Abstract Rationale Sleep apnea diagnosis in infants requires polysomnography (PSG), but timely PSG is not readily available in many settings. Prior research assesses correlation with oxygen desaturation index (ODI) on pulse oximetry study and apnea hypopnea index (AHI) on PSG, but data is limited in infants. This pilot study prospectively evaluates AHI/OAHI correlation with ODI and how these correlations vary for various definitions of oxyhemoglobin desaturation in infants. Methods Infants aged 12 months undergoing PSG were prospectively enrolled between 11/1/2024 and 8/1/2025 for concurrent separate oximetry recording. Correlation between ODI from oximetry report and both AHI and obstructive AHI (OAHI) from PSG were determined. Different ODI definitions were included based on desaturation (3 or 4%) and duration (1, 5, or 10 seconds). Exclusion criteria included cyanotic congenital heart disease, neuromuscular disease, severe neurologic conditions, inadequate oximetry study data, and significant non-apneic hypoxemia. Seven PSG’s were split-night studies with diagnostic portion on room air, followed by supplemental oxygen titration. Spearman correlation analysis was performed for the entire study and for the diagnostic portion. Results There were 19 patients included, with average age at PSG 7.5±2.9 months (range 2.3-11.6 months). The respective median AHI and OAHI were 8.1 events/hour (IQR 5.8-9.5) and 6.3 events/hour (IQR 3.7-7.0) for the entire PSG, and 8.9 events/hour (IQR 6.9-13.4) and 6.5 events/hour (IQR 3.7-10.0) for the diagnostic portion. For the entire PSG, the ODI correlation was significant with AHI using either an ODI definition of 3% for 5 seconds (r = 0.46, p = 0.049) or 4% for 5 seconds (r = 0.5, p = 0.03). For the diagnostic portion, there was significant correlation between AHI as well as OAHI and ODI of 1 and 5 seconds, but the strongest correlation was with 3% desaturation for 1 second, both for AHI (r = 0.51, p = 0.02) and OAHI (r = 0.57, p = 0.01). There were no significant correlations between AHI/OAHI and ODI of 10 seconds. Conclusion There is significant correlation between ODI from oximetry report and AHI/OAHI from PSG in infants, supporting the use of oximetry as a potential screening evaluation for sleep apnea. ODI with longer desaturation durations tended to underestimate AHI/OAHI. Supplemental oxygen use may impact correlation as shown by a lack of OAHI and AHI correlation with ODI at short duration. Further research is needed to elucidate the mechanism underlying the effect of supplemental oxygen on accuracy of oximetry study. This abstract is funded by: The Frances R Luther Charitable Trust and the Cincinnati Children’s Hospital Medical Center Research Fund
Ignatiuk et al. (Fri,) studied this question.