Sleep Disorders
Conditions
Keywords
Primary Insomnia, Insomnia, Placebo-Controlled, Double-Blind, Polysomnography, PET Studies, Escitalopram, Zolpidem
Brief summary
This study will compare the symptoms, experiences, and laboratory sleep characteristics of young adults with and without insomnia.
Detailed description
The overall aim of this research study is to compare the symptoms, experiences and laboratory sleep characteristics of young adults with and without insomnia. Insomnia is a pattern of difficulty falling asleep, staying asleep or feeling poorly rested despite an adequate amount of time for sleep, which occurs nearly every night for one month or longer. For those with insomnia, we will look at the effects of an intervention with one of two medications (escitalopram or zolpidem) or an inactive pill (placebo). This intervention will be followed by re-evaluation of symptoms, experiences and laboratory sleep characteristics. The three hypotheses being investigated are: compared to control subjects, those with insomnia will demonstrate affective disturbance and heightened arousal; the different medications will have different degrees of effect on the two dimensions being measured (affective disturbance and heightened arousal); and PET scans will reveal different patterns of activity in the brains of groups of people with insomnia. We will specifically focus on the syndrome of Primary Insomnia (PI), defined by DSM-IV as insomnia that lasts for at least one month and causes significant impairment or distress. PI excludes insomnia that occurs exclusively during the course of another sleep, mental, substance-induced, or medical disorder. Insomnia is a significant public health problem because of its prevalence, morbidity, and the risk it poses for the development of subsequent mental disorders, particularly depressive and anxiety disorders. Understanding the psychobiology of primary insomnia is a critical step toward addressing questions regarding its relationships with mood and anxiety disorders. Our model of insomnia builds on two major concepts running through previous insomnia research, affective disturbance and heightened arousal, as driving factors for the sleep-wake disturbances that define PI. Implicit in this model is that individuals with PI have different degrees of each dysfunction, which accounts for their heterogeneity of clinical symptoms. Contemporary theories of affect structure suggest that these two dimensions may be orthogonal in pure form, but are nevertheless related in clinical conditions characterized by mixed anxiety-depression, such as PI. Measures of affective disturbance and arousal in this study will include questionnaires, diary-based assessments, and physiological measures. Pharmacological treatment probes may help to further distinguish the roles of affective disturbance and heightened arousal in insomnia. We will use a benzodiazepine receptor agonist (BzRA), zolpidem, and an antidepressant, escitalopram. BzRA potentiate the effects of GABA (1), but have minimal direct activity at any other receptor types. They are efficacious treatments for insomnia (2-4), but have little effect on mood. We chose zolpidem because it is relatively specific for hypnotic versus anxiolytic or other actions (5), because it is the most widely-prescribed BzRA hypnotic, and because it is well-tolerated (6). Clinically-effective doses of even nonsedating antidepressants can also improve symptoms in PI (7), suggesting that direct sedation is not their only mechanism for improving insomnia. We chose escitalopram because it is neither strongly alerting nor sedating in clinical and polysomnographic studies. This sleep-neutral profile will allow us to use it for its effects on affective disturbance, rather than its nonspecific sedating properties. Escitalopram's specific effect on serotonin reuptake blockade and its lack of affinity for Bz receptors distinguish it from zolpidem as a pharmacologic probe. Functional neuroimaging studies in wakefulness and sleep may also help to identify the substrate of affective disturbance and heightened arousal in insomnia. Affective disturbance in the form of MDD is associated with alterations in both regional deactivation patterns during NREM sleep, and regional activation patterns during REM sleep. These observations suggest that a sleep-wake functional neuroimaging paradigm in insomnia patients, in conjunction with behavioral measures, may help to identify which brain systems mediate heightened arousal and affective disturbance, and how these systems interact.
Interventions
The benzodiazepine receptor agonist (BzRA), zolpidem was given in an initial dose of 5 mg by mouth every night, 30 minutes prior to bedtime. The dose was increased to a maximum of 10 mg after the first week if there was no improvement in overall symptoms (CGI score of 4 or \>). The dose was decreased to 5 mg if side effects occurred.
The antidepressant, escitalopram was initiated at 5 mg by mouth every night, 30 minutes prior to bedtime. If there were no side effects, the dose was increased every four days until the target dose of 20 mg (maximum dose) was reached by day 13. If significant side effects appeared, the highest tolerated dose was used.
A placebo capsule was given with instructions to take it every night by mouth, 30 minutes prior to bedtime.
Sponsors
Study design
Eligibility
Inclusion criteria
* Physically healthy * Meets DSM-IV criteria for primary insomnia * For subjects interested in PET study only: right-handedness
Exclusion criteria
* Currently taking antidepressants, antianxiety medications or medications for sleep disorders * Currently experiencing symptoms of psychiatric disorders such as major depressive disorder, bipolar disorder, generalized anxiety disorder * Significant or unstable acute or chronic medical conditions, such as seizure disorder, tumor, liver disease, active peptic ulcer disease, arthritis, irritable bowel disease * Meets DSM-IV criteria for sleep apnea or periodic limb movement disorder
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Change in Pittsburgh Sleep Quality Index | post treatment minus baseline assessment battery. This averaged 100 days. | Self-report measure of sleep quality developed at University of Pittsburgh by Daniel J. Buysse, M.D. The PSQI total score ranges from 0 to 21 with 0 being marvelous sleep and 21 being horrid sleep. The difference score, reported below, is the total score after at least 5 weeks of treatment in one of the three arms, minus the baseline total score. A negative score means that the sleep of the participant improved. |
| Change in Diary Sleep Efficiency | post treatment minus baseline. This averaged 69 days. | The change in self-report sleep efficiency calculated from 7-day sleep diary (DSE): Sleep efficiency is the percent of (time spent asleep divided by the amount of time between good night time and final awakening). It ranges from 0 (no sleep at all) to 100 (asleep the second your head hits the pillow until you wake up in the morning and get out of bed). Participants report the time they go to bed, how long they think it takes them to fall asleep, how many minutes they are awake during the night, and then what time they finally wake up in the morning. These values are used to calculate the diary sleep efficiency for each night and then we averaged these across the 7 days of diary collected pre and post treatment. The values below are post treatment DSE minus pre treatment DSE. A positive number means that the DSE was higher (better) post treatment. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Change in PSG Sleep Efficiency for the Second Night in the Sleep Lab at Each Timepoint | post treatment minus baseline PSG sleep studies. This averaged 70 days | Change in PSG Sleep Efficiency (SE) between post-treatment and baseline: Sleep efficiency is the percent of time spent asleep divided by the total sleep recording period in the sleep lab. This value is calculated using the results of the polysomnographic sleep study. It ranges from 0 (no sleep at all) to 100 (asleep the second the sleep recording starts (GNT) until the sleep recording ends (GMT) in the morning). The values below are post treatment SE minus pre treatment SE. A positive number means that the SE was higher (better) post treatment. |
Countries
United States
Participant flow
Recruitment details
Recruitment period lasted from 1/1/2002 to 12/1/2007. Subjects were recruited from the general public via tv, newspaper, and radio advertisements and via mass mailings.
Pre-assignment details
Subjects who did not meet DSM-IV criteria for primary insomnia, had apnea or periodic limb movement disorder, or had an exclusionary psychiatric diagnosis (generalized anxiety disorder, major depressive disorder, substance abuse) were excluded.
Participants by arm
| Arm | Count |
|---|---|
| Experimental: Zolpidem The benzodiazepine receptor agonist (BzRA), zolpidem was given in an initial dose of 5 mg by mouth every night, 30 minutes prior to bedtime. The dose was increased to a maximum of 10 mg after the first week if there was no improvement in overall symptoms (CGI score of 4 or \>). The dose was decreased to 5 mg if side effects occurred. | 26 |
| Experimental: Excitalopram The antidepressant, escitalopram was initiated at 5 mg by mouth every night, 30 minutes prior to bedtime. If there were no side effects, the dose was increased every four days until the target dose of 20 mg (maximum dose) was reached by day 13. If significant side effects appeared, the highest tolerated dose was used. | 23 |
| Placebo Comparator: Placebo A placebo capsule was given with instructions to take it every night by mouth, 30 minutes prior to bedtime. | 20 |
| Total | 69 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 | FG002 |
|---|---|---|---|---|
| Overall Study | Withdrawal by Subject | 6 | 3 | 1 |
Baseline characteristics
| Characteristic | Experimental: Excitalopram | Placebo Comparator: Placebo | Experimental: Zolpidem | Total |
|---|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical Between 18 and 65 years | 23 Participants | 20 Participants | 26 Participants | 69 Participants |
| Age, Continuous | 34.49 years STANDARD_DEVIATION 9.34 | 37.46 years STANDARD_DEVIATION 8.63 | 38.07 years STANDARD_DEVIATION 9.07 | 36.67 years STANDARD_DEVIATION 9.05 |
| Region of Enrollment United States | 23 participants | 20 participants | 26 participants | 69 participants |
| Sex: Female, Male Female | 12 Participants | 11 Participants | 15 Participants | 38 Participants |
| Sex: Female, Male Male | 11 Participants | 9 Participants | 11 Participants | 31 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — | — / — |
| other Total, other adverse events | 0 / 26 | 2 / 23 | 0 / 20 |
| serious Total, serious adverse events | 0 / 26 | 0 / 23 | 1 / 20 |
Outcome results
Change in Diary Sleep Efficiency
The change in self-report sleep efficiency calculated from 7-day sleep diary (DSE): Sleep efficiency is the percent of (time spent asleep divided by the amount of time between good night time and final awakening). It ranges from 0 (no sleep at all) to 100 (asleep the second your head hits the pillow until you wake up in the morning and get out of bed). Participants report the time they go to bed, how long they think it takes them to fall asleep, how many minutes they are awake during the night, and then what time they finally wake up in the morning. These values are used to calculate the diary sleep efficiency for each night and then we averaged these across the 7 days of diary collected pre and post treatment. The values below are post treatment DSE minus pre treatment DSE. A positive number means that the DSE was higher (better) post treatment.
Time frame: post treatment minus baseline. This averaged 69 days.
Population: The number of subjects with sleep diaries at baseline and after at least 5 weeks of treatment. Not all of these participants 'completed' the protocol
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Zolpidem | Change in Diary Sleep Efficiency | 3.20 diff score of diary Sleep Efficiency | Standard Deviation 11.62 |
| Escitalpram | Change in Diary Sleep Efficiency | 0.61 diff score of diary Sleep Efficiency | Standard Deviation 11.65 |
| Placebo | Change in Diary Sleep Efficiency | 4.87 diff score of diary Sleep Efficiency | Standard Deviation 11.26 |
Change in Pittsburgh Sleep Quality Index
Self-report measure of sleep quality developed at University of Pittsburgh by Daniel J. Buysse, M.D. The PSQI total score ranges from 0 to 21 with 0 being marvelous sleep and 21 being horrid sleep. The difference score, reported below, is the total score after at least 5 weeks of treatment in one of the three arms, minus the baseline total score. A negative score means that the sleep of the participant improved.
Time frame: post treatment minus baseline assessment battery. This averaged 100 days.
Population: Number of subjects who had total PSQI scores after at least 5 weeks of treatment. The PSQI requires that all questions be answered for a total score to be calculated. This analysis includes participants who did not complete the protocol and also omits those who has missing total scores due to missing items on the PSQI
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Zolpidem | Change in Pittsburgh Sleep Quality Index | -0.86 difference score of PSQI total | Standard Deviation 4.51 |
| Escitalpram | Change in Pittsburgh Sleep Quality Index | -3.15 difference score of PSQI total | Standard Deviation 4.52 |
| Placebo | Change in Pittsburgh Sleep Quality Index | -3.32 difference score of PSQI total | Standard Deviation 2.63 |
Change in PSG Sleep Efficiency for the Second Night in the Sleep Lab at Each Timepoint
Change in PSG Sleep Efficiency (SE) between post-treatment and baseline: Sleep efficiency is the percent of time spent asleep divided by the total sleep recording period in the sleep lab. This value is calculated using the results of the polysomnographic sleep study. It ranges from 0 (no sleep at all) to 100 (asleep the second the sleep recording starts (GNT) until the sleep recording ends (GMT) in the morning). The values below are post treatment SE minus pre treatment SE. A positive number means that the SE was higher (better) post treatment.
Time frame: post treatment minus baseline PSG sleep studies. This averaged 70 days
Population: Number of subjects who had polysomnography at Week 9 and at pretreatment. Some of the participants who 'completed' the protocol did not have follow up sleep studies.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Zolpidem | Change in PSG Sleep Efficiency for the Second Night in the Sleep Lab at Each Timepoint | 0.43 difference score for SE | Standard Deviation 6.23 |
| Escitalpram | Change in PSG Sleep Efficiency for the Second Night in the Sleep Lab at Each Timepoint | 1.16 difference score for SE | Standard Deviation 12.94 |
| Placebo | Change in PSG Sleep Efficiency for the Second Night in the Sleep Lab at Each Timepoint | 0.32 difference score for SE | Standard Deviation 10.5 |