then given 0.9 mg/kg of Escherichia coli endotoxin intravenously over 30 mins. Partial liquid ventilation, conventional mechanical ventilation, or high-frequency oscillatory ventilation was continued for an additional 4 hrs before the animals were killed. A group of animals not challenged with endotoxin underwent conventional ventilation for 4.5 hrs, serving as the control group (control, n = 6). Lungs were removed and samples were frozen at -70 degrees C. Representative samples were stained for histology. A visual count of neutrophils per high power field (hpf) was performed in five randomly selected fields per sample in a blinded fashion by light microscopy. Lung samples were homogenized in triplicate in phosphate buffer, ultrasonified, freeze thawed, and clarified by centrifugation. Supernatants were analyzed for myeloperoxidase (MPO) activity by spectrophotometry with odianisidine dihydrochloride and hydrogen peroxide at 460 nm. Measurements and Main Results: Histologic analysis of lung tissue obtained from control animals showed normal lung architecture. Specimens from the PLV and HFOV groups showed a marked decrease in alveolar proteinaceous fluid, pulmonary vascular congestion, edema, necrotic cell debris, and gross inflammatory infiltration when compared with the CMV group. Light microscopy of lung samples of animals supported with PLV and HFOV had significantly lower neutrophil counts when compared with CMV (PLV, 4 +/- 0.3 neutrophils/hpf; HFOV, 4 +/- 0.5 neutrophils/hpf; CMV, 10 +/- 0.9 neutrophils/hpf; p<.01). In addition, MPO activity from lung extracts of PLV and HFOV animals was significantly lower than that of CMV animals (PLV, 61 +/- 13.3 units of MPO activity/lung/kg; HFOV, 43.3 +/- 6.8 units of MPO activity/lung/kg; CMV, 140 +/- 28.5 units of MPO activity/lung/kg; p<.01). MPO activity from lungs of uninjured control animals was significantly lower than that of animals in the PLV, HFOV, and CMV groups (control, 2.2 +/- 2 units of MPO activity/lung/kg; p<.001). Conclusions: Partial liquid ventilation decreases pulmonary neutrophil accumulation, as shown by decreased neutrophil counts and MPO activity, in an experimental animal model of ALI induced by systemic endotoxemia. The attenuation in pulmonary leukostasis in animals treated with PLV is equivalent to that obtained by a ventilation strategy that targets lung recruitment, such as HFOV.