Inaccurate pulse oximeter readings tied to less supplemental oxygen for darker-skinned ICU patients

Patients with darker skin who received less accurate readings of their oxygen levels using pulse oximeters — the ubiquitous devices clamped on hospitalized patients’ fingers — also received less supplemental oxygen during ICU stays, according to a study published Monday.

The new research in the journal JAMA Internal Medicine adds to the evidence that the faulty readings in darker-skinned patients can affect their care and may be one factor explaining racial disparities, such as higher rates of limb loss and death for Black and Hispanic ICU patients.

Meanwhile, a different study published last week in the BMJ suggests that imperfect readings from pulse oximeters may impact care of Black patients broadly, not merely those who are critically ill. An analysis of medical records of tens of thousands of general and surgical patients in Veterans Health Administration medical centers showed that Black patients were more likely than white patients to have hidden hypoxia — undetected low oxygen levels linked to higher rates of death and organ failure.

Both studies were conducted using databases of hospital patient data collected through 2019, meaning neither study involved Covid-19 patients. It’s been known for decades that the devices are less accurate in patients with darker skin and those wearing nail polish, but new interest and a stream of research about potential racial bias in the devices has been sparked by the racial disparities seen in Covid deaths and treatment. The measurement of oxygen levels using the devices has played a critical role in determining which Covid patients are admitted to the hospital and given supplemental oxygen and other therapies. The devices, invented in the 1970s, were tested on largely white populations.

“This is telling us what we see as disparities could be due to technology that is not optimized for all populations,” said Leo Anthony Celi, a co-author of the JAMA Internal Medicine paper, and an ICU physician and a principal research scientist at the Institute for Medical Engineering and Science at MIT who helped create the large public database of ICU patients used in the study. “We’re seeing the downstream effect. It performs poorly as soon as you apply it outside the demographic it was designed for.”

Thomas Valley, a pulmonologist at the University of Michigan who is part of a team that previously published one of the pivotal studies showing pulse oximeters work less well in darker-skinned patients, said the JAMA Internal Medicine study provided important new evidence that the lack of precision in pulse oximeters may result in poorer outcomes in those with darker skin. “When we created our study, I never thought we’d be able to put our finger on the actual harm,” said Valley, who was not involved in the new work published in JAMA Internal Medicine. “This study shows we are treating people differently because of occult hypoxemia.”

A number of papers in recent months have shown that pulse oximeters work less precisely in those with darker skin. Celi’s group wanted to investigate whether that issue could be contributing to known racial disparities in care, such as higher mortality rates from Covid and lower rates of admission for specialized cardiac care services. “We see these outcomes, but why are they happening?” asked Eric Gottlieb, the study’s lead author, a nephrologist, and a clinical scientist at the Massachusetts Institute of Technology. “If you think someone’s O2 is a little higher than it really is and you give them less oxygen, does that turn into a disparity?”

The study of more than 3,000 ICU patients found that Black, Hispanic, and Asian patients had lower blood oxygen levels — measured with the gold-standard blood draw from an artery — compared to levels detected with pulse oximeters.  They also received less supplemental oxygen for a given blood oxygen level compared with white patients, an average difference of about 0.2 to 0.3 liters per minute. A separate analysis by the group of whether Black, Hispanic, and Asian patients received fewer doses of a drug not tied to oxygen levels, heparin, showed there was no racial disparity, suggesting that the difference in measured oxygen levels was the cause of the disparity in oxygen administration.

The second paper in BMJ analyzed more than 30,000 readings of oxygen levels taken by pulse oximeters and arterial blood draws within minutes of each other. The study authors found that 15.6% of white patients had hidden hypoxemia while 19.6% of Black patients did.

The study also answered critics who had said differences in oxygen levels might occur because blood draws in previous comparisons had been made 10 minutes after pulse oximeter readings; this study found differences even in blood draws taken two or five minutes later. It also showed that repeat pulse oximeter readings were more unreliable for Black patients.

The study did not find a significant difference between White and Hispanic patients as other studies had, possibly because of how the VA encodes for Hispanic ethnicity in medical records. Populations of Asian, Native American/Alaska Native, and Native Hawaiian/Pacific Islander veterans were too small to analyze. Other researchers have suggested that any dark-skinned patients, regardless of race, are vulnerable to inaccuracies in their readings.

The authors estimated that hidden hypoxia could be occurring up to 80,000 times a year in Black patients in the VA system. “These are patients that are not being intensively monitored, are not being given remdesivir [a treatment for Covid-19], are not being upgraded,” said Theodore J. Iwashyna, a professor of medicine and public health at Johns Hopkins University and the study’s senior author. “I think that’s too many.”

“This is hundreds and thousands of patients being affected, and that’s just at one hospital system,” added Valeria Valbuena, a general surgery resident at the University of Michigan and the study’s lead author. “This is literally affecting millions of patients.”

In their study, Valbuena and Iwashyna noted that the VA, the nation’s largest integrated health system, could be a major driver of improving the devices if it insisted on buying only devices that worked equally well on veterans of all skin colors.

The critical question of how large a role imprecision in the devices plays in health disparities is being debated. Some doctors and device manufacturers have pushed back against the criticism of pulse oximeters, arguing that patients may be undergoing unnecessary and painful arterial blood draws because of concerns over accuracy of the devices, and others have questioned the high cost of redesigning and replacing current devices.

“Every 10 years this issue raises its ugly head and nothing gets done,” Valbuena said. “People say, ‘It’s going to be cumbersome.’ That suggests that because the problem is in Black patients, it’s not worth fixing.”

Gottlieb and Celi agreed that the devices should be improved because so many clinicians rely on them and because decisions on whether to admit people into hospitals, supply oxygen, or provide higher levels of care are partly based on slight differences in oxygen levels. “The positive side of this is it gives you something you can fix,” Gottlieb said. “You can simply say, ‘Let’s make the pulse oximeter better.’”

The FDA announced on June 21 that it will convene its medical devices advisory committee later this year to discuss concerns about how well the devices work in patients with darker skin.

In the meantime, many physicians are relying more on arterial blood gas measurements, which are more painful to collect, and telling darker-skinned patients who rely on home pulse oximeters to call their physicians at any sign of shortness of breath, regardless of what the devices indicate.