RESUMO
OBJECTIVE: To examine the clinical applicability of substituting central venous oxygen saturation (ScvO2) for mixed venous oxygen saturation (SmvO2) in monitoring global tissue oxygenation. DESIGN: Prospective clinical investigation. SETTING: University hospital. PARTICIPANTS: Seventy-three adult patients. INTERVENTIONS: Venous oxygen saturation was recorded, and oxygen saturation difference between SmvO2 and ScvO2 (DeltaSmvcv) was calculated in 2 groups of patients (group I, sepsis patients [n = 41], and group II, general anesthesia for cardiovascular surgery patients [n = 32]) during initial placement of pulmonary artery catheters. MEASUREMENTS AND MAIN RESULTS: Patients were classified as follows: class A, patients having a DeltaSmvcv >-5%; class B, patients having a DeltaSmvcv between -5% and +5%; and class C, patients having a DeltaSmvcv >+5 %. Statistically significant differences were observed in cardiac index, oxygen delivery index, and oxygen extraction ratio between class A and B in both groups. Class C of group II showed the worst correlation between SmvO2 and ScvO2 and had significantly lower arterial carbon dioxide tension values than class A and B. CONCLUSION: Pulmonary artery blood sampling should not be replaced with central venous blood. Hypocapnia and increased oxygen extraction ratio seem to be the major factors that worsen the relationship between ScvO2 and SmvO2.