To examine whether the tip of the femoral vein catheter used for sampling femoral venous PO2 during cycling exercise is contaminated by skin or saphenous vein blood, we studied 6 healthy volunteers [21.7 +/- 0.7 (SD) yr] during three identical incremental exercise tests while breathing room air on the same day. Femoral venous blood was sampled simultaneously from two catheters inserted into the femoral vein but advanced in opposite directions (7 cm distally and 5 cm proximally). Blood sampling for measurements of PO2, PCO2, pH, hemoglobin concentration, and oxyhemoglobin saturation was done simultaneously from both catheters in duplicate at rest, at 60% of maximum workload (60% W), and at maximum symptom-limited exercise (100% W). Temperature was measured with a thermistor probe placed in the proximal catheter. At rest, distal PO2 was significantly lower than that measured proximally (24.9 +/- 4.3 vs 30.8 +/- 6.1 mmHg, respectively; P < 0.004), but no differences were found during exercise (60% W, 23.6 +/- 3.4 vs. 24.5 +/- 3.6 mmHg; 100% W, 26.0 +/- 3.6 vs. 25.5 +/- 2.8 mmHg, respectively). Comparison of blood temperatures between proximal and distal sites of sampling in two subjects showed negligible differences. Intrasubject coefficient of variation of distal femoral venous PO2 over the three bouts of exercise was 11.5% (2.9 mmHg) at rest, 5.9% (1.4 mmHg) at 60% W, and 5.6% (1.5 mmHg) at 100% W. Mean differences in distal PO2 between duplicate samples were 0.5 +/- 1.4 mmHg at rest, 0.1 +/- 0.8 mmHg at 60% W, and 0.6 +/- 0.9 mmHg at 100% W.(ABSTRACT TRUNCATED AT 250 WORDS)