Daniel J. Buysse

Circadian rhythms in human performance and mood under constant conditions

This study explored the relationship between circadian performance rhythms and rhythms in rectal temperature, plasma cortisol, plasma melatonin, subjective alertness and well‐being. Seventeen healthy young adults were studied under 36 h of «unmasking» conditions (constant wakeful bedrest, temporal isolation, homogenized «meals») during which rectal temperatures were measured every minute, and plasma cortisol and plasma melatonin measured every 20 min. Hourly subjective ratings of global vigour (alertness) and affect (well‐being) were obtained followed by one of two performance batteries. On odd‐numbered hours performance (speed and accuracy) of serial search, verbal reasoning and manual dexterity tasks was assessed. On even‐numbered hours, performance (% hits, response speed) was measured at a 25–30 min visual vigilance task. Performance of all tasks (except search accuracy) showed a significant time of day variation usually with a nocturnal trough close to the trough in rectal temperature. Performance rhythms appeared not to reliably differ with working memory load. Within subjects, predominantly positive correlations emerged between good performance and higher temperatures and better subjective alertness; predominantly negative correlations between good performance and higher plasma levels of cortisol and melatonin. Temperature and cortisol rhythms correlated with slightly more performance measures (5/7) than did melatonin rhythms (4/7). Global vigour correlated about as well with performance (5/7) as did temperature, and considerably better than global affect (1/7). In conclusion: (1) between‐task heterogeneity in circadian performance rhythms appeared to be absent when the sleep/wake cycle was suspended; (2) temperature (positively), cortisol and melatonin (negatively) appeared equally good as circadian correlates of performance, and (3) subjective alertness correlated with performance rhythms as well as (but not better than) body temperature, suggesting that performance rhythms were not directly mediated by rhythms in subjective alertness.

Sleep and morningness-eveningness in the ‘middle’ years of life (20–59y)

SUMMARY The following four issues were assessed in a group of 110 adults between the age of 20 and 59y: (1) the effect of age (regarded as a continuous variable) on polysomnographic sleep characteristics, habitual sleep‐diary patterns, and subjective sleep quality; (2) the effects of age on morningness‐eveningness; (3) the effects of morningness‐eveningness on sleep, after controlling for the effects of age; and (4) the role of morningness‐eveningness as a mediator of the age and sleep relationship. Increasing age was related to earlier habitual waketime, earlier bedtime, less time in bed and better mood and alertness at waketime. In the laboratory, increasing age was associated with less time asleep, increased number of awakenings, decreased sleep efficiency, lower percentages of slow‐wave sleep (SWS) and rapid eye movement (REM) sleep, higher percentages of Stage 1 and 2, shorter REM latency and reduced REM activity and density. Increasing age was also associated with higher morningness scores. After controlling for the effects of age, morningness was associated with earlier waketime, earlier bedtime, less time in bed, better alertness at waketime, less time spent asleep, more wake in the last 2 h of sleep, decreased REM activity, less stage REM (min and percentage), more Stage 1 (min and percentage) and fewer minutes of Stage 2. For one set of variables (night time in bed, waketime, total sleep time, wake in the last 2 h of sleep and minutes of REM and REM activity), morningness‐eveningness accounted for about half of the relationship between age and sleep. For another set of variables (bedtime, alertness at waketime, percentages of REM and Stage 1), morningness‐eveningness accounted for the entire relationship between age and sleep. In conclusion, age and morningness were both important predictors of the habitual sleep patterns and polysomnographic sleep characteristics of people in the middle years of life (20‐59 y).

Symptoms of stress and depression as correlates of sleep in primary insomnia.

Objective: Previous studies have not evaluated the clinical correlates of the electroencephalographic spectral profile in patients with insomnia. In the preliminary study described here, we evaluated the extent to which symptoms of stress and depression are associated with subjective sleep complaints and quantitative measures of sleep in individuals with chronic insomnia. Methods: Subjects were 14 healthy adults who met criteria for primary insomnia as specified in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders. Measures of stress, depression, and subjective sleep quality were collected before subjects participated in a two-night laboratory sleep series. We hypothesized that elevated symptoms of stress and depression would be associated with subjective sleep complaints and electroencephalographic evidence of hyperarousal during sleep. Hyperarousal during sleep was defined as decreases in delta power and elevations in alpha and beta power throughout non–rapid eye movement sleep, and symptoms of stress were defined as the tendency to experience stress-related intrusive thoughts and the interaction between intrusion tendency and the number of recent stressful events (subjective stress burden). Results: A stronger tendency to experience stress-related intrusive thoughts was associated with greater sleep complaints and a trend toward higher beta power, whereas increases in subjective stress burden were associated with decreases in delta power. In addition, elevations in subclinical symptoms of depression were associated with greater sleep complaints and elevations in alpha power. Conclusions: Observed relationships among symptoms of stress, depression, subjective sleep complaints, and electroencephalographic power may be relevant to the discrepancy between subjective and objective measures of sleep in patients with insomnia and may be more broadly applicable to sleep complaints in association with stressful life events and major depression.

Electroencephalographic sleep correlates of episode and vulnerability to recurrence in depression

The study of electroencephalogram (EEG) sleep in depressed patients before and after treatment with psychotherapy can distinguish episode-related and persistent biological features. With longitudinal follow-up, we can also assess whether EEG sleep measures are associated with recurrence of depression. In the current study, we examined EEG sleep during the depressed state and during symptomatic remission after treatment with interpersonal psychotherapy in 42 outpatients with major depression. Analyses included both visually-scored and computer-analyzed measures. Patients showed significant increases in sleep latency (p = .01) and rapid eye movement (REM) latency (p = .04) from baseline to remission, as well as a decrease in REM sleep percent (p = .03). Total delta EEG counts decreased from baseline to remission (p = .03), specifically in the second nonrapid eye movement (NREM) period (p = .03); as a result, the relative distribution of delta activity shifted toward sleep onset (i.e., increased delta sleep ratio; p = .03). Automated REM counts also decreased from depression to remission (p = .006). Compared to patients who remained well through one year of follow-up, those who suffered a recurrence of depression had less delta EEG activity at baseline and remission (p = .01), particularly in the lowest delta frequency band of 0.5-1.0 Hz. Specific components of sleep (total delta activity, delta ratio, REM activity) constitute episode-related biological features. Other components (slowest delta activity) may represent vulnerability factors for recurrence.

Sleep as a mediator of the stress-immune relationship.

Objective To evaluate the role of sleep in the relationship of intrusive thoughts/avoidance behaviors to natural killer cell (NKC) number and function. Method Twenty-nine individuals seeking treatment for bereavement-related depression were studied in the sleep laboratory. Background and clinical variables, including the Impact of Event Scale (IES) and the Hamilton Rating Scale for Depression (HRSD), were administered during the week preceding a 3-night sleep study. Blood samples were collected upon awakening after the second or third night of sleep. Results Greater frequency of intrusive thoughts and avoidance behaviors was associated with more time spent awake during the first non-rapid eye movement period (NREM-1) and lower NKC number (p values < .01). Greater time spent awake during NREM-1 was associated with lower NKC numbers (p < .05). Regression analyses revealed that the significant relationship between symptoms of intrusion/avoidance and NKC number was no longer significant when time spent awake during NREM-1 was entered into the regression equation. Time spent awake during NREM-1 accounted for 12% of the variance in NKC number (p < .05), whereas intrusion/avoidance accounted for 7% of the variance in NKC number (NS). Conclusions These results suggest that EEG-assessed sleep may be a significant correlate of the stress-immune relationship.

Pretreatment REM sleep and subjective sleep quality distinguish depressed psychotherapy remitters and nonremitters

BACKGROUND We compared pretreatment subjective and electroencephalographic sleep measures among depressed patients who remitted with psychotherapy alone and those who did not remit. METHODS Patients were 111 midlife women with recurrent major depressive disorder. Baseline psychiatric ratings and sleep studies were conducted prior to treatment with weekly interpersonal psychotherapy. Remission was defined as a score of < or = 7 for 3 consecutive weeks on the Hamilton Depression Rating Scale. Clinical and sleep measures were compared between remitters (n = 62) and nonremitters (n = 49) using t tests and random regression. Linear discriminant function analyses were used to categorize remitters and nonremitters on the basis of sleep measures. RESULTS Treatment nonremitters had significantly worse subjective sleep quality and significantly elevated phasic REM sleep as measured by multivariate and univariate analyses. The linear accumulation of REM activity during sleep occurred at a significantly higher rate in nonremitters than in remitters. Linear discriminant function analyses based on subjective sleep quality and REM activity correctly identified 68.3% of nonremitters and 68.5% of remitters. CONCLUSIONS These findings highlight the role of subjective and REM sleep measures as correlates of short-term psychotherapy treatment response in major depressive disorder. Disturbed sleep may be a physiological indicator of increased limbic and brain stem arousal.

Social rhythm stability following late-life spousal bereavement : associations with depression and sleep impairment

The aim of this study was to investigate changes in social rhythm stability and sleep in spousally bereaved subjects (n = 94) and in nonbereaved elderly control subjects (n = 45). Social rhythm stability and activity level were measured with a diary-like instrument, the Social Rhythm Metric (SRM). We observed that spousal bereavement, per se, was not associated with a lower social rhythm stability or activity level except in the presence of a major depressive episode. We also observed an inverse correlation between severity of depression and social rhythm stability, and a positive correlation between depression and both subjective and objective measures of sleep impairment. Higher social rhythm stability was correlated with better sleep in subjects with high activity levels, but not in subjects with low activity levels. Longitudinal data, including pre-bereavement assessment of social rhythm stability, are necessary to ascertain directional effects, i.e., whether loss of spouse occasions disruption of social rhythms or whether such disruption precedes sleep impairment and depression.

The sleep of healthy people--a diary study

To provide baseline data for various research studies at the University of Pittsburgh over a 10-year period, 266 healthy subjects (144 male, 122 female, aged 20–50 years) meeting certain criteria each completed a 14-night sleep diary. For each night, the diary allowed the subjective measurement of bedtime, wake time, time in bed (TIB), sleep efficiency, number of minutes of wake after sleep onset (WASO), alertness on awakening, and percentage of mornings needing an alarm (or a person functioning as one). Weeknight versus weekend night differences in TIB (TIBdiff), weekday alertness, and reliance on alarms were examined as possible indicators of sleep debt. In addition, general descriptive data were tabulated. On average, bedtimes were at 23:48 and wake times at 07:23, yielding a mean TIB of 7 hours 35 minutes. As expected, bedtimes and wake times were later on weekend nights than on weeknights. Bedtimes were 26 minutes later, wake times 53 minutes later, yielding a mean weekend TIB increase of 27 minutes. Overall, subjects perceived their sleep latency to be 10.5 minutes, reported an average of one awakening during the night (with an average of 6.4 minutes of WASO), had a diary sleep efficiency of 96.3%, and awoke with an alertness rating of 69.5%. These variables differed little between weeknight and weekend nights. Subjects used an alarm (or a person functioning as an alarm) on 60.9% nights overall, 68.3% on weeknights, 42.5% on weekends. When TIBdiff was used as an estimate of sleep debt (comparing subjects with TIBdiff >75 minutes with those with a TIBdiff < 30 minutes), the group with more “catch-up sleep” on weekends had shorter weeknight TIB durations (by about 24 minutes) and relied more on an alarm for weekday waking (by about 22%), indicating the possible utility of these variables as sleep debt indices. (Chronobiology International, 17(1), 49–60, 2000)

Sleep and Circadian Rhythms in Four Orbiting Astronauts

This experiment measured the sleep and circadian rhythms of four male astronauts aboard a space shuttle (STS-78) orbiting the Earth for 17 days. The space mission was specially scheduled to minimize disruptions in circadian rhythms and sleep so that the effects of space flight and microgravity per se could be studied. Data were collected in 72-h measurement blocks: one block 7 days before launch, one early within the mission (3 days after launch), one late in the mission (12 days after launch), and one 18 days after landing. Within each measurement block, all sleep was recorded both polysomnographically and by sleep diary. Core body temperature was sampled every 6 mins. Actillumes© were worn continuously. All urine samples were collected separately. Performance was assessed by a computerized test battery (3/day) and by end-of-shift questionnaires (1/day); mood and alertness were measured by visual analogue scales (5/day). Circadian rhythms in orbit appeared to be very similar in phase and amplitude to those on the ground, and were appropriately aligned for the required work/rest schedule. There was no change from early flight to late flight. This was also reflected in mood, alertness, and performance scores, which were satisfactory at both in-flight time points. However, in-flight sleep showed a decreased amount of sleep obtained (mean = 6.1 h), and all four astronauts showed a decrease in delta sleep. No further degradation in sleep was seen when early flight was compared to late flight, and no other sleep parameters showed reliable trends.

Are age differences in sleep due to phase differences in the output of the circadian timing system

Our aim was to evaluate whether age-related changes in the phase of the output of the circadian timing system (CTS) can explain age differences in habitual bedtime/wake time and in sleep consolidation parameters. Analyses focused on a group of healthy elderly people (older than 70 years) with no sleep problems and with similar subjective sleep quality as a young control group. The 2-week sleep diary data and 24h laboratory temperature recordings were examined for 70 subjects (22 young men [YM], 19 old men [OM], 29 old women [OW]). Polysomnographic (PSG) sleep data recorded during temperature data acquisition were also available for 62 subjects. These analyses made use of our recently developed technique to demask temperature rhythm data. As expected, compared to the young subjects, older subjects showed earlier habitual bedtime and wake time, more disturbed sleep, and a tendency for an earlier minimum of the circadian temperature rhythm. Despite sleep consolidation differences, the groups showed very similar habitual phase-angle differences (interval between the time occurrence of the fitted temperature minimum and habitual wake time). Both elderly and young subjects woke up on average 3 h after the temperature minimum. After controlling for the effects of age group, habitual bedtime and wake time were related to clock time phase of the circadian temperature rhythm, with an earlier phase associated with earlier habitual bedtime and wake time. None of the sleep consolidation parameters were linked to the temperature phase angle. In conclusion, sleep consolidation changes associated with healthy aging do not appear to be related to changes in the phase-angle difference between the output signal from the CTS and sleep.