Inhalers, nebulizers, antibiotics, and steroids – these are some of the most common tools in our pulmonary arsenal that we deploy on a daily basis. But, there is no treatment more fundamental to a pulmonary practitioner than oxygen. So how is it that something that naturally occurs and comprises 21% of ambient air has become so medicalized?
It is difficult (perhaps impossible) to find a pulmonologist or a hospitalist who has not included the phrase “obtain ambulatory saturation to qualify the patient for home oxygen” in at least one of their progress notes on a daily basis. Chronic obstructive pulmonary disease (COPD) is the most common reason for the prescription of long-term oxygen therapy (LTOT), a large industry tightly regulated by the Centers for Medicare & Medicaid Services (CMS).
The evidence for the use of LTOT in patients with COPD dates back to two seminal papers published in 1980 and 1981. The British Medical Research Council Working Party conducted the BMRC trial, in which 87 patients with a Pao2 of 40 mm Hg to 60 mm Hg, CO2 retention, and a history of congestive heart failure were randomized to treatment with 15 hours per day of home oxygen therapy, starting at 2 L and titrating to Pao2 of 60 mm Hg vs. standard therapy without oxygen (Lancet. 1981;1[8222]:681-6). There was an impressive 22% mortality benefit at 3 years.
Another study published around the same time, the Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease (NOTT) trial (Ann Intern Med. 1980;93[3]:391-8) directly compared continuous 24-hour to nocturnal home oxygen therapy in patients with COPD and severe hypoxemia with a Pao2 less than 55 mm Hg. Again, there was an impressive mortality benefit in favor of continuous home oxygen with a 9% and 18% mortality difference at 1 and 2 years of enrollment, respectively.
Afterward, it became universally accepted dogma that patients with COPD and severe hypoxemia stood to substantially benefit from LTOT. For years, it was the only therapy associated with a mortality reduction. The LOTT study (Albert RK, et al. N Engl J Med. 2016;375[17]:1617-27) included 768 patients with stable COPD and a resting or nocturnal Spo2 of 89% to93%, as well as patients with moderate exercise-induced desaturation (Spo2 of greater than or equal to 80% and less than 90% for greater than or equal to 10 seconds during the 6-minute walk test). Half of these patients received oxygen for 24 hours per day, during sleep, or during exercise (depending on when desaturation would occur) and half received no oxygen. There was no difference in time to death or first hospitalization or in rates of hospitalization or exacerbation. There was also no difference between groups in quality of life, lung function, or distance walked in 6 minutes.
The INOX (Lacasse Y, et al. N Engl J Med. 2020;383[12]:1129-38) trial, in which 243 patients with oxygen saturation less than 90% for at least 30% of the night were assigned to receive nocturnal vs sham oxygen, found similar results. There was no difference in the composite outcome of all-cause mortality and progression to 24-7 oxygen requirement (according to the criteria originally defined by NOTT). A 2022 systematic review and meta-analysis including six studies designed to assess the role of LTOT in patients with COPD and moderate desaturation, including LOTT and INOX, found no benefit to providing LTOT (Lacasse Y, et al. Lancet Respir Med. 2022;10[11]:1029-37).
Based on these studies, a resting Spo2 of 88% seems to be the threshold below which LTOT improves outcomes. CMS lists four classes of patients eligible for LTOT: (1) Patients with Pao2 < 55 mm Hg or pulse oximetry less than or equal to 88% at rest or (2) during sleep or (3) during exercise, and (4) patients with Pao2 > 55 mm Hg but less than or equal to 59 mm Hg or pulse oximetry of 89% who have lower extremity edema, evidence of pulmonary hypertension, or erythrocythemia (Centers for Medicare & Medicaid Services. Medicare Coverage Database. 2021;100-103:240.2. These criteria reflect the inclusion criteria of the BMRC trial and NOTT.
COPD management has changed significantly in the 40 years since NOTT was published. In the early 1980s, standard of care included an inhaled beta-agonist and oral theophylline. We now prescribe a regimen of modern-day inhaler combinations, which can lead to a mortality benefit in the correct population. Additionally, rates of smoking are markedly lower now than they were in 1980. In the Minnesota Heart Survey, the prevalence of being an ever-smoking man or woman in 1980 compared with 2009 dropped from 71.6% and 54.7% to 44.2% and 39.6%, respectively (Filion KB, et al. Am J Public Health. 2012;102[4]:705-13). Treatment of common comorbid conditions has also dramatically improved.
A report containing all fee-for-service data published in 2021 by CMS reported oxygen therapy accounted for 9.8% of all DME costs covered by CMS and totaled approximately $800,000,000 (Centers for Medicare & Medicaid Services. FFS Data. 2021. This represents a significant financial burden to our health system and government.
Two of the eligible groups per CMS (those with isolated ambulatory or nocturnal hypoxemia) do not benefit from LTOT in RCTs. The other two groups are eligible based on trial data from a small number of patients who were studied more than 40 years ago. These facts raise serious questions about the cost-efficacy of LTOT.
So where does this leave us?
There are significant barriers to repeating large randomized oxygen trials. Due to broad inclusion criteria for LTOT by CMS, there are undoubtedly many people prescribed LTOT for whom there is minimal to no benefit. Patients often feel restricted in their mobility and may feel isolated being tethered to medical equipment. It is good practice to think about LTOT the same way we do any other therapy we provide - as a medicine with associated risks, benefits, and costs.
Despite its ubiquity, oxygen remains an important therapeutic tool. Still, choosing wisely means recognizing that not all patients who qualify for LTOT by CMS criteria will benefit.
Drs. Kreisel and Sonti are with the Division of Pulmonary, Critical Care, and Sleep Medicine, MedStar Georgetown University Hospital, Washington, DC.