The purpose of this study was to evaluate the accuracy of the SaO2 measured by the Nellcor N-10 portable pulse oximeter, a small, lightweight, battery-operated device. Method: Our study population consisted of 30 normal, healthy Caucasian volunteers. The accuracy of the Nellcor N-10 was evaluated under conditions of normoxemia (SaO2 > 95%) and hypoxemia (SaO2 75-95%). Hypoxemia was produced by a rebreathing technique with CO2 absorbent. After a subject's SaO2 stabilized at the desired level, an arterial blood sample was obtained and analyzed on an IL282 CO-Oximeter for SaO2, carboxyhemoglobin (COHb), and methemoglobin (metHb). The N-10 SaO2 was recorded when the arterial sample was drawn. Because the N-10 measures the SaO2 of hemoglobin capable of binding oxygen (normal hemoglobin, norm Hb-SaO2), the IL282 SaO2 reading, which represents all forms of hemoglobin (total hemoglobin SaO2: totHb-SaO2) was converted to IL282 normHb-SaO2, ie, IL282 normHb-SaO2 = [(IL282 totHb-SaO2) ò (100 - COHb - metHb)] [100]. Results: A total of 60 data sets were collected. The Nellcor N-10 SaO2 range during the study was 77-100%. The mean (± SD) N-10 SaO2 was 90 ± 6%, the mean IL282 totHb-SaO2 was 89 ± 6%, and the mean IL282 normHb-SaO2 was 92 ± 6%. The mean COHb was 1.9 ± 0.4%, and the mean metHb was 0.6 ± 0.2%. The difference between the N-10 SaO2 and IL282 totHb-SaO2 and the difference between the N-10 SaO2 and the IL282 normHb-SaO2 were statistically significant. There was a 0.96 correlation between the N-10 SaO2 and the IL282 totHb-SaO2 and between the N-10 SaO2 and the IL282 normHb-SaO2. Conclusions: Although the differences between the Nellcor N-10 SaO2 and the IL282 SaO2 were statistically significant, in most cases the difference was too small to be clinically important. The IL282 SaO2 and the N-10 SaO2 correlated well. When COHb and metHb levels are low, as they were in this normal population, the Nellcor N-10 pulse oximeter demonstrates acceptable accuracy in the SaO2 range of 77-100%.