We noted a rise in acetaldehyde levels in clinical samples of venous whole blood containing ethanol that did not occur in samples from teetotalers. Experiments were performed to define the mechanism involved in acetaldehyde production. The addition of 0.10% ethanol to whole blood produced an immediate increase in acetaldehyde due to acetaldehyde in the stock solution followed by a subsequent increase that became statistically significant by 48 hr. Separation of blood into components documented that the increase in acetaldehyde was associated with the red cell but not plasma fraction. Incubation of isolated hemoglobin with ethanol produced a rise in acetaldehyde levels. Incubation of oxygenated whole blood with ethanol produced a linear increase in aceteldehyde, whereas nitrogen-exposed blood produced no increase. The rise of acetaldehyde in the presence of ethanol was dependent on the concentration of oxygenated hemoglobin A(0). Addition of inhibitors of catalase, alcohol dehydrogenase, and glycolytic enzymes (aminotriazole, azide, pyrazole, sodium fluoride, sodium citrate, and iodoacetate) did not inhibit the rise of acetaldehyde, but addition of the hemoglobin ligand cyanide abolished the rise in acetaldehyde. Kinetic analysis with oxygenated whole blood plus inhibitors revealed a K-m of 2.5 mm and V-max of 1.42 mu M/min. We conclude that oxyhemoglobin contributes to the metabolism of ethanol to acetaldehyde. These findings may explain in part the high levels of acetaldehyde found in red cells compared with plasma. The results also have implications for the optimum storage of blood samples for acetaldehyde analysis.