Laser-based pulse oximetry eliminates pigmentation effects on oxygen saturation measurements: A pilot study

Abstract

Background Almost since its introduction, pulse oximetry has been plagued by inaccuracy associated with pigmentation, whether from fingernail polishes or melanin. The presence of melanin in the optical path of a pulse oximetry sensor has been shown to artifactually increase oxygen saturation measurements which can result, clinically, in occult hypoxemia and misdiagnoses. Methods This report describes the theoretical basis for this inaccuracy and presents results from both a benchtop study and a clinical study testing the effects of pigmentation on conventional LED-based pulse oximetry compared to laser-based pulse oximetry. The clinical portion of this study was performed on 18 consenting participants, nine darkly pigmented and nine lightly pigmented, to assess the ability of laser-based pulse oximetry to eliminate this dangerous pigmentation bias. The clinical study directly compared oxygen saturation readings on laser-based pulse oximeters to readings performed on two different LED-based pulse oximeters. All measurements were compared to invasive reference laboratory measurements performed on arterial blood samples. We hypothesized that monochromatic light sources used in laser-based pulse oximetry would make this new technology insensitive to pigmentation bias. Results The clinical portion of this study showed significantly greater (p < 0.001) measurement error (variance) for the two LED-based pulse oximeters (5.48 and 5.47) compared to laser-based pulse oximetry (3.54), when analyzed for all participants. The bias differences in oxygen saturation measurements by the LED-based pulse oximeters, when made on lightly versus darkly pigmented participants compared to invasive reference measurements, accounts for most of the increase in measurement error. Conclusions By combining theoretical development, experimental benchtop testing, and a clinical study, this research explains and demonstrates that the wide spectral bandwidth of LEDs is the root cause of pigmentation bias in commercially available LED-based pulse oximetry and validates the ability of narrow-band laser-based pulse oximetry to eliminate this pigmentation bias.