All-night EEG spectral analysis as a tool for the prediction of clinical response to antidepressant treatment.

Earlier investigations have suggested that variables derived from quantified electroencephalographic (EEG) sleep analysis might predict good clinical response in an early phase of antidepressant treatment. In this report we evaluate the predictive value of all-night sleep EEG spectral analysis during the washout period before treatment. We compared the spectral EEG sleep profiles of major depressed inpatients divided into two groups according to an improvement > or = 50% on the Hamilton Rating Scale for Depression. Findings in this population demonstrate the presence of specific characteristics of the responder group compared with the nonresponder group. Delta band relative power was increased in the former group, while theta, alpha, and beta relative power were decreased. All the bands showed decrease in absolute power in the responder group. These results can be interpreted as enhanced sleep intensity in the responder group. All-night sleep EEG spectral variables are valid baseline markers of the functional differences between treatment responders and nonresponders and thus might permit prediction of clinical outcome.

A quantitative evaluation of the relationships between growth hormone secretion and delta wave electroencephalographic activity during normal sleep and after enrichment in delta waves.

The existence of a relationship between growth hormone (GH) release and slow-wave sleep (SWS), often studied in the past using conventional scoring of sleep stages, remains controversial. In the present study, this relationship was reevaluated by spectral analysis of the sleep electroencephalogram (EEG) and deconvolution analysis of the plasma GH concentrations during normal nocturnal sleep and after enrichment in SWS by means of ritanserin, a selective 5-HT2 receptor antagonist. Eight healthy male subjects each participated in two randomized night studies after having received either a placebo or a 5-mg dose of ritanserin. They were subjected to 8 hours of polysomnography, including spectral analysis of the sleep EEG. Plasma GH levels were measured at 10-minute intervals. The mean delta absolute power and the mean GH secretory rates were significantly higher under ritanserin than under placebo for the first 3 hours after sleep onset (+24% and +29%, respectively). Their nocturnal profiles were significantly and positively correlated in all subjects (average r = 0.710 under placebo, 0.567 under ritanserin; p < 0.0001 in both cases). GH secretory pulses were found to be coincident with delta activity peaks in both directions. The amount of GH secreted during significant GH pulses was correlated with the amount of concomitant delta wave activity (r = 0.803 under placebo, r = 0.764 under ritanserin, p < 0.0001). Similarly, the amount of delta wave activity found during delta wave peaks was correlated with the amount of GH secreted concomitantly (r = 0.715 under placebo, r = 0.723 under ritanserin: p < 0.0001). These results demonstrate a close temporal and quantitative relationship between GH secretion and delta wave activity, which may be evidence of common stimulatory mechanisms.

Temporal relationship between prolactin secretion and slow-wave electroencephalic activity during sleep.

It is well established that plasma prolactin (PRL) concentrations exhibit a sleep-dependent pattern, with the highest levels occurring during sleep and the lowest during waking. Still, controversy exists concerning an association between rapid eye movement (REM) and non-REM sleep cycles and plasma PRL pulses. These studies were all based on conventional scoring of sleep stages. In the present study, plasma PRL concentrations were analyzed at 10-minute intervals in 10 subjects during the night when sleeping. PRL secretory rates were calculated by a deconvolution procedure. Spectral parameters of sleep electroencephalographic (EEG) recordings were analyzed together with PRL secretion using cross-correlation. Slow-wave activity of the EEG and PRL secretion ran parallel in all individuals. Conversely, alpha and beta bands and the EEG mean frequency were inversely proportional to PRL secretion. In 9 of the 10 subjects studied, PRL secretion was concomitant with delta waves or lagged behind by 10-20 minutes, depending on subjects, with maximum cross-correlation coefficients ranging between 0.40 and 0.67. This temporal relationship between PRL secretion and delta waves was further assessed by a pulse-by-pulse analysis based on the calculation of probability levels after computer simulations. Nine of the 10 subjects displayed significant concomitance between delta wave activity and PRL secretory oscillations. These results demonstrate that PRL secretion during sleep is coupled to delta waves in young healthy men.

Changes in EEG power density during sleep laboratory adaptation.

First- and second-night effects on the electroencephalogram (EEG) were investigated by means of polygraphic sleep recordings and all-night spectral analysis. Eighteen normal subjects were studied for three consecutive nights in a hospital sleep laboratory. Visual sleep scoring showed that there was a first-night effect in normal subjects similar to that reported previously [increased wakefulness; decreased total sleep time, sleep efficiency, and rapid eye movement (REM) sleep]. Spectral analysis of the sleep EEG revealed important changes, most of which occurred in REM sleep. Increased delta, theta, and beta1 power densities accompanied by decreased mean frequency were seen in REM sleep in the second night. On the basis of REM sleep deprivation results previously published, our data suggest that the second night could be affected by partial REM sleep deprivation that occurred in the first night. Delta and theta power density values decreased in the first non-rapid eye movement episode of nights 1 and 2; this could result from increased REM sleep pressure. The overall consistency of spectral data in the first and second night with REM sleep findings derived from visual scoring in the first night lends further support to this hypothesis. The sleep disturbance experienced during the first night in a sleep laboratory may be a useful and valid model of transient insomnia. Therefore, we conclude that data from all nights recorded should be included in assessing a subject's sleep.

Sleep stage scoring using the neural network model: comparison between visual and automatic analysis in normal subjects and patients.

In this paper, we compare and analyze the results from automatic analysis and visual scoring of nocturnal sleep recordings. The validation is based on a sleep recording set of 60 subjects (33 males and 27 females), consisting of three groups: 20 normal controls subjects, 20 depressed patients and 20 insomniac patients treated with a benzodiazepine. The inter-expert variability estimated from these 60 recordings (61,949 epochs) indicated an average agreement rate of 87.5% between two experts on the basis of 30-second epochs. The automatic scoring system, compared in the same way with one expert, achieved an average agreement rate of 82.3%, without expert supervision. By adding expert supervision for ambiguous and unknown epochs, detected by computation of an uncertainty index and unknown rejection, the automatic/expert agreement grew from 82.3% to 90%, with supervision over only 20% of the night. Bearing in mind the composition and the size of the test sample, the automated sleep staging system achieved a satisfactory performance level and may be considered a useful alternative to visual sleep stage scoring for large-scale investigations of human sleep.

First-night effect in normal subjects and psychiatric inpatients.

The goal of the present study was to evaluate the first-night effect in psychiatric inpatients using large subject samples (n > 30) in order to obtain a good statistical evaluation. Thirty-two normal subjects and 94 psychiatric inpatients (38 depressives and 56 insomniacs) were studied for three consecutive nights in the hospital sleep laboratory. Our results showed clearly that there was a first-night effect in normal subjects, similar to that reported in previously published data, characterized by a longer rapid eye movement (REM) sleep latency (p < 0.05), increased wakefulness (p < 0.01) and total sleep time (p < 0.02) and a decreased sleep efficiency (p < 0.01). REM sleep latency and stage REM in the first third of the night were still altered in the second night. Both clinical groups had a less marked first-night effect than normal subjects, showing alterations only observed in REM sleep (p < 0.01) (decreased REM sleep, longer REM sleep latency, increased REM sleep gravity center). However, the first-night effect was more pronounced in insomniacs than in depressed patients. No statistical differences between the second and third nights' recordings were found in sleep parameters. It is suggested that first-night data should not be simply discarded but could be used in subsequent analyses.

Temporal link between plasma thyrotropin levels and electroencephalographic activity in man.

Plasma thyrotropin (TSH) levels have been previously shown to be associated with the internal sleep structure determined by conventional scoring of sleep stages. This temporal relationship was re-evaluated using spectral analysis of the sleep electroencephalogram (EEG). Eight healthy male subjects underwent two randomized night studies after having received either placebo or 5 mg ritanserin, a selective 5-HT2 receptor antagonist known to increase slow-wave sleep. Delta relative power and TSH levels, determined at 10 min intervals, were found to be inversely related with an average cross-correlation coefficient highly significant (P < 0.0001) in both experimental conditions. Alpha slow-wave index, an estimator of awakenings, and TSH pulses exhibited a significant temporal association in both conditions. These results demonstrate that TSH fluctuations are linked to the sleep EEG activity in man.

Growth hormone response to clonidine and the cortisol response to dexamethasone in depressive patients.

The dexamethasone suppression test (DST) and the clonidine stimulation test (CST) were studied in 47 depressed patients. Issues addressed included (1) the usefulness of both tests as markers of major depression; (2) the relationship between the two tests and the pathophysiology underlying this relationship; and (3) the psychopathological correlates of both tests. The widely reported link between abnormal DST results and melancholic depression was confirmed. The DST and the CST showed extensive overlap, suggesting a relationship in major depression between the biological abnormalities indexed by each test (i.e., hypothalamic-pituitary-adrenal axis and noradrenergic system). Finally, the psychopathological correlates of various subgroups categorized on the basis of test responses (e.g., normal test results, blunted CST results, and both abnormal DST and CST results) confirmed significant differences between groups on two items of the Hamilton Rating Scale for Depression: psychic anxiety and somatic anxiety. The results suggest that particular patterns of neuroendocrine abnormalities may be associated with particular profiles of depressive symptomatology.

An architecture for EEG signal processing and interpretation during sleep (ESPIS).

The project's aim is to develop a dedicated workstation in order to process multiple channels of electrophysiological signals in real-time during sleep. In ESPIS we are aiming to define both an architecture and an environment for EEG signal interpretation in medicine based on computer science gold standards (Unix, XWindow, Motif). Signal processing and pattern recognition analysis are provided by parallel processing on a specific developed acquisition architecture (DSP) based on transputers. The main result is a high performance prototype demonstrating signal interpretation during sleep which has already been tested in a medical environment. The overall specifications allow this biomedical device to be extended to other types of medical signals.

[Neuro-electrophysiologic studies in abstinent and depressed alcoholic patients treated with tianeptine]. / Etudes neuro-électrophysiologiques chez le malade alcoolique abstinent et déprimé traité par la tianeptine.

A concerted study of the clinical and electrophysiologic effects of tianeptine was conducted in alcoholic patients hospitalized for 5 weeks for alcohol withdrawal cures and subsequent depression. Because of the well known manifestations of infraclinical cognition impairment, sleep disorders and greater susceptibility to undesirable effects of psychotropic drugs, which hinder health care in this type of patient, the authors investigated changes in cognitive functions and the effect of tianeptine on sleep organization and daytime vigilance. Results after 4 weeks treatment (3 times 12.5 mg/day) included: besides its antidepressant effect, tianeptine reduces the manifestations of anxiety, without sedation effects, either clinical or electrophysiologic; tianeptine has no deleterious effect on cognitive functions, on the contrary, it appears to favour recovery when they are impaired; tianeptine does not modify sleep structure, notably in paradoxal sleep; tianeptine is an antidepressant which has a good acceptability, even for a population at "risk".

Neural network model: application to automatic analysis of human sleep.

We describe an approach to automatic all-night sleep analysis based on neural network models and simulated on a digital computer. First, automatic sleep stage scoring was performed using a multilayer feedforward network. Second, supervision of the automatic decision was achieved using ambiguity rejection and artifact rejection. Then, numerical analysis of sleep was carried out using all-night spectral analysis for the background activity of the EEG and sleep pattern detectors for the transient activity. Computerized analysis of sleep recordings may be considered as an essential tool to describe the sleep process and to reflect the dynamical organization of human sleep.

5-HT2 receptors are partially involved in the relationship between renin release and delta relative power.

A strong relationship was previously described between the nocturnal oscillations of plasma renin activity (PRA) and the sleep cycles, with levels of PRA that increase during non rapid eye movement sleep and decrease during rapid eye movement sleep. This study was designed to determine whether ritanserin, a 5-hydroxytryptamine-2 (5-HT2) receptor antagonist known to increase slow wave sleep both in human and in animals and to decrease plasma renin activity response to serotonergic stimulation in the rat, would uncouple this relationship. Eight subjects underwent two randomized night studies after having received either placebo or 5 mg ritanserin administered in the morning. They were subjected to 8 hour polysomnography, including spectral analysis of the electroencephalogram and to continuous blood sampling. Blood was sampled from 2300 to 700h every 10 min and plasma renin activity (PRA) was measured by radioimmunoassay of angiotensin 1. The nocturnal profiles were analysed using the pulse detection program ULTRA. Ritanserin produced the expected increase in slow wave sleep (SWS) duration (132 +/- 10 min under ritanserin vs 72 +/- 9 min under placebo; p < 0.001) and a significant increase in delta relative power (69 +/- 2% under ritanserin vs 60 +/- 2% under placebo; p < 0.01). The mean overnight PRA levels had a tendency to decrease under ritanserin (1.66 +/- 0.34 ngAngl/ml per h under ritanserin vs 1.48 +/- 0.31 ngAngl/ml per h under placebo; p = 0.08). Individual PRA oscillations were preserved and remained strongly associated with delta power oscillations. PRA peak levels were similar in both experimental conditions, but the absolute amplitude of the oscillations was decreased under ritanserin (1.50 +/- 0.36 ngAngl/ml per h vs 1.04 +/- 0.14 ngAngl/ml per h; p < 0.05). These results demonstrate that ritanserin, at a dose that augments delta power, only weakly affects renin release, which suggests that 5-HT2 receptors are only partially involved in the processes coupling renin release and SWS and that other mechanisms probably control the sleep-associated variations in PRA.

Slow wave electroencephalic activity parallels renin oscillations during sleep in humans.

Previous studies have demonstrated that the nocturnal oscillations of plasma renin activity (PRA) exactly reflect rapid eye movement (REM) non-REM (NREM) sleep alternation with levels of PRA that increase during NREM sleep and decrease during REM sleep. These studies were based exclusively on conventional scoring of sleep stages. In the present study, we used spectral analysis of the sleep EEG to determine the variations in the different EEG frequency bands, together with PRA profiles. Eight male volunteers participated in a 1 night study. They were subjected to 8 h polysomnography including spectral analysis of the EEG, and to blood sampling every 10 min. Delta relative power and Sleep Intensity Index and PRA oscillations ran parallel in all individuals. An increase in slow waves was associated with an increase in PRA, whereas a decrease was associated with a decrease in PRA. Cross-correlation coefficients were significant and ranged between 0.34 and 0.74. Conversely, theta, alpha and beta bands and the EEG mean frequency were inversely proportional to PRA, with lower cross-correlation coefficients. These results may give further support to the hypothesis of a common mechanism controlling both SWA and renin release from the kidney.

Temporal relationships between pulsatile cortisol secretion and electroencephalographic activity during sleep in man.

A temporal link between slow wave sleep and low or decreasing cortisol release has been previously demonstrated. This relationship was re-evaluated in 15 healthy male subjects using spectral analysis of their sleep electroencephalogram (EEG). EEG activity in the delta, theta, alpha and beta bands was cross-correlated with cortisol secretory rates at 10-min intervals. For the period of pulsatile cortisol secretion, an inverse relationship was found with the delta band with an average cross-correlation coefficient of -0.505 (P < 0.0001). Variations in cortisol secretory rates coincided with or anticipated opposite variations in delta wave activity by 10 or 20 min. A significant positive correlation was found with theta activity, but alpha and beta bands did not elicit any systematic association with cortisol profiles. These results demonstrate a temporal association between cortisol secretory pulses and delta wave activity in man, suggesting the existence of a central control common to both variables.

A double-blind, placebo-controlled evaluation of the effects of orally administered venlafaxine on sleep in inpatients with major depression.

Venlafaxine, a member of a novel chemical class, phenethylamines, is a new antidepressant that inhibits neuronal uptake of serotonin, norepinephrine, and dopamine (in decreasing order of potency) at doses of 75 to 375 mg per day. Depression and antidepressant drugs are known to modify human sleep patterns. Our objective in this double-blind, placebo-controlled study was to assess the effects of venlafaxine on polysomnographic variables by comparing the effects of venlafaxine and placebo on sleep (hypnographic and all-night electroencephalographic [EEG] spectral analysis) and clinical measures (Hamilton Rating Scale for Depression [HAM-D], Montgomery-Asberg Depression Rating Scale [MADRS], and Clinical Global Impressions [CGI]) in inpatients with major depression (DSM-III-R). Following a 7- to 13-day placebo washout period, patients were randomly assigned to receive either placebo or venlafaxine (maximum dose 225 mg/day) for up to 29 days. Sleep evaluations took place at baseline (3 nights immediately before entering the double-blind phase), after 1 week of treatment, and after 1 month of treatment. Sleep stage parameters and all-night spectral parameters were first tested by analysis of variance for repeated measures and then, if indicated, by two-tailed Student t-test. The results on psychiatric rating scales showed improvement from baseline in both treatment groups at all time points, with improvement tending to be greater in the venlafaxine group. Venlafaxine induced a decrease of sleep continuity (decreased total sleep time and increased wake time), an important increase in the onset latency of rapid eye movement (REM) sleep, and a decrease in total REM sleep duration. All-night sleep EEG frequency structure was not modified significantly by venlafaxine treatment as compared with placebo. In conclusion, venlafaxine, despite its novel chemical structure, shows a sleep profile comparable with that of most classical antidepressants.