Background Pulse oximeters are a standard non-invasive tool to measure blood oxygen levels, and are used in multiple healthcare settings. It is important to understand the factors affecting their accuracy to be able to use them optimally and safely. This analysis aimed to explore the association of the measurement error of pulse oximeters with systolic BP, diastolic BP and heart rate (HR) within ranges of values commonly observed in clinical practice.

Methods The study design was a retrospective observational study of all patients admitted to a large teaching hospital with suspected or confirmed COVID-19 infection from February 2020 to December 2021. Data on systolic and diastolic BPs and HR levels were available from the same time period as the pulse oximetry measurements.

Results Data were available for 3420 patients with 5927 observations of blood oxygen saturations as measured by pulse oximetry and ABG sampling within 30 min. The difference in oxygen saturation using the paired pulse oximetry and arterial oxygen saturation difference measurements was inversely associated with systolic BP, increasing by 0.02% with each mm Hg decrease in systolic BP (95% CI 0.00% to 0.03%) over a range of 80–180 mm Hg. Inverse associations were also observed between the error for oxygen saturation as measured by pulse oximetry and with both diastolic BP (+0.03%; 95% CI 0.00% to 0.05%) and HR (+0.04%; 95% CI 0.02% to 0.06% for each unit decrease in the HR).

Conclusions Care needs to be taken in interpreting pulse oximetry measurements in patients with lower systolic and diastolic BPs, and HRs, as oxygen saturation is overestimated as BP and HR decrease. Confirmation of the oxygen saturation with an ABG may be appropriate in some clinical scenarios.