Background Administration of oxygen in the delivery room is informed by oxygen saturation (SpO₂). An oxygen saturation range of 60%–70% is the threshold for administering oxygen in the first minutes after birth. The accuracy of newer generation oximeters to measure SpO₂ has not been compared against the ‘gold standard’, direct arterial blood oxygen saturation (SaO₂) when SaO₂ is low. The aim of this study was to determine the accuracy and precision of Nellcor and Masimo oximeters to measure SpO₂ when SaO₂ <70%.

Method Prospective observational study in ventilated anaesthetised newborn lambs with an indwelling carotid artery catheter. Ventilation was adjusted to achieve hypoxaemia. Nellcor (Oxi-Max 600 with Max-N sensor) and Masimo (Rad 4 with low noise optical probe (LNOP) sensor) sensors were applied to the right forelimb (preductal). An arterial blood sample was collected at 1–5 min intervals when the animal was hypoxic. The displayed SpO₂ was recorded. We used Bland–Altman analysis to determine precision and accuracy of each oximeter when SaO₂ <70%.

Results 17 lambs were studied, 165 measurements were obtained, 123 were SaO₂ <70%. The mean difference (±1.96 SD) Nellcor SpO₂-SaO₂ when SaO₂ <70% was 17% (−12% to 46%). The mean difference (±1.96 SD) Masimo SpO₂-SaO₂ when SaO₂ <70% was 13% (−19% to 45%).

Conclusions At SaO₂<70%, both monitors overestimated oxygen saturation (SpO₂) compared with the gold standard. Both oximeters were equally inaccurate when SaO₂ was low.