[The meaningfulness of central venous blood samples. Central venous versus mixed venous O2 status].

Both mixed-venous and "central-venous" oxygen status (O2 partial pressure [pO2], O2 saturation [sO2], O2 concentration [cO2], hemoglobin concentration [cHb]) are often considered to adequately represent total-body oxygen supply. Since modern technology has made continuously in vivo measurement possible, mixed-venous O2 saturation (svO2) and partial O2 saturation (psO2) have become extensively used for that purpose. Both venous sites of measurement are used as diagnostic adjuncts regarding hemodynamic status. However, both are associated with certain problems. There is a lack of any clear definition of a "central-venous" site of the catheter tip (right atrium, superior vena cava, inferior vena cava). Instead, the location of the catheter tip depends upon the approach to the central venous system. One must also be aware of significant migrational tendencies of the catheter tip. Thus, "central-venous" samples can only represent the situation in a single portion of the circulation, whereas the respective organ areas may vary considerably with the same catheter from one time point to another. Furthermore, the state of the coronary circulation never can be evaluated. The frequently postulated correlation of mixed-venous and "central-venous" values decreases with increasing deterioration of the cardiovascular system, especially in patients with high cardiovascular risks. The main overall parameters of mixed-venous oxygen status are oxygen content (cvO2), cardiac output (C.O.), and oxygen consumption (QO2) of the tissues. The relation between arterial and mixed-venous O2 status is given by Fick's principle: caO2-cvO2 = QO2/C.O. From this, it becomes obvious that a relation between cvO2 and C.O. may only be presumed if QO2 and caO2 remain constant. Evaluation of O2 availability (AO2) using cvO2 or the determining components of cvO2 seems reasonable only if the modulating influences of C.O. and QO2 are taken into consideration. Therefore, any empirically deduced relation between svO2 and C.O. (regardless of the origin, i.e., linear or nonlinear) must be viewed as being accidental. This implies that svO2 and cvO2 per se are only partly helpful in evaluating the hemodynamic situation. The best parameters for interpreting the cardiovascular situation seem to be arteriovenous O2 difference (avDO2) and C.O. In particular, the avDO2 can be accepted as an integrating parameter for AO2 and QO2, and may therefore serve as a diagnostic indication of cardiovascular imbalance.