Oxygen for Moderate COPD?

A Sliding Scale

Nicholas Gross, MD, PhD


June 27, 2017

A Randomized Trial of Long-term Oxygen for COPD With Moderate Desaturation

Long-Term Oxygen Treatment Trial Research Group, Albert RK, Au DH, et al
N Engl J Med. 2016;375:1617-1627.


It is well-known that patients with severe chronic obstructive pulmonary disease (COPD) are subject to premature mortality. An important trial showed that administering oxygen for 18 or more hours per day to patients with COPD with severe oxygen desaturation dramatically improved their survival.[1] Subsequent trials were performed to determine whether a survival benefit could also be obtained by oxygen therapy for patients with COPD with less severe hypoxemia. Those trials were inconclusive, largely because they were underpowered.[2,3]

Nevertheless, further studies were validated not only because there were indications that a subset of patients with moderately severe COPD might have benefitted by oxygen supplementation but also because patients with COPD with moderate hypoxemia were subject to acute exacerbations requiring hospitalization; the hope was that oxygen supplementation to patients with moderate hypoxemia could reduce the rate of hospitalization and be cost-effective, as well as improving clinical status.[4]

In addition, a trial with that aim might alternatively show that long-term oxygen was being overprescribed to patients who did not benefit from it.

Study Summary

To address the need to clarify which patients with less-than-severe hypoxemia would benefit from long-term oxygenation, a large clinical trial, the Long-Term Oxygen Treatment Trial (LTOTT), was performed.

Patients with COPD and moderate hypoxemia (oxyhemoglobin saturation as measured by pulse oximetry [Spo2], 89% - 93%) during a standard 6-minute walk test were randomly prescribed oxygen at 2 L/min, with the goal being to maintain Spo2 at or above 90% at rest and during the walk test. A similar group of patients with COPD, the control group, received ambient air and no oxygen supplementation. The primary outcome was a composite of time to death or first hospitalization. Other outcomes were incidence of acute exacerbations, development of severe hypoxemia either at rest or during a 6-minute walk, and two health status questionnaires: the Quality of Well-Being Scale and the St George's Respiratory Questionnaire.

Seven hundred thirty-eight patients with only mild or moderate hypoxemia were equally divided into supplemental oxygen and no supplementation and enrolled and followed for up to 6 years. At conclusion, the results showed no significant differences in the primary outcome between the two groups. Time to death or time to first hospitalization or first acute exacerbation were virtually identical. Similarly, quality of life, lung function, 6-minute walk distance, and functional status were not different between the two groups.

There were, however, some adverse events related to the use of oxygen supplementation. These were reports of tripping over equipment and six instances of fires or burns. The authors concluded that oxygen supplementation did not provide sustained benefit of any kind for patients with COPD who have only mild or moderate hypoxemia at rest or on effort. The conclusion is that long-term oxygen inhalation is not beneficial in patients with COPD who have less than severe arterial hypoxia at rest.


In and before the 1980s, there was no therapy that was known to prolong life in patients with severe COPD and resting hypoxemia. The finding that oxygen supplementation for 18 hours or more per day dramatically prolonged life of patients with severe COPD was seminal, if not sensational.[1] It was appropriate therefore to determine whether patients with less severe hypoxemia, who often have arterial CO2 retention, multiple comorbidities, and acute exacerbations an average of once or more per year, might also benefit from long-term oxygen supplementation. In practice, the possibility of clinical benefit has often been taken for granted, even when the patients did not correspond to those in the seminal long-term oxygen therapy trial. This has resulted in the widespread administration of supplemental oxygen to patients whose COPD did not correspond to that of those patients who benefited so much in the original 1980 trial.

Administration of long-term oxygen therapy is estimated to cost $2 billion/year.[5] Attempts to determine the benefit or otherwise from oxygen supplementation since the 1980s have been inconclusive; a larger, longer trial was required to determine whether any aspect of health, from prolonged survival to lung function and health status, was improved. The answer is almost certainly no. A limitation is that the trial was not blinded as to oxygen administration. However, it is hard to imagine how that could have affected the outcome.

Where in the scale of severity of COPD should one now commence oxygen administration? One answer is: Patients with resting hypoxemia of partial pressure of oxygen dissolved in the blood less than 55 mm Hg or amount of oxygen bound to hemoglobin in arterial blood less than 88% at rest and breathing room air. Whether arterial desaturation only on exercise is beneficial remains to be determined.