Volume 20, Issue 2 , Pages 133-135, April 2006
Is Supplemental Oxygen Necessary?
Article Outline
DR SINHA et al1 have shown that even short-term exposure to 100% inspired oxygen may cause a prolonged decrement in postoperative pulmonary function. The authors provide an acceptable explanation for their findings by using proven physiologic principles. Based on their observations, the authors conclude that provision of 50% inspired oxygen is preferable to a higher FIO2 after cardiopulmonary bypass. Inherent in this suggestion is the basic assumption that some degree of supplementation of inspired oxygen is desirable and/or necessary. Is that so?
The rationale for supplementation of inspired oxygen is well known to nearly all healthcare providers. Whether provided in emergency care situations, coronary care units, post-anesthesia recovery areas, or operating rooms, supplemental oxygen is administered in an attempt to provide a safety net and to prevent detrimental hypoxemia and tissue hypoxia. Undoubtedly, the physiologic basis for this clinical decision is based on interpretation of the O2-hemoglobin (Hgb) dissociation curve (Fig 1). I was taught, as were my teachers, that “small changes in oxygen tension will cause a dramatic and marked decrement in oxygen saturation” and further “such a dramatic decrease in oxygen saturation likely will cause tissue hypoxia.” Few would dispute the accuracy of these assumptions. However, it is indisputable that there is little or no evidence to support the second part of this assumption. In fact, whether one believes in “intelligent design” or evolution, to accept the accuracy of the assumption would be to admit a gross flaw in design of the Hgb molecule or that the molecule evolved a trait with little or no survival benefit. I believe the clue to solution of this conundrum rests in the methodology used to construct the O2-Hgb dissociation curve. By using a tonometer with known PO2 values, saturation and/or content of blood samples was determined, resulting in the well-known relationship shown in Figure 1. However, it is important to understand that the only organ in humans that acts as a tonometer is the lung (ie, only in lung does PO2 determine oxygen content/saturation). Everywhere else, oxygen extraction by the tissues determines the resultant O2 content/saturation and PO2 is the dependent variable. Had we been taught the relationship between PO2 and O2 content/saturation as a physiologic function, we might draw a different conclusion (Fig 2). Based on this relationship, we would conclude that large amounts of oxygen might be extracted from the Hgb molecule, with minimal change in PO2. Or, as stated by Comroe,2 “… the steep middle and lower parts protect the tissues by enabling them to withdraw large amounts of O2 from blood for relatively small decreases in PO2.” Thus, the Hgb molecule serves to protect the organism by maintaining the diffusion gradient from blood to the tissues, even when the Hgb is markedly desaturated.
Fig 1.
The normal oxyhemoglobin dissociation curve with oxygen saturation dependent on the partial pressure of oxygen.
Fig 2.
The same data as in Figure 1 presented to represent function of the Hgb molecule at the tissue level.
Let us then address the original question. What benefit is provided by maintaining arterial hemoglobin saturation greater than 90% (PaO2 >60 mmHg)? Clearly, we cannot claim a benefit based on increased oxygen delivery (the product of arterial oxygen content and cardiac output) because the increase in oxygen content provided by a few percentage points in arterial oxygen saturation is insignificant relative to Hgb concentration, cardiac output, and the arterial oxygen saturation existent when the patient breathes room air. Therefore, the only remaining rationale must be based on incorrect interpretation of the O2-Hgb curve. For years, this argument has been met with a response similar to “even if there is no proven benefit, supplementation of inspired oxygen should occur until adverse effects of such therapy are proven.” In other words, the converse of evidence-based medicine. In the absence of supplemental oxygen, arterial hypoxemia would be detected with a pulse oximeter and would indicate the presence of lung dysfunction. Although arterial hypoxemia might be reversed with supplementation of inspired oxygen, presence of the primary pulmonary derangement would be masked or made worse. In the absence of supplemental oxygen therapy, the clinician likely would direct appropriate therapy toward the primary pulmonary defect, be it hypoventilation, V/Q mismatching, right-to-left intrapulmonary shunt, or diffusion defect.3, 4
Although not specifically noted, the authors provide compelling evidence for omission of postoperative supplemental inspired oxygen. For any given right-to-left intrapulmonary shunt fraction, PaO2/FIO2 increases with decrease in FIO2 when saturation falls below 100% (Fig 3). In other words, the Hgb molecule serves to prevent a continued fall in PaO2, as alveolar oxygen tension decreases due to decreasing inspired oxygen concentration. Therefore, the reported PaO2/FIO2 of 300 mmHg to 500 mmHg indicates that the PaO2 of the authors’ patient’s breathing room air would have been 75 to 80 mmHg!
Fig 3.
PaO2/FIO2 as a function of FIO2, with right-to-left intrapulmonary shunting of blood of 20% to 50%.
If it were not for the physiologically unfounded fear of the “slippery slope” of the oxyhemoglobin dissociation curve, supplementation of inspired oxygen appropriately would be relegated to the treasure trove of physiologically unsound maneuvers, subsequently proven to be of no benefit. Intermittent positive-pressure breathing, blow gloves, incentive spirometry, and leeches all, at one time, were believed to be logical therapeutic maneuvers and were held in great favor until proven to be physiologically unsound. Perhaps, in my lifetime, investigations such as that by Sinha et al1 will lead to the appropriate application of oxygen therapy.
References
- Effect of lung ventilation with 50% oxygen in air or nitrous oxide versus 100% oxygen on oxygenation index after cardiopulmonary bypass . J Cardiothorac Vasc Anesth . 2006;20:136–142
- . Physiology of Respiration . (ed 2).. Chicago, IL: Yearbook Medical Publishers; 1974; p 188
- Supplemental oxygen impairs detection of hypoventilation by pulse oximetry . Chest . 2004;126:1552–1558
- . Has oxygen administration delayed appropriate respiratory care? Fallacies regarding oxygen therapy . Respir Care . 2003;48:611–620
PII: S1053-0770(06)00042-5
doi:10.1053/j.jvca.2006.01.009
© 2006 Elsevier Inc. All rights reserved.
Volume 20, Issue 2 , Pages 133-135, April 2006
