Relationship Between Pulse Oximetry and Arterial Oxygen Saturation During Rest and Exercise at Extreme Altitude

Manferdelli, Giorgio, Denis J. Wakeham, Keval D. Patel, Eric T. Hedge, Michael G. Leahy, Abel Thomas, James Berry, Tony G. Babb, Peter Hackett, Renie Guilliod, Benjamin D. Levine, Christopher M. Hearon, Jr., and Andrew R. Tomlinson. Relationship between pulse oximetry and arterial oxygen saturation during rest and exercise at extreme altitude. High Alt Med Biol. 00:00-00, 2026.

Pulse oximetry (SpO₂) is commonly used to guide oxygen use, including at high altitude, but its accuracy during severe hypoxemia remains uncertain. This study aimed to compare arterial oxygen saturation (SaO₂) and SpO₂ while using graded levels of supplemental oxygen at simulated extreme altitude in a hypobaric chamber.

Four healthy, unacclimatized adults completed rest and 60 Watts of exercise at different flow rates of supplemental oxygen (6, 4, 2, 1, and 0 l/min nominal flow) at 282 mmHg barometric pressure (Pb; equivalent to 8,100 m) and 253 mmHg Pb (equivalent to 8,848 m) with simultaneous measurement of arterial blood gases (radial artery catheter) and SpO₂ (forehead pulse oximetry). Exercise was only performed at 282 mmHg.

Across all altitudes and conditions, SaO₂ and SpO₂ demonstrated a very strong correlation (all r² ≥ 0.95, p < 0.001), with SpO₂ showing a consistent positive bias (pooled: 1.7% [-4.4% to 7.7%]; 282 mmHg rest: 0.4% [-4.5% to 5.3%]; 282 mmHg exercise: 3.1% [-2.2% to 8.5%]; 253 mmHg rest: 2.9% [4.9% to 10.6%]). The magnitude of the SpO₂ bias was higher when only SaO₂ <70% was considered (pooled: 5.2%). The root mean square error was 2.6 for SaO₂ ≥70% and increased to 6.2 when SaO₂ was <70%.

Although SpO₂ represents a precise noninvasive measure for SaO₂ across a wide range of oxygen saturations (70%-100%), SpO₂ has lower precision and accuracy during severe hypoxemia (<70% SaO₂).