Sleep disturbance in PTSD and other anxiety-related disorders: an updated review of clinical features, physiological characteristics, and psychological and neurobiological mechanisms

Abstract

The current report provides an updated review of sleep disturbance in posttraumatic stress disorder and anxiety-related disorders. First, this review provides a summary description of the unique and overlapping clinical characteristics and physiological features of sleep disturbance in specific DSM anxiety-related disorders. Second, this review presents evidence of a bidirectional relationship between sleep disturbance and anxiety-related disorders, and provides a model to explain this relationship by integrating research on psychological and neurocognitive processes with a current understanding of neurobiological pathways. A heuristic neurobiological framework for understanding the bidirectional relationship between abnormalities in sleep and anxiety-related brain pathways is presented. Directions for future research are suggested.

Fig. 1

Important wake, NREM sleep, and REM sleep-regulating brain structures and/or nuclei, and associated neuromodulatory milieus. Wake-promoting regions, in green 1a–i; NREMS (including SWS) promoting regions in blue 2a–b; REM sleep-regulating regions in orange (3a–d). Wake (green): 1a: lateraldorsal tegmentum (LDT; Ach); 1b: pedunculopontine tegmentum (PPT; Ach); 1c: locus coeruleus (LC; NE); 1d: dorsal raphe nucleus (DRN; 5HT); 1e: tuberomammillary nucleus (TMN; Hist); 1f: hypocretin neurons of Lateral Hypothalamus (LH; Hct); 1g: cholinergic neurons of basal forebrain (BF; Ach); 1h: ventrolateral periaqueductal gray (vlPAG; DA); 1i: ventral tegmental area (VTA, DA). NREMS (blue): 2a: ventrolateral preoptic nucleus (VLPO; GABA, galanin; human homolog: intermediate nucleus (IN)); 2b: parafacial zone (PZ; GABA). REMS (orange): 3a: sublaterodorsal tegmental nucleus (SDT; Glu); 3b: ventral gigantocellular reticular nuclei (mediate REM-related muscle atonia; GABA/glycine); 3c: MCH neurons of lateral hypothalamus (LH; MCH/GABA); 3d: ventrolateral periaqueductal gray (vlPAG; GABA), *which is REM-suppressing. **LDT and PPT Ach neurons are also active during and may promote REMS, but may not play a central regulatory role

Fig. 2

Polysomnographic characteristics of different sleep stages in humans, along with a list of their associated clinical features and proposed or evidence-based functions selected for their pertinence to anxiety disorders. Some studies have studied the effects of sleep overall (especially sleep duration) on performance in various domains, such as executive cognitive function and emotion regulation, whereas other studies have examined the role of specific sleep stages on performance of various functions more specifically

Fig. 3

Important brain regions involved in fear, threat, and anxiety expression and modulation. Neuroimaging research indicates that the amygdala and insula are involved in the expression of fear, threat and anxiety. The dorsal anterior cingulate cortex (dACC) is more involved in the processing and expression of anxiety and fear, whereas the rostral ACC (rACC) is more involved in their modulation. The medial prefrontal cortex (mPFC) is involved in their modulation, especially the ventromedial PFC (vmPFC)

Fig. 4

Schematic diagram including the psychological and neurocognitive factors that link sleep disturbance and anxiety-related disorders in a bidirectional fashion. Sleep disturbance, regardless of cause, may result in maladaptive sleep-related compensatory behaviors, resulting in sleep timing that is out-of-synchrony with sleep drive, and therefore wake time in bed. This results in hyperarousal (i.e., cognitive and emotional) in the sleep environment, promoting a vicious cycle of sleep disturbance, compensatory behaviors, hyperarousal cognitions and anxious emotions. Sleep disturbance also has deleterious effects on cognitive functions, due to insufficient sleep-dependent recovery of neural substrates that carry out said functions during wake; this results in widespread deficits including deficits in emotion regulation (i.e., reduced ability to modulate anxious emotions). Sleep disturbance also disrupts processes thought to be carried out during sleep, such as adaptive memory functions (e.g., fear extinction retention; safety learning); this also promotes or maintains anxiety/fear emotions. Anxious emotions and cognitive hyperarousal are mutually reinforcing, and feed back into the sleep disturbance cycle. The specific role of REMS and NREMS in recovery and sleep-dependent processes is a topic of investigation. The nature of mechanistic connections between anxiety-related events such as nightmares and panicked awakenings, and NREM or REM sleep-dependent processes, remains to be demonstrated with empirical evidence. At the very least, their promotion of cognitive hyperarousal and anxious emotions can be inferred from their contribution to sleep disruption and the fearful emotions they generate. A neuromodulatory milieu involving heightened arousal signals, e.g., norepinephrine, and/or hypocretin, drives and underlies this relationship

Fig. 5

Heuristic neurobiological framework for the sleep disturbance and anxiety disorder relationship. Abnormal activation in wake, NREMS, and/or anxiety regions may generate sleep disturbance in anxiety-related disorders and promote a bidirectional sleep-disturbance-anxiety relationship. a Sleep disturbance emerges from hyperactivation in intrinsic wake circuitry. For example, here, excess arousal in NE-generating LC (green with orange halo) results in inhibition of VLPO/Intermediate Nucleus (IN) (blue) and hyperactivation of cerebral cortex. b Sleep disturbance emerges from dysfuncton in central (NREM) sleep-promoting region. For example, here failure of VLPO GABA signaling (blue, with orange “X”) disinhibits arousal centers (green). c Sleep disturbance emerges from hyperactivation of fear/threat/anxiety regions. For example, here amygdala hyperactivation (red, with orange halo) sends excitatory inputs to LC, which then inhibits VLPO-mediated sleep promotion and sends excitatory inputs to cerebral cortex, resulting in cortical hyperarousal, as in a. A common node in these examples is LC and NE signaling, but aberrant activity in other wake, NREMS, REMS, and anxiety regions may produce analogous effects