Oxygen Induced Hypercapnia

During the resuscitation of hypoxemic patients, as tensions are high, it is often easy to overlook the possible deleterious effects of oxygen therapy. However, during the resuscitation of a patient presenting with an acute exacerbation of COPD, this oversight could be detrimental. While using a carefully titrated supplemental oxygen regimen for patients in a suspected acute exacerbation of underlying COPD has become common practice, I have found it is often not being practiced commonly. As clinicians, we are trained to titrate the level of supplemental oxygen to a measured pulse oximetry reading of 88-92% in these patients. We give just enough oxygen to prevent hypoxemia without causing dangerous levels of hypercarbia. But Why? What does the literature show? 

Historical Background:

An article published in 1949 described oxygen-induced neurological symptoms, including one fatal coma in patients with cyanosis due to COPD.¹They designed a study of 4 patients with COPD and cyanosis and provided them with Oxygen therapy. They observed that the intracranial pressure would increase during oxygen therapy and return to baseline when oxygen therapy was discontinued. Investigators hypothesized that oxygen intoxication could have led to an accumulation of carbon dioxide (CO₂) in the body and cautioned against the use of oxygen in these patients. 

That same year, an article published in the Lancet by described a hypercapnic coma during oxygen therapy, which clinically improved after oxygen therapy was discontinued.²This brought rise to the “hypoxic drive theory,” suggesting there are 2 central drivers of the respiratory effort.Hypercarbia and hypoxemia. The thought here is that COPD patients are chronically slightly hypercarbic, so they no longer respond to elevated blood CO₂as a stimulus. Their only trigger for a positive respiratory drive is hypoxemia. This suggest that providing uncontrolled amounts of supplemental oxygen to a patient with COPD will remove the patient’s hypoxic respiratory drive causing hypoventilation, worsening hypercapnia and ultimate leading to fulminant respiratory failure.⁴

Hypoxic Drive

The first study to really investigate this theory was done in 1980 by Aubier, et al.³They looked at 22 ICU patients with COPD in respiratory failure and measured the minute ventilation and PaCO2 in all patients first while breathing room air and then while breathing supplemental oxygen.

What they discovered is that, when placed on supplemental oxygen, patients had an initial drop in minute ventilation (V_E). That is, they started to hypo-ventilate. At the same time, the amount of dissolved CO₂in the blood started to rise. However, the minute ventilation soon recovered to near baseline levels, while the PaCO₂continued to increase. So, there was really no direct correlation between the patients change in minute ventilation and the observed increase in PaCO₂. 

A second study published by Aubier in 1980 again showed a small, but significant decrease in minute ventilation.⁴However, tidal volume in this study did not change between patients treated with oxygen vs room air. Remember, minute ventilation is calculated by multiplying the tidal volume (V_T) by the respiratory rate (ƒ). 

V_E= V_Tx ƒ

Therefore, the change in minute ventilation was felt to be secondary to the decreased in inspiratory flow rate between the two groups. In this study, respiratory drive was determined to by mouth occlusion pressure and this was actually increased above that of normal subjects in both groups. So, the authors concluded that while minute ventilation did change, this really had no significant effect on the patients overall respiratory drive. 

These early studies showed hypoventilation due to loss of hypoxic respiratory drive, while plausible, is certainly not the major contributor to the development of hypercarbia after oxygen administration.⁵So, something else must be going on. 

The Haldane Effect

The ability of deoxygenated Hb to bind CO₂is much higher than that of oxygenated Hb. where HbCO₂ is carbaminohemoglobin and HbO₂ is oxyhemoglobin.

HbCO₂ ↔ O₂ × HbO₂ + PaCO₂

Providing supplemental oxygen shifts the equilibrium between deoxygenated and oxygenated Hb more towards the oxygenated form. While this would intuitively be a good thing in a hypoxemic patient, this also reduces the amount of CO₂that can be bound to Hb. That CO₂ is then dissolved in the blood, resulting in an increased PaCO₂. This is known as the Haldane effect.⁵ 

This increase in PaCO₂is normally excreted through a compensatory increase in minute ventilation (or hyperventilation). However, patients with very severe COPD are often unable to adequately increase their minute ventilation resulting in an increase in PaCO₂. Early studies on this phenomenon suggests that the Haldane effect only explained about 25% of the total increase in PaCO₂due to oxygen administration³. So, there must be something more to blame still.

Ventilation-Perfusion (V/Q) Mismatch

Under normal physiology, alveolar ventilation and perfusion should be pretty well matched. Two extremes of ventilation-perfusion (V/Q) mismatch may occur.⁵

First, when there is poor ventilation of an alveolus but there is adequate perfusion. This occurs when oxygen rich air is not getting to the capillary beds of the alveolus even though there may be a good supply of red blood cells sitting in the capillary beds waiting for it to arrive. This is called shunting. Think of red blood cells as the box cars of a train. They show up to the station and there are no passengers to get on. The train wasted all that time and energy and have nothing to transport. It shunted all of those resources to a station with no benefit. 

The second way in which VQ mismatch occurs is when there is adequate ventilation but there is no perfusion. This occurs when there is plenty of air flowing into the alveolus, but the oxygen can’t be transported into the bloodstream. We call this dead space ventilation. Think of the oxygen molecules as people waiting for a train, but the train never comes. 

The body has protective mechanisms to optimize the V/Q ratio.⁵When oxygen tension within an alveolus (PAO₂) decreases, the pulmonary capillaries supporting this particular alveolus to vasoconstrict. This is an attempt to minimize shunting, a mechanism called hypoxic pulmonary vasoconstriction.

The alveolar partial pressure of oxygen (PAO₂) is the strongest mediator for hypoxic pulmonary vasoconstriction.⁵Providing a patient with a higher than needed fraction of inspired oxygen (FiO₂) will increase oxygen tension in alveoli that have a low level of ventilation and inhibit hypoxic pulmonary vasoconstriction. In alveoli in other parts of the lung that have more severely impaired ventilation but adequate perfusion, a worsening of the V/Q mismatch will occur.⁵

A study published in 2000 also studied the V/Q mismatch during oxygen therapy.⁶They looked at 22 patients presenting with an Acute exacerbation of COPD who received an IV infusion of an inert gas. Investigators measured the makeup of expired air and ABG readings in each patient while breathing room air, and then while breathing 100% supplemental O₂. They classified patients into 2 distinct groups based on what happened to their CO₂levels while on supplemental oxygen. They called them ‘retainers’ if their CO₂increased by more than 3mmHg while on supplemental O₂and as ‘non-retainers’ if it did not. What they found was quite interesting. V/Q mismatch increased significantly in both groups. Minute ventilation decreased significantly in the retainers but not in the non-retainers. Finally, alveolar dead space ventilation increased significantly in the retainer group but was unchanged in the non-retainers. So, Since V/Q mismatch increased in both groups, the authors concluded that V/Q mismatch could not be the cause of observed CO₂retention. Further, because minute ventilation decreased in the retainer group but remained stable in the non-retainers, they proposed that oxygen induced hypoventilation could be to blame. This contradicts the study from Aubier that we looked at earlier.³Investigators attempted to explain this contradiction, suggesting that perhaps the decrease in observed minute ventilation from the 1980 study just wasn’t quite large enough to account for the level of observed CO₂ retention. However, this conclusion is likely poorly interpreted because although overall ventilation decreased in the retainer group, ventilation to alveoli with higher V/Q mismatch was increased, leading to increased alveolar dead space ventilation in the retainer group. It is plausible that although V/Q mismatch increased in both groups, it only caused hypercarbia in the retainer group because of the increased dead space ventilation!⁵

To back this up, A 1996 study published in critical care medicine compared data derived from a computer model of the pulmonary circulation with data from a case series of patients with COPD to evaluate the specific factors contributing to CO₂retention due to oxygen therapy.⁷They confirmed the supplemental oxygen alters hypoxic pulmonary vasoconstriction and modulates the Haldane effect, resulting in changes in physiologic dead space. Most significantly, this increase in physiologic dead space was sufficient enough to account for the oxygen-induced hypercapnia that has been described in literature ever since the late 1940’s.

Conclusion

So, what do we do with this knowledge? Well, literature has shown us that patients most susceptible to oxygen-induced hypercapnia are those with the most severe levels of hypoxemia on presentation.^3,6So, how can we avoid worsening hypercarbia without withholding oxygen therapy in the setting of hypoxemia? 

3 studies were published between 2002 and 2011 on the topic of pre-hospital administration of high-flow oxygen concentrations in COPD patients^{8, 9, 10}. They showed that Increased oxygen flow was associated with increasing risk of death, assisted ventilation or respiratory failure with an odds ratio of 1.2 for every 1 L/min of oxygen flow used.

National guidelines for the management of the acute exacerbation of COPD recommend an initial FiO₂ of no more than 2 L/min on a nasal cannula. These studies showed that oxygen therapy with an FiO₂ in excess of this is commonly used, may worsen hypercapnia and is associated with a higher mortality. Recognizing patients at risk and understanding this disease process is the first step in improving the acute care of these patients. 

Carefully titrating oxygen to achieve a pulse oximetry reading of 88% to 92%in acute COPD exacerbations has been shown to avoid hypoxemia and reduce the risk of oxygen-induced hypercapnia.¹¹ This results in better outcome when compared to higher oxygen saturations.¹²

It is important to point out that this strategy only applies to patients with known or suspected respiratory failure due to an acute COPD exacerbation. You should always approach your patients with an evidence based mindset and remember that education is not just the learning of facts, but training the mind to think. Einstein said that…

Author: Nicholas McManus, DO

References

1. Davies CE, Mackinnon J. Neurological effects of oxygen in chronic cor pulmonale. Lancet. 1949;16:883–885. [PubMed]
2. Donald K. Neurological effects of oxygen. Lancet. 1949;16:1056–1057.
3. Aubier M et al. Effects of the administration of O2 on ventilation and blood gases in patients with chronic obstructive pulmonary disease during acute respiratory failure. Am Rev Respir Dis.1980;122(5):747–754.
4. Aubier M, Murciano D, Fournier M, Milic-Emili J, Pariente R, Derenne JP: 
   Central respiratory drive in acute respiratory failure of patients with chronic obstructive pulmonary disease. Am Rev Respir Dis 1980, 122:191-199. PMID: 67746339
5. Abdo WF, Heunks LMA: Oxygen-induced hypercapnia in COPD: myths and facts. Critical Care 2012, 16:323. PMID: 23106947
6. Robinson et al. The role of hypoventilation and ventilation-perfusion redistribution in oxygen-induced hypercapnia during acute exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2000;161(5):1524–1529.
7. Hanson CW, Marshall BE, Frasch HF, Marshall C. Causes of hypercarbia with oxygen therapy in patients with chronic obstructive pulmonary disease. Crit Care Med. 1996;16:23–28. doi: 10.1097/00003246-199601000-00007. [PubMed] [CrossRef]
8. Denniston AK, O’Brien C, Stableforth D. The use of oxygen in acute exacerbations of chronic obstructive pulmonary disease: a prospective audit of pre-hospital and hospital emergency management. Clin Med. 2002;16:449–451. [PMC free article] [PubMed]
9. Durrington HJ, Flubacher M, Ramsay CF, Howard LS, Harrison BD. Initial oxygen management in patients with an exacerbation of chronic obstructive pulmonary disease. QJM. 2005;16:499–504. doi: 10.1093/qjmed/hci084. [PubMed] [CrossRef]
10. Wijesinghe M, Perrin K, Healy B, Hart K, Clay J, Weatherall M, Beasley R. Pre hospital oxygen therapy in acute exacerbations of chronic obstructive pulmonary disease. Intern Med J. 2011;16:618–622. doi: 10.1111/j.1445-5994.2010.02207.x. [PubMed] [CrossRef]
11. Austin MA, Wills KE, Blizzard L, Walters EH, Wood-Baker R. Effect of high flow oxygen on mortality in chronic obstructive pulmonary disease patients in prehospital setting: randomised controlled trial. BMJ. 2010;16:c5462. doi: 10.1136/bmj.c5462. [PMC free article] [PubMed] [CrossRef]
12. O’Driscoll BR, Howard LS, Davison AG. BTS guideline for emergency oxygen use in adult patients. Thorax. 2008;16(Suppl 6):vi1–68. [PubMed]