2000 OPEN FORUM Abstracts
DIGITAL PASS-THROUGH PULSE OXIMETER PROBES ARE NOT ACCURATE OR PRECISE WHEN USED AS A REFLECTANCE DEVICE
Lori Ingalls, Christine Fogarty, Rebecca Fasnacht, Aleisha Kuehne, Sivorn Ouk, Kory Bowe, Jeff Ward RRT, James Pringnitz, RRT, Fred Helmholz, MD, David Plevak, MD. Rochester Community & Technical College-Mayo Clinic, Rochester, MN
Objective: Digital pulse oximetry probes are designed to be placed on a finger tip using a pass-through technique (PTT). We noticed that patients in our institution were being monitored by forehead placement of these probes, utilizing reflectance methodology (RM). We wished to determine the precision and accuracy of the digital pulse oximeter probes in estimating oxygen saturation (SpO2) when utilizing RM. Methods: Forty adult hospital patients were studied. The pulse oximeters (Mallincrodt-Nellcor® with D-25 single-patient use probes), and co-oximeters (Chiron Diagnostics) were calibrated using standard protocols. Two digital pulse oximeter probes were simultaneously applied to each patient. One probe, secured to the forehead, used RM to estimate Sp02. A second probe was secured to a fingertip and utilized the PTT. Prior to placement of the probes, each probe site was rubbed with an alcohol swab for 30 seconds to improve local circulation. The forehead probe was secured by a strip of tape. The finger probe was placed on the hand opposite the radial artery cannula. If no cannula was present, a radial artery aspirate was performed, and the finger probe was placed on the hand opposite the aspiration. After application to the patient, the probes were allowed to stabilize for 1 minute. As an arterial blood sample was being drawn from the patient, values were recorded from both pulse oximeters. These readings were compared to the "gold standard" co-oximeter oxyhemoglobin values (SaO2). Accuracy and precision for both pulse oximeter methods were determined and Bland-Altman plots were used to provide a visual display of the comparison.
Results: The mean difference from SaO2 values for the RM was 4.58% (accuracy). The standard deviation of theses differences (Sa02 minus RM) was 9.98% (precision). Fig. 1 displays a Bland-Altman plot of these data. In contrast, the mean difference from Sa02 for PTT was 2.18%. The standard deviation of these differences (Sa02 minus PT) was 3.23%.
Conclusions: The digital pulse oximeter probe is not accurate or precise when used as a reflectance probe.
Fig. 1. (See Original for Figure)