Critical assessment of use of central venous oxygen saturation as a mirror of mixed venous oxygen in severely ill cardiac patients.

Abstract

Bedside catheterization of the right heart with a percutaneously introduced flowdirected catheter was carried out in 24 critically ill patients; 71 determinations of mixed venous oxygen saturation (MVo 2 ) were obtained. A second catheter was inserted either into a central intrathoracic vein or the right atrium for simultaneous sampling of central venous (CVo 2 ) or right atrial (RAo 2 ) oxygen saturation. For the group as a whole, mean CVo 2 (57.9% ± 15.25) was significantly greater than mean MVo 2 (53.3% ± 14.84) ( P <0.001), but there was no significant difference and correlation was good between changes in central venous compared to changes in mixed venous oxygen saturation. On the other hand, there was no significant mean difference as well as an excellent correlation between individual values of RAo 2 and MVo 2 (r=+0.95). Patients with heart failure or shock showed poor correlation between CVo 2 and simultaneously determined MVo 2 . Furthermore, subjects with shock showed a mean CVo 2 (58.0%±13.05) that was significantly greater than mean MVo 2 (47.5% ± 15.11). In contrast, there was no significant difference between mean RAo 2 and mean MVo 2 and excellent correlation of individual values in patients with either heart failure or shock. Although CVo 2 is a poor reflection of MVo 2 in subjects with severe heart failure or shock, there was a better correlation between changes in CVo 2 with corresponding changes in MVo 2 . In addition, RAo 2 correlated closely with corresponding values of MVo 2 . The reversal of the normal relationship between CVo 2 and MVo 2 under these circumstances is compatible with the thesis that low output states are attended by redistribution of blood flow away from femoral, splanchnic, and renal circulation with preferential preservation of cerebral blood flow.