In acute respiratory distress syndrome (ARDS), ECMO is the final appro
ach to maintain adequate blood gases because of its ability to provide
complete gas exchange. In newborns, veno-venous access proved to be s
uperior to veno-arterial. Veno-venous ECMO for adults, however, is not
established yet. Thus we designed such a system and tested it in shee
p. Methods: In sheep (N = 7, mean weight 94 +/- 2 kg) a 28 F catheter
was advanced through the right jugular vein into the caudal caval vein
with a second aperture in the cranial caval vein. A second withdrawal
catheter (20 F) was placed in cranial direction. Oxygenated blood was
reinjected through a 20 F catheter, positioned through the left jugul
ar vein into the right atrium. Postoperatively, the ECMO system was te
sted in conscious, spontaneously breathing sheep with an extracorporea
l oxygen and blood flow (EBF, EOF) of 61/min. ARDS was then induced by
pressure controlled ventilation, and ECMO was started again. After in
dividual adjustment of EBF and EOF, both were increased by 2 1/min eac
h. At all time points, hemodynamic readings and blood gases were taken
. Results: In healthy sheep, ECMO led to a significant decrease in res
piratory minute volume from 12.4 +/- 0.2 to 1.4 +/- 0.4 1/min, while b
lood gases remained unchanged. After development of ARDS any spontaneo
us pulmonary gas exchange could be excluded. ECMO provided complete ga
s exchange. A further increase in EOF significantly lowered PaCO2 (42.
9 +/- 6.9 vs. 34.3 +/- 5.2 mmHg), whereas an increase in EBF significa
ntly improved PaO2 (88.9 +/- 15.7 vs. 106.1 +/- 18.5 mmHg). Changes in
EBF never had any effect on systemic or pulmonary blood pressure. No
significant recirculation occured. Discussion: The efficacy of this EC
MO system is proven by the fact that conscious sheep had nearly no nee
d to breathe, and adequate blood gases were maintained after lung inju
ry was sustained. These effects were accompanied by extremely stable s
ystemic and pulmonary hemodynamics.