Artifacts in fetal pulse oximetry: incomplete sensor-to-skin contact

Our purpose was to investigate apposition of reflectance sensors as a possible source of artifact in fetal pulse oximetry.

A laboratory model was created to examine the effect of varying the distance of a reflectance probe from a subject's finger. The setup used two wavelengths (660 and 940 nm) and pulsatile (AC) and nonpulsatile signals (DC). An investigator's finger was stabilized in a purposely built mold from where the distance between skin and sensors could be varied in 1 mm steps.

The AC and DC components of the signal react differently to increasing the distance between sensor and skin. The overall effect is that the ratio [formula: see text] increases with skin to sensor distance, which results in falsely low readings of oxygen saturation.

Poor sensor-to-skin contact in reflectance pulse oximetry may cause false readings despite the presence of good waveforms and hence go unrecognized. Probe and software design needs to be able to identify and exclude such artifact before this technique can become useful for intrapartum fetal monitoring.