In general, criteria for ECMO are the existence of severe hypoxemia not responding to maximal conventional therapy. Most if not all ventilatory and nonventilatory strategies, including ECMO, focus primarily on the maintenance of satisfactory arterial oxygenation (often defined as an Sao2 > 90%), regardless of the adequacy of tissue oxygenation. It is often assumed that acute hypoxemia is poorly tolerated by critically ill patients because of increased risk of tissue hypoxia and death. It must be emphasized however, that tissue oxygenation is determined not only by Sao2 but also by hemoglobin concentration, cardiac output, oxygen affinity of hemoglobin, oxygen extraction, and metabolic demand of the body.2–3 Therefore, because of the presence of several factors that determine oxygen delivery and consumption, Sao2 alone cannot be expected to be a sensitive index of tissue oxygenation. To date, no large prospective randomized trial has been published to evaluate the relationship between acute hypoxemia, tissue oxygenation, and clinical outcome in critically ill patients and to demonstrate to what extent could such a relationship be modified by other parameters of oxygen delivery and consumption such as hemoglobin affinity and oxygen demand of the body.