Painstaking titration of oxygen flows with elaborate precision is often performed in carbon dioxide retainers with chronic obstructive pulmonary disease (COPD). The driving principle behind this line of thinking is the tradition-borne belief that supplemental oxygen may inhibit the hypoxic respiratory drive and lead to hypoventilation with rise in the PaCO2 levels. It has even been reasoned that hypoxia was a respiratory stimulant in these patients through the sino-aortic nerve activity.1 How important is the contribution of the hypoxic drive in patients with respiratory failure, especially among those with hypercapnia? Does supplemental oxygen really suppress ventilation in spontaneously breathing patients with COPD and lead to a significant rise in CO2 levels? Could other mechanisms including an increase in the dead space and the loss of hypoxic pulmonary vasoconstriction be more likely to induce hypercapnia?
Anatomical dead space
The anatomical dead space comprises of the airways that allow passage of air to the respiratory bronchioles and alveoli. Gas exchange does not occur in this part of the respiratory system. The volume of the anatomical dead space is approximately 150 ml. Dead space volume / tidal volume (Vd/Vt) represents the proportion of dead space relative to the tidal volume. With low tidal volume breaths, the relative proportion of the dead space increases (high Vd/Vt), and may lead to hypercapnia if the minute ventilation is not increased.
Alveolar dead space
The alveolar dead space refers to a reduction in perfusion to ventilated alveoli (high V/Q).2 Thus, alveolar dead space representing areas of adequate ventilation but impaired perfusion is in direct contrast to areas of the lung that are poorly ventilated, with normal perfusion (low V/Q, leading to shunt) In most parenchymal disease processes, the mechanism for hypercapnia is an increase in the alveolar dead space.
Physiological dead space
The combination of anatomical and alveolar dead spaces comprises of the physiological dead space.2
Physiological dead space (Vd phys) = Anatomical dead space + alveolar dead space
Oxygen administration and respiratory drive: the evidence
Aubier et al. analyzed the breathing pattern of 20 patients with COPD during the acute phase of respiratory failure; in 12 patients, further analysis was carried out after recovery from the acute phase.3 The results were compared with those of normal subjects in a similar age group. The minute ventilation was similar during the acute and chronic phase and was comparable to those of normal subjects. The respiratory drive, measured using mouth occlusion pressure, was fivefold higher during the acute phase compared to normal subjects. Following supplemental oxygen, the mouth occlusion pressure decreased by 40%, suggesting a significant reduction of the respiratory drive. Supplemental oxygen at 5l/min for 30 minutes resulted in a 14% reduction in the minute ventilation, due to a reduced respiratory rate, indicating lower inspiratory flows. The PaCO2 rose after oxygen administration, although the rise was not commensurate with the decrease in minute ventilation, suggesting other possible mechanisms for the hypercapnia. The authors proposed that an increase in dead space could explain hypercapnia that followed oxygen supplementation.
In a later study, Aubier et al. evaluated changes in ventilation and arterial blood gases on 22 patients with acute infective exacerbation of COPD, following switch over from room air to 100% oxygen.4 Patients were hypoxic (mean PaO2 38 ± 2 mm Hg) and hypercapnic 65 ± 3 mm Hg while breathing room air. Respiratory parameters were monitored throughout the 15-minute period of breathing 100% oxygen. An initial decrease in minute ventilation was observed following administration of 100% oxygen; the lowest value was noted between 20–180 seconds of oxygen inhalation. Following the initial decrease, the minute ventilation gradually increased and plateaued off about 12 minutes after commencement of 100% oxygen administration. The plateau level of minute ventilation at the end of the 15-minute study period was 93 ± 6% of the baseline level. However, the PaCO2 continued to rise during the entire study period and increased by a mean of 23 ± 5 mm Hg (Fig 1). The PaO2 increased from a mean baseline level of 38 ± 2 mm Hg to 225 ± 23 mm Hg.
No significant change was observed in the tidal volume or respiratory rate with inhalation of 100% oxygen. The rise in PaCO2 was disproportionately high and could not be explained by the marginal decrease in minute ventilation. During acute exacerbation of COPD, a marked increase in the respiratory drive is commonly observed. The above studies suggest that although the respiratory drive may decrease marginally following oxygen supplementation, it still remains higher than normal. The decrease in respiratory drive alone does not explain the rise in PaCO2 levels.3
The Haldane effect
Hemoglobin carries carbon dioxide as carbamino compounds. According to the Haldane effect, the carbon dioxide dissociation curve is shifted to the right as the oxyhemoglobin level rises.5 Thus, oxygenated hemoglobin carries less carbon dioxide. In the normal lungs, as the hemoglobin gets oxygenated to oxyhemoglobin, the carbon dioxide is released and expired out. Consider the case of a patient with COPD. Oxygen administration would lead to release of carbon dioxide from the hemoglobin as expected. However, the expiration of carbon dioxide released from the hemoglobin remains inadequate as there is no proportional increase in ventilation in COPD patients. This leads to an increase in the PaCO2 levels.6
Loss of the hypoxic pulmonary vasoconstrictor effect
When the partial pressure of oxygen in the alveoli drops, the smooth muscle of the pulmonary circulation contract, resulting in vasoconstriction.7 This is a reflex mechanism to reduce intrapulmonary shunt in the presence of hypoxia. The vasoconstrictor response of the pulmonary vessels in response to hypoxia is in contrast to the hypoxia-induced vasodilator effect that occurs in the systemic circulation.8 On administration of supplemental oxygen, the alveolar partial pressure of oxygen rises. The rise in oxygen level leads to inhibition of the hypoxic pulmonary vasoconstrictor response, leading to a redistribution of blood flow from better ventilated to relatively less ventilated areas of the lung, with a regional reduction in the V/Q ratio. In contrast, in the better-ventilated areas of the lung (from which the blood was diverted away), the perfusion is reduced relative to ventilation leading to a higher V/Q ratio and an increase in the dead space ventilation. Aubier et al. also observed a significant increase in the dead space ventilation from 77 ± 2 to 82 ± 2 following 15 minutes of breathing 100% oxygen in their study.6 Both the above mechanisms may contribute to a rise in PaCO2 levels.
Robinson et al. evaluated the mechanism of oxygen-induced hypercapnia among 22 patients with acute exacerbation of COPD.9 Patients were allowed to breathe air initially, followed by 100% oxygen through a nasal mask for at least 20 minutes. Ventilation, V/Q ratio, and cardiac output were measured while breathing air and 100% oxygen. Patients were classified as retainers if the PaCO2 rose by more than 3 mm Hg and non-retainers if the PCO2 rise was less than 3 mm Hg. Among 12 patients in the retainer group the PaCO2 level rose by 8.3 ± 5.6; in the non-retainer group, the PaCO2 change was –1.3 ± 2.2 mm Hg. The minute ventilation decreased from 9.0 ± 1.5 to 7.2 ± 1.2 L/min in the retainer group after 20 minutes of breathing 100% oxygen. However, the rise in CO2 seemed disproportionate to the decrease in minute ventilation. The V/Q mismatch was of the same magnitude in both the groups, likely due to inhibition of the hypoxic pulmonary vasoconstrictor response. A significant increase in dead space ventilation was observed in the retainer group, which may have resulted in the rise in PCO2.
Hanson et al. compared data derived from a simulated computer model of the pulmonary circulation with data from a series of patients with COPD.10 The model with simulated oxygen therapy generated data comparable to COPD patients treated with supplemental oxygen. The authors observed that oxygen administration led to an increase in the physiological dead space resulting from inhibition of the hypoxic pulmonary vasoconstrictor response and through the Haldane effect. They concluded that the increase in physiological dead space alone was sufficient to explain the rise in PaCO2 observed in patients who received supplemental oxygen.
- There is a modest rise in PaCO2 levels in patients with acute exacerbation of COPD who receive supplemental oxygen
- There is a marginal decrease in minute ventilation associated with oxygen administration; however, the rise in PaCO2 level cannot be explained solely on the basis of hypoventilation
- Furthermore, the minute ventilation returns to baseline levels with minutes, although the PaCO2 levels continue to remain high, suggesting an alternate mechanism for the hypercapnia
- A more plausible mechanism for the rise in PaCO2 levels is through inhibition of hypoxic pulmonary vasoconstriction due to the higher alveolar oxygen tension resulting from supplemental oxygen. Release of hypoxic pulmonary vasoconstriction results in redistribution of blood to less ventilated areas, with an increase in the dead space and hypercapnia
- The Haldane effect results in reduced carbon dioxide carriage as the hemoglobin gets oxygenated in the lung. However, the carbon dioxide cannot be eliminated adequately as the ventilation remains low, contributing to the hypercapnia
- It is important not to withhold oxygen in patients who are hypoxic based on an irrational fear of suppressing the “hypoxic drive”
1. Davies CE, Mackinnon J. Neurological effects of oxygen in chronic cor pulmonale. Lancet. 1949;2(6585):883-885, illust. doi:10.1016/s0140-6736(49)91459-2
2. Intagliata S, Rizzo A, Gossman W. Physiology, Lung Dead Space. In: StatPearls. StatPearls Publishing; 2021. Accessed January 2, 2022. http://www.ncbi.nlm.nih.gov/books/NBK482501/
3. 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(2):191-199. doi:10.1164/arrd.1922.214.171.124
4. Aubier M, Murciano D, Milic-Emili J, 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. doi:10.1164/arrd.19126.96.36.1997
5. Teboul JL, Scheeren T. Understanding the Haldane effect. Intensive Care Med. 2017;43(1):91-93. doi:10.1007/s00134-016-4261-3
6. Abdo WF, Heunks LM. Oxygen-induced hypercapnia in COPD: myths and facts. Crit Care. 2012;16(5):323. doi:10.1186/cc11475
7. Lumb AB, Slinger P. Hypoxic pulmonary vasoconstriction: physiology and anesthetic implications. Anesthesiology. 2015;122(4):932-946. doi:10.1097/ALN.0000000000000569
8. Farhi LE, Sheehan DW. Pulmonary circulation and systemic circulation: similar problems, different solutions. Adv Exp Med Biol. 1990;277:579-586. doi:10.1007/978-1-4684-8181-5_65
9. Robinson TD, Freiberg DB, Regnis JA, Young IH. 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. doi:10.1164/ajrccm.161.5.9904119
10. 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;24(1):23-28. doi:10.1097/00003246-199601000-00007
Top 35 critical care blogs from feedspot: https://blog.feedspot.com/critical_care_blogs/