Autonomic activation at bedtime predicts subsequent spindle density

Materials and methods: All subjects (n= 49, 21 females, age: 27.40±11.31) underwent a night of polysomnographic recordings at home. Sleep Efficiency (SE, time asleep/time in bed) was computer and adopted as objective sleep quality proxy. The slow (10-13 Hz) and the fast (13-16 Hz) spindles were automatically detected from N2 NREM sleep stage on central (C3, C4) and frontal (F3, F4) electrodes. For each spindle parameter (amplitude, duration, intensity, density) a weighted average was calculated between homologous electrodes. Time and frequency domain analysis of HRV were conducted on the electrocardiogram (ECG) signal characteristics of the 5 minutes before the sleep onset. The parameter extracted were SDNN (standard deviation of normal to normal RR intervals), RMSSD (root mean square of successive RR differences), pNN50 (percentage of successive RR that differ by more than 50 ms), and …