BACKGROUND Central venous oxygen saturation (Scv(O(2))) has been used as a surrogate marker for mixed venous oxygen saturation (Sv(O(2))). Femoral venous oxygen saturation (Sfv(O(2))) is sometimes used as a substitute for Scv(O(2)). The purpose of this study is to test the hypothesis that these values can be used interchangeably in a population of patients who are critically ill. METHODS We conducted a survey to assess the frequency of femoral line insertion during the initial treatment of patients who are critically ill. Scv(O(2)) vs Sfv(O(2)) STUDY: Patients with femoral and nonfemoral central venous catheters (CVCs) were included in this prospective study. Two sets of paired blood samples were drawn simultaneously from the femoral and nonfemoral CVCs. Blood samples were analyzed for oxygen saturation and lactate. RESULTS One hundred and fifty physicians responded to the survey. More than one-third of the physicians insert a femoral line at least 10% of the time during the initial treatment of patients who were critically ill. Scv(O(2)) vs Sfv(O(2)) STUDY: Thirty-nine patients were enrolled. The mean Scv(O(2)) and Sfv(O(2)) were 73.1% +/- 11.6% and 69.1% +/- 12.9%, respectively (P = .002), with a mean bias of 4.0% +/- 11.2% (95% limits of agreement: -18.4% to 26.4%). The mean serum lactate from the nonfemoral and femoral CVCs was 2.84 +/- 4.0 and 2.72 +/- 3.2, respectively (P = .15). CONCLUSIONS This study revealed a significant difference between paired samples of Scv(O(2)) and Sfv(O(2)). More than 50% of Scv(O(2)) and Sfv(O(2)) values diverged by > 5%. Sfv(O(2)) is not always a reliable substitute for Scv(O(2)) and should not routinely be used in protocols to help guide resuscitation.