Can overnight portable pulse oximetry be used to stratify obstructive sleep apnea risk in infants? A correlation analysis.

ABSTRACT

INTRODUCTION: There is limited evidence on the accuracy of oximetry in the evaluation of infant obstructive sleep apnea (OSA). We aimed to determine the utility of overnight oximetry to stratify infants at risk for OSA, to determine urgency for definitive screening with an overnight in-laboratory polysomnogram (PSG). METHODS: Retrospective single-institution cohort study of infants undergoing PSG and a separate overnight oximetry over an 8-year period. Correlations, using oximetry in both in-hospital (attended) or at-home (unattended) settings, for ODI410 (decrease in oxygen saturation ≥ 4% from baseline, duration ≥ 10 seconds) and ODI40 (duration > 0 second) with the obstructive apnea-hypopnea index (AHIo) were obtained. The area under the curve was calculated, and sensitivity and specificity values have been presented as receiver operating characteristic curves. RESULTS: Thirty-eight infants were included. The mean (SD) age (months) was 5.7 (3.9) at diagnostic PSG and 5.5 (3.7) at the time of oximetry. The mean AHIo for the entire cohort was 6.7 (6.2). The mean (SD) ODI40 was 8.6 (9.0) and the mean (SD) ODI410 was 5.4 (5.1).The correlation between ODI and AHIo was statistically significant for the cohort (ODI40 vs. AHIo [r = .59, P < .001] and ODI410 vs AHIo [r = .55, P = .0003]). Using an ODI40 cutoff of 3, the sensitivity, specificity, negative predictive value and positive predictive value for diagnosing OSA was 86%, 40%, 50%, and 80% respectively for an AHIo greater than 2, and 100%, 35%, 100%, and 58% respectively for an AHIo greater than or equal to 5. CONCLUSION: There is a significant positive correlation between the ODI4 obtained from oximetry and the AHIo obtained from PSG in infants at risk for OSA. An ODI40 greater than 3 may be useful to stratify infants at risk for moderate to severe OSA when used in attended (in-hospital) or unattended (in-home) settings.