Stahl's Essential Psychopharmacology: Neuroscientific Basis and Practical Applications, 4th Ed.

Chapter 11. Disorders of sleep and wakefulnessand their treatment

   Neurobiology of sleep and wakefulness

    The arousal spectrum

    The sleep/wake switch


   Insomnia and hypnotics

    What is insomnia?

    Chronic treatment for chronic insomnia?

    Benzodiazepine hypnotics

    GABAA positive allosteric modulators (PAMs) as hypnotics

    Psychiatric insomnia and the GABAA PAMs

    Melatonergic hypnotics

    Serotonergic hypnotics

    Histamine H1 antagonists as hypnotics

    Dopamine agonists and α2δ ligands for insomnia associated with restless legs syndrome (RLS)

    Behavioral treatments of insomnia

    Who cares about slow-wave sleep?

    Orexin antagonists as novel hypnotics

   Excessive daytime sleepiness (hypersomnia) and wake-promoting agents

    What is sleepiness?

    What’s wrong with being sleepy?

    Mechanism of action of wake-promoting agents


This chapter will provide a brief overview of the psychopharmacology of disorders of sleep and wakefulness. Included here are short discussions of the symptoms, diagnostic criteria, and treatments for disorders that cause insomnia, excessive daytime sleepiness, or both. Clinical descriptions and formal criteria for how to diagnose sleep disorders are mentioned here only in passing. The reader should consult standard reference sources for this material. The discussion here will emphasize the links between various brain circuits and their neurotransmitters and disorders that cause insomnia or sleepiness. The goal of this chapter is to acquaint the reader with ideas about the clinical and biological aspects of sleep and wakefulness, how various disorders can alter sleep and wakefulness, and how many new and evolving treatments can resolve the symptoms of insomnia and sleepiness.

The detection, assessment, and treatment of sleep/wake disorders are rapidly becoming standardized parts of a psychiatric evaluation. Modern psychopharmacologists increasingly consider sleep to be a psychiatric “vital sign,” requiring routine evaluation and symptomatic treatment whenever sleep disorders are encountered. This is similar to the situation of pain (Chapter 10), which is also increasingly being considered as another psychiatric “vital sign.” That is, disorders of sleep (and pain) are so important, so pervasive, and cut across so many psychiatric conditions that the elimination of these symptoms – no matter what psychiatric disorder may be present – is increasingly recognized as necessary in order to achieve full symptomatic remission for the patient.

Many of the treatments discussed in this chapter are covered in previous chapters. For details of mechanisms of insomnia treatments that are also used for the treatment of depression, the reader is referred to Chapter 7. For those insomnia treatments that are benzodiazepines and share the same mechanism of action with various benzodiazepine anxiolytics, the reader is referred to Chapter 9. For various hypersomnia treatments, especially stimulants, the reader is referred to Chapter 12 on ADHD and to Chapter 14 on drug abuse, which also discuss the use and abuse of stimulants. The discussion in this chapter is at the conceptual level, and not at the pragmatic level. The reader should consult standard drug handbooks (such as Stahl’s Essential Psychopharmacology: the Prescriber’s Guide) for details of doses, side effects, drug interactions, and other issues relevant to the prescribing of these drugs in clinical practice.

Neurobiology of sleep and wakefulness

The arousal spectrum

Although many experts approach insomnia and sleepiness by emphasizing the separate and distinct disorders that cause them, many pragmatic psychopharmacologists approach insomnia or excessive daytime sleepiness as important symptoms that cut across many conditions and that occur along a spectrum from deficient arousal to excessive arousal (Figure 11-1). In this conceptualization, an awake, alert, creative and problem-solving person has the right balance between too much and too little arousal (baseline brain functioning in gray at the middle of the spectrum in Figure 11-1). As arousal increases beyond normal, during the day there is hypervigilance (Figure 11-1); if this increased, arousal occurs at night and there is insomnia (Figure 11-1, and overactivation of the brain in red at the right-hand side of the spectrum in Figure 11-2). From a treatment perspective, insomnia can be conceptualized as a disorder of excessive nighttime arousal, with hypnotics moving the patient from too much arousal to sleep (Figure 11-2).

Figure 11-1. Arousal spectrum of sleep and wakefulness. One’s state of arousal is more complicated than simply being “awake” or “asleep.” Rather, arousal exists as if on a dimmer switch, with many phases along the spectrum. Where on the spectrum one lies is influenced in large part by five key neurotransmitters: histamine (HA), dopamine (DA), norepinephrine (NE), serotonin (5HT), and acetylcholine (ACh). When there is good balance between too much and too little arousal (depicted by the gray [baseline] color of the brain), one is awake, alert, and able to function well. As the dial shifts to the right there is too much arousal, which may cause hypervigilance and consequently insomnia at night. As arousal further increases this can cause cognitive dysfunction, panic, and in extreme cases perhaps even hallucinations. On the other hand, as arousal diminishes, individuals may experience inattentiveness, cognitive dysfunction, sleepiness, and ultimately sleep.

Figure 11-2. Insomnia: excessive nighttime arousal? Insomnia is conceptualized as being related to hyperarousal at night, depicted here as the brain being red (overactive). Agents that reduce brain activation, such as positive allosteric modulators of GABAA receptors (e.g., benzodiazepines, “Z drugs”), histamine 1 antagonists, and serotonin 2A/2C antagonists, can shift one’s arousal state from hyperactive to sleep.

On the other hand, as arousal diminishes, symptoms crescendo from mere inattentiveness to more severe forms of cognitive disturbances until the patient has excessive daytime sleepiness with sleep attacks (Figure 11-1, and hypoactivation of the brain in blue at the left-hand side of the spectrum in Figure 11-3). From a treatment perspective, sleepiness can be conceptualized as a disorder of deficient daytime arousal, with wake-promoting agents moving the patient from too little arousal to awake with normal alertness (Figure 11-3).

Figure 11-3. Excessive daytime sleepiness: deficient daytime arousal. Excessive sleepiness is conceptualized as being related to hypoarousal during the day, depicted here as the brain being blue (hypoactive). Agents that increase brain activation, such as the stimulants, modafinil, and caffeine, can shift one’s arousal state from hypoactive to awake with normal alertness.

Note in Figure 11-1 that cognitive disturbance is the product of both too little and too much arousal, consistent with the need of cortical pyramidal neurons to be optimally “tuned,” with too much activity making them just as out of tune as too little. Note also in Figures 11-1 through 11-3 that the arousal spectrum is linked to the actions of five neurotransmitters shown in the brains represented in these figures (i.e., histamine, dopamine, norepinephrine, serotonin, and acetylcholine). Sometimes these neurotransmitter circuits as a group are called the ascending reticular activating system, because they are known to work together to regulate arousal. This same ascending neurotransmitter system is blocked at several sites by many agents that cause sedation. Actions of sedating drugs on these neurotransmitters are discussed in Chapter 5 on antipsychotics and illustrated in Figure 5-38Figure 11-1 also shows that excessive arousal can extend past insomnia to panic, hallucinations, and all the way to frank psychosis (far right-hand side of the spectrum).

The sleep/wake switch

We have discussed how the ascending neurotransmitter systems from the brainstem regulate a cortical arousal system on a smooth continuum like a rheostat on a lighting system or a volume button on a radio. There is another set of circuits in the hypothalamus that regulate sleep and wake discontinuously, like an on/off switch. Not surprisingly, this circuitry is called the sleep/wake switch (Figure 11-4). The “on” switch is known as the wake promoter and is localized within the tuberomammillary nucleus (TMN) of the hypothalamus (Figure 11-4A). The “off” switch is known as the sleep promoter and is localized within the ventrolateral preoptic (VLPO) nucleus of the hypothalamus (Figure 11-4B).

Figure 11-4. Sleep/wake switch. The hypothalamus is a key control center for sleep and wake, and the specific circuitry that regulates sleep/wake (i.e., whether the dimmer switch is set all the way to the left for sleep or is somewhere else along the continuum for wake) is called the sleep/wake switch. The “off” setting, or sleep promoter, is localized within the ventrolateral preoptic nucleus (VLPO) of the hypothalamus, while “on” – the wake promoter – is localized within the tuberomammillary nucleus (TMN) of the hypothalamus. Two key neurotransmitters regulate the sleep/wake switch: histamine from the TMN and GABA from the VLPO. (A) When the TMN is active and histamine is released to the cortex and the VLPO, the wake promoter is on and the sleep promoter inhibited. (B) When the VLPO is active and GABA is released to the TMN, the sleep promoter is on and the wake promoter inhibited. The sleep/wake switch is also regulated by orexin/hypocretin neurons in the lateral hypothalamus (LAT), which stabilize wakefulness, and by the suprachiasmatic nucleus (SCN) of the hypothalamus, which is the body’s internal clock and is activated by melatonin, light, and activity to promote either sleep or wake.

Two other sets of neurons are shown in Figure 11-4 as regulators of the sleep/wake switch: orexin-containing neurons of the lateral hypothalamus (LAT) and melatonin-sensitive neurons of the suprachiasmatic nucleus (SCN). The lateral hypothalamus serves to stabilize and promote wakefulness via a peptide neurotransmitter known by two different names: orexin and hypocretin. These lateral hypothalamic neurons and their orexin are lost in narcolepsy, especially narcolepsy with cataplexy. New hypnotics on the horizon (orexin antagonists) block the receptors for these neurotransmitters and are discussed later in this chapter. The SCN is the brain’s internal clock, or pacemaker, and regulates circadian input to the sleep/wake switch in response to how it is programmed by hormones such as melatonin and by the light/dark cycle. Circadian rhythms and the SCN are discussed in Chapter 7 on antidepressants and illustrated in Figures 7-39 to 7-42.

The circadian wake drive is shown in Figure 11-5 over two full 24-hour cycles. Also shown in Figure 11-5 is the ultradian sleep cycle (a cycle faster than a day, showing cycling in and out of REM and slow-wave sleep several times during the night). Homeostatic sleep drive, illustrated as well in Figure 11-5, increases the drive for sleep as the day goes on, presumably due to fatigue, and diminishes at night with rest. The novel neurotransmitter adenosine is linked to homeostatic drive, and appears to accumulate as this drive increases during the day, and to diminish at night. Caffeine is now known to be an antagonist of adenosine, and this may explain in part its ability to promote wakefulness and diminish fatigue, namely by opposing endogenous adenosine’s regulation of the homeostatic sleep drive.

Figure 11-5. Processes regulating sleep. Several processes that regulate sleep/wake are shown here. The circadian wake drive is a result of input (light, melatonin, activity) to the suprachiasmatic nucleus. Homeostatic sleep drive increases the longer one is awake and decreases with sleep. As the day progresses, circadian wake drive diminishes and homeostatic sleep drive increases until a tipping point is reached and the ventrolateral preoptic sleep promoter (VLPO) is triggered to release GABA in the tuberomammillary nucleus (TMN) and inhibit wakefulness. Sleep itself consists of multiple phases that recur in a cyclical manner; this process is known as the ultradian cycle, and is depicted at the top of this figure.

Two key neurotransmitters regulate the sleep/wake switch: histamine from the TMN and GABA (γ-aminobutyric acid) from the VLPO. Thus, when the sleep/wake switch is on, the wake promoter TMS is active and histamine is released (Figure 11-4). This occurs both in the cortex to facilitate arousal, and in the VLPO to inhibit the sleep promoter. As the day progresses, circadian wake drive diminishes and homeostatic sleep drive increases (Figure 11-5); eventually a tipping point is reached, and the VLPO sleep promoter is triggered, the sleep/wake switch is turned off, and GABA is released in the TMN to inhibit wakefulness (Figure 11-4).

Disorders characterized by excessive daytime sleepiness can be conceptualized as the sleep/wake switch being off during the daytime. Wake-promoting treatments such as modafinil given during the day tip the balance back to wakefulness by promoting the release of histamine from TMN neurons. The exact mechanism of this enhancement of histamine release by modafinil or stimulants is not known, but is currently hypothesized to be related in part to a downstream consequence of the actions of wake-promoting drugs on dopamine neurons, especially by blocking the dopamine transporter DAT.

On the other hand, disorders characterized by insomnia can be conceptualized as the sleep/wake switch being on at night. Insomnia can be treated either by agents that enhance GABA actions, and thus inhibit the wake promoter, or by agents that block the action of histamine released from the wake promoter and act at postsynaptic H1 receptors.

Disorders characterized by a disturbance in circadian rhythm can be conceptualized as either “phase delayed,” with the wake promoter and sleep/wake switch being turned on too late in a normal 24-hour cycle, or “phase advanced,” with the wake promoter and sleep/wake switch being turned on too early in a normal 24-hour cycle. That is, individuals who are phase delayed, including many depressed patients and many normal adolescents, still have their sleep/wake switch off when it is time to get up (see discussion in Chapter 7 and Figure 7-39). Giving such individuals morning light and evening melatonin can reset the circadian clock in the SCN so that it wakes the person up earlier. Other individuals may be phase advanced, including many normal elderly people. Giving these individuals evening light and morning melatonin can reset their SCNs so that the sleep/wake switch stays off a bit longer, returning the patient to a normal rhythm.


Histamine is one of the key neurotransmitters regulating wakefulness, and is the ultimate target of many wake-promoting drugs (via downstream histamine release) and sleep-promoting drugs (antihistamines). Histamine is produced from the amino acid histidine, which is taken up into histamine neurons and converted to histamine by the enzyme histidine decarboxylase (Figure 11-6). Histamine’s action is terminated by two enzymes working in sequence: histamine N-methyl-transferase, which converts histamine to N-methyl-histamine, and MAO-B, which converts N-methyl-histamine into N-MIAA (N-methyl-indole-acetic acid), an inactive substance (Figure 11-7). Additional enzymes such as diamine oxidase can also terminate histamine action outside of the brain. Note that there is no apparent reuptake pump for histamine. Thus, histamine is likely to diffuse widely away from its synapse, just like dopamine does in prefrontal cortex.

Figure 11-6. Histamine is produced. Histidine (HIS), a precursor to histamine, is taken up into histamine nerve terminals via a histidine transporter and converted into histamine by the enzyme histidine decarboxylase (HDC). After synthesis, histamine is packaged into synaptic vesicles and stored until its release into the synapse during neurotransmission.

Figure 11-7. Histamine’s action is terminated. Histamine can be broken down intracellularly by two enzymes. Histamine N-methyl-transferase (histamine NMT) converts histamine into N-methyl-histamine, which is then converted by monoamine oxidase B (MAO-B) into the inactive substance N-methyl-indole-acetic acid (N-MIAA).

There are a number of histamine receptors (Figures 11-8 through 11-11). The postsynaptic histamine 1 (H1) receptor is best known (Figure 11-9A) because it is the target of “antihistamines” (i.e., H1antagonists) (Figure 11-9B). When histamine itself acts at H1 receptors, it activates a G-protein-linked second-messenger system that activates phosphatidyl inositol, and the transcription factor cFOS, and results in wakefulness, normal alertness, and pro-cognitive actions (Figure 11-9A). When these H1 receptors are blocked in the brain, they interfere with the wake-promoting actions of histamine, and thus can cause sedation, drowsiness, or sleep (Figure 11-9B).

Figure 11-8. Histamine receptors. Shown here are receptors for histamine that regulate its neurotransmission. Histamine 1 and histamine 2 receptors are postsynaptic, while histamine 3 receptors are presynaptic autoreceptors. There is also a binding site for histamine on NMDA receptors – it can act at the polyamine site, which is an allosteric modulatory site.

Figure 11-9. Histamine 1 receptors. (A) When histamine binds to postsynaptic histamine 1 receptors, it activates a G-protein-linked second-messenger system that activates phosphatidyl inositol and the transcription factor cFOS. This results in wakefulness and normal alertness. (B) Histamine 1 antagonists prevent activation of this second messenger and thus can cause sleepiness.

Histamine 2 (H2) receptors, best known for their actions in gastric acid secretion and the target of a number of anti-ulcer drugs, also exist in the brain (Figure 11-10). These postsynaptic receptors also activate a G-protein second-messenger system with cAMP, phosphokinase A, and the gene product CREB. The function of H2 receptors in brain is still being clarified, but apparently is not linked directly to wakefulness.

Figure 11-10. Histamine 2 receptors. Histamine 2 receptors are present both in the body and in the brain. When histamine binds to postsynaptic histamine 2 receptors it activates a G-protein-linked second-messenger system with cAMP, phosphokinase A, and the gene product CREB. The function of histamine 2 receptors in the brain is not yet elucidated but does not appear to be directly linked to wakefulness.

A third histamine receptor is present in brain, namely the H3 receptor (Figures 11-8 and 11-11). Synaptic H3 receptors are presynaptic (Figure 11-11A) and function as autoreceptors (Figure 11-11B). That is, when histamine binds to these receptors, it turns off further release of histamine (Figure 11-11B). One novel approach to new wake-promoting and pro-cognitive drugs is to block these receptors, thus facilitating the release of histamine, allowing histamine to act at H1 receptors to produce the desired effects (Figure 11-11C). Several H3 antagonists are in clinical development.

There is a fourth type of histamine receptor, H4, but these are not known to occur in the brain. Finally, histamine acts also at NMDA (N-methyl-D-aspartate) receptors (Figure 11-8). Interestingly, when histamine diffuses away from its synapse to a glutamate synapse containing NMDA receptors, it can act at an allosteric modulatory site called the polyamine site, to alter the actions of glutamate at NMDA receptors (Figure 11-8). The role of histamine and function of this action are not well clarified.

Histamine neurons all arise from a single small area of the hypothalamus known as the tuberomammillary nucleus (TMN), which is part of the sleep/wake switch illustrated in Figure 11-4. Thus, histamine plays an important role in arousal, wakefulness, and sleep. The TMN is a small bilateral nucleus that provides histaminergic input to most brain regions and to the spinal cord (Figure 11-12).

Figure 11-11. Histamine 3 receptors. Histamine 3 receptors are presynaptic autoreceptors (A), which means that when histamine binds to these receptors it turns off further histamine release (B). Antagonists of these receptors, which are in development, therefore disinhibit histamine release (C) and may hypothetically enhance alertness and cognition.

Figure 11-12. Histaminergic projections from the hypothalamus. In the brain, histamine (HA) is produced solely by cells in the tuberomammillary nucleus (TMN) of the hypothalamus (Hy). From the TMN, histaminergic neurons project to various brain regions including the prefrontal cortex (PFC), the basal forebrain (BF), the striatum (S) and nucleus accumbens (NA), the amygdala (A) and hippocampus (H), brainstem neurotransmitter centers (NT), and spinal cord (SC).

Insomnia and hypnotics

What is insomnia?

Insomnia has many causes, including both sleep disorders and psychiatric disorders. Insomnia can also contribute to the onset, exacerbation, or relapse of many psychiatric disorders and is linked to various dysfunctions in many medical illnesses. Primary insomnia may be a condition with too much arousal both at night and during the day, and thus may be a form of insomnia where the patient is not sleepy during the day despite having poor sleep at night. Primary insomnia may also be a symptom that can progress to a first major depressive episode. Thus, is insomnia a symptom or a disorder? The answer appears to be “yes, both.”

Chronic treatment for chronic insomnia?

A major reconceptualization of insomnia has recently occurred among experts, with a newly formed consensus that insomnia can be chronic and that it may need to be treated chronically. This is a departure from the position held by many sleep experts in the past – that insomnia was treated by attacking its underlying cause, and not by giving chronic “symptomatically masking” treatment with hypnotics. The old guidelines recommending short-term use of hypnotics for insomnia were the product of safety concerns for hypnotics identified first during the barbiturate era and then during the benzodiazepine era.

Other problems associated with long-term use of hypnotics have to do with use of drugs whose half-lives are not ideal for use as hypnotics (Figure 11-13AB, and C). That is, many agents used as hypnotics, particularly in the past, have half-lives that are too long (Figure 11-13A and B). This can cause drug accumulation and hip fractures from falls, especially in the elderly, when such agents are used every night (Figure 11-13A). Long half-life can also cause next-day carryover effects and sedation and memory problems from residual daytime drug levels (Figure 11-13A and B). Other agents used as hypnotics have half-lives that are too short, and their effects can wear off before it is time to wake up, causing insufficient sleep maintenance and nocturnal awakenings, as well as restless and disturbed sleep in some patients (Figure 11-13C). More recently, however, the hypnotics given most frequently for chronic use are those that have optimized half-lives targeting rapid onset of action, and plasma drug levels above the minimally effective concentration, but only until it is time to wake up (Figure 11-13D). Perhaps no therapeutic area of psychopharmacology is as critically dependent upon plasma drug levels, and thus the pharmacokinetics of the drug, as is the use of hypnotics. This fact may be related to the nature of the arousal system and of the sleep/wake switch, which requires pharmacologic action to a degree sufficient to reach the critical tipping point that trips the switch “off” to allow sleep, but only at night.

Figure 11-13. Half-lives of hypnotics. The half-lives of hypnotics can have an important impact on their tolerability and efficacy profiles. (A) Hypnotics with ultra-long half-lives (greater than 24 hours: for example, flurazepam and quazepam) can cause drug accumulation with chronic use. This can cause impairment that has been associated with increased risk of falls, particularly in the elderly. (B) Hypnotics with moderate half-lives (15–30 hours: estazolam, temazepam, most tricyclic antidepressants, mirtazapine, olanzapine) may not wear off until after the individual needs to awaken and thus may have “hangover” effects (sedation, memory problems). (C) Hypnotics with ultra-short half-lives (1–3 hours: triazolam, zaleplon, zolpidem, melatonin, ramelteon) can wear off before the individual needs to awaken and thus cause loss of sleep maintenance. (D) Hypnotics with half-lives that are short but not ultra-short (approximately 6 hours: zolpidem CR and perhaps low doses of trazodone or doxepin) may provide rapid onset of action and plasma levels above the minimally effective concentration only for the duration of a normal night’s sleep.

Other reasons for short-term restrictions on benzodiazepine hypnotics (Figure 11-14) in the past had to do with their long-term effects, including loss of efficacy over time (tolerance) and withdrawal effects, including rebound insomnia in some patients worse than their original insomnia (Figure 11-15A). Recent investigations have shown that some non-benzodiazepine hypnotics may not have these problems (Figure 11-15B). These include the GABAApositive allosteric modulators (PAMs), sometimes also called “Z drugs” (because they all start with the letter Z: zaleplon, zolpidem, zopiclone) (Figure 11-16). Perhaps the best long-term studies have been done with eszopiclone, which shows little or no tolerance, dependence, or withdrawal with use for many months (Figure 11-15B). This is probably also the case for long-term use of zolpidem, zolpidem CR, and the melatonergic agent ramelteon, as well as for “off-label” use of the sedating antidepressant trazodone, none of which have restrictions against chronic use. For these reasons, it is now recognized that chronic insomnia may need chronic treatment with certain hypnotics.

Figure 11-14. Benzo hypnotics. Five benzodiazepines that are approved in the United States for insomnia are shown here. These include flurazepam and quazepam, which have ultra-long half-lives; triazolam, which has an ultra-short half-life; and estazolam and temazepam, which have moderate half-lives.

Figure 11-15. Long-term effects of hypnotics. (A) Short-term, benzodiazepines can be efficacious for treating insomnia. With long-term use, however, benzodiazepines may cause tolerance and, if discontinued, withdrawal effects that may include rebound insomnia. (B) Positive allosteric modulators (PAMs) at GABAA receptors are efficacious for insomnia in the short term, and in the long term do not seem to cause tolerance or withdrawal effects.

Figure 11-16. GABAA positive allosteric modulators (PAMs). Several GABAA PAMs, or “Z drugs,” are shown here. These include racemic zopiclone (not available in the United States), eszopiclone, zaleplon, zolpidem, and zolpidem CR. Zaleplon, zolpidem, and zolpidem CR are selective for GABAA receptors that contain the α1 subunit; however, it does not appear that zopiclone or eszopiclone have this same selectivity.

Benzodiazepine hypnotics

There are at least five benzodiazepines approved specifically for insomnia in the US (Figure 11-14), although there are several others in different countries. Various benzodiazepines developed for the treatment of anxiety disorders are also frequently used to treat insomnia. Benzodiazepine anxiolytics are discussed in Chapter 9 and their mechanism of action is illustrated in Figure 9-23. Because benzodiazepines do not have ideal half-lives for many patients (Figure 11-13AB, and C), and can cause long-term problems (Figure 11-15A), they are generally considered second-line agents for use as hypnotics. However, when first-line agents fail to work, benzodiazepines still have a place in the treatment of insomnia, particularly for insomnia associated with various psychiatric and medical illnesses.

GABAA positive allosteric modulators (PAMs) as hypnotics

These hypnotics act at GABAA receptors to enhance the action of GABA by binding to a site different from where GABA itself binds to this receptor. Benzodiazepines are classified as GABAA PAMs (discussed in Chapter 9 and illustrated in Figure 9-23). Barbituate hypnotics are yet another type of GABAA PAM. However, not all GABAA PAMs are the same, since there are important differences in the ways in which various drugs bind to the GABAAreceptor, which impacts both the safety and the efficacy of various classes of GABAA PAMs.

That is, the GABAA PAMs zaleplon, zolpidem, and zopiclone (Figure 11-16) appear to bind to the GABAA receptor in a way that does not cause a high degree of tolerance to their therapeutic actions, dependence, or withdrawal upon discontinuation from long-term treatment. By contrast, benzodiazepines (Figure 11-14) bind in a manner that changes the conformation of the GABAA receptor such that tolerance generally develops, as well as some degree of dependence and withdrawal, especially for some patients and for some benzodiazepines. Furthermore, for some Z drugs, there is specificity for the α1subtype of GABAA receptor (Figure 11-16). GABAA receptor subtypes are introduced in Chapter 9 and illustrated in Figure 9-21. There are six different subtypes of α subunits for GABAA receptors, and benzodiazepines bind to four of them (α1, α2, α3, and α5) (Figure 11-14), as do zopiclone and eszopiclone (Figure 11-16). The α1 subtype is known to be critical for producing sedation and thus is targeted by every effective GABAA PAM hypnotic. The α1 subtype is also linked to daytime sedation, anticonvulsant actions, and possibly to amnesia. Adaptations of this receptor with chronic hypnotic treatments that target it are thought to lead to tolerance and withdrawal. The α2 and α3 receptor subtypes are linked to anxiolytic, muscle relaxant, and alcohol-potentiating actions. Finally, the α5 subtype, mostly in the hippocampus, may be linked to cognition and other functions. Zaleplon and zolpidem are α1 selective (Figure 11-16). The functional significance of selectivity is not yet proven, but may contribute to the lower risk of tolerance and dependence of these agents.

Modifications of two of the Z drugs, zolpidem and zopiclone, are available for clinical use. For zolpidem, a controlled-release formulation known as zolpidem CR (Figure 11-16) extends the duration of action of zolpidem immediate-release from about 2–4 hours (Figure 11-13B) to a more optimized duration of 6–8 hours, improving sleep maintenance (Figure 11-13D). An alternative dosage formulation of zolpidem for sublingual administration with faster onset and given at a fraction of the usual nighttime dose is also available for middle-of-the-night administration for patients who have middle insomnia. For zopiclone, a racemic mixture of both R and S zopiclone, there is the introduction of the S enantiomer, eszopiclone (Figure 11-16). Clinically meaningful differences between the active enantiomer and the racemic mixture are debated.

Psychiatric insomnia and the GABAA PAMs

In many ways, the introduction of the Z drugs has contributed to the reconceptualization of the treatment of chronic insomnia. That is, optimized pharmacokinetic durations of action (Figure 11-13D) coupled with studies establishing safety in long-term use without a high incidence of tolerance or dependence (Figure 11-15B) have opened the door to the treatment of chronic insomnia chronically. However, most studies of hypnotics are in primary insomnia, not in insomnia associated with psychiatric disorders, leading to fewer clear guidelines as to how to use hypnotics to treat insomnia in conditions such as depression, anxiety disorders, bipolar disorder, etc.

Investigators are currently beginning to address the appropriate use, including long-term use, of concomitant hypnotics for various psychiatric disorders. For example, recent studies have shown that hypnotics may enhance remission rates both for patients with major depression who have insomnia and for patients with generalized anxiety disorder (GAD) who have insomnia (Figure 11-17). Not only do the symptoms of insomnia improve as expected when patients with GAD or major depression are treated with eszopiclone added to an SSRI (e.g., fluoxetine or escitalopram), but so do the other symptoms of GAD or depression, leading to higher remission rates (Figure 11-17). Whether this applies to all Z drugs, or indeed to any hypnotic of any mechanism that is successful in improving insomnia added to any antidepressant for these conditions, is not yet known. Whether treating insomnia will also help prevent future episodes of depression or GAD is also not known, but considering that insomnia is perhaps the most frequent residual symptom after treating depression with an antidepressant (discussed in Chapter 7 and illustrated in Figure 7-5), it makes intuitive sense to utilize hypnotics as augmenting agents to first-line treatments for depression or anxiety disorders, and if necessary to utilize hypnotics chronically to eliminate symptoms of insomnia in these conditions.

Figure 11-17. Treating psychiatric insomnia. Insomnia is a common residual symptom of psychiatric disorders, including depression and generalized anxiety disorder (GAD). Recent findings suggest that remission rates may be increased in depression or GAD with insomnia when a hypnotic is added to first-line antidepressant treatment, and that this is attributable not only to improvement in insomnia but also to improvement in other symptoms.

Melatonergic hypnotics

Melatonin is the neurotransmitter secreted by the pineal gland, and acts especially in the suprachiasmatic nucleus (SCN) to regulate circadian rhythms (discussed in Chapter 7 and illustrated in Figures 7-42Athrough 7-42D). Figures 7-42A and 7-42B show the effects of light and dark, respectively, on melatonin. The melatonergic antidepressant agomelatine (Figures 7-417-42D11-18) shifts circadian rhythms in depressed subjects with phase delay (Figure 7-42C). Melatonin itself, as well as selective melatonin receptor agonists such as ramelteon or tasimelteon (Figure 11-18), has similar actions on shifting circadian rhythms in individuals without depression but who have phase delay (many normal teenagers) or phase advance (many normal elderly people), or in those experiencing jet lag from travel-induced shifts in circadian rhythms. It is also known that melatonin and selective melatonin receptor agonists (Figure 11-18) are effective hypnotics for sleep onset. Melatonin is available over the counter in the US, in doses that are not always reliable. Commercially available melatonin in a controlled-release formulation is available outside the US.

Figure 11-18. Melatonergic agents. Endogenous melatonin is secreted by the pineal gland and mainly acts in the suprachiasmatic nucleus to regulate circadian rhythms. There are three types of receptors for melatonin: melatonin 1 and 2 (MT1 and MT2), which are both involved in sleep, and melatonin 3, which is actually the enzyme NRH:quinine oxidoreductase 2 and not thought to be involved in sleep physiology. There are several different agents that act at melatonin receptors, as shown here. Melatonin itself, available over the counter, acts at MT1 and MT2 receptors as well as at the melatonin 3 site. Both ramelteon and tasimelteon are MT1 and MT2 receptor agonists and seem to provide sleep onset though not necessarily sleep maintenance. Agomelatine is not only an MT1 and MT2 receptor agonist, but is also a 5HT2C and 5HT2B receptor antagonist and is available as an antidepressant in Europe.

Melatonin acts at three different sites, not only melatonin 1 (MT1) and melatonin 2 (MT2) receptors, but also at a third site, sometimes called the melatonin 3 site, which is now known to be the enzyme NRH:quinone oxidoreductase 2, and which is probably not involved in sleep physiology (Figure 11-18). MT1-mediated inhibition of neurons in the SCN could help to promote sleep by decreasing the wake-promoting actions of the circadian “clock” or “pacemaker” that functions there, perhaps by attenuating the SCN’s alerting signals, allowing sleep signals to predominate and thus inducing sleep. Phase shifting and circadian rhythm effects of the normal sleep/wake cycle are thought to be primarily mediated by MT2 receptors which entrain these signals in the SCN.

Ramelteon is an MT1/MT2 agonist marketed for insomnia, and tasimelteon, another MT1/MT2 agonist, is in clinical testing (Figure 11-18). These agents improve sleep onset, sometimes better when used for several days in a row. They are not known to help sleep maintenance, but will induce natural sleep in those subjects who suffer mostly from initial insomnia.

Serotonergic hypnotics

One of the most popular hypnotics among psychopharmacologists is the antidepressant trazodone. This sedating antidepressant with a half-life of only about 6–8 hours was recognized long ago by clinicians as being highly effective as a hypnotic when given at a lower dose than that used as an antidepressant, and by giving it just once a day at night (see discussion in Chapter 7 and Figures 7-47 through 7-50). In fact, although trazodone was never officially approved as a hypnotic, nor marketed as a hypnotic, it nevertheless accounts for up to half of all prescriptions for hypnotics.

How does trazodone work? In Chapter 7, trazodone’s mechanism as an antidepressant is discussed and illustrated (Figures 7-47 through 7-50). It is clear that to act as an antidepressant, the dose of trazodone must be sufficiently high to recruit not only its most potent 5HT2A antagonist properties, but also its serotonin reuptake blocking properties (Figures 7-48 and 7-49). At these doses, trazodone can be quite sedating, because its H1antihistamine and α1 antagonist properties are also recruited. Contributions of H1 antagonism and α1-adrenergic antagonism to sedation are discussed in Chapter 5 and illustrated in Figure 5-38.

By trial and error, if not by serendipity, clinicians discovered that trazodone’s half-life is actually an advantage when this drug is administered as a hypnotic (Figure 7-50), because its daytime sedating effects, which are so evident when administering high doses twice daily for depression, can be greatly diminished by giving this short-half-life agent only at night and by lowering its dose (Figure 7-48). However, in doing so, trazodone loses its serotonin reuptake blocking properties, and thus its antidepressant actions, yet retains α1 blocking actions, as well as H1 antagonist and 5HT2A antagonist actions (Figure 7-48).

Histamine H1 antagonists as hypnotics

It is widely appreciated that antihistamines are sedating. Antihistamines are popular as over-the-counter sleep aids (especially those containing diphenhydramine/Benadryl or doxylamine) (Figure 11-19). Because antihistamines have been widely used for many years, there is the common misperception that the properties of classic agents such as diphenhydramine apply to any drug with antihistaminic properties. This includes the idea that all antihistamines have “anticholinergic” side effects such as blurred vision, constipation, memory problems, dry mouth; that they cause next-day hangover effects when used as hypnotics at night; that tolerance develops to their hypnotic actions; that they cause weight gain.

Figure 11-19. Diphenhydramine. Diphenhydramine is a histamine 1 receptor antagonist commonly used as a hypnotic. However, this agent is not selective for histamine 1 receptors and thus can also have additional effects. Specifically, diphenhydramine is also a muscarinic 1 receptor antagonist and thus can have anticholinergic effects (blurred vision, constipation, memory problems, dry mouth).

It now seems that these ideas about antihistamines arise from the fact that most agents with potent antihistamine properties, from diphenhydramine, to tricyclic antidepressants (discussed in Chapter 7 – see Figures 7-62through 7-70), mirtazapine (also discussed in Chapter 7 – see Figure 7-45), quetiapine (discussed in Chapter 5 as “baby bear” – see Figures 5-47 through 5-50) and many others, are not selective for H1 receptors at normal therapeutic doses, and that many of the undesirable properties classically associated with antihistamines are probably due to other receptor actions, not to H1 antagonism per se. In particular, diphenhydramine and many of the agents classified as antihistamines are also potent antagonists of muscarinic receptors (Figure 11-19), so it is not generally possible to separate the antihistamine actions of such agents from their antimuscarinic actions in clinical use. The same is true for most tricyclic antidepressants, which have antimuscarinic and α1-adrenergic blocking properties in addition to their antihistaminic properties (Figures 7-627-687-69).

Some interesting findings are beginning to emerge from clinical investigations of H1 selective antagonists as hypnotics. The prototype of this approach is very low doses of the tricyclic antidepressant doxepin (Figure 11-20). Because of the very high affinity of doxepin for the H1 receptor, it is possible to make it into an H1 selective antagonist just by lowering the dose (Figure 11-20). This agent is so selective at low doses that it is even being used as a PET ligand to label CNS H1 receptors selectively. At doses a small fraction of those necessary for its antidepressant actions, doxepin can occupy a substantial number of CNS H1 receptors (e.g., at 1–6 mg of doxepin as a hypnotic compared to 150–300 mg of doxepin as an antidepressant) (Figure 11-20). Furthermore, doxepin is actually a mixture of two chemical forms, one of which (and its active metabolites) has a shorter half-life (8–15 hours) than the other, which has a traditional long tricyclic antidepressant half-life of 24 hours. Functionally, the mixture of the two agents means that nighttime administration yields substantially less residual plasma drug levels in the morning compared to tricyclics with a 24-hour half-life, thus reducing daytime carryover effects.

Figure 11-20. Doxepin. Doxepin is a tricyclic antidepressant (TCA) that, at antidepressant doses (150–300 mg/day), inhibits serotonin and norepinephrine reuptake and is an antagonist at histamine 1, muscarinic 1, and α1-adrenergic receptors. At low doses (1–6 mg/day), however, doxepin is quite selective for histamine 1 receptors and thus may be used as a hypnotic.

Although it is not surprising that very low doses of doxepin that selectively antagonize H1 histamine receptors are effective hypnotics, early clinical testing is revealing that long-term administration of doxepin provides rapid sleep induction with all-night sleep maintenance but without next-day carryover effects, development of tolerance to its hypnotic efficacy, or weight gain. Eliminating α1-adrenergic and muscarinic cholinergic blockade may explain the lack of anticholinergic side effects, and the lack of development of tolerance to hypnotic actions. Although agents with H1 antagonist properties can cause weight gain, apparently H1 selective antagonism without 5HT2Cantagonism may not be associated with weight gain. These mechanisms are discussed in relation to weight gain in Chapter 5 (illustrated in Figures 5-36 and 5-41 through 5-44) and in Chapter 14 (illustrated in Figure 14-21).

Dopamine agonists and α2δ ligands for insomnia associated with restless legs syndrome (RLS)

A common cause of insomnia that is neither primary insomnia nor insomnia secondary to a psychiatric illness is insomnia secondary to restless legs syndrome (Table 11-1). Rather than using traditional sedative hypnotics for insomnia secondary to RLS, first-line treatment is with dopamine agonists such as ropinirole or pramipexole, and second-line treatment is with α2δ ligands such as gabapentin or pregabalin (Table 11-1).

Table 11-1 Restless legs syndrome (RLS) versus periodic limb movement disorder (PLMD)

Clinically diagnosed as urge to move the legs, worse during inactivity, relieved in part by movement, worse in the evening

Can prevent or delay onset of sleep; disrupt sleep if RLS returns; tiredness or sleepiness next day

May be idiopathic or symptomatic (i.e., associated with pregnancy, end-stage renal disease, fibromyalgia, iron deficiency, arthritis, peripheral neuropathy, radiculopathy)

Can be triggered by alcohol, nicotine, caffeine

Majority of RLS patients have PLMD, but only a few PLMD patients have RLS; both may be linked to dopamine or iron deficiency

RLS patients should have iron, iron stores and ferritin levels (ferritin is a cofactor of tyrosine hydroxylase, which synthesizes dopamine)

RLS is not PLMD, which occurs during sleep and is diagnosed by polysomnogram; has no urge sensation to move while awake

Treatments for RLS


·        Dopamine agonists (ropinirole, pramipexole) (may cause somnolence, nausea)

·        Iron replacement

·        Levodopa (fast onset but short acting – may just delay onset of RLS until later at night unless redosed)


·        Gabapentin/pregabalin especially if RLS painful; low-potency opioids (propoxyphene, codeine)

·        Benzodiazepines or GABAA PAMs

Behavioral treatments of insomnia

One should not forget improvement of sleep hygiene (Table 11-2), nor cognitive behavioral approaches (Table 11-3), as adjunctive treatments of insomnia of any cause, and as first-line treatments for primary insomnia, as these can be quite effective in selected patients with various types of insomnia.

Table 11-2 Good sleep hygiene

Avoid naps

Use the bed for sleeping, not reading, TV, etc.

Avoid alcohol, caffeine, and nicotine before sleep

Avoid strenuous exercise before sleep

Limit time in bed to sleep time (get up if not asleep within 20 minutes and return to bed when sleepy)

Don’t clock-watch

Adopt regular sleep/wake habits

Avoid bright light late at night, and expose self to light in the morning

Table 11-3 Behavioral treatments for insomnia

Stimulus control therapy (go into bedroom only when drowsy)

Progressive muscular relaxation (tense and relax muscles top to bottom of body)

Sleep restriction (consolidate sleep by progressive lengthening of sleep time)

Imagery training (try to stay awake as long as possible)

Biofeedback (learn to recognize arousal with feedback from skin, muscle and brain monitors)

Cognitive therapy (eliminate faulty beliefs and attitudes about sleep)

Sleep hygiene education (understand interaction between lifestyle, environment, and sleep)

Who cares about slow-wave sleep?

The exact function of stage 3 and 4 sleep (delta or slow-wave sleep) remains under active investigation. Not all patients with insomnia have a deficiency of slow-wave sleep, and not all patients with a deficiency of slow-wave sleep have insomnia. However, some empiric clinical observations suggest that a deficiency of slow-wave sleep can contribute to a sense of lack of restorative sleep, and daytime fatigue. Patients with pain syndromes and a deficiency of slow-wave sleep can experience enhanced daytime subjective experience of their pain; patients with depression and a deficiency of slow-wave sleep can have enhanced symptoms of fatigue, apathy, and cognitive dysfunction. Thus, sufficient restorative slow-wave sleep seems intuitively like a good thing to have, but the proof of how much is enough, and what the implications are of too little slow-wave sleep, remain elusive.

Some agents such as serotonergic antidepressants (SSRIs, SNRIs), stimulants, and stimulating antidepressants (e.g., NDRIs) can all interfere with slow-wave sleep, and a limited number of agents are known to enhance slow-wave sleep. These include α2δ ligands (e.g., gabapentin and pregabalin), the GABA reuptake inhibitor tiagabine, and 5HT2A/2C antagonists including trazodone and GHB (the GABAB-enhancing agent γ-hydroxybutyrate, also known as sodium oxybate). Augmenting the treatment of fatigue and pain with slow-wave sleep-enhancing agents can sometimes reduce these symptoms.

Orexin antagonists as novel hypnotics

Orexin neurons are localized exclusively in certain hypothalamic areas (lateral hypothalamic area, perifornical area, and posterior hypothalamus) (Figure 11-21). These orexin neurons make the neurotransmitters orexin A and orexin B, which are released from their neuronal projections all over the brain, but especially in the monoamine neurotransmitter centers in the brainstem (Figure 11-21). The postsynaptic actions of the orexins are mediated by two receptors called orexin 1 and orexin 2 (Figure 11-22). The neurotransmitter orexin A interacts with both orexin 1 and 2 receptors but orexin B interacts only with orexin 2 receptors (Figure 11-22). Notably, orexin 1 receptors are particularly highly expressed in the brainstem locus coeruleus, site of noradrenergic neurons; orexin 2 receptors are highly expressed in the TMN (tuberomammillary nucleus), site of histamine neurons. It is believed that the effect of orexin on wakefulness is largely mediated by activation of the TMN histaminergic neurons that express orexin 2 receptors. Presumably orexin 2 receptors therefore play a pivotal role, with orexin 1 receptors playing an additional role in sleep/wake regulation. Orexins mediate behaviors in addition to wakefulness and vigilance; they also regulate feeding behavior and reward, perhaps particularly through orexin 1 receptors.

Figure 11-21. Orexin projections from the hypothalamus. The neurotransmitter orexin (also called hypocretin) is made by cells located in the hypothalamus, specifically in the lateral hypothalamic area (LHA) and the perifornical and posterior hypothalamus (PH). Orexin A and orexin B produced by these cells are released at various brain areas, including monoamine neurotransmitter centers in the hypothalamic tuberomammillary nucleus (TMN; for histamine) and in the brainstem such as the ventral tegmental area (VTA; for dopamine), the locus coeruleus (LC; for norepinephrine), the pedunculopontine tegmental and laterodorsal tegmental nuclei (PPT/LDT; for acetylcholine), and raphe nucleus (for serotonin).

Figure 11-22. Orexin receptors. Orexin neurotransmission is mediated by two types of postsynaptic G-protein-coupled receptors, orexin 1 (Ox1R) and orexin 2 (Ox2R). Orexin A is capable of interacting with both Ox1R and Ox2R, whereas orexin B binds selectively to Ox2R. Binding of orexin A to Ox1R receptors leads to increased intracellular calcium as well as activation of the sodium/calcium exchanger. Binding of orexin A and B to Ox2R leads to increased expression of NMDA (N-methyl-D-aspartate) glutamate receptors as well as inactivation of G-protein-regulated inward rectifying potassium channels (GIRK). Ox1R are found particularly highly expressed in the noradrenergic locus coeruleus, whereas Ox2R are highly expressed in the histaminergic tuberomammillary nucleus (TMN).

Lack of orexins is associated with narcolepsy. Pharmacologic blockade of orexin receptors has been investigated not only as a novel hypnotic mechanism, but also for weight loss and to treat drug abuse. Specifically, both the dual orexin receptor antagonists (DORAs) for both orexin 1 and 2 receptors, and single orexin receptor antagonists (SORA1s and SORA2s), selective either for orexin 1 receptors or for orexin 2 receptors, have been developed (Figure 11-23) and are being extensively tested at this time. DORAs such as almorexant, SB-649868, and suvorexant (also known as MK-4305) have preliminary evidence of efficacy in the treatment of insomnia, and some are advancing in clinical trials, particularly suvorexant. Other DORAs include MK-6096, DORA 1, DORA 5, and DORA 22. Single-action SORAs are also in development with SORA1 agents (e.g., SB-334867, SB674042, SB408124, SB410220) proving not to be particularly robust in treating insomnia, but with promising preliminary preclinical results for SORA2 agents (e.g., EMPA, JNJ10394049).

Figure 11-23. Orexin receptor antagonists. Several orexin receptor antagonists are currently in testing as hypnotics. Single orexin receptor antagonists (SORAs) work selectively at orexin 1 receptors (SORA1) or at orexin 2 receptors (SORA2). Dual orexin antagonists (DORAs) that bind both orexin 1 and orexin 2 receptors are also being studied.

Notably, the localization of orexin 1 and 2 receptors, coupled with the lack of preclinical effects of some orexin 1 antagonists on sleep, suggests that the wake-promoting effects of orexins are mediated mainly by orexin 2 receptors or a combination of orexin 1 and 2 receptors (Figure 11-24). Thus, hypnotics are either DORAs targeting both receptors or SORA2s targeting orexin 2 receptors. SORA1s targeting orexin 1 receptors are in development as possible treatments to reduce craving for drugs or food (see discussion in Chapter 14).

Figure 11-24. Blockade of orexin receptors. Blockade of orexin receptors by SORAs and DORAs is hypothesized to lead to blockade of the excitatory effects of orexin neurotransmitters, thus promoting sleep.

To date, the DORA suvorexant appears to improve both the initiation and maintenance of sleep in human subjects, without the side effects expected of a benzodiazepine or Z-drug hypnotic, namely lacking dependence, withdrawal, rebound, unsteady gait, falls, confusion, amnesia, or respiratory depression (Figures 11-23 and 11-24). So far the theoretical possibility that DORAs could cause a reversible form of narcolepsy with hypnagogic hallucinations, sleep paralysis, and cataplexy has not been observed. It appears that acute, short-lived, and intermittent temporary blockade of orexin receptors is well tolerated without the induction of a narcolepsy-like syndrome.

Excessive daytime sleepiness (hypersomnia) and wake-promoting agents

What is sleepiness?

Sleepiness is a term that is sometimes used synonymously with hypersomnia. Here we will discuss the symptom of excessive daytime sleepiness, its causes, and especially its treatment with three wake-promoting agents: caffeine, modafinil, and stimulants. The most common cause of sleepiness is sleep deprivation, and the treatment is sleep, not drugs. Other causes of excessive daytime sleepiness are various sleep disorders, psychiatric disorders, medications, and medical disorders (Table 11-4). Although society often devalues sleep and can often imply that only wimps complain of sleepiness, it is clear that excessive daytime sleepiness is not benign, and in fact can even be lethal. That is, loss of sleep causes performance decrements equivalent to intoxication with alcohol and, not surprisingly, traffic accidents and fatalities. Thus, sleepiness is important to assess even though patients often do not complain about it when they have it. Comprehensive assessment of patients with sleepiness requires additional information be obtained from the patient’s partner, particularly the bed partner. Most conditions can be evaluated by patient and partner interviews, but sometimes augmented with subjective ratings of sleepiness such as the Epworth Sleepiness Scale, as well as objective evaluations of sleepiness such as overnight polysomnograms, plus next-day multiple sleep latency testing and/or maintenance of wakefulness testing (Table 11-5).

Table 11-4 What causes sleepiness?

Sleep deprivation

Sleep disorders


Obstructive sleep apnea (OSA)

Restless legs

Periodic limb movement disorder (PLMD)

Circadian rhythm disorders (shift work, jet lag, delayed sleep)

Primary hypersomnia

Poor sleep hygiene

Psychiatric illness

Psychiatric and other medications

Substance use/abuse

Medical disorders


Insulin resistance/diabetes

Table 11-5 How is sleepiness evaluated?

Subjective method

Epworth Sleepiness Scale

·        8 questions self rated on a 0–3 scale

Objective method

Multiple Sleep Latency Test (MSLT)

·        Nocturnal polysomnogram

·        Five daytime nap opportunities lying in a quiet, dark room at 2-hour intervals – told not to oppose sleep

·        Score time to sleep onset defined by EEG

o   max time 20 minutes

o   wake patient 15 minutes from sleep onset

Maintenance of Wakefulness Test (MWT)

·        Nocturnal polysomnogram

·        Five daytime nap opportunities lying in a quiet, dark room at 2-hour intervals – instructed to resist sleep

·        Often the morning after an overnight polysomnogram

What’s wrong with being sleepy?

Patients with excessive daytime sleepiness have problems with cognitive functioning. For example, when patients with narcolepsy or sleep deprivation try to perform cognitive testing, with great effort they can often activate their dorsolateral prefrontal cortex (DLPFC) normally, but cannot sustain it, yet when narcoplesy patients take a stimulant or modafinil they are able to sustain the activation of their DLPFC and also to sustain their cognitive performance without decrement. Presumably, this improvement is the result of optimizing and increasing the actions of dopamine in DLPFC brain circuits.

Mechanism of action of wake-promoting agents


This drug is a proven wake-promoting agent whose exact molecular mechanism of action remains debated. It is known to activate relatively selectively neurons in the wake-promoting TMN and the lateral hypothalamus, and this leads to the release of both histamine and orexin. However, the activation of the lateral hypothalamus and release of orexin do not appear to be necessary for the action of modafinil, since modafinil still promotes wakefulness in patients who have loss of hypothalamic orexin neurons in narcolepsy. The activation of TMN and lateral hypothalamic neurons may be secondary and downstream actions resulting from modafinil’s effects on dopamine neurons.

The most likely modafinil binding site is probably the dopamine transporter (DAT or DA reuptake pump) (Figure 11-25). Although modafinil is a weak DAT inhibitor, the concentrations of the drug achieved after oral dosing are quite high, and sufficient to have a substantial action on DAT. In fact, the pharmacokinetics of modafinil suggest that this drug acts via a slow rise in plasma levels, sustained plasma levels for 6–8 hours, and incomplete occupancy of DAT, all properties that could be ideal for enhancing tonic dopamine activity to promote wakefulness (Figure 11-25) rather than phasic dopamine activity to promote reinforcement and abuse (see discussion in Chapter 14 on substance abuse). Once dopamine release is activated by modafinil, and the cortex is aroused, this can apparently lead to downstream release of histamine from the TMN and then further activation of the lateral hypothalamus with orexin release to stabilize wakefulness. The same appears to occur after administration of the stimulants amphetamine and methylphenidate.

Figure 11-25. Modafinil. The precise mechanism of action of modafinil is yet to be fully elucidated. It is known to bind to the dopamine transporter (DAT) and in fact requires its presence. Modafinil’s low affinity for the DAT has led some to question whether its binding there is relevant; however, because plasma levels of modafinil are high, this “compensates” for the low binding affinity. It is believed that the increase in synaptic dopamine following blockade of DAT leads to increased tonic firing and downstream effects on neurotransmitters including those involved in wakefulness, such as histamine and orexin/hypocretin.

A newer wake-promoting agent is the R enantiomer of modafinil, called armodafinil (Nuvigil). Armodafinil has a later time to peak levels, a longer half-life, and higher plasma drug levels 6–14 hours after oral administration than the marketed form of modafinil, which is a racemic mixture of R plus S modafinil. The pharmacokinetic properties of armodafinil could theoretically improve the clinical profile of modafinil, with greater activation of phasic dopamine firing, possibly eliminating the need for a second daily dose, as is often required with racemic modafinil.


The two principal stimulants used as wake-promoting agents are methylphenidate and amphetamine, especially d-amphetamine. Many forms of these stimulants are now available and are reviewed in detail in Chapter 12 on attention deficit hyperactivity disorder (ADHD) and in Chapter 14 on substance abuse. Amphetamine is known to be a competitive inhibitor and substrate for the dopamine transporter (DAT) and also a dopamine releaser and inhibitor of the vesicular monoamine transporter (VMAT2) within presynaptic dopamine nerve terminals. Methylphenidate is also known to be an inhibitor of DAT which acts not unlike the NDRI (norepinephrine–dopamine reuptake inhibitor) antidepressants discussed in Chapter 7. The mechanism of methylphenidate is also discussed in more detail in Chapter 12. At the doses used to treat sleepiness and ADHD, both of which are much lower than doses used by stimulant addicts, the agents amphetamine and methylphenidate also block the norepinephrine transporter (NET), especially in controlled-release formulations. Basically, the stimulants as dosed to treat sleepiness or ADHD enhance the synaptic availability of dopamine and norepinephrine, and thereby improve wakefulness.


Caffeine is an incredible over-the-counter drug, popular in many beverages, but how does it work? Originally thought to work as an inhibitor of the enzyme phosphodiesterase, it now appears to act mostly as an antagonist of endogenous neurotransmitters called purines, of which an important one is adenosine, at purine receptors (Figure 11-26). Certain purine receptors are functionally coupled with dopamine receptors such that the actions of dopamine at D2 receptors (Figure 11-26A) are antagonized when adenosine is binding to its receptor (Figure 11-26B). Not surprisingly, therefore, when an antagonist of adenosine such as caffeine is present, this indirectly promotes the actions of dopamine (Figure 11-26C).

Figure 11-26. Caffeine. Caffeine is an antagonist at purine receptors, and in particular adenosine receptors. (A) These receptors are functionally coupled with certain postsynaptic dopamine receptors, such as dopamine 2 (D2) receptors, at which dopamine binds and has a stimulatory effect. (B) When adenosine binds to its receptors, this causes reduced sensitivity of D2 receptors. (C) Antagonism of adenosine receptors by caffeine prevents adenosine from binding there, and thus can enhance dopaminergic actions.


Gamma-hydroxybutyrate or GHB is also known as sodium oxybate and as Xyrem. This agent is approved for the treatment of excessive daytime sleepiness associated with narcolepsy, as well as for cataplexy. It appears to promote wakefulness by its profound actions on slow-wave sleep at night, making the patient more rested and therefore more alert the next day. Because of its abuse potential and colorful history, it is scheduled as a controlled substance and its supplies are tightly regulated through a central pharmacy in the US. It has been labeled a “date rape” drug by the press, as it has occasionally been used with alcohol for this purpose. Because it profoundly increases slow-wave sleep and the growth hormone surge that accompanies slow-wave sleep, it was used (abused) by athletes as a performance-enhancing drug especially in the 1980s when it was sold over the counter in health-food stores. GHB is used in some European countries as a treatment for alcoholism. Because of the observed enhancement of slow-wave sleep, GHB was recently developed for the treatment of narcolepsy and cataplexy. It is sometimes used “off label” to treat refractory cases of fibromyalgia (see Chapter 10 for discussion of pain syndromes such as fibromyalgia and their treatment).

GHB is actually a natural product present in the brain, with its own GHB receptors upon which it acts (Figure 11-27). GHB is formed from the neurotransmitter GABA, and also acts at GABAB receptors as a partial agonist (Figure 11-27).

Figure 11-27. Sodium oxybate. Gamma-hydroxybutyrate (GHB, also called sodium oxybate), is formed from the neurotransmitter GABA and acts as a partial agonist at GABAB receptors. It is approved for use both in cataplexy and for excessive sleepiness, and appears to enhance slow-wave sleep.


The neurobiology of wakefulness is linked to an arousal system that utilizes the five neurotransmitters histamine, dopamine, norepinephrine, acetylcholine, and serotonin as components of the ascending reticular activating system. Sleep and wakefulness are also regulated by a hypothalamic sleep/wake switch, with wake-promoter neurons in the tuberomammillary nucleus that utilize histamine as neurotransmitter, and sleep-promoter neurons in the ventrolateral preoptic nucleus that utilize GABA as neurotransmitter, both stabilized by the peptide neurotransmitters orexin A and B. The synthesis, metabolism, receptors, and pathways for the neurotransmitter histamine are reviewed in this chapter, as are the pathways for the orexin neurons and the distribution of their receptors. Insomnia is also briefly reviewed, as are the mechanisms of action of several hypnotics, from the benzodiazepines to the popular “Z drugs” that act as positive allosteric modulators or PAMs for GABAA receptors. Other hypnotics include trazodone, melatonergic hypnotics, and antihistamines, including novel dual orexin receptor antagonists (DORAs) currently in clinical testing.

Excessive daytime sleepiness is also briefly reviewed, as are the mechanisms of action of the wake-promoting drugs modafinil, caffeine, and stimulants. The actions of GHB (γ-hydroxybutyrate) plus a number of novel sleep- and wake-promoting drugs in clinical development are also reviewed.