COMMENTARY

Oxygen Targets in COVID-19 With ARDS: What's Optimal?

Aaron B. Holley, MD

Disclosures

April 17, 2020

The novel coronavirus (COVID-19) pandemic has strained medical resources in countries across the world. Early data from Wuhan, China, show that more than a third of patients with this infection (41.3%) will require oxygen and more than 1 in 20 (6.1%) will require mechanical ventilation. A somewhat smaller portion (3.4%) will develop acute respiratory distress syndrome (ARDS).

Given its apparent transmissibility and the lack of an effective vaccine or treatment, projected infection rates and requirements for respiratory support may overwhelm the health systems of affected countries.

Before discussing what we've learned about oxygen requirements for these patients, I need to throw out several disclaimers.

First, the data we have on disease presentation, behavior, and clinical course are evolving faster than most of us can keep up. What is true today may not be true tomorrow. This piece is likely to be relevant for only a short period of time.

Second, although COVID-19 is already straining the resources required to provide respiratory support to patients, it's unlikely that the oxygen supply will be affected, at least not in hospitals located in developed countries. Outcomes for critically ill patients, however, are directly affected by oxygen targets. And because the devices required for oxygen delivery have been limited in some areas, oxygen targets may determine the distribution of limited resources.

Now that we have that out of the way, let's discuss what we know about oxygen targets for critically ill patients without COVID-19. Like many topics, this is something we've been debating for decades, with an uptick in data and recommendations over the past 5 years. The reason for the debate is simple: Oxygen is required for aerobic metabolism and sustained organ function, so in theory, more should be better. In reality, though, this is not true. Humans do not walk around in a hyperoxygenated state, and too much oxygen has harmful effects. These include, but are not limited to, unwanted vasoconstriction, free radical development with subsequent damage to tissues, and nitrogen washout in the lungs with resulting atelectasis and de-recruitment.

So, where's the sweet spot and what do the data say? In 2016 the Oxygen-ICU randomized clinical trial reported an improvement in outcomes with a conservative oxygen strategy. That strategy targeted a PaO2 of 70-100 mm Hg, or an arterial oxyhemoglobin saturation (SpO2) between 94% and 98%. The comparison group targeted a more liberal goal of an SpO2 of 97%-100%. The authors concluded that the conservative protocol resulted in lower ICU mortality. As a result, guidelines developed by an international panel of experts updated their oxygen targets. A 2018 systematic review and meta-analysis more or less confirmed the benefits of a more conservative oxygen strategy.

Fast-forward to now, and forget about COVID-19 for a second. In March 2020, two new randomized clinical trials examining oxygen therapy were published in the New England Journal of Medicine. A large retrospective study was published in CHEST in late 2019. Each analyzed oxygen targets in the critically ill. The first found no difference in a conservative versus liberal approach, the second found potential harm with the conservative approach, and the third reported that an oxygen saturation range of 94%-98% was optimal.

Where Does This Leave Us?

Let's deal with the randomized clinical trials first. The trial that found a conservative strategy to be harmful enrolled the sickest patients (all had ARDS with a mean PaO2/FiO2 ~ 120) and used a lower oxygen target (SpO2 88%-92%) in the conservative arm. The comparison arm in the trial that found equivalence didn't target a specific, liberal oxygen strategy but rather "usual care." For the usual-care group, there were no specific threshold limits for FiO2 or SpO2. Patients in the usual-care group still spent less time at FiO2 of 0.21 and more time with SpO2 > 97% when compared with the conservative group, but perhaps the differences weren't large enough to affect outcomes...?

Summarizing all of these data, I think we can draw some general conclusions.

First, an oxygen saturation target as low as SpO2 88% is probably contraindicated, particularly for patients with moderate to severe ARDS.

Second, the equivocal study need not scare us away from limiting SpO2 to below 96%. This certainly seems safe and it may still prove to be beneficial. Taking into account the different designs, patient populations, and oxygen strategies, it's not surprising that the older 2016 randomized clinical trial and the two published in 2020 had slightly different results. Like targets for glucose, positive end expiratory pressure (PEEP), and mean arterial pressures (MAPs), we'll refine what we know about the optimal SpO2 as more data accumulate.

What does all of this mean for patients with COVID-19? Experts have issued guidelines for management of critically ill patients with the disease. They recommend starting oxygen at an SpO2 of 90% and maintaining it no higher than 96%.

Essentially, this is a target of 90%-96%. Based on what we know from non-COVID-19 data, I can get behind that. I'd be leery of letting any COVID-19 patient with ARDS drop below 90%, though. Setting a lower limit of 92% gives you a little breathing room.

Aaron B. Holley, MD, is an associate professor of medicine at Uniformed Services University and program director of pulmonary and critical care medicine at Walter Reed National Military Medical Center. He covers a wide range of topics in pulmonary, critical care, and sleep medicine.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....