Effects of Skin Pigmentation on Pulse Oximetry: A Pilot Study

Background Pulse oximetry is an essential, noninvasive tool in critical care, yet mounting evidence indicates that readings may be inaccurate in patients with darker skin pigmentation. These discrepancies can delay recognition of hypoxemia and contribute to racial disparities in clinical outcomes. Objective To quantify the difference between oxygen saturation as measured by pulse oximetry and arterial blood gas measurements across a spectrum of skin pigmentation using objective measures of melanin and skin tone. Methods A prospective observational study was conducted in an adult critical care unit. Skin pigmentation was measured using reflectance spectrophotometry. Simultaneous pulse oximetry and arterial blood gas readings were collected during routine arterial blood gas sampling. Bias (pulse oximetry value – arterial blood gas value) was calculated for each measurement pair. Multivariate linear regression models were used to assess the association between skin pigmentation and measurement bias, adjusting for potential confounders such as age, sex, and perfusion index. Results Forty-two measurement pairs were analyzed. A weak and nonsignificant relationship was found between the individual typology angle and melanin index and the difference between pulse oximetry and arterial blood gas measures. Conclusions Individual typology angle and melanin index are not strong predictors of the difference in oxygen readings between pulse oximeters and arterial blood gas measurements, even in critically ill patients. The weak fit of all models and the lack of statistical significance indicate that other unmeasured factors may play a more substantial role in explaining the variability.