0456 Natural History of Insomnia: Sleep Reactivity Predicts New-Onset Acute Insomnia

Introduction

Greater vulnerability to stress-related sleep disturbance (i.e., sleep reactivity) is a risk factor for chronic insomnia (CI). What has not been investigated is whether greater sleep reactivity, as assessed by the Ford Insomnia Response to Stress Test (FIRST), predicts the onset of acute insomnia (AI), and more, whether greater sleep reactivity predicts the transition from AI to CI.

Methods

A national cohort of 1,222 good sleeper subjects (68% female; mean age=53.2 years) were prospectively assessed to estimate the incidence of AI and CI. The FIRST was completed at baseline and sleep diaries were completed on a daily basis for a period of one year. Subjects were categorized based on their FIRST scores (high, FIRST>16; low, FIRST≤16). Subjects were also grouped based on whether they developed AI (two consecutive weeks with a frequency of ≥ 3 nights per week of sleep initiation or maintenance problems) or maintained good sleep (GS; n=896). For those subjects that transitioned to AI (n=326), they were also grouped based on whether or not they developed CI (insomnia ≥ 3 nights/week for at least three months; n=23). Chi-square analyses were performed to determine if higher FIRST scores at baseline predicted the incidence of AI or CI.

Results

32.5% of subjects in the high FIRST group met criteria for AI at some point during the one-year interval, whereas 22.5% of subjects in the low FIRST group experienced AI (χ 2=15.2, p<.001). In contrast, FIRST did not predict CI status (low FIRST, 8.5% CI, high FIRST, 5.6% CI; χ 2=1.1, p=.30).

Conclusion

Greater sleep reactivity predicted incident AI but not the onset of CI. While these findings suggest that sleep reactivity may be a predisposing factor for AI, data are not consistent with previous findings showing FIRST scores are predictive of the development of CI. It’s possible that the present study was underpowered to detect these differences, given that the incidence of CI was low (less than 2% of the total sample). Additional analyses are ongoing to evaluate the temporal association between stressful life events and AI in subjects with high and low FIRST scores.

Support

Perlis: NIH R01AG041783, K24AG055602

Generalized covariance-adjusted canonical correlation analysis with application to psychiatry

The lack of control over covariates in practice motivates the need for their adjustment when measuring the degree of association between two sets of variables, for which canonical correlation is traditionally used. In most studies however, there is also a lack of control over the attributes of responses for the sets of variables of interest. In particular, a portion of the response variable may be continuous and the other discrete. For such settings, the traditional partial canonical correlation approach is restrictive, since a covariate-adjustment for a set of continuous variables is assumed. By ignoring the assumption of continuous variates and proceeding with a partial canonical correlation analysis in the presence of continuous and discrete variates, results in canonical correlation estimates that are not consistent. In this paper we generalize the traditional partial canonical correlation approach to covariate-adjustment by allowing the response variables to contain continuous, as well as discrete, variates. The methodology is illustrated with a psychiatric application for examining which sleep variables relate to which depressive symptoms, as measured by commonly used constructs that presents with both continuous and discrete outcomes.

Behavioral treatment of insomnia: a clinical case series study

There is substantial experimental evidence that behavioral treatment of insomnia produces significant clinical improvement and that treatment gains tend to be maintained over time. Less clear is whether behavioral treatment is effective as it is plied in clinical settings. In this clinical case series study, we evaluated 47 patients with primary insomnia. It was found that patients were, on average, 43% improved. This average corresponded to a 65% reduction in sleep latency, a 46% decrease in number of awakenings per night, a 48% reduction in wake time after sleep onset, and a 13% increase in total sleep time. These results suggest that behavioral treatment for insomnia is as effective in clinical settings as it is as under clinical trial conditions.

Longitudinal sleep EEG power spectral analysis studies in adolescents with minor head injury

Power spectral analysis (PSA) was used to evaluate the longitudinal overnight electroencephalographic (EEG) sleep recordings of eight subjects, between the ages of 15 and 19 years, who had sustained a minor head injury (MHI). Recordings occurred within 72 h, 6 weeks, and 12 weeks following MHI. A conditioning night preceded the first study night during which EEG electrodes were in place and subjects slept at least 7.5 h with a mean sleep efficiency of 91%. PSA was performed on four channels of EEG data recorded from fronto-temporal (F3-T3, F4-T4), and temporal (T3-T5, T4-T6) electrodes. The three waveforms associated with sleep, Delta, Theta, and Alpha-1 were all significantly elevated within 72 h post-MHI. Over time all wave forms decreased in mean log power. Theta in rapid eye movement (REM) sleep Cycle 1 decreased significantly within 6 weeks postinjury. The greatest number of significant changes, over the 12-week period were recorded during the non-REM (NREM) sleep. Changes included (1) significant decreases in mean log power of Theta and Alpha-1 in Cycle 1 from fronto-temporal leads; (2) significant decreases in Delta, Theta, and Alpha-1 in Cycle 2 from fronto-temporal leads, and (3) significant decreases in Delta and Theta for consistency during Cycle 2 from temporal leads. The intrusion of Theta into the first REM cycle within 6 weeks and its subsequent decrease within 6 weeks suggested the initiation of recovery toward baseline values. This was followed by decreased levels of Theta power during NREM Cycles 1 and 2, and Delta power in Cycle 2, both of which approached their lowest levels within 12 wks. The decrease in Alpha-1 power occurred last. Alpha-1 remained elevated through both Cycles 1 and 2 of the 6th week and then showed a precipitous decrease between the sixth and twelfth week. These findings suggested that following MHI, the brain has a specific sequence of recovery as illustrated by Delta, Theta, and Alpha-1 powers requiring different time frames to reach their lowest levels.

Enhancement of sleep by microinjection of triazolam into the medial preoptic area

A growing literature suggests that the basal forebrain may contain structures involved in the regulation of sleep. As part of a series of studies designed to locate the site(s) of hypnotic action of benzodiazepines, we have injected triazolam into the medial preoptic area (MPA) of the hypothalamus of rats. Total sleep time was increased, due primarily to an increase in non-rapid eye movement (non-REM) sleep and a trend toward a decrease in intermittent waking time. A previous study from this laboratory reported that injections into the raphe nucleus decreased sleep, whereas injections at adjacent sites were without effect. These studies indicate the selectivity with which different brain regions respond to triazolam in terms of actions on sleep.

Periodic cessation of respiratory effort during sleep in adult rats

Using a noninvasive technique which measures respiration as a function of chest and abdominal movement, it was found that adults rats have periodic cessations of respiratory effort during the daytime. In a preliminary study, male Fischer-344 rats had respiratory pauses of 2.4-2.6 seconds duration, which tended to be more frequent in older (22-month) compared to young adult (3-month) rats. Respiratory rate was lower and respiratory volume greater in the 22-month-old animals. In a more detailed study of 3-month-old male Sprague-Dawley rats, respiratory pauses were divided into two types: those preceded by a large inspiration ("sighs") and those which were not. The latter, which appear to be most analogous to human apneas of clinical interest, occurred in all animals studied, with a frequency of 13 to 26 events during six hour recordings. These pauses varied in duration from 2.0 to 6.1 seconds and were most frequent and longest in REM sleep.

Arousal induced by injection of triazolam into the dorsal raphe nucleus of rats

In order to explore the possible sites at which benzodiazepines alter sleep, we have performed sleep studies following administration of 0.5 microgram of triazolam into the dorsal raphe nucleus of rats. Triazolam significantly increased sleep latency and decreased non-rapid eye movement (REM) sleep, with an effect greatest in the first 2 hours after injection. Total REM sleep time was not significantly affected, although there was a modest trend toward reduction in the first 2 hours. In contrast to the decreased sleep resulting from injection into the dorsal raphe nucleus, triazolam did not significantly alter sleep in animals in whom it was injected into surrounding areas. Similarly, the low dose employed in this study did not significantly affect sleep when injected into the lateral ventricle. These data are reminiscent of studies showing transient decreases in sleep following lesions of the dorsal raphe nucleus.

Sleep and benzodiazepine receptor sub-types

CL 218,872, which preferentially binds to benzodiazepine (BZ) type 1 receptors, and flurazepam, which is thought to bind to both types 1 and 2, were given alone and in combination to rats. CL 218,872 had little effect on sleep latency, but significantly increased total sleep time. As expected, flurazepam both shortened sleep latency and prolonged total sleep. Pretreatment with CL 218,872 had no effect on these alterations in sleep induced by flurazepam. The implications for possible significance of sub-typing of BZ receptors are discussed.

Sleep induction by an adrenal steroid in the rat

The ring A reduced metabolites of deoxycorticosterone and progesterone, 3 alpha, 5 alpha-tetrahydrodeoxycorticosterone (THDOC) and 3 alpha-hydroxy-5 alpha-dihydroprogesterone (3 alpha-OH-DHP) have been shown to be potent barbiturate-like ligands of the benzodiazepine receptor complex. The former has also been reported to have anxiolytic effects in mice and rats. In the present study, sleep recordings were obtained on rats given 5 and 10 mg/kg of these steroids alone and in combination with flurazepam. THDOC, but not 3 alpha-OH-DHP, had potent dose-dependent sleep-inducing properties and increased nonREM sleep. Flurazepam had similar hypnotic effects and also reduced REM sleep. There were no significant interactions between THDOC and flurazepam, except in the case of REM latency, which tended to increase when the two compounds were given together. In summary, THDOC, a mineralocorticoid metabolite found in brain, has sedative properties and could conceivably play a role in stress responses.