Systemic venous oxygen saturation is clinically used as an indicator of a satisfactory oxygen supply demand balance on cardiopulmonary bypass (CBP). Cerebral desaturation has been associated with postoperative cognitive dysfunction and has an incidence of 17% to 23% on bypass. We tested the hypothesis that systemic venous saturation did not correlate with jugular bulb venous saturation. Blood was drawn from the radial artery, jugular bulb catheter, and venous return line for determination of pH, oxygen tension and saturation, and carbon dioxide tension at four times during bypass: warm 1 (following initiation of CPB); cold 1 (stable hypothermia); cold 2 (hypothermia prior to rewarm); and warm 2 (nasopharyngeal temperature 36 degrees C to 37 degrees C). Correlations of jugular bulb and systemic venous saturation at cold 1 were r = 0.29, r2 = 0.08, and p = 0.0005, and at warm 2 were r = 0.22, r2 = 0.05, and p = 0.007. We conclude that systemic saturation is a poor indicator of cerebral saturation. The poor association of jugular and systemic pump venous saturations underscores our inability to evaluate adequacy of cerebral perfusion. Jugular saturation is lower than pump venous return blood, especially at times of lower oxygen delivery, thus either continuous invasive or noninvasive evaluation of cerebral oxygenation is required to evaluate the adequacy of cerebral perfusion.