Comparability of pulse oximeters used in sleep medicine for the screening of OSA

Abstract

Obstructive sleep apnea syndrome (OSA) is a frequent clinical picture. It is characterized by repetitive respiratory arrest with a consecutive decrease in arterial oxygen saturation (SaO2). In clinical practice, the number of desaturations per hour, oxygen desaturation index (ODI), is used as an important diagnostic criterion. Medical literature, however, mentions different threshold values that are defined as pathological. By means of systematic comparative measurements, the study presented here will examine to what extent the diagnosis and the quantification of OSA severity are affected by the device-specific measurement technique, thus impacting the predictive value of nighttime pulse oximetry in outpatient OSA screening. Different pulse oximeters commonly used in clinical practice were analyzed comparatively regarding technical parameters, temporal dynamics and the reproducibility of measuring results. The measurements were executed simultaneously and time synchronized in a reference group of five test subjects (four males, one female, average age 33.0 ± 9.4 years), in a group of five patients (all males, average age 51.8 ± 18.4 years) and using a simulator (pulse oximeter simulator index 2). All devices underestimate the simulator's predetermined oxygen desaturation of 10%. The dispersion of values is high. The device-specific characteristics have a significant influence on the collected data. The fundamental weakness of the systems lies in the reproducibility of measuring results (this only seems adequate at a signal resolution in steps of 0.1%) as well as the differing temporal dynamics. In the synchronous use of different systems on patients for the purpose of a direct comparison of devices, the dispersion of values is serious, reaching a fluctuation range of up to factor 1.42. In measuring dynamic events (apneas), different pulse oximeters do not record identical values. This is due to the different internal signal processing of the devices. Without prior knowledge of the pulse oximeter used and the chosen device settings, meaningful interpretation of the measured desaturations is, therefore, ambiguous. Accordingly, different devices require different threshold values in determining the ODI. Standardized technical parameters and the standardization of signal processing are imperative for outpatient screening of sleep-related breathing disorders (SRBD) via pulse oximetry.