Does the color of skin affect pulse oximetry?
Myron Yaster MD and Charles Cote MD
I first became aware of pulse oximetry in the early 1980s when Dr. William New Jr., a professor at Stanford University, presented grand rounds at Johns Hopkins on his research on how to measure blood oxygen levels in a non-invasive manner and presented his new product, the N-100 pulse oximeter and his new company Nellcor, Inc. to us. (The name Nellcor was built using the initials of the three co-founders William New, Jack Lloyd, and James Corman). The N-100 was revolutionary not only for its ability to measure oxygen saturation but its use of variable tones to alert us to falling oxygen saturation levels. Importantly, despite holding the patent for this change in tone with saturation values, they felt so strongly about the importance of this unique warning system that they chose to share this without remuneration with future manufacturers of pulse oximeters. Alas, the company Nellcor Inc. no longer exists as an independent company and, after several corporate acquisitions and mergers, it is now owned by Medtronic. In 2000 Masimo sued Nellcor for patent infringement and over a period of several years Nellcor and Masimo established a royalty deal, which ended the legal dispute and any ongoing infringement lawsuits. Another patent infringement suit between Masimo and Apple resulted in removal of pulse oximetry from Apple’s I-watch.
As a quick review: A standard pulse oximeter passes two wavelengths of light through tissue to a photodetector. Taking advantage of the pulsate flow of arterial blood, it measures the change in absorbance during systole as the finger volume changes and the detectors sense a change over the course of a cardiac cycle, allowing it to determine the absorbance due to arterial blood alone; this remarkable modification allows, exclusion of absorbance due to venous blood, skin, bone, muscle, and fat. Measurements are artifactually lower in patients wearing green or blue nail polish but unaffected by red or blue nail polish. Intravenous dyes (methylene blue and cardio green) cause a similar artifactually low measurement. The two oximeter wavelengths measure the quantities of bound (oxygenated) and unbound (non-oxygenated) hemoglobin, and from their ratio, the percentage of bound hemoglobin is computed; these wavelengths are in the same range of absorption of green and blue colors accounting for the measurement artifacts.
Pathologies such as carbon monoxide poisoning, methemoglobinemia, and cyanide poisoning also affect pulse oximetry accuracy, while patient movement, poor probe placement, low blood flow states, and ambient light can cause artifacts. What about skin color? In today’s PAAD, Starnes et al.1 in a letter to the editor of the New England J of Medicine review this issue and how 2 different pulse oximeters (Masimo and Nelcor) performed. Dr. Charles Coté pioneered studies of pulse oximetry in children2,3 and I’ve asked him to help me in today’s PAAD. Myron Yaster MD
Letter to the editor
Starnes JR, Welch W, Henderson CC, Hudson S, Risney S, Nicholson GT, Doyle TP, Janssen DR, Londergan BP, Parra DA, Slaughter JC, Aliyu MH, Graves JA, Soslow JH. Pulse Oximetry and Skin Tone in Children. N Engl J Med. 2025 Mar 6;392(10):1033-1034. doi: 10.1056/NEJMc2414937. Epub 2025 Feb 12. PMID: 39938116.
“In the Pulse Oximetry and Skin Tone in Children (POSTer-Child) study, Starnes et al. enrolled 320 patients younger than 21 years of age undergoing cardiac catheterization in 2024. Skin tone was measured with the use of a spectrophotometer. SpO2 was recorded with the use of two pulse oximeters (Nellcor and Masimo) at the exact time of blood sampling for measurement of fractional saturation by co-oximetry. Pulse-oximetry bias (SpO2−SaO2), precision (standard deviation of bias), and accuracy root mean square error (ARMS) were calculated (for all three measures, higher values indicate worse pulse-oximeter performance), as was the percentage of patients with occult hypoxemia (SaO2 of <88% when SpO2 is ≥92%). A total of 48 of 319 patients (15.0%) identified as Black, and 44 (13.8%) identified as Hispanic.”1
Ok, what did they find? “Average bias was 1.32 percentage points for the Nellcor device and 1.88 percentage points for the Masimo device (Table 1). Bias was higher among children with darker skin (individual typology angle [ITA] category 5 or 6) than among those with lighter skin (ITA category 1 or 2) for both pulse oximeters (P<0.001). Precision and ARMS were also higher for children with darker skin. Occult hypoxemia was present in 4 of 56 children (7%) in ITA category 5 or 6 for the Nellcor device and in 5 of 60 children (8%) for the Masimo device, as compared with 0 of 81 children and 3 of 88 children (3%), respectively, in ITA category 1 or 2.”1 In plain English: pulse oximetry overestimated oxygen saturation in darker pigmented skin patients which can lead to undertreatment of hypoxemia and this overestimation was worse with the Masimo device.
Gudelunas et al.4 studied 146 healthy adults also with Nellcor and Masimo devices and analyzed 9763 matched readings during stable oxygen saturation values ranging from 68-100%. They similarly found that “Skin pigment, PI (perfusion index), and degree of hypoxemia significantly contributed to errors (bias) in both pulse oximeters; the P value of the relationship to mean bias or median absolute bias was <.00001. More importantly they found that “the combined frequency of missed diagnosis of hypoxemia (pulse oximeter readings 92%-96% when arterial oxygen saturation was <88%) in low perfusion conditions was 1.1% for light, 8.2% for medium, and 21.1% for dark skin. They concluded that “Low peripheral perfusion combined with darker skin pigmentation leads to clinically significant high-reading pulse oximeter errors and missed diagnoses of hypoxemia. Darkly pigmented skin and low perfusion states are likely the cause of racial differences in pulse oximeter performance in retrospective studies.”
Pulse Oximetry and Skin Tone in Children
Aside from product differences are any of you surprised by these results? We aren’t and thought this was settled science particularly at low saturations (<80%).5,6 Indeed, we are pretty sure this is a standard question in U.S. Anesthesiology/Pediatric Anesthesiology Board examinations. Certainly the 20% error rate during low perfusion states in darkly pigmented patients, often the time we need the most accurate information, is a major and likely underappreciated concern. We thought this would be a good review and discussion point for all of you, particularly those who work in a teaching environment. Finally, can anything be done to improve the reliability of pulse oximetry in darker skinned patients? Surprisingly, the short answer is NO. A recent paper and editorial in JAMA depressingly laid out several solutions that are unlikely to be taken by manufacturers or the FDA.7,8
Send your thoughts and comments to Myron who will post in a Friday reader response.
PS: For more information see: Dorsey D, Negussie F, Igaga E, et al. The open oximetry project: safe and accurate pulse oximeters for all skin tones. APSF Newsletter. 2025:17–19. https://www.apsf.org/wp-content/uploads/newsletters/2025/4001/APSF4001-a06-open-oximetry.pdf
References
1. Starnes JR, Welch W, Henderson CC, et al. Pulse Oximetry and Skin Tone in Children. The New England journal of medicine 2025;392(10):1033-1034. (In eng). DOI: 10.1056/NEJMc2414937.
2. Coté CJ, Goldstein EA, Cote MA, Hoaglin DC, Ryan JF. A single-blind study of pulse oximetry in children. Anesthesiology 1988;68(2):184-188.
3. Coté CJ, Goldstein EA, Fuchsman WH, Hoaglin DC. The effect of nail polish on pulse oximetry [see comments]. AnesthAnalg 1988;67(7):683-686.
4. Gudelunas MK, Lipnick M, Hendrickson C, et al. Low Perfusion and Missed Diagnosis of Hypoxemia by Pulse Oximetry in Darkly Pigmented Skin: A Prospective Study. Anesthesia and analgesia 2024;138(3):552-561. (In eng). DOI: 10.1213/ane.0000000000006755.
5. Feiner JR, Severinghaus JW, Bickler PE. Dark skin decreases the accuracy of pulse oximeters at low oxygen saturation: the effects of oximeter probe type and gender. Anesthesia and analgesia 2007;105(6 Suppl):S18-s23. (In eng). DOI: 10.1213/01.ane.0000285988.35174.d9.
6. Bickler PE, Feiner JR, Severinghaus JW. Effects of skin pigmentation on pulse oximeter accuracy at low saturation. Anesthesiology 2005;102(4):715-9. (In eng). DOI: 10.1097/00000542-200504000-00004.
7. Shachar C, Drabo EF, Iwashyna TJ, Ferryman K. Addressing Racial and Ethnic Bias in Pulse Oximeters-A Wicked Problem. Jama 2025;333(7):563-564. (In eng). DOI: 10.1001/jama.2024.25443.
8. Ferryman K, Crews DC, Drabo EF, Iwashyna TJ, Kane O, Jackson JW. Adherence to FDA Guidance on Pulse Oximetry Testing Among Diverse Individuals, 1996-2024. Jama 2025;333(7):631-632. (In eng). DOI: 10.1001/jama.2024.26473.