Central venous oxygen saturation (ScvO 2) >70% or mixed venous oxygen saturation (SvO 2) >65% is recommended for both septic and non-septic patients. Although it is the task of experts to suggest clear and simple guidelines, there is a risk of reducing critical care to these simple recommendations. This article reviews the basic physiological and pathological features as well as the metrological issues that provide clear evidence that SvO 2 and ScvO 2 are adaptative variables with large inter-patient variability. This variability is exemplified in a modeled population of 1,000 standard ICU patients and in a real population of 100 patients including 15,860 measurements. In these populations, it can be seen how optimizing one to three of the four S(c)vO 2 components homogenized the patients and yields a clear dependency with the fourth one. This explains the discordant results observed in large studies where cardiac output was increased up to predetermined S(c)vO 2 thresholds following arterial oxygen hemoglobin saturation, total body oxygen consumption needs and hemoglobin optimization. Although a systematic S(c)vO 2 goal-oriented protocol can be statistically profitable before ICU admission, appropriate intensive care mandates determination of the best compromise between S(c)vO 2 and its four components, taking into account the specific constraints of each individual patient.