Sleep spindle and psychopathology characteristics of frequent nightmare recallers.

Frequent nightmare recallers show abnormalities in non-REM (NREM) sleep, specifically in sleep spindles—brain oscillations linked to memory consolidation and psychopathology. Compared to controls, the nightmare group had lower density of slow spindles across most EEG sites, higher density of fast spindles in frontal regions, and elevated fast spindle frequency in central regions. These differences remained after controlling for depression. Correlational analyses revealed that higher psychopathology scores were associated with higher slow spindle densities and slower spindle frequencies only in the nightmare group.

Study at a glance

Abstract

Idiopathic nightmares are a common disturbance of rapid eye movement sleep (REM) sleep, but studies of comorbid pathologies and sleep architecture suggest that non-REM (NREM) sleep is also affected. Sleep spindles are a NREM sleep characteristic associated with both pathophysiology and sleep-dependent memory consolidation, yet they have not been evaluated in frequent nightmare recallers. The morning naps of 38 participants with frequent idiopathic nightmares (mean age: 23.7 ± 3.78 years) and 25 age- and sex-matched controls (23.9 ± 3.65 years) were recorded and their sleep evaluated. A custom spindle detector assessed NREM sleep stage 2 (N2) sleep spindles on six electroencephalogram (EEG) derivations (F3, F4, C3, C4, O1, and O2) for density (number spindles/N2 time), mean frequency, and amplitude. Total spindles (10-16 Hertz (Hz) range), slow spindles (10-12.79 Hz), and fast spindles (12.8-16 Hz) were all assessed separately. Compared with the Control group, the Nightmare group had longer N2 sleep latency and a marginally greater %N2 sleep. The Nightmare group also had a lower than normal density of slow spindles in most EEG derivations, a higher density of fast spindles in frontal derivations, and an elevated fast spindle oscillatory frequency-"faster fast" spindles-mainly in central derivations. These differences withstood controls for pre-existing group differences in depression. Correlational analyses demonstrated a further pattern of group differences by which higher pathology scores were associated with higher slow spindle densities and slower spindle frequencies for the Nightmare but not the Control group. A similar pattern was observed for some dream content measures, ie, the Nightmare group showed positive correlations of slow spindle density with dreamed fear and word count and negative correlations with dreamed positive emotion. Conversely, the Control group showed opposite trends. Results thus demonstrate abnormalities in the composition of N2 sleep-and especially in N2 spindles-among frequent nightmare recallers and link these abnormalities to both trait (psychopathology) and state (dream content) factors. Spindle findings for psychopathology resemble, but are not identical with, previous findings for patients with major depression, social anxiety, and schizophrenia and are thus consistent with an explanation implicating spindles as trait markers of psychopathology. Correlational analyses go beyond a trait explanation to suggest several possible state-based explanations involving memory consolidation mechanisms, specifically, the possibility that spindles index either emotional or verbal task-based processes.