Hyperoxic ventilation: Is it good or is it bad in other conditions?

A few days ago I performed anesthesia on a 68 y/o morbidly obese female with a history of HTN.  She was having a total shoulder arthroplasty.  Prior to the induction of anesthesia, an interscalene catheter was placed with US guidance.  GETA was induced, and the patient was position in a sitting position.  Unfortunately, shortly after the patient was placed in the sitting position, she developed a low blood pressure.  A small amount of ephedrine was given due to a low heart rate.  Her next blood pressure was not measurable.  More Ephedrine was given and the next blood pressure was 50s/x.  Phenylephrine was given and blood pressure came back up. 

During the time it took to get her blood pressure back up to a normal level, about 6 minutes passed.  Since the patient was in the sitting position, there was concern that the blood pressure at the level of the brain was significantly less than the pressure measured at the arm.  There have been a series of published case reports of patients suffering irreversible brain damage after relative hypotension in the sitting position.

I wondered if providing hyperoxic ventilation would mitigate this risk compared to normoxic ventilation.  Hyperoxic ventilation has been shown as discussed in the previous two posts on this topic, to ameliorate the effects of severe dilutional anemia in both pigs and humans.  In healthy volunteers, Weiskopf et al. performed hemodilution on volunteers down to a Hgb of 5 g/dL where subtle neurocognitive deficits were detected.  These changes could be reverse with hyperoxic ventilation [1]. This demonstrates that, given significant anemia, if adequate volume is maintained, changes to neurons due to lack of oxygen can be resolved by the  dissolved O2 content in arterial blood.  Meier et al. [2] looked at the effects of hyperoxic ventilation in pigs who were bled until there were clear manifestations of VO2 dependency on DO2 (i.e. shock).  After one hour in this state, the pigs were paritally resuscitated with a colloid solution.  One group received hyperoxic ventilation (achieving a PaO2 of 500 mmHg).  The other group was maintained on RA. All of the animals died in the control group, while 75% of the other animals survived that were given HV.  The authors noted that in their study, DO2 was not increased in the group that was given FiO2 of 1.0, however, they speculated that since the O2 that did arrive at the tissues was more usable (greater utilizability of O2) because it was not bound to Hgb.  The authors point out that in previous studies of hemorrhagic shock, HV failed to improve survival without resuscitation. The concept of greater O2 usability is explored carefully in the Part I of this series on hyperoxic ventilation.

In another study [3], Methemoglobinemia was induced in pigs and once again, a control group was given RA while the experimental group was given FiO2 1.0. Although survival time was increased by 100 min in the group given HV, the difference in survival was not different at 6 hours.  The authors concluded that at physiological Hgb levels, HV fails to provide the same benefit as that in severe anemia.

To understand why oxygen may be beneficial when given in the setting of severe isovolemic anemia or in partially resuscitated shock, but not in methemoglobinemia, it was must be understood how HV affects the vasculature.  At normal Hgb levels i.e. ~9 g/dL or so, hyperoxia causes arteriolar vasoconstriction and microcirculatory failure.  Arteriolar vasoconstiction can be devastating in patients who already are suffering from peripheral vasoconscriction from hemorrhagic shock without fluid resuscitation, and this explains why oxygen therapy is not useful and probably harmful in the setting of unresuscitated hypovolemic shock.  Of course, this is directly pertinent to anesthesiologists who are accustomed to providing 100% oxygen prior to induction even to patients whose volume status may be questionable. 

In isovolemic anemia (when the Hgb declines to about 7 g/dL), the compensatory changes are a decrease in systemic SVR by vasodilation, improved rheologic properties, with increased blood flow velocity.  The vasodilation is thought to be related to increased release of NO from endothelial cells damaged by increases in shear stress on the vessel walls. This vasodilational state is reveresed by the hyperoxia.  Thus, oxygen has the potential to not only improve CaO2, but also MAP.  This is has been shown in multiple studies looking at isovolemic anemia with and without hyperoxia. .   The negative vasoconstrictive effects of oxygen are well known and prompted a review of the literature on its routine use in acute MI patients [4]. Only one study could be found that met the inclusion criteria, and this study found that mortality was higher in patients given O2 compared to those who did not receive O2 (11.3% vs. 3.9%).  This increase in mortality did not reach statistical significance.  The authors felt that oxygen therapy was probably not helpful, and may be harmful for patients suffering acute myocardial ischemia.  In the cath lab direct measurements of the coronary vasculature found that after institution of hyeroxic ventilation, the small coronary vessels had an increase in resistance of 40% with a 30% decrease in coronary blood flow.

Fortunately, anesthetics are direct vasodilators.  Therefore, giving 100% oxygen during a general anesthetic with propofol or potent vapors, are unlikely to result in ischemia.  In fact, providing oxygen, can improve hemodynamics as it has been shown to increase SVR and mean aortic pressure improving CBF in the face of hypotension induced by sodium nitroprusside.

So, given the above scenario of significant hypotension after induction of anesthesia with propofol and breathing inhalational agents, it should be expected that hyperoxic ventilation should counteract the vasodilatory properties of the anesthetic agents, improve mean aortic pressure benefitting coronary perfusion; and therefore, would be potentially beneficial and likely not harmful.  Furthermore, after induction a typical reduction in FRC and increased atelectasis results in lower PaO2/FiO2 ratios and this is particularly true in obese patients. So, achieving very high PaO2 is not likely possible during many general anesthetics.  However, in the setting of uncompensated hypovolemic shock, giving 100% oxygen prior to fluid resuscitation could be harmful as peripheral vasoconstriction would be aggravated.




1. Weiskopf RB et al. Oxygen reverses deficits of cognitive function and memory and increased heart rate induced by acute severe isovolemic anemia. Anesthesiology 2002;96:871
2. Meier J, Kemming GI, Kisch-Wedel H, Blum J, Pape A, Habler OP. HV ventilatino reduces 6-hr mortality after partial fluid resuscitation from hemorrhagic shock.  Shock 2004;22(3):240-47.
3. Meier J, Pape A, et al. Hyperoxia in lethal methemoglobinemia: effects on oxygen transport, tissue oxygenation, and survival in pigs. Crit Care Med 2005;33:1582-8.
4. Moradkhan R and Sinoway LI.  Revisiting the role of oxygen therapy in cardiac patients. J Amer Coll Card 2010;56(13):1013-16.
5. Suttner SW et al. The influence of hyperoxic ventilation during sodium nitroprusside-induced hypotension on skeletal muscle tissue oxygen tension.   Anesthesiology 2002;96:1103-8.