Wearable in-ear pulse oximetry validly measures oxygen saturation between 70% and 100%: A prospective agreement study

Abstract

Objectives Postoperative monitoring outside intensive and post-anaesthesia care units is seldom, partly due to lack of suitable and approved systems. We therefore aim to validate the oxygen saturation (SpO 2 ) and pulse rate measurement of the in-ear sensor c-med° alpha with a reference pulse oximeter. Methods This prospective agreement study was conducted in 12 healthy (ASA 1) adult (18–50 years) volunteers according to the EN ISO 80601-2-61. The sitting volunteers were equipped with the finger pulse oximeter Rad-5 and two c-med° alpha sensors in each ear. The inspiratory oxygen content was reduced via a tight-fitting breathing mask to achieve five defined plateaus with stable SpO 2 between 99% and 70%. The deviation of the SpO 2 and pulse rate measurements of the c-med° alpha from those of the Rad-5 was calculated using the mean square error (A rms ). Bias and limits of agreement between both devices were calculated using the Bland-Altman technique. The precision was compared based on the repeatability coefficients. Results The c-med° alpha measured SpO 2 had an A rms = 1.9% relative to the Rad-5, a non-significant bias (−0.1% (-0.2% to 0.0%)), levels of agreement from −4.0% to 3.8%, and the same repeatability coefficient (0.8% vs. 0.8%). The c-med° alpha measured pulse rate did not deviate from the one measured with the certified finger pulse oximeter (bias: 0.1 min −1 (0 to 0.1 min −1 ), level of agreement: −3.6 to 3.7 min −1 , A rms : 1.8 min −1 ). Conclusions The c-med° alpha fulfils the EN ISO 80601-2-61 standard and is sufficiently accurate for measuring SpO 2 and pulse rate in healthy adults at rest. Trial registration EUDAMED No. CIV-21-03-036033