The pulse in reflectance pulse oximetry: modeling and experimental studies.

OBJECTIVE: Reflectance pulse oximetry permits the use of alternative monitoring sites such as the face or torso, and is the approach commonly employed in fetal pulse oximetry systems. The purpose of this study is to investigate the impact of assumptions about the nature of arterial pulsatility on the calibration of such systems.
METHODS: Monte Carlo simulations of reflectance pulse oximetry were run on a six-layer tissue model, varying depth and magnitude of the arterial pulse. SpO2 readings on and off the femoral artery obtained during desaturation studies in newborn piglets were compared to predictions. Results. Monte Carlo simulation results clarified the difference between deep and shallow pulsatility found with photon diffusion models, agreeing with earlier in vivo observations. Significant overestimation of SpO2 <75% and slight underestimation >75% is expected if a sensor is placed on a highly pulsatile site. The on- and off-artery SpO2 readings recorded during desaturation in the newborn piglet follow the model predictions.
CONCLUSIONS: The sensitivity of reflectance pulse oximetry calibration to the depth and magnitude of arterial pulsatility reinforces the observation that monitoring site selection is of importance in optimizing reflectance pulse oximetry performance, particularly fetal pulse oximetry. Sites with palpable pulsatility should be avoided.