Comparison of real beat-to-beat signals with commercially available 4 Hz sampling on the evaluation of foetal heart rate variability

Abstract

Evaluation of foetal heart rate (FHR) variability is an essential part of foetal monitoring, but a precise quantification of this parameter depends on the quality of the signal. In this study, we compared real FHR beat-to-beat signals with 4Hz sampling provided by commercial foetal monitors on linear and nonlinear indices and analysed their clinical implications. Simultaneous acquisition of beat-to-beat signals and their 4Hz sampling rate counterparts was performed using a scalp electrode, during the last hour of labour in 21 fetuses born with an umbilical artery blood (UAB) pH≥7.20 and 6 born with an UAB pH<7.20. For each case, the first and last 10min segments were analysed, using time and frequency domain linear, and nonlinear FHR indices, namely mean FHR, low frequency, high frequency, approximate, sample and multiscale entropy. Significant differences in variability indices were found between beat-to-beat and 4Hz sampled signals, with a lesser effect seen with 2Hz sampling. These differences did not affect physiological changes observed during labour progression, such as decreased entropy and linear time domain indices, and increased frequency domain indices. However, significant differences were found in the discrimination between fetuses born with different UAB pHs, with beat-to-beat sampling providing better results in linear indices and 4Hz sampling better results in entropy indices. In conclusion, different FHR sampling frequencies can significantly affect the quantification of variability indices. This needs to be taken into account in the interpretation of FHR variability and in the development of new equipment.