Rapid eye movement sleep changes during the adaptation night in combat veterans with posttraumatic stress disorder

Rapid eye movement sleep changes during the adaptation night in combat veterans with posttraumatic stress disorder

Rapid Eye Movement Sleep Changes during the Adaptation Night in Combat Veterans with Posttraumatic Stress Disorder Richard J. Ross, William A. Ball, L...

28KB Sizes 0 Downloads 86 Views

Rapid Eye Movement Sleep Changes during the Adaptation Night in Combat Veterans with Posttraumatic Stress Disorder Richard J. Ross, William A. Ball, Larry D. Sanford, Adrian R. Morrison, David F. Dinges, Steven M. Silver, Nancy B. Kribbs, Francis D. Mulvaney, Philip R. Gehrman, and David E. McGinnis Background: Hyperarousal in posttraumatic stress disorder (PTSD) is manifested during sleep as well as waking. Elevated rapid eye movement sleep (REMS) phasic activity, likely signifying central nervous system alerting, has been identified in PTSD. The authors reasoned that PTSD compared to control subjects would show particularly increased REMS phasic activity on the first night of polysomnography, with adaptation to a novel environment. Methods: First-night polysomnograms of 17 veterans with PTSD were compared with those of 11 control subjects. Sleep was also studied in subsets of both groups over two nights. Results: On the first night, the PTSD subjects had a higher density of rapid eye movements in the first REMS period. This measure was increased on the first compared to the second night, but there was no interaction effect between night and group. Conclusions: REMS changes are again demonstrated in veterans with PTSD. Introduction to a novel environment activated a REMS phasic process, but not differentially in PTSD compared to control subjects. Biol Psychiatry 1999;45:938 –941 © 1999 Society of Biological Psychiatry Key Words: Hyperarousal, posttraumatic stress disorder, rapid eye movement sleep, phasic activity

Introduction

P

osttraumatic stress disorder (PTSD) may involve a disturbance of rapid eye movement sleep (REMS) mechanisms (Ross et al 1989). This hypothesis follows

From the Psychiatry Service, Philadelphia Veterans Affairs Medical Center, Philadelphia, Pennsylvania (RJR, FDM, PRG, DEM); Department of Psychiatry, University of Pennsylvania School of Medicine (RJR, WAB, ARM, DFD, NBK) and Department of Animal Biology, University of Pennsylvania School of Veterinary Medicine (ARM, LDS), Philadelphia, Pennsylvania; and Coatesville Veterans Affairs Medical Center, Coatesville, Pennsylvania (SMS). Address reprint requests to Richard J. Ross, MD, PhD, Behavioral Health Service (116A), Philadelphia VA Medical Center, University & Woodland Aves., Philadelphia, PA 19104. Received March 5, 1998; revised June 23, 1998; accepted June 26, 1998.

© 1999 Society of Biological Psychiatry

from the observation that the repetitive, stereotypical anxiety dream characterizes PTSD with high sensitivity and specificity (Ross et al 1989), as dreaming occurs predominantly during REMS. We previously have demonstrated elevations of REMS phasic activity, including rapid eye movements and leg muscle twitch bursts, in a group of Vietnam veterans with PTSD, on the second night of recorded sleep (Ross et al 1994a, 1994b). Evidence from animal experiments relates REMS phasic events to the central nervous system mechanisms of alerting (Sanford et al 1993). We were interested in determining whether REMS rapid eye movements and leg muscle twitches, along with the occurrence of anxiety dreams, would be increased in PTSD subjects on the first night in the sleep laboratory, when “vigilance” during sleep could be heightened. Sleeping in an unfamiliar setting might be expected to powerfully activate REMS phasic processes in PTSD subjects, who manifest hyperarousal (American Psychiatric Association 1994). First (adaptation)-night data generally are not reported because they are considered unrepresentative of typical sleep patterns. The so-called “first-night effect” includes increases in sleep latency (time between lights out and sleep onset) and REMS latency (from sleep onset to the first REMS period), as well as decreases in sleep efficiency (time spent asleep/total recording period) and REMS percentage (total REMS time/time spent asleep) (Agnew et al 1966). The potential of a first-night stimulus for producing elevations of REMS phasic activity has not been investigated. Because they manifest hypervigilance, we wondered whether a “first-night effect” would be particularly prominent in subjects with PTSD (Woodward et al 1996).

Methods and Materials Seventeen Vietnam War combat veterans (42.3 6 2.3 years, mean age 6 SD) were recruited from the psychiatry clinic of the Philadelphia Veterans Affairs Medical Center and the residential 0006-3223/99/$19.00 PII S0006-3223(98)00233-9

REM Sleep Changes during Adaptation in PTSD

PTSD treatment unit at the Coatesville VAMC (Coatesville, PA). All were physically healthy, and none had abused any substance during the preceding 5 weeks or had taken a psychotropic drug over the prior 2 weeks. They were administered components of the SCID-NP-V (Spitzer et al 1987); all had current PTSD. Eleven had lifetime major depression, and 5 had current major depression; 3 had generalized anxiety disorder; 12 had agoraphobia; 4 had panic disorder; and 1 had hypomania. Thirteen had a history of an alcohol use disorder. Eleven physically and mentally healthy men (aged 43.1 6 2.9 years), including 2 Vietnam War veterans, were recruited from among hospital employees and through newspaper advertisements. None reported any dyssomnia or parasomnia. Subjects were treated in accordance with the standards of the Philadelphia and Coatesville VA Medical Centers. Written informed consent was obtained. Remuneration was provided. Polysomnography was conducted at the Philadelphia VAMC. The subjects retired at their usual bedtimes and either woke spontaneously or were aroused at their typical waking times. They attended their usual daytime activities, including treatment at the Coatesville VAMC. A urine toxicology screen was performed on each study night. This report describes differences in sleep between 17 subjects with PTSD and 11 normal control subjects on the first night of polysomnography and dream mentation during this night from 9 of the PTSD subjects and 6 of the control subjects. Also, we explored differences in sleep between a subset of 9 subjects with PTSD and a subset of 7 control subjects, all studied for 2 nights within a 2-week period. Data from the second night were included in previous reports (Ross et al 1994a, 1994b). Sleep was recorded on a Grass polygraph at 10 mm/sec. Electroencephalographic activity from the left central (C3) and midline occipital (Oz) leads was referenced to linked mastoid leads. Chin muscle activity, bilateral anterior tibialis muscle activity, and the electrocardiogram were recorded. Electrooculographic activity recorded between a point above the nasion and points below the left and right outer canthi indicated vertical, oblique, and horizontal movements (Boukadoum and Ktonas 1986). In 12 subjects, a nasal thermistor registered nasal airflow, and no subject showed other evidence of sleep apnea. We developed a self-administered questionnaire to assess dream recall, dream quality (pleasant, unpleasant, or neutral), and “nightmares” (Hartmann 1984). The records were scored according to the criteria of Rechtschaffen and Kales (1968) by a technician blind to the subjects’ identity. We tabulated: total recording period, latency to stage 2 sleep onset, sleep period time, time spent asleep, and sleep efficiency; percentages of stages 1, 2, 3, and 4 sleep and slow-wave sleep (stages 3 1 4); and REMS latency, total REMS time, number of REMS periods, average REMS period duration, REMS percentage, REM activity (number of rapid eye movements), average REM density (REM activity/ total REMS time), REM density in the first REMS period, average REM activity (REM activity/time spent asleep), and the REMS phasic leg activity (RPLA) index of twitch bursts in either anterior tibialis muscle .1.5 sec in duration and .1/2 the amplitude of the presleep ankle flexion response (Ross et al 1994b). First-night sleep measures were analyzed using two-tailed t

939

BIOL PSYCHIATRY 1999;45:938 –941

Table 1. Adaptation Night Sleep Continuity and Sleep Architecture Measuresa

Total recording period (min) Sleep latency (min) Sleep period time (min) Time spent asleep (min) Sleep efficiency (%) Stages (%) 1 2 3 4 314 REMS a b

PTSD (n 5 17)

Normal (n 5 11)

374.7 (63.8) 18.8 (26.5) 360.5 (74.1) 342.2 (71.3) 90.9 (8.1)

386.4 (53.7) 23.5 (36.9) 368.6 (80.0) 336.6 (74.2) 86.3 (13.4)

11.3 (7.0) 64.4 (9.8) 4.8 (4.9) 0.4 (1.2) 5.2 (5.9) 19.1 (5.5)b

11.4 (6.6) 65.9 (9.4) 8.8 (8.2) 0.9 (2.0) 9.7 (9.6) 13.0 (5.8)

Mean (SD) given for all variables. t(df 5 26) 5 2.81, p , .01, two-tailed t test.

tests; alpha was set at .022, according to the modified Bonferroni test for planned comparisons with a familywise error rate of .20 (Keppel 1982). Assumption testing was done using the Kolmogorov–Smirnov test for normality and the Levene Median test for equal variance. Where the normality test failed, a Mann–Whitney U test was used. The “first-night effect” was explored using two (group) 3 two (night) mixed factors analysis of variance (ANOVA) with repeated measures on night. Differences among means were evaluated with alpha set at .025, according to the modified Bonferroni test for planned comparisons with a familywise error rate of .20. Where the normality test failed, the data were logarithmically transformed, as determined using a range procedure (Winer 1971).

Results With the exception of REMS measures, the sleep measures under investigation were equivalent in the two groups (Table 1). Compared to the control subjects, the PTSD subjects had a significantly higher average REMS period duration, REMS percentage, REM density in the first REMS period, and average REM activity (Table 2). In the questionnaire, 8 of the 9 subjects with PTSD reported having dreamed; 3 recalled a dream’s content, which was unpleasant, actually nightmarish, in 2 cases and neutral in 1. While 4 of the 6 control subjects reported dreaming, the dreams that were recalled, by 2, were either pleasant or neutral. The across-night data are shown in Table 3. On the first compared to the second night, there was an increase in REMS latency and decreases in total REMS time and REMS percentage; REM density in the first REMS period was increased. There was no main effect of night on sleep latency, time spent asleep, sleep efficiency, and average REM density. The PTSD subjects showed significant elevations of REMS percentage, average REM density,

940

R.J. Ross et al

BIOL PSYCHIATRY 1999;45:938 –941

Table 2. Adaptation Night REMS Measuresa

REMS latency (min) Total REMS time (min) Number of REMS periods Average REMS period duration (min) REMS (%) Average REM density REM density in 1st REMS period Average REM activity RPLA index

PTSD (n 5 17)

Normal (n 5 11)

83.7 (50.4) 64.6 (20.6) 3.3 (0.9) 26.9 (5.7) 19.1 (5.5) 6.8 (3.4) 6.2 (3.7) 1.4 (0.9) 5.2 (5.9)

97.6 (39.3) 46.0 (27.2) 3.1 (0.8) 19.3 (9.5) 13.0 (5.8) 4.7 (2.3) 2.9 (1.8) 0.6 (0.4) 1.7 (2.2)

Analysis t U U t t t t U U

5 5 5 5 5 5 5 5 5

20.776, ns 140.5, ns 102.5, ns 2.66, p , .014 b 2.81, p , .010 b 1.75, ns 2.82, p , .010 b 151.5, p , .007 c 123.5, ns

a

Mean (SD) given for all variables. Two-tailed t test (df 5 26). c Mann–Whitney U test. b

and REM density in the first REMS period. There were no significant interaction effects between night and group.

Discussion An increase in REM density in chronic combat-related PTSD has been reported by Ross et al (1994a) and Mellman et al (1993). We now report increases in first REMS period REM density and average REM activity in subjects with chronic combat-related PTSD compared to normal control subjects on the adaptation night. In keeping with our second-night findings (Ross et al 1994a), no differences in measures of sleep continuity and non-REMS architecture were demonstrated on the adaptation night. Only REMS changes distinguished the first night of sleep in these veterans with PTSD from that of normal control subjects. This observation further supports the hypothesis that some REMS dysregulation, specifically, underlies the subjective sleep disturbance in PTSD (Ross et al 1989). The concurrent diagnoses in many of our PTSD subjects must be considered potential confounders (Ross et al 1994a, 1994b); studies in subjects with uncomplicated diagnoses are required to resolve this point.

Although even people who often experience nightmares rarely report them in a sleep laboratory (Hartmann 1984), 2 of our PTSD subjects (but no control subject) recalled dream mentation that met the criteria for a nightmare; given the evidence for some nonspecific correlation between REM activity and the intensity of sleep mentation (Rechtschaffen 1973), the elevated first REMS period REM density we observed might reflect a neural substrate. This view is consistent with the correspondence on positron-emission tomography scanning between cortical areas involved in rapid eye movements during REMS and areas activated during rapid eye movements in waking (Hong et al 1995). The typical “first-night effect” was manifested as an increase in REMS latency and decreases in total REMS time and REMS percentage. Contrary to our prediction, no differential “first-night effect” in the PTSD subjects compared to the control group was observed. To our knowledge, the current finding that REM density in the first REMS period was elevated by a first-night stimulus is the first demonstration that introduction to a novel environment can activate REMS phasic processes. The absence of an interaction between study night and subject group suggests that other

Table 3. Sleep Measures across Nights 1 and 2a PTSD (n 5 9)

Sleep latency (min) Time spent asleep (min) Sleep efficiency (%) REMS latency (min) Total REMS time (min) REMS (%) Average REM density REM density in 1st REMS period a b

Normal (n 5 7)

Night 1

Night 2

Night 1

Night 2

26.6 (34.5) 333.6 (70.7) 89.1 (10.5) 90.8 (62.5) 65.3 (22.7) 20.1 (7.1) 7.4 (3.5) 6.5 (3.5)

13.6 (9.8) 359.7 (71.2) 90.3 (9.0) 37.1 (31.5) 97.8 (19.3) 27.5 (4.4) 6.2 (3.6) 5.4 (5.3)

23.5 (43.7) 337.6 (94.2) 84.1 (16.0) 89.4 (38.0) 48.6 (32.2) 13.6 (6.6) 3.5 (1.3) 2.5 (1.7)

10.6 (7.7) 382.3 (55.1) 91.8 (10.6) 62.7 (6.2) 77.4 (18.9) 19.9 (3.4) 3.1 (2.1) 2.0 (2.0)

Mean (SD) given for all variables. Logarithmic transformation.

Night effect (df 5 1,14) F F F F F F F F

5 5 5 5 5 5 5 5

0.18, nsb 0.13, nsb 1.16, nsb 10.97, p , .006 21.68, p , .001 21.69, p , .001 1.61, ns 8.14, p , .014 b

Group effect (df 5 1,14) F F F F F F F F

5 5 5 5 5 5 5 5

0.58, 0.26, 0.24, 0.51, 3.54, 8.61, 6.79, 7.58,

nsb nsb nsb ns ns p , .011 p , .021 p , .017 b

REM Sleep Changes during Adaptation in PTSD

probes of a possibly supersensitive REMS phasic event generator should be used to characterize PTSD. An aversive emotional stimulus, rather than simply a novel, potentially arousing one, may be necessary. Supported by the Department of Veterans Affairs Medical Research Service (RJR) and USPHS Grant MH42903 (ARM). The authors thank Richard L. Horner, PhD, for help with the statistical analysis. Presented in part at the eighth annual meeting of the Association of Professional Sleep Societies, Boston, June 4 –9, 1994.

References Agnew HW, Webb WB, Williams RL (1966): The first night effect: An EEG study of sleep. Psychophysiology 2:263–266. American Psychiatric Association (1994): Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Press. Boukadoum AM, Ktonas PY (1986): EOG-based recording and automated detection of sleep rapid eye movements: A critical review, and some recommendations. Psychophysiology 23: 598 – 611. Hartmann E (1984): The Nightmare: The Psychology and Biology of Terrifying Dreams. New York: Basic Books. Hong CC, Gillin JC, Dow BM, Wu J, Buchsbaum MS (1995): Localized and lateralized cerebral glucose metabolism associated with eye movements during REM sleep and wakefulness: A positron emission tomography (PET) study. Sleep 18:570 –580. Keppel G (1982): Design and Analysis: A Researcher’s Handbook. Englewood Cliffs, NJ: Prentice-Hall, pp. 146 –150.

BIOL PSYCHIATRY 1999;45:938 –941

941

Mellman TA, Kulick-Bell R, Kumar A, Nolan B (1993): Sleep and arousal regulation in combat-related PTSD. Sleep Res 22:153. Rechtschaffen A (1973): The psychophysiology of mental activity during sleep. In: McGuigan FJ, Schoonover RA, editors. The Psychophysiology of Thinking. New York: Academic Press, pp. 153–205. Rechtschaffen A, Kales A (1968): A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Washington, DC: US Government Printing Office. Ross RJ, Ball WA, Sullivan KA, Caroff SN (1989): Sleep disturbance as the hallmark of posttraumatic stress disorder. Am J Psychiatry 146:697–707. Ross RJ, Ball WA, Dinges DF, Kribbs NB, Morrison AR, Silver SM, et al (1994a): Rapid eye movement sleep disturbance in posttraumatic stress disorder. Biol Psychiatry 35:195–202. Ross RJ, Ball WA, Dinges DF, Kribbs NB, Morrison AR, Silver SM, et al (1994b): Motor dysfunction during sleep in posttraumatic stress disorder. Sleep 17:723–732. Sanford LD, Morrison AR, Ball WA, Ross RJ, Mann GL (1993): The amplitude of elicited PGO waves: A correlate of orienting. Electroencephalogr Clin Neurophysiol 86:438 – 445. Spitzer RL, Williams JBW, Gibbon M (1987): Structured Clinical Interview for DSM-III-R—Non-patient Version (SCIDNP-V). New York: Biometrics Research Department, New York State Psychiatric Institute. Winer BJ (1971): Statistical Principles in Experimental Design, 2nd ed. New York: McGraw Hill. Woodward SH, Bliwise DL, Friedman MJ, Gusman FD (1996): First night effects in post-traumatic stress disorder inpatients. Sleep 19:312–317.