Anthony M. Vintzileos, MD, professor and chair, Department of Obstetrics, Gynecology and Reproductive Sciences, UMDNJ-Robert Wood Johnson Medical School

The aim of this research was to determine the feasibility of non-invasive fetal pulse oximetry in the human fetus by using transabdominal continuous-wave near-infrared spectroscopy. First, a fetal heart rate was obtained by using a fetal heart rate monitor. We recorded the spectroscopy data at optical densities at 735 and 850 nm in six pregnant women at greater than 36 weeks gestation and the fetal heart rate for approximately one to 10 minutes in each patient. The mean oxygen saturation values of each of the six individual fetuses ranged from 50% to 74% [overall mean saturation 61% ± 14.8% (SD)].

These results suggested that transabdominal fetal pulse oximetry is feasible for human patient application since measured values were similar to those obtained by using transvaginal (internal) fetal pulse oximetry.

The introduction of pulse oximetry (arterial oxygen saturation monitoring) has dramatically improved patient care in many fields, including anesthesiology, intensive care and newborn intensive care. In obstetrics, fetal heart rate (FHR) monitoring, although an indirect measure of fetal oxygenation, continues to be the standard for antepartum and intrapartum assessment of fetal well-being. Intrapartum fetal oxygen saturation monitoring is a relatively new technique that gives objective and reliable information regarding fetal arterial oxygen status, and it has been recently introduced for use in the intrapartum period when electronic FHR monitoring reveals non-reassuring FHR patterns. However, the current fetal pulse oximeters can only be used during the intrapartum period (during labor) and require a cervical dilation of at least 2 cm and ruptured membranes. Noninvasive transabdominal fetal pulse oximetry has never before been performed on the human fetus, although the noninvasive nature of such a tool would be ideal for evaluating fetal oxygenation not only during labor, but also during the antepartum period. For transabdominal fetal pulse oximetry, continuous-wave near infrared (NIR) technology offers the advantage of a large quantity of photons penetrating the maternal tissue and reaching the fetus. Another requirement is the use of transmittance probes with wavelengths as high as 735 - 850 nm, as compared to 660 - 690 nm for adult surface oximetry or 720 - 750 nm for transvaginal pulse oximetry.

Optical Devices Inc., in collaboration with the Department of Biochemistry and Biophysics at the University of Pennsylvania and the Department of Optical Engineering at Huazhong University of Science and Technology in China, have obtained promising results using this technology for transabdominal fetal pulse oximetry in various experimental models. As a primary investigator, over the last several months our Division of Maternal-Fetal Medicine at RWJMS has embarked on a research project with Optical Devices, Inc. to examine, for the first time, the feasibility of non-invasive fetal pulse oximetry in the human fetus by using continuous-wave near-infrared spectroscopy. We reported the results of this effort at the 24th annual meeting of the Society for Maternal-Fetal Medicine last February, and a paper emanating from this research has been recently published by the American Journal of Obstetrics and Gynecology.

The transabdominal fetal pulse oximeter used in these studies has high tissue penetration and improved signal to noise ratio. Three wavelength light emitting diodes (LED) serve as light sources at 735, 805, and 850 nms. The light penetrates the fetal head through the maternal abdomen, and some portion of the light from the tissue, including the fetus, migrates back to a detector (photo-multiplier) by the well known photon migration process. The signal voltage from the photo-multiplier was amplified and changed to a digital signal through an analog/digital board and sent to a computer to be saved as a file. The data analysis was based on the modified Beer-Lambert law with optical densities of 735 and 850 nm used to obtain the concentration changes of the oxy-hemoglobin (oxy-HB) and deoxy-hemoglobin (deoxy-HB). The saturation was expressed as oxygen saturation % = 100* oxy-HB/ (oxy-HB+deoxy-HB). This instrument was applied to six pregnant women with a gestational age greater than 36 weeks who visited the hospital for a regular checkup or nonstress testing. All the fetuses were healthy and were subsequently born vaginally without any complication. They all were normally grown and had 1 and 5 minute Apgar scores greater than 7. Patients used either the semi-Fowler or sitting position. An FHR monitor was used to obtain the FHR signal. To minimize stray light, which causes the detector to be saturated, we covered the optodes’ location with a thick black cloth so that the procedure could be carried out in room light. After optimizing the light intensity and detection sensitivity of the three wavelengths, we recorded the spectroscopy data and the FHR for about one to 10 minutes for each patient. The FHR information was recorded and stored into the spectroscopy data file for later analysis.

The results showed that the mean fetal oxygen saturation values for each of the six fetuses ranged from 50% to 74%. The overall mean saturation was 61% ± 14.8% (SD). These results are very compatible and similar to the expected fetal oxygen saturation values found after invasive (internal) fetal pulse oximetry.

Enlarge the image Figure 1: Fourier transformed signals of the three light intensities and FHR from the ultrasound monitor. (A peak at 2.4 Hz NIR signal for all three wavelengths is the same as that of FHR indicating that the 2.4 Hz frequency carries the FHR signal.)

The implications of this research are very significant in many ways. First, although the intrapartum (internal) use of continuous fetal oxygen saturation may address the high false positive results of using fetal heart rate monitoring during labor, this method requires a transcervical application. Also, the problem of high false positive results still remains during the antepartum period when FHR monitoring or fetal biophysical profiles are used to evaluate fetal status. Therefore, our non-invasive technique to evaluate fetal oxygenation status in an objective and accurate manner will be beneficial in both antepartum as well as intrapartum periods. Reliable transabdominal monitoring of fetal oxygen saturation is expected not only to improve, but also to revolutionize, our ability to accurately predict fetal hypoxia/acidemia in utero, especially during the antepartum evaluation of high risk fetuses prior to labor.

One of the main differences between our fetal pulse oximeter and the adult pulse oximeters is that our technology allows us to penetrate deep tissues, whereas the adult devices measure surface tissue and therefore can only penetrate to a depth of less than 1 cm. By using this technology, we developed an experimental, non-invasive, no-risk device for monitoring fetal arterial saturation from the maternal abdomen, and we demonstrated that this machine can successfully measure fetal signals. We have shown that the transabdominally obtained fetal arterial oxygen saturation ranges between 50% and 74% (mean 61%). These values are within the fetal arterial saturation values that have been obtained transvaginally. Ours is the first report of measurement of fetal oxygen saturation values in non-laboring patients. We have demonstrated that transabdominal fetal pulse oximetry is, indeed, feasible for human patient application. Future studies should explore the clinical usefulness of transabdominal fetal pulse oximetry.