1030: Occult Hypoxemia and Pulse Oximetry Performance Across Skin Pigmentation Groups

Effect of skin tone on the accuracy of the estimation of arterial oxygen saturation by pulse oximetry: a systematic review

“…Evaluation of skin tone only occurred in three studies. 18 , 19 , 20 The proportion of participants with darker skin tones or of non-White ethnicities was particularly low in earlier studies ( Supplementary Table S2 ). One study comprised 100% Black participants.…”

“…For patients with dark skin tones, four studies reported overestimation of SaO 2 by pulse oximetry, 2 , 19 , 56 one reported underestimation, 17 and one (a conference abstract) reported increased errors. 18 The magnitude of bias varied between studies, particularly in relation to SaO 2 ; lower SaO 2 was associated with a greater degree of overestimation bias. Bias of up to 10.4 (6.8)% was reported in profoundly hypoxaemic participants (SaO 2 approximately 75%) using an old device no longer in service (Biox Ohmeda IIA ear probe [Ohmeda, Boulder, CO, USA]).…”

“…Two studies ( n =221) were industry-funded retrospective analyses of historic laboratory data and it was not possible to determine if data from studies already included in the review were contained within these. 17 , 18 All of the studies involved participants breathing ≤21% oxygen in order to evaluate the effect of lowering blood oxygenation. A range of pulse oximeters were evaluated in this category; many early devices are now obsolete including the multiple wavelength Hewlett-Packard 47201A.…”

“…The two studies summarising collated data from numerous laboratory studies and analysed by manufacturers (Medtronic [Dublin, Ireland] and Masimo [Irvine, CA, USA]) failed to demonstrate a difference in bias related to skin tone. 17 , 18 …”