Hyperoxia-induced Decrease in Organ Blood Flow

We thank Dr. Her for his comments on our article.1He notes that a previous human study by Floyd et al.  demonstrated that the magnitude of the decrease in cerebral blood flow (CBF) induced by 100% oxygen administration is more profound than that of the increase in arterial oxygen content,2which seems to support the notion that hyperoxia can induce tissue hypoxia. This study, which used arterial spin labeled-perfusion magnetic resonance imaging, reported a 29-33% decrease in CBF during 100% oxygen breathing. If the 30% reduction were accurate, then Dr. Her may have a point. However, this reduction in CBF is significantly greater than what has been observed in other studies, in which the reduction has been on the order of 5–15%.3–5Unfortunately the investigators in the Floyd et al.  study made unwarranted assumptions. For instance, they assumed that T1 for blood was the same during air and oxygen breathing despite published data to the contrary.6This component of the lumped constant in their calculation of CBF leads to a 36% difference based on the T1 of blood with air versus  oxygen breathing. Further, they assumed T1 in tissue to be similar to that in blood and suggest that violations of this assumption would cause only a small error. This is clearly not the case and is the reason for the overestimation of changes in CBF found in their study. In a recent article by Bulte et al. ,5in which 100% oxygen reduced CBF less than 10%, the changes in relaxation times of blood and tissue with increased Fio²are confounders of the arterial spin labeled technique, and that failure to accurately account for them when calculating perfusion will lead to gross overestimation of hyperoxia-induced blood flow changes.

We agree with Dr. Her that acute hypocapnia can be associated with markers of ischemia, such as impaired psychomotor performance, at least in the arterial Pco²range 20–25 mmHg.7However, our original manuscript included data showing that during spontaneous breathing, the tendency of 100% oxygen administration to cause hypocapnia is either very small or does not exist.1Tissue Po²regulates blood flow such that, although oxygen administration does induce vasoconstriction, there is a monotonically increasing  relationship between arterial and tissue Po².8–10This has also been observed in the retina.11In a pig study, the administration of 100% oxygen reduced retinal blood flow by 62% but increased periarteriolar and intervascular Po².12It is therefore unlikely that hyperoxia could contribute to retinal ischemia induced by an increase in intraocular pressure, and indeed there is direct evidence to the contrary.13 

Oxygen administration also reduces coronary blood flow; however, in the study cited by Dr. Her,14there is no evidence that the reduced coronary blood flow induced either ischemia or myocardial hypoxia.

With regard to Dr. Her’s comment on oxygen and postoperative wound infections, evidence suggests an inverse relationship between tissue Po²and infection rate,15and the bulk of evidence supports the use of supplemental perioperative oxygen to reduce wound infection rate.16,17The study by Pryor and colleagues cited by Dr. Her,18which failed to demonstrate a beneficial effect, has been criticized on methodological grounds.19,20 

In summary, peripheral blood flow is regulated to maintain tissue oxygenation in the face of alterations in oxygen delivery. There is no evidence that the autoregulatory decrease in tissue blood flow during hypoxia induces tissue ischemia or hypoxia.

*Center for Hyperbaric Medicine & Environmental Physiology, Duke University Medical Center, Durham, North Carolina. moon0002@mc.duke.edu