How much oxygen do I give?

  • The objective of oxygen therapy is to give the patient as much oxygen as is required to return the PaO2 to what is normal for the particular patient.

There is no secret to this – you give as much oxygen as is required to return the PaO2 to what is normal for the particular patient. You perform a therapeutic maneuver – giving Oxygen, and you measure the result, by performing serial blood gases. There is nothing to be gained by giving too much oxygen, and a huge amount to be lost by not giving enough.

The initial inspired concentration of oxygen depends on the clinical circumstances – if the patient is only mildly hypoxemic, saturating in the late 80s, then small amounts of supplemental oxygen given by nasal cannulae are all that is necessary. However, if the patient is in-extremis, then always start with 100% (or thereabouts) and work downwards.

 Shouldn’t I be careful about the amount of oxygen that I give COPD patients?

There is a universal misnomer that if you give too much oxygen to patients with COPD that they stop breathing, and hence medical and nursing students are often taught that COPD patients should not be given more than 28% oxygen because their respiratory drive is oxygen dependent (due to chronic CO2 retention) and they will lose their stimulus to breath. Physicians will cite rising CO2 levels in patients treated with oxygen as evidence of this.

There is a fundamental flaw in this theory: throughout this tutorial we have discussed the mechanisms by which oxygen is prevented from entering the blood. It is the blood oxygen content that is important, not the inspired fraction. Patients, depending on the extent of disease, will have differing extents of ventilation-perfusion mismatch and diffusion defects: the patient needs enough inspired oxygen to return the PaO2 to what is normal for them, and the way to establish this is by starting high and working downwards with serial blood gases.

We know that high CO2 levels are well tolerated by the body, but hypoxia is not: withholding oxygen therapy for fear of hypercarbia is negligent. It is not clear that such hypercarbia results, in any case, from hypoventilation: a number of studies (1;2)  have demonstrated that the increase in PaCO2 after administration of oxygen is due mainly to an increase in the ratio of dead space to tidal volume (Vd/Vt). This is probably due to reversal of hypoxic pulmonary vasoconstriction. Moreover, the increase in oxygenated hemoglobin leads to an increase in CO2 release by way of the Haldane effect.


   (1)    Crossley DJ, McGuire GP, Barrow PM, Houston PL. Influence of inspired oxygen concentration on deadspace, respiratory drive, and PaCO2 in intubated patients with chronic obstructive pulmonary disease. Crit Care Med 1997; 25(9):1522-1526.

   (2)    Sassoon CS, Hassell KT, Mahutte CK. Hyperoxic-induced hypercapnia in stable chronic obstructive pulmonary disease. Am Rev Respir Dis 1987; 135(4):907-911.


Copyright 2002 All rights reserved