The contribution of arterio-venous extracorporeal lung assist to gas exchange in a porcine model of lavage-induced acute lung injury Jo ¨ rg Brederlau 1 , Ralf Muellenbach 1 , Markus Kredel 1 , Ulrich Schwemmer 1 , Martin Anetseder 1 , Clemens Greim 2 and Norbert Roewer 1 1 Department of Anaesthesiology, Wu ¨ rzburg University Hospital, Wu ¨ rzburg, Germany; 2 Department of Anaesthesiology, Fulda Hospital, Fulda, Germany This prospective large-animal study was performed to evaluate the contribution of arterio-venous extracorpor- eal lung assist (AV-ECLA) to pulmonary gas exchange in a porcine lavage-induced acute lung injury model. Fifteen healthy female pigs, weighing 50.39 /3.8 kg (mean9 /SD), were included. After induction of general anaesthesia and controlled ventilation, an arterial line and a pulmonary artery catheter were inserted. Saline lung lavage was performed until the PaO 2 decreased to 519 /16 mmHg. After a stabilization period of 60 min, the femoral artery and vein were cannulated and a low-resistance membrane lung was interposed. Under apnoeic oxygenation, varia- tions of sweep-gas flow were performed every 20 min in order to evaluate the membrane lung’s efficacy, in terms of carbon dioxide (CO 2 ) removal and oxygen (O 2 ) uptake. Although AV-ECLA is highly effective in eliminating CO 2 , if combined with apnoeic oxygenation, normocapnia was not achievable. AV-ECLA’s contribution to oxygenation during severe hypoxemia was antagonized by a signifi- cant increase in the pulmonary shunt fraction. Perfusion (2006) 21, 277  284. Introduction Pump-driven extracorporeal lung assist (ECLA) is regarded as a rescue therapy in adults presenting with acute respiratory distress syndrome (ARDS). 1,2 However, thus far, no reduction in mortality could be demonstrated, and several side-effects hamper its benefits, eg, pump-induced traumatization of blood cells and activation of the coagulation system, plasma-leakage of oxygenator membranes, infec- tions, technical difficulties and logistic problems associated with patient transport for diagnostic and therapeutic interventions. The idea of pumpless arterio-venous extracorpor- eal lung assist (AV-ECLA) is 40 years old. 3 It did not gain clinical relevance because of the high resis- tance of the oxygenators available. Today, AV-ECLA consists of a low-resistance membrane lung inter- posed into a simple arterio-venous shunt between the femoral artery and vein. 4 The system’s minimal total pressure gradient allows an avoidance of an extracorporeal blood pump and uses the patient’s heart as the driving force. With a cardiac index (CI)  /2 L/min/m 2 , and a mean arterial pressure (MAP)  /70 mmHg, a blood flow of 1  2.5 L/min through the extracorporeal lung is achievable. AV-ECLA is contraindicated in patients with severe left ventricular dysfunction and peripheral vascular disease. With a left-to-right shunt of 25  30% of CI through the membrane, CO 2 elimination was shown to reach 95%. 5 Hemodynamics remained stable under experimental conditions, 6 and influence on arterial oxygenation seems to be minimal. Limiting factors are the oxygen-carrying capacity of the blood and the maximal achievable blood flow through the membrane lung. 7 Nevertheless, early clinical observations reported improvements in oxyge- nation besides sufficient CO 2 elimination, and im- ply ranking AV-ECLA as a rescue therapy in severe ARDS. 8,9 The current study was performed to evaluate AV-ECLA’s impact on gas exchange in adult pigs with severe lavage-induced acute lung injury. We hypothesized that AV-ECLA will not affect arterial oxygen partial pressure. Address for correspondence: Dr med Jo ¨ rg Brederlau, Klinik und Poliklinik fu ¨r Ana ¨sthesiologie, Zentrum Operative Medizin, Universita ¨tsklinikum Wu ¨ rzburg, Oberdu ¨ rrbacher Str. 6, D-97080 Wu ¨ rzburg. E-mail: brederlau_j@klinik.uni-wuerzburg.de Perfusion 2006; 21: 277  284 # 2006 SAGE Publications 10.1177/0267659106074769