Home Oxygen for COPD: What the Trials Actually Show

Home oxygen extends survival in COPD only for one well-defined group: people whose blood oxygen is severely low at rest. Two trials from around 1980, the American Nocturnal Oxygen Therapy Trial (NOTT) and the British Medical Research Council (MRC) study, established that steady supplemental oxygen reduced mortality in patients with severe resting hypoxemia. When the 2016 Long-Term Oxygen Treatment Trial (LOTT) tested oxygen in patients with only moderate desaturation, it found no benefit for survival, hospitalization, or quality of life. The evidence tracks how low the oxygen level sits at rest, not whether a fingertip monitor occasionally dips.

The two trials that drew the line

Both foundational trials were conducted in the late 1970s and enrolled patients who were already severely hypoxemic, and that shared starting point is the key to reading them.

NOTT: more oxygen beat less oxygen

The Nocturnal Oxygen Therapy Trial, reported in the Annals of Internal Medicine in 1980, enrolled 203 patients with COPD and documented hypoxemia. Entry required an arterial oxygen tension (PaO2) at or below 55 mmHg, or at or below 59 mmHg together with a sign of strain on the right side of the heart: ankle edema, a raised red-cell count (hematocrit of 55 percent or more), or electrocardiographic evidence of pulmonary hypertension. Patients were randomly assigned to continuous oxygen, used for about 18 hours a day, or to nocturnal oxygen only, about 12 hours. Mortality in the nocturnal-only group ran close to twice that of the continuous group. NOTT did not test whether to give oxygen; it tested how much. Within a severely hypoxemic population, more hours saved more lives.

MRC: oxygen beat no oxygen

The Medical Research Council Working Party trial, published in The Lancet in 1981, asked the more basic question. Its 87 patients had chronic bronchitis or emphysema with severe hypoxemia, carbon dioxide retention, and a history of heart failure. They received either oxygen for at least 15 hours a day or no supplemental oxygen at all. Over five years, 19 of 42 oxygen-treated patients died compared with 30 of 45 controls, and the survival curves did not separate until roughly 500 days had passed. Oxygen worked, but slowly, and only in people who were profoundly hypoxemic to begin with.

Why the threshold is physiological, not arbitrary

The cutoff sits where it does because of the shape of the oxyhemoglobin dissociation curve. Hemoglobin stays nearly fully saturated across a wide range of arterial oxygen tensions, then gives up its oxygen steeply once PaO2 falls below roughly 60 mmHg, near an oxygen saturation of 90 percent. Above that shoulder, adding oxygen raises the number on a monitor but changes little about delivery to tissue. Below it, small further drops in PaO2 translate into large drops in the oxygen actually carried to organs. Sustained severe hypoxemia at that level drives constriction of the lung's blood vessels, strain on the right heart, and a compensatory rise in red-cell mass. Reversing that cascade is what the trials of that era were doing, which is why long-term oxygen eligibility is anchored near a resting PaO2 of 55 mmHg, or an SpO2 at or below about 88 percent.

LOTT tested the gray zone and found nothing

For decades the open question was whether patients who fall short of those strict criteria, the much larger group with only moderate desaturation, might benefit too. The Long-Term Oxygen Treatment Trial, published in the New England Journal of Medicine in 2016, was designed to answer it. Its 738 participants had stable COPD with moderate resting desaturation (SpO2 of 89 to 93 percent) or desaturation only during exertion (SpO2 falling below 90 percent on a six-minute walk test). They were randomized to long-term supplemental oxygen or to none. The result was flat: no significant difference in the time to death or first hospitalization (hazard ratio 0.94, with a 95 percent confidence interval of 0.79 to 1.12). Oxygen also failed to improve quality of life, lung function, walking distance, or the rate of COPD exacerbations.

What the guidelines took from this

Eligibility for long-term oxygen therapy still rests on those older thresholds. Guideline bodies recommend continuous oxygen for a resting PaO2 at or below 55 mmHg (or SpO2 at or below 88 percent), and for patients in the 56 to 59 mmHg range who also show cor pulmonale, edema, or polycythemia, the same secondary criteria NOTT used. LOTT did not overturn that framework. It mapped its outer edge, showing that extending oxygen to milder desaturation adds cost and burden without measurable benefit.

Reading the numbers, not the anxiety

The three trials tell a coherent story. Supplemental oxygen is a treatment for severe, sustained hypoxemia, not a response to a single low reading and not a way to buy reassurance from a higher number on a fingertip monitor. An oximeter that dips into the low 90s overnight or during a walk can feel alarming, yet LOTT showed that treating that pattern with oxygen neither extended life nor reduced hospital stays. The distinction carries weight because home oxygen has real costs: equipment that tethers a person to the house, fire and burn risk near open flames, lost mobility, and the standing suggestion that someone is sicker than the evidence says. The trials never claimed oxygen is unimportant; they showed that its proven benefit is concentrated in a group defined by objective, resting measurements. This article is educational and not medical advice, and any decision about home oxygen depends on individual testing and belongs with a person's own clinicians.