ABSTRACT: Dissolved oxygen (i.e., unbound to hemoglobin) is often neglected as a determinant of central venous oxygen saturation (ScvO2) in review articles and textbooks. These statements may lead to potential misinterpretation of SCvO2 value across FiO2 changes. In this study, we aimed to explore the influence of PaO2 and FiO2 on ScvO2 in ventilated critically ill patients. This was a prospective observational study in two surgical intensive care units. Mechanically ventilated and sedated patients with cardiac output and ScvO2 monitoring and PaO2/FiO2 > 200 with inspiratory oxygen (FiO2) ≤ 0.5 were enrolled (cohort [ScvO2]). A second cohort of brain-injured patients with jugular venous oxygen saturation monitoring was studied to assess the application of the results to regional circulation (cohort [SjvO2]). Central venous oxygen saturation was measured at baseline FiO2 and at FiO2 = 1. We finally estimated the participation of the dissolved oxygen (PadissolvO2) to the ScvO2 variations. Twenty patients formed the cohort ScvO2 and eight formed the cohort SjvO2. Central venous oxygen saturation rose from 71% (69%–76%) to 84% (78%–88%) after increasing FiO2, whereas PaO2 rose from 100 (85–124) mmHg to 387 (360–449) mmHg. The rise of ScvO2 was mostly ascribable to the dissolved oxygen. The increase of ScvO2 was not explained by changes in cardiac output or hemoglobin levels. Jugular venous oxygen saturation rose from 71% (58%–78%) to 83% (78%–89%) after increasing FiO2. Arterial dissolved oxygen level can significantly influence the ScvO2 value. Therefore, PaO2 should not be overlooked while considering the ScvO2 value as a therapeutic goal. Interpretation of ScvO2 variations in response to a therapeutic challenge (i.e., fluid challenge, inotropic drug initiation) should be performed at constant FiO2.