Nightmares before military deployment, that is, a period of increased risk of trauma exposure, increase the risk for PTSD symptom development in response to deployment. Furthermore, sleep apnea has been related to higher PTSD scores in PTSD patients. These observations indicate that disturbed sleep may influence PTSD symptomatology. Evidence from previous studies show that sleep promotes generalization of extinction of memory in healthy humans (Pace-Schott et al., 2009
; Spoormaker et al., 2010
). This indicates that disturbed sleep may directly contribute to PTSD development by means of disrupting the beneficial process of sleep with regard to fear extinction. Nightmares predominantly occur during REMS (Nielsen & Levin, 2007
), while they may also occur during nonrapid eye movement sleep in PTSD patients (Hefez, Metz, & Lavie, 1987
). REMS in particular seems to have an effect on fear extinction (Spoormaker et al., 2010
). The observed relationship between nightmares and the risk of developing PTSD symptoms may suggest that disturbed REMS is a risk factor for PTSD.
Besides, nightmares have been associated with increased noradrenalin levels (Raskind et al., 2007
). Furthermore, noradrenergic activity may be involved in PTSD development (Southwick et al., 1999
). Thus, alternatively, the development of PTSD and the occurrence of nightmares may be epiphenomena, both induced by increased noradrenalin levels, and may not be causally linked. Our results suggest that nightmares are a trait, making an individual more vulnerable to developing PTSD symptoms. Our study also shows that insomnia symptoms fail to predict PTSD symptoms at 6 months post-deployment when pre-deployment mood and anxiety complaints are taken into account. These results differ from the previous studies in which a positive association between the development of PTSD and insomnia symptoms before or directly after trauma exposure (Bryant et al., 2010
; Koren et al., 2002
; Mellman, David, et al., 1995
) was observed. This may be explained by the fact that these studies did not correct for mood and anxiety complaints. The relationship between insomnia symptoms and mood and anxiety complaints is complex and may be bidirectional: insomnia symptoms may contribute to mood and anxiety complaints and—vice versa—insomnia symptoms may be moderated by mood or anxiety complaints (Abad & Guilleminault, 2005
). Therefore, it is possible that mood and anxiety complaints caused by insomnia increase the risk of developing PTSD after trauma and are therefore mediators in the relation between insomnia symptoms and PTSD development. Alternatively, mood and anxiety complaints may be confounding factors, causing both insomnia symptom complaints and PTSD development. In the current design, we could not differentiate whether mood and anxiety complaints are confounding factors or mediators in the relation between insomnia symptoms and PTSD development.
OSAS was not more prevalent in patients with PTSD compared to TCs and HCs, contrary to previous uncontrolled studies that suggested indices between 60 and 90% in PTSD (Krakow, Germain, et al., 2000
; Krakow et al., 2004
; Yesavage et al., 2010
In these studies, screening instruments for detecting OSA may have been more sensitive than in our study, especially because some studies defined a cutoff of five events per hour. Another explanation for the high incidence of OSA in some previous studies is that the usage of benzodiazepines was not discontinued before sleep recordings, which increases the occurrence of OSA (Dolly & Block, 1982
). In our study, participants with regular benzodiazepine usage were excluded, and participants with habitual benzodiazepine usage were refrained from sleep medication in the sleep laboratory. Finally, our study group consisted of middle-aged veterans, whereas other studies included either female PTSD patients or elderly veterans with PTSD. The incidence of OSA may be different in other populations with PTSD. As none of the previous studies included a control group, it cannot be concluded that the incidence of OSA is elevated in PTSD. Our study underlines the importance of controlled studies to determine whether OSA is more prevalent in PTSD than in matched controls.
Nonetheless, we did observe a positive correlation between the apnea–hypopnea index per hour and more severe PTSD. Possibly, PSTD is a risk factor for OSAS, leading to a higher incidence of OSAS in severe PTSD. Alternatively, PTSD patients who happen to suffer from OSA as well may experience symptom increases due to disturbed sleep. One uncontrolled study has suggested a positive effect on PTSD symptoms after treating OSAS with continuous positive airway pressure (Krakow, Lowry, et al., 2000
). Possibly, when sleep is important for recovery, and is compromised by arousals due to obstructive events, OSAS may intervene with response to treatment. It would be advisable to screen for OSAS in case of snoring and other indicators of OSAS, especially in therapy resistant patients.
In this dissertation, we also report a putative working mechanism for how sleep disturbances may influence daytime complaints. We found that GH levels during the night were decreased in PTSD patients, compared to HCs. Reduced GH secretion correlated with awakenings during the night. A regression analyses with delayed recall as dependent showed that both sleep fragmentation and GH secretion were significant predictors for memory retention of a declarative memory task. This indicated that sleep-dependent memory consolidation is disturbed in PTSD due to decreased nocturnal GH secretion and more interrupted sleep. More research is warranted to confirm these novel findings.
GH receptors are present in the hippocampus (Lai, Emtner, Roos, & Nyberg, 1991
). It is suggested that GH stimulates neuroplasticity in the hippocampus (Kim, Grover, Bertolotti, & Green, 2010
). A recent functional MRI study in healthy volunteers showed decreased hippocampus activation and decreased performance on a memory task after a night of experimentally induced sleep fragmentation (Van der Werf et al., 2009
). Thus, sleep fragmentation can have an effect on hippocampal function. In patients with PTSD, structural and functional changes in the hippocampus also have been reported (Bremner, 2006
). Interestingly, insomnia complaints in PTSD are inversely related to volume on the CA3 area of the hippocampus (Neylan et al., 2010
). Decreased GH secretion may hypothetically be related to decreased hippocampus activity and possibly also hippocampal volume in PTSD. The effect of GH on neurons of the hippocampus during sleep deprivation has been shown in a recent preclinical study (Kim et al., 2010
). This study shows that in the absence of GH during sleep deprivation N
-aspartate (NMDA) receptor-mediated synaptic currents decreased in hippocampal neurons. Moreover, NMDA receptor loss was observed, as was a decline in long-term potentiation. These processes normalized when GH injections were administered during sleep deprivation. Reduced GH secretion may possibly be related to decreased hippocampal functioning in patients with PTSD. The relationship between treatment response to a selective serotonin reuptake inhibitor (SSRI) and neurogenesis in the hippocampus was reported in a preclinical study (Santarelli et al., 2003
). In combat-related PTSD, a relationship between treatment response and hippocampal volume has also been suggested (Vermetten, Vythilingam, Southwick, Charney, & Bremner, 2003
). Future research should indicate whether GH secretion is related to hippocampus volume or functioning in PTSD. Furthermore, the relationship between hippocampal volume, neurogenesis, and treatment response remains to be elucidated.