765 - Utility of Umbilical Pulse Oximetry in Air-Breathing Late Preterm & Term Newborn Infants

Publication Number: 765.7060

Valerie Elberson, John R. Oishei Children's Hospital, Buffalo, NY, United States; Mausma I. Bawa, University At Buffalo, Buffalo, NY, United States; Rocco Paluch, Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo, Buffalo, NY, United States; Munmun Rawat, University at Buffalo, Buffalo, NY, United States; Praveen Chandrasekharan, Jacobs School of Medicine and Biomedical Sciences at the University at Buffalo, Buffalo, NY, United States

Background: A pulse oximeter estimates the percentage of hemoglobin carrying oxygen by detecting pulsatility in the blood vessel (Chan. 2013). With good technique, a pulse oximeter (ox) will accurately display the oxygen saturation (SpO2) & heart rate (HR) within 1-2 min. Skin color affects pulse ox readings, showing discrepancies between light & dark (Vesoulis. 2022; Sjording. 2020). Acrocyanosis at birth could further interfere with the readings. Previously, in an ovine model, we tested the concept of umbilical pulse ox (PAS2023, EAPS2024). Although umbilical cord pulse ox is post-ductal, could it help predict the SpO2 & HR in the first 10 min after birth as skin is not involved?

Objective: Our objectives were: i) to assess the feasibility of placing an umbilical pulse ox probe post umbilical cord clamping & ii) to assess the agreement between pre & umbilical SpO2 & HR.

Design/Methods: After IRB approval, we conducted this feasibility study. A right hand & umbilical pulse ox probes were placed in newborn airbreathing infants after clamping the umbilical cord (Fig1a). Massimo probes were used, and simultaneous SpO2 & HR readings were collected. Newborn demographics, number of observations & missed readings with both pulse ox probes were collected. A difference in SpO2 ≤10% and HR ≤10 bpm was considered a positive correlation. Chi-square for categorical variables, receiver operator curve (ROC) with area under the curve (AUC), bland-altman (BA) plot, & correlation coefficients (R) were used to assess the agreement between preductal SpO2 (pSpO2) & umbilical SpO2 (uSpO2), preductal HR (pHR) & umbilical HR (uHR) .

Results: Newborn characteristics are shown in Fig1b. It was feasible to safely secure the pulse ox probe on the umbilical cord & record readings. The output from pulse ox (N=154 points) was significantly higher with umbilical pulse ox compared to preductal: pSpO2 (74%) vs. uSpO2 (95%) p < 0.001, pHR (69%) vs. uHR (92%) p < 0.001. The BA plot for pSpO2 & uSpO2 (R=0.71) with a bias of -2.6(-5.2, 0.01) & AUC of 0.907 (Fig2a&2b). The BA plot for pHR & uHR (R=0.30) with a bias of -17.7 (95% CI -29,-6) & AUC of 0.842, p< 0.01 (Fig3a&3b) suggesting agreement between models.

Conclusion(s): Umbilical pulse oximetry is a novel concept that has never been evaluated. When measured simultaneously, umbilical pulse ox had significantly more readings of SpO2 and HR than preductal pulse ox. This pilot study demonstrated that umbilical pulse ox is feasible with proper technique & could be a reasonable alternative. However, because preductal pulse ox was not accurate, studies with larger sample sizes are needed & are ongoing.

Figure 1
Figure 1a – Shows the preductal pulse oximeter probe and umbilical pulse oximeter probe, as pointed out by the (red) arrows, a newborn infant with the monitor displaying preductal SpO2 (pSpO2), heart rate (pHR), and the interrupted arrows pointing at umbilical SpO2 (uSpO2) & umbilical HR (uHR). Figure 1b – Shows the characteristics of the newborn. N-number. C – Cesarean section, V- vaginal delivery.

Figure 2
Figure 2a – Shows the bland-altman (BA) plot between preductal SpO2 (pSpO2) & umbilical SpO2 (uSpO2) with bias -2.6 (95% confidence interval -5.20,0.01) showing agreement between the two models. Figure 2b – Shows the receiver operator curve for preductal SpO2 (pSpO2) & umbilical SpO2 (uSpO2) with significant area under the curve (AUC=0.907,p < 0.01).

Figure 3
Figure 3a – Shows the bland-altman (BA) plot between preductal heart rate (pHR), & umbilical HR (uHR) with bias -17.7 (95% confidence interval -29,-6) showing agreement between the two models. Figure 3b – Shows the receiver operator curve for preductal heart rate (pHR), & umbilical HR (uHR) with a significant area under the curve (AUC=0.842,p < 0.01).