乌鸦传媒

Overestimates in pulse oximetry measurements among Black and Hispanic patients were tied to delays in eligibility for COVID therapies, researchers found.

Compared with white individuals, pulse oximetry overestimated oxygen saturation by an average of 1.7% (0.5-3.0) in Asian patients, 1.2% (95% CI 0.6-1.9) in Black patients, and 1.1% (95% CI 0.3-1.9) for non-Black Hispanic patients, reported Ashraf Fawzy, MD, of Johns Hopkins University in Baltimore, and colleagues in .

But only Black and Hispanic patients were less likely to be recognized as being eligible for recommended COVID-19 therapies as a result:

• Black: HR 0.71 (95% CI 0.63-0.83)
• Non-Black Hispanic: HR 0.77 (95% CI 0.66-0.89)
• Asian: HR 0.97 (95% CI 0.62-1.5)

Almost 24% had unrecognized COVID-treatment eligibility, with 54.8% of those patients being Black and 27.1% being Hispanic, the authors noted. Compared with white patients, Black patients had a median 1 hour difference in time to recognition of eligibility.

"To our knowledge, the performance of pulse oximetry in patients with COVID-19 and its potential association with clinical decision-making remains unexplored and can help identify factors that may explain the disproportionate COVID-19 mortality experienced by patients of certain racial and ethnic minority groups," Fawzy's group wrote.

Use of supplemental oxygen or a pulse oximetry reading of 94% or less is a cutoff for remdesivir (Veklury) eligibility among hospitalized patients with COVID. Similarly, dexamethasone is also used for patients requiring supplemental oxygen.

"An overestimation of oxygen saturation levels could be associated with an underappreciation of clinical risk as presented by these calculators" that predict clinical risk, the authors noted.

They examined data on 7,448 patients from March 4, 2020 to Nov. 12, 2021 captured in the , which included five hospitals in the Johns Hopkins Health System. Patients self-reported race and ethnicity during hospital registration.

Overall, 1,213 of these patients had both their oxygen saturation and pulse oximetry levels measured concurrently during their hospital stay. Black and non-Black Hispanic patients tended to be younger with longer hospital stays, and have more oxygen saturation measurements per person than white patients. Also, "non-Black Hispanic patients had a smaller proportion of women and fewer comorbidities," according to the authors.

Interestingly, Fawzy's group found that median oxygen saturation for Asian and Black patients was consistently lower than pulse oximetry for readings ranging from 88% to 96%, while oxygen saturation was consistently higher than pulse oximetry for white patients. This meant that pulse oximetry consistently overestimated true oxygen saturation for Black and Asian patients, and underestimated it for white patients.

Occult hypoxemia, defined as oxygen saturation below 88% with concurrent pulse oximetry measurements of 92% to 96%, occurred in 3.7% of samples in Asian patients, 3.7% in Black patients, 2.8% in non-Black Hispanic patients, and 1.7% in white patients. Overall, 30.2% of Asian patients had occult hypoxemia during their hospital stay, as did 28.5% of Black patients, 29.8% of non-Black Hispanic patients, and 17.2% of white patients, the authors noted.

In an , Valeria Valbuena, MD, of the University of Michigan in Ann Arbor, and colleagues, noted the recently uncovered racial disparities in pulse oximetry measurement, namely, "a of the pulse oximeter is that patients with darker skin ... are more likely to experience occult hypoxemia," they wrote.

The editorialists added that until this flaw is corrected through "market pressure," clinicians should undertake strategies to take these limitations into account, such as "lowering the threshold for suspected disease and obtaining more arterial blood gases."

"Although the device measurement error is real and based purely on optics, the decision to do nothing about a faulty device is a human one, and one that can and should be corrected," Valbuena and colleagues wrote.

Among the study limitations, according to Fawzy's group, were that results may not be generalizable to "healthy individuals or those with less acute illness" and that treatment delay estimates may be affected by how often they were measured.

Comment