Repeatability and accuracy of fingertip pulse oximeters for measurement of hemoglobin oxygen saturation in arterial blood and pulse rate in anesthetized dogs breathing 100% oxygen.

Abstract

To evaluate the repeatability and accuracy of fingertip pulse oximeters (FPO) for measurement of hemoglobin oxygen saturation in arterial blood and pulse rate (PR) in anesthetized dogs breathing 100% O2. 29 healthy client-owned anesthetized dogs undergoing various surgical procedures. In randomized order, each of 7 FPOs or a reference pulse oximeter (PO) was applied to the tongue of each intubated anesthetized dog breathing 100% O2. Duplicate measurements of oxygen saturation (Spo2) and PR were obtained within 60 seconds of applying an FPO or PO. A nonparametric version of Bland-Altman analysis was used. Coefficient of repeatability was the interval between the 5th and 95th percentiles of the differences between duplicate measurements. Bias was the median difference, and the limits of agreement were the 5th and 95th percentiles of the differences between each FPO and the PO. Acceptable values for the coefficient of repeatability of Spo2 were ≤ 6%. Agreements were accepted if the limits of agreement had an absolute difference of ≤ ± 3% in Spo2 and relative difference of ≤ ± 10% in PR. Coefficient of repeatability for Spo2 was acceptable for 5 FPOs, but the limits of agreement for Spo2 were unacceptable for all FPOs. The limits of agreement for PR were acceptable for 2 FPOs. Results suggested that some FPOs may be suitable for accurately monitoring PRs of healthy anesthetized dogs breathing 100% O2, but mild underestimation of Spo2 was common.