Can Nap Polysomnography Predict Obstructive Sleep Apnea Syndrome in Snoring Children? - A Pilot Study

Abstract

SESSION TITLE: Sleep Disorders Posters I: Diagnosis SESSION TYPE: Original Investigation Poster PRESENTED ON: Wednesday, October 28, 2015 at 01:30 PM - 02:30 PM PURPOSE: The studies on nap polysomnography (nap-PSG) were very scarce. Previous recommendations were against the use of nap-PSG due to its poor sensitivity. However, the diagnostic criteria for childhood obstructive sleep apnea (OSA) changed periodically to increase sensitivity. Newer devices in polysomnography were also expected to yield more sensitivity and specificity. We aimed to evaluate the value of nap-PSG as a tool for diagnosis of childhood OSA, when compared to standard overnight polysomnography (standard-PSG). METHODS: Snoring patients with regular nap schedule, excluding those with medication affecting sleep, were included. Patients underwent standard-PSG first and then nap-PSG within a month. Both tests were done with the same recording devices and in the same manner. Patient's sleep behavior was observed by a sleep technician. Patients were categorized into primary snoring, mild OSA, moderate OSA and severe OSA. Sleep parameters between nap-PSG and standard-PSG were compared. RESULTS: Ten patients, median (ranged) aged was 5 (2-7) years old, 7 male, median BMI was 16.3 (12.9-22.2), were recruited. Median (ranged) apnea-hypopnea index (AHI) and arousal index of nap-PSG were 5.25 (0-43.6) and 23.95 (4.0-43.4) respectively, whereas those of standard-PSG were 5.6 (0.4-15.5) and 14.8 (9.6-21.0) respectively. There was no statistical significant difference between two groups. Sensitivity and specificity for nap-PSG in diagnosis of OSA were 88.9% and 100%. Measurement of agreement among diagnosis revealed a good agreement (P-value < 0.01, Kappa = 0.61). Scatter plot showed a monotonic relationship between AHI from both tests. Spearman's rank correlation coefficient revealed a high positive correlation (P-value < 0.01, correlation coefficient = 0.833). However, Bland-Altman plot's average discrepancy and limits of agreement were large (Mean = -7.6, SD = 11.30). CONCLUSIONS: Nap-PSG showed a promising trend as an alternative diagnostic tool for standard-PSG. Bland-Altman plot's disagreement might be because of small sample size. Pre versus Post nap-PSG analysis should be done to eliminate first night effect. Nap-PSG may need a new AHI cut-off point. CLINICAL IMPLICATIONS: Nap-PSG may help predict OSA in snoring children or screen in clinical urgency case. The cost in diagnosis of OSA may be lower. Furthermore, this study will help us understand physiology of nap in OSA and non OSA children. DISCLOSURE: The following authors have nothing to disclose: Chula Kooanantkul, Teeradej Kuptanon, Aroonwan Preutthipan No Product/Research Disclosure Information