Abstract Introduction Slow Wave Activity (SWA; EEG power 0.5–4 Hz) reflects sleep pressure and homeostatic regulation. In healthy sleep, SWA declines across the night as sleep pressure dissipates. Pediatric sleep-disordered breathing, including obstructive sleep apnea (OSA), may disrupt this process, yet traditional severity metrics such as Apnea–Hypopnea Index (AHI) fail to capture neurophysiological impact. We hypothesized that SWA dynamics, modeled as exponential decay, would reveal region-specific disruptions associated with respiratory event severity. Methods We analyzed overnight polysomnography from 52 children (ages 4–11 years) with varying OSA severity (AHI range: 0 to 36 event/hour). High-density EEG was segmented into artifact-free NREM epochs, and SWA was computed for frontal and posterior channels. SWA decline was modeled using an exponential function: SWA(t)=A·exp(−Bt)+C, where A(0) = initial amplitude, B (0) = decay rate, C = asymptote. SWA sampling strategies to fit the exponential function included uniform and non-uniform techniques. Both approaches considered windows of analysis of 20, 30, and 60 minutes. Linear regression models predicted apnea-hypopnea index (AHI), obstructive apnea index (OAI), and hypopnea index (HI) from regional SWA decay parameters (for frontal and posterior sites), controlling for age and gender (18 models: 3 window sizes × 2 sampling methods × 3 outcomes). Results Non-uniform sampling outperformed uniform sampling across all outcomes. Models using 20–30 min non-uniform windows explained up to 56% of variance in HI (Adjusted R² = 0.47–0.56), whereas OAI models showed negligible explanatory power. Posterior decay rate (back B) was the most consistent predictor: slower SWA dissipation (greater B) was strongly associated with higher AHI and HI (β = –7.55 for HI, p 0.001). Associations with frontal parameters were weak and inconsistent. Hypopneas exhibited stronger associations with SWA disruption than apneas, suggesting partial airway obstruction exerts greater impact on sleep homeostasis. Conclusion Modeling of SWA decline reveals clinically relevant disruptions in pediatric SDB beyond conventional metrics. Slower posterior SWA decay indicates impaired homeostatic recovery and correlates with apnea severity, particularly hypopneas. Non-uniform sampling enhances model performance, supporting its use in future research. SWA-based metrics may complement AHI to better characterize neurophysiological burden and guide treatment decisions in pediatric populations. Support (if any)
Garcia-Molina et al. (Fri,) studied this question.