Sleep-amount differentially affects fear-processing neural circuitry in pediatric anxiety: A preliminary fMRI investigation

Christina O Carlisi,Amanda E Guyer,Kevin Hilbert,Monique Ernst

doi:10.3758/s13415-017-0535-7

Abstract

Insufficient sleep, as well as the incidence of anxiety disorders, both peak during adolescence. While both conditions present perturbations in fear-processing-related neurocircuitry, it is unknown whether these neurofunctional alterations directly link anxiety and compromised sleep in adolescents. Fourteen anxious adolescents (AAs) and 19 healthy adolescents (HAs) were compared on a measure of sleep amount and neural responses to negatively valenced faces during fMRI. Group differences in neural response to negative faces emerged in the dorsal anterior cingulate cortex (dACC) and the hippocampus. In both regions, correlation of sleep amount with BOLD activation was positive in AAs, but negative in HAs. Follow-up psychophysiological interaction (PPI) analyses indicated positive connectivity between dACC and dorsomedial prefrontal cortex, and between hippocampus and insula. This connectivity was correlated negatively with sleep amount in AAs, but positively in HAs. In conclusion, the presence of clinical anxiety modulated the effects of sleep-amount on neural reactivity to negative faces differently among this group of adolescents, which may contribute to different clinical significance and outcomes of sleep disturbances in healthy adolescents and patients with anxiety disorders.

Highlights

Adolescence is a period of heightened risk for the onset of psychiatric problems, anxiety (Beesdo, Knappe, & Pine, 2009; Pine, Cohen, Johnson, & Brook, 2002; Pine & Fox, 2015)
Three hypotheses guided this work: (1) ventromedial PFC (vmPFC) activation would be reduced in anxious relative to healthy adolescents during negative emotion processing; (2) sleep amount would have a stronger negative impact on the neural correlates of emotion processing, including hippocampus, amygdala, insula, and medial PFC (mPFC) in anxious versus healthy adolescents; and (3) the functional connectivity of these regions would be influenced by sleep differently in anxious compared to healthy adolescents, yet specific predictions are limited due to a lack of existing data to drive expectations
The analyses revealed a correlation between dorsal anterior cingulate cortex (dACC)–dorsomedial PFC (dmPFC) connectivity and sleep amount that was negative in anxious adolescents (AAs) (r = −.736, p < .01), and positive in healthy adolescents (HAs) (r = .533, p < .05; see Fig. 4a)

Summary

Introduction

Adolescence is a period of heightened risk for the onset of psychiatric problems, anxiety (Beesdo, Knappe, & Pine, 2009; Pine, Cohen, Johnson, & Brook, 2002; Pine & Fox, 2015). Sleep deprivation has been suggested to induce abnormal DMN activation or to negatively impact effective allocation of brain resources to task-relevant demands in a way that may explain enhanced emotionality associated with sleep deprivation (Regen et al, 2016) Based on this background, this study assessed whether the presence of an anxiety disorder moderates sleep effects on emotion processing in adolescents. Three hypotheses guided this work: (1) vmPFC activation would be reduced in anxious relative to healthy adolescents during negative emotion processing; (2) sleep amount would have a stronger negative impact on the neural correlates of emotion processing, including hippocampus, amygdala, insula, and mPFC in anxious versus healthy adolescents; and (3) the functional connectivity of these regions would be influenced by sleep differently in anxious compared to healthy adolescents, yet specific predictions are limited due to a lack of existing data to drive expectations. Adolescents completed the Childhood Depression Inventory (CDI; Kovacs, 1978) and the StateTrait Anxiety Inventory (STAI), trait version (Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983)

Participants

Results

Discussion

Limitations

Conclusion

Compliance with ethical standards