Two in-line oxygen saturation monitors, the CDI 100 and OxySat II, were evaluated in the clinical setting. Eighty-seven venous blood samples were drawn during 20 elective cardiopulmonary bypass procedures. Monitor readings were compared to OSM III co-oximeter values. The results revealed that saturation (%) determination was biased, -3.16 +/- 2.21 SD for the CDI 100 and -0.34 +/- 2.17 SD for the OxySat II. Hemoglobin (g/dl) and hematocrit (%) measurement, available only for the CDI 100, resulted in a bias of +5.54 +/- 5.68 SD and +1.94 +/- 1.78 SD, respectively. It was concluded that both monitors operated within clinically acceptable limits, with a more favorable outcome for the OxySat II.