Since mixed venous oxygen saturation (SvO2) depends on O2-supply and O2-consumption, its measurement is said to indicate tissue O2-balance and to be suitable for ensuring tissue oxygenation in critically ill patients. Blood for SvO2 determinations should be drawn exclusively from the pulmonary artery, because mixing of systemic venous blood is incomplete in the right atrium and ventricle. SvO2 can be determined in vitro and in vivo. Current in vivo techniques determine oxygen saturation from the relation between light emitted into the blood stream and that reflected from blood cells, the use of 3 instead of 2 wavelengths greatly improving the accuracy of the method. For reasonable interpretation of SvO2 it is necessary to consider the concentration of haemoglobin and its derivatives, shifts of the O2-dissociation curve, intracardiac left-to-right shunts, and systemic arteriovenous shunts. Disturbances of macrocirculation (e.g. vascular surgery) and microcirculation (e.g. sepsis) also impair the diagnostic value of SvO2 determinations. The critical value of SvO2 appears to depend on the prevailing disease. Patients with chronically impaired O2 transport appear to tolerate very low SvO2 values better than acutely ill patients, presumably due to adaptive changes in the former group. Central venous oxygen saturation may indicate directional changes of the SvO2, but does not estimate the real SvO2-value. The hypothesis that continuous SvO2 measurements improve prognosis or lower treatment costs has not yet been confirmed. Measurements of mixed venous oxygen saturation may improve monitoring and treatment of critically ill patients in selected cases; however, these measurements are not suitable to indicate reliably the status of tissue oxygenation under all conditions.

In clinical use, the concentrations of local anaesthetics at the site of action are unknown. With the method described here, concentrations of local anaesthetics can be predetermined and kept constant at the site of action. In six volunteers, a blister was raised on the ventral surface of the forearm. After removal of the epithelium, the blister base was rinsed continuously with carbogenated Tyrode's solution with and without increasing concentrations of bupivacaine (Carbostesin) for 15 min each. The effects of bupivacaine were determined by changes in the perception (tactile sensation) of drops falling on the blister base from increasing heights. The minimal height at which the drops were just perceived characterized the threshold of perception. With increasing bupivacaine concentrations, threshold increased until the drops were no longer perceived, at a median concentration of 2.48 mmol litre-1 (range 1.24-3.10 mmol litre-1). After the blister base was rinsed with Tyrode's solution, threshold of perception reached baseline values, which was in accordance with an intact blister base.