Get access

Oxygen: the poison is in the dose


  • Editor's note: This analytical review by Robert M. Winslow, MD, is being published posthumously. The topic and content reflect Winslow's lifelong interest in oxygen delivery mechanisms and their physiologic consequences. The monograph benefitted from numerous discussions and review by Drs Franklin Bunn, Kim Vandergriff, Jack Levin, Marcos Intaglietta, Michele Samaja, and Abdu Alayash, among others. Although Winslow was best known to the transfusion medicine community for his work with hemoglobin-based oxygen carriers, his original work in several related areas and his conclusions regarding the physiologic control of oxygen delivery should be of great interest to the readers of TRANSFUSION. We anticipate that this review will be a major contribution to a complex literature and appropriately, considering Winslow's decades of contributions to this field, will constitute a worthy epitaph for his distinguished scientific career. Because Robert Winslow's untimely death precluded his usual meticulous attention to the final details and prevented any response to reviewers' suggested revisions, the final manuscript process was organized by Nancy J. Winslow, prepared for publication by Harvey G. Klein, and accepted for publication without revision.—Paul M. Ness, MD.

Harvey G. Klein, MD, on behalf of the late Robert M. Winslow, MD, National Institutes of Health, Building 10, Room 1C-711, Bethesda, MD 20892; e-mail address:


Cell-free hemoglobin (Hb) has been blamed for a spectrum of problems, including vasoconstriction pancreatitis, myocardial infarction, and pulmonary hypertension in hemolytic anemia, malaria, and sickle cell anemia, and from Hb-based oxygen carriers (HBOCs). Toxicities have been attributed to scavenging of nitric oxide (NO). However, while NO scavenging may explain many in vitro effects, and some effects in animal models and clinical trials with HBOCs, key inconsistencies in the theory require alternative explanations.

This review considers the hypothesis that cell-free Hb oversupplies oxygen to tissues, leading to oxygen-related toxicity, possibly through formation of reactive oxygen species and local destruction of NO. Evidence for this hypothesis comes from various sources, establishing that tissue oxygen levels are maintained over very narrow (and low) levels, even at high oxygen consumption. Tissue is normally protected from excessive oxygen by its extremely low solubility in plasma, but introduction of cell-free Hb, even at low concentration, greatly augments oxygen supply, engaging protective mechanisms that include vasoconstriction and ischemia.

The requirement to limit oxygen supply by cell-free Hb suggests novel ways to modify it to overcome vasoconstriction, independent of the intrinsic reaction of Hb with NO. This control is essential to the design of a safe and effective cell-free HBOC.