(1) Describe the incidence of nightmares and dreams during PSG

(2) Identify PSG characteristics of nightmares and dreams

(3) Identify clinical correlates

Polysomnography (PSG) is uncommonly indicated for dreams and nightmares, and they are rarely captured during otherwise clinically indicated PSG.

Intake and post-PSG forms were reviewed for ninety-three patients. Questionnaires included Epworth Sleepiness Scale (ESS), Insomnia Severity Index (ISI), Nightmare Disorder Index (NDI), and a structured post-PSG dreams and nightmares questionnaire. Dreams were self-reported. Nightmares were differentiated from bad dreams if the former caused a patient-reported awakening.

Overall, 87.1% were Active Duty and 89.2% reported the presence or absence of dreams. In total, 50.5% (47) reported no dreams, 22.6% (21) reported a dream, 2.2% (2) a bad dream, and 14% (13) a nightmare. NDI differed significantly between groups (6 ± 5 vs. 7 ± 6 vs. 12 ± 0 vs. 11 ± 5, respectively) (p=0.029). The groups were not significantly different in sex, age, BMI, diagnosis of depression/anxiety, PTSD, TBI, chronic pain, prescription of sleep aid/psychoactive medication, ESS/ISI values, or REM percentage.  Autonomic symptoms were associated with 75% (9) of nightmares, and 5 (41.7%) reported a trauma-related nightmare (TRN). TRN contents included one combat, one sexual assault, two accidents, and one death. TRN was an accurate replay for 2 and a mix of replay/non-replay events for 2. 

Our cohort uniquely demonstrated that almost half of first-time PSG patients at a military sleep center experience a dream or nightmare during their PSG. Higher NDI scores may predict who is more likely to suffer from a bad dream or nightmare in the sleep lab, aiding selection for further investigations. Assessing for the occurrence of dreams or nightmares during PSG is likely indicated.