427 - Oxygen Saturation Histograms as a Tool to Quantify Hypoxemia Burden and Predict NICU Course among Very-Low-Birthweight Premature Infants

Poster Number: 427
Publication Number: 427.223

Liron Borenstein-Levin, Rambam Medical Center, Haifa, HaZafon, Israel; Viktoria Leikin Zach, Rambam Medical Center, Haifa, Hefa, Israel; Irina Raizberg, Rambam Medical Center, Haifa, Hefa, Israel; Gil Dinur, Rambam Medical Center, Haifa, Hefa, Israel; Yoav Littner, RAMBAM MEDICAL CENTER, HAIFA, Hefa, Israel; Amir M. Kugelman, Rambam Medical Center, Haifa, Haifa, Tel Aviv, Israel; Ori Hochwald, Rambam Medical Center, Haifa, Israel, Haifa, Hefa, Israel

Background: Premature infants experience frequent desaturation events during the first weeks of life. Oxygen saturation (SpO2) histograms, easily presented by the bedside monitor, graphically describe the amount of time the infant spent in each SpO2 value, and oxygen stability in different time periods

Objective: a) To compare the time the infant spent< SpO2 90% as documented by the SpO2 histogram, and the number of desaturations episodes reported by the bedside team. and b) To assess whether the time the infant spent at SpO2 < 90% during the first week of life affects future diagnosis of PDA requiring intervention and BPD at 36 weeks.

Design/Methods: All infants < 1500g who were born during the study period and required respiratory support during the first day of life were included in the study. 24-hours SpO2 histograms were documented once daily in predefined time. Demographic data, prematurity complications and number of desaturation episodes (SpO2 < 90%) were documented from the chart. Time spent in SpO2 < 90% was calculated from the histograms. 

Results: 73 VLBW infants with a mean (SD) gestational age of 28.5 (2.1) weeks and birth weight of 1000 (200) g participated in the study. In 1793 histograms documented during the study period, a strong Pearson correlation (r=0.79) was found between the time the infants spent in SpO2 < 90% as presented by the SpO2 histogram and number of desaturations episodes below 90%, documented by the bedside team. However, for each value of desaturation episodes documented by the team, the time the infant spent < 90% largely differed (Figure 1). This difference is probably due to different episodes’ lengths, reflected in the histogram, but not in the chart. 
Infants with BPD (defined as O2 requirement at 36 weeks corrected age) spent significantly more time at SpO2 < 90% during the first week of life as compared to infants with no BPD (median [IQR] 11.1 [8.3,14.1] % vs. 1.7 [0.1,3.4] % respectively, P< 0.0001). Similarly, infants with PDA requiring pharmacological or surgical treatment spent 10.7 [6.2,12.2] % in SpO2 < 90% as compared to 1.7 [0.14,3.3] % among infants with no PDA.Conclusion(s): SpO2 histogram can serve as an adjunctive tool to assess oxygenation status of VLBW infants. It was found to better demonstrate the oxygenation instability when compared to the charting by the bedside team. Time spent in SpO2 < 90% is suggested as a marker for future prematurity morbidities, such as BPD and significant PDA, and perhaps allow earlier intervention to be considered.
Figure 1Pearson correlation revealed strong correlation (r=0.79) between the number of desaturation episodes (SpO2 < 90%) documented at the bedside chart, and the time the infant spent in SpO2 < 90%. However, major differences in time spent < 90% for each number of episodes was observed. for example: for days with 10 documented desaturation episodes, the time spent at SpO2 < 90% ranged from 1% to 18%