Nonbenzodiazepine Hypnotics

By | March 23, 2015

Non-benzodiazepine hypnotics were introduced in most markets in the early 1990s. To date, three non-benzodiazepine hypnotics have been launched for insomniazolpidem, zopiclone, and zaleplon (Wyeth and King Pharmaceuticals’ Sonata). Characteristics that distinguish these drugs from traditional benzodiazepines (discussed later) are their increased receptor-binding specificity, favorable pharmacokinetics, and overall broader range of safety. More specifically, non-benzodiazepine hypnotics have minimal residual effects on daytime performance (e.g., cognitive functioning, memory, alertness), low abuse potential, very few clinically significant drug interactions, and little or no effect on sleep architecture.

However, like most traditional benzodiazepine hypnotics, non-benzodiazepine hypnotics are indicated only for the short-term treatment of insomnia; to date, there are few available data from long-term, well-controlled trials with which to determine these agents’ safety and efficacy when used chronically (i.e., longer than three months). Like the benzodiazepine hypnotics, non-benzodiazepine hypnotics are classified as controlled substances (i.e., “Scheduled” drugs), meaning that they have the potential to lead to dependence and abuse; this status limits the frequency and duration of prescriptions for these agents.

Mechanism of Action

Unlike traditional benzodiazepines, which act nonselectively at two central receptor binding sites — benzodiazepine type-1 (or omega 1) and benzodiazepine type-2 (or omega 2) — located on the gamma-aminobutyric acid (gamma aminobutyric acid)-A receptor complex, non-benzodiazepine hypnotics interact preferentially with benzodiazepine-1 receptors, which researchers believe are responsible for these agents’ hypnosedative effects. Researchers hypothesize that activity at the benzodiazepine-2 receptors may be responsible for traditional benzodiazepines’ negative effects on psychomotor performance and memory, as well as their habit-forming effects.


Zolpidem (Sanofi-Synfhelabo’s Ambien / Stilnox, Fujisawa’s Myslee, generics) has been marketed for the short-term treatment of insomnia in Europe since 1988 and in the United States since 1993. (Note: Sanofi-Synthelabo acquired sole rights to the compound from Pharmacia [formerly Searle] in 2002.) In Japan, the drug was launched by its Japanese licensee Fujisawa (as Myslee) in late 2000. Zolpidem is by far the leading prescription sleep aid on the market today.

Zolpidem is an imidazopyridine hypnotic that has a rapid onset of action and a short elimination half-life (approximately 2.5 hours). The drug elicits its hypnotic effects via selective binding to the benzodiazepine-1 subtype of the benzodiazepine receptor located on the alpha subunit of the gamma aminobutyric acid-A receptor complex. Unlike traditional benzodiazepines, zolpidem has little activity at the benzodiazepine-2 subtype of the benzodiazepine receptor; researchers suspect that this lack of activity may explain zolpidem’s apparent lack of myorelaxant and anticonvulsant effects commonly associated with the less selective benzodiazepines. Also unlike the traditional benzodiazepines, zolpidem does not appear to disturb normal sleep architecture.

More than 15 years of postmarketing experience support zolpidem’s efficacy and safety as a hypnotic agent. The drug has been shown to effectively decrease sleep latency and increase total sleep time in patients with transient, short-term, and chronic insomnia with a low occurrence of adverse events and minimal effects on memory and next-day functioning. The most common adverse events associated with zolpidem are nausea, dizziness, malaise, nightmares, agitation, and headache (occurring in less than 10% of patients). The drug has also been shown to have a lower propensity than traditional benzodiazepines to induce tolerance, dependence, and rebound insomnia (i.e., worsening of sleep below baseline levels).

In an early trial conducted in chronic insomniacs (n = 75), subjects were given 10 mg or 15mg zolpidem or placebo in a randomized, double-blind manner for a period of 35 nights followed by a 3-night period during which they received single-blind placebo. The specific objectives of the study were to (1) evaluate the efficacy and safety of the two doses of zolpidem versus placebo, (2) determine whether patients develop tolerance to the hypnotic effect of zolpidem when given the drug for 35 consecutive nights, (3) determine whether patients have psychomotor performance deficits during the 35-night treatment period, and (4) evaluate sleep following abrupt discontinuation of zolpidem after 35 consecutive nights of treatment. The primary outcome measures were latency to persistent sleep (as measured by polysomnographic recordings) and sleep efficiency (total sleep time divided by time in bed, multiplied by 100).

Final analysis of the data revealed that sleep latency was significantly shorter than in the placebo group at weeks 2 through 5 for the 10 mg zolpidem group and at weeks 2 through 6 for the 15mg group. Overall, mean latency to persistent sleep for the two zolpidem groups was 15 to 30 minutes shorter than for the placebo group during the five weeks of active treatment. Similar results were seen for sleep efficiency — that is, sleep efficiency was significantly higher in the 10 mg and 15mg zolpidem groups compared with placebo at weeks 2 through 5 and 2 through 6, respectively. Mean percent sleep efficiency for the 10 mg zolpidem group was typically 6-8% higher than for the placebo group, representing an increase of 25-40 minutes. In this study (as in later studies), the 10 mg zolpidem dose seemed to be ideal because the 15mg dose was associated with a higher incidence of undesirable effects (i.e., shortened rapid eye movement (rapid eye movement) sleep at weeks 3 and 4, and worsened sleep quality on the first night post-treatment).

Further results that favored zolpidem 10 mg showed no evidence of impaired cognitive or motor functioning based on Digit Symbol Substitution Test (DSST) and Digit Symbol Copying Test (DSCT) scores (which are standard tools used to assess residual effects of a drug), decreased ability to concentrate, or feelings of morning sleepiness. Zolpidem did not change the amount of time spent in slow-wave (stages 3 and 4) sleep or rapid eye movement sleep compared with placebo, and no evidence of rebound worsening with regard to latency to persistent sleep or other sleep parameters during any of the nights of the withdrawal period compared with baseline was observed.

A more recent study with zolpidem in chronic insomniacs investigated the drug’s efficacy and tolerability when used on a nonnightly basis for a period of eight weeks. In this double-blind, placebo-controlled study (n = 163), patients were randomized to zolpidem 10 mg or placebo and instructed to take medication when they needed it, but at least three and no more than five times per week. The primary outcome variables were the Investigator and Patient Global Ratings.

At study completion, the mean Patient Global Ratings (including four individual items) were significantly better for zolpidem than for placebo at each of the four study periods (weeks 1 and 2, weeks 3 and 4, weeks 5 and 6, weeks 7 and 8). During each study period, approximately twice as many zolpidem-treated patients reported improved sleep as those taking placebo. Similarly, mean Investigator Global Rating scores for the zolpidem-treated group were significantly better at each time point than those for the placebo-treated group. On pill nights, total sleep time was significantly higher for the zolpidem-treated group than for the placebo-treated group (mean total sleep time 414 minutes versus 362 minutes, respectively) and sleep latency was significantly shorter for the zolpidem-treated group than for the placebo-treated group (mean sleep latency 37 minutes versus 52 minutes, respectively) during each study period. In addition, mean rebound insomnia measures for nights when patients did not use zolpidem after using it the night before, during all study periods, were comparable to mean baseline values (although in week 1, sleep latency was 14 minutes longer than the mean baseline sleep latency), suggesting no discontinuation effects in patients taking zolpidem. Additional 12-week data presented at the 2004 meeting of the Associated Professional Sleep Societies (APSS) from a trial with a like design (n = 199) revealed similar results.

The two previously discussed placebo-controlled trials as well as several other trials conducted over the past decade (citations provided previously) support zolpidem’s safety and efficacy in individuals who suffer from insomnia.


Zopiclone (Aventis’s Imovane / Amoban, Chugai’s Amban, generics) was first introduced in Denmark in 1987, making it the first of the three newer, non-benzodiazepine hypnotics to reach the market. The drug is now available in most major markets (with the exception of the United States) for the short-term treatment of insomnia.

Zopiclone is a cyclopyrrolone derivative with a pharmacological profile similar to that of the benzodiazepines, but with a novel chemical structure. It has a rapid onset of action and an elimination half-life of approximately five to seven hours (longer in the elderly and in those with hepatic insufficiency). Unlike zolpidem, which binds preferentially to the alpha 1 subunit of the gamma aminobutyric acid-A receptor, zopiclone, like traditional benzodiazepines, is aspecific in its binding. Nevertheless, zopiclone has exhibited a low propensity to induce dependence and / or abuse or withdrawal effects compared with the traditional benzodiazepines, possibly because it produces a more limited change in gamma aminobutyric acid-A receptor subunit expression.

As is the case with zolpidem, more than 15 years of postmarketing experience support zopiclone’s efficacy and safety as a hypnotic agent. The drug has been shown to effectively decrease sleep latency and increase total sleep time in patients with insomnia with a low occurrence of adverse events and minimal effects on memory and next-day functioning. The most notable AE associated with zopiclone is the drug’s bitter, metallic aftertaste reported by some patients; this particular AE contributes to zopiclone’s less popular status compared to zolpidem.

In clinical comparisons with traditional benzodiazepines, zopiclone has proven at least as effective as the short- and intermediate-acting benzodiazepines triazolam, midazolam, and temazepam and the long-acting benzodiazepines flurazepam, nitrazapem, and flunitrazepam, with an overall reduced occurrence of residual next-day effects such as motor impairment (e.g., morning coordination), daytime sedation, and restlessness (especially compared to the long-acting benzodiazepines). However, some data suggest that, in elderly insomniacs, zopiclone is slightly less effective than the long-acting benzodiazepines flunitrazepam and nitrazepam in improving sleep parameters. (Importantly, flunitrazepam’s efficacy is offset by its poor tolerability in elderly patients.)

In one of the largest published double-blind, placebo-controlled studies with zopiclone (n = 1,507), individuals with insomnia (86% of whom suffered for >4 weeks) were administered zopiclone (7.5 mg), flunitrazepam (1 mg), triazolam (0.25 mg), or placebo for a period of four weeks; a 3-day washout period preceded treatment and a 14-day washout period during which patients were observed followed treatment. The primary variables that were assessed were quality of sleep and daytime well-being, which were recorded daily by the patients using the Visual Analogue Scale (VIS-A for evening assessment and VIS-M for morning assessment). Physicians also used the Clinical Global Impressions (CGI) Scale for weekly evaluations of treatment efficacy. Treatment “responders” were defined as those who had (1) improvement in at least three sleep quality parameters given in the VIS-M (either shortening of sleep latency by >15 minutes or prolongation of total sleep time by >20%, or reduction of the number of nocturnal awakenings to three or less), and (2) no impairment in daytime well-being indicated in the VIS-A (a feeling of freshness on waking up in the morning as well as no impairment of daytime well-being as a result of tiredness or anxiety).

At study completion, 1,315 (87%) of the 1,507 patients who were initially randomized to treatment remained in the study. Total response (quality of sleep + daytime well-being) was significantly higher in the zopiclone group compared with the placebo group during the whole four-week treatment period. With the exception of triazolam, all treatments tended to produce a greater response in patients who had been suffering from insomnia for less than one year than in those with insomnia of one year or longer. In addition, severe insomniacs tended to respond better to treatment than those with slight insomnia, especially in the zopiclone group. Interestingly, the individual treatments’ effects on daytime well-being had more to do with any ultimate differences in total response scores among treatment groups than did their effects on quality of sleep. More specifically, zopiclone provided a “clear and constant improvement” over placebo on daytime well-being scores, whereas improvements with triazolam and flunitrazepam were much less notable; in contrast, all three treatments produced comparable improvements over placebo with regard to overall sleep quality (sleep latency, total sleep time, waking periods). These findings emphasize the importance of looking at factors other than just those relating to sleep quality when determining a drug’s effectiveness in insomniacs.

Aside from a bitter taste reported by some patients taking zopiclone, adverse events were not significantly different in the treatment groups and the placebo group in the previously discussed study. In addition, no rebound insomnia was observed in any of the treatment groups. (The authors note that rebound insomnia has been reported for all three drugs elsewhere.)


Zaleplon (Wieth and King Pharmaceuticals’ Sonata) was launched for the short-term treatment of insomnia in the United States in late 1999 and in several European markets by the end of 2000. (Note: King Pharmaceuticals acquired U.S. rights to zaleplon from Elan in June 2003.) In February 2001, U.S. regulatory approval was granted for an extension of zaleplon’s label to include five-week use. In Japan, zaleplon has been in Phase III trials for several years.

Zaleplon is an ultra-short-acting pyrazolopyrimidine hypnotic that binds selectively to the benzodiazepine-1 site on the gamma aminobutyric acid-A receptor complex. Because of its short elimination half-life (1 — 1.5 hours), zaleplon can be used with more flexibility than most other sedative hypnotics. More specifically, individuals suffering from insomnia can take zaleplon either at bedtime or in the middle of the night (i.e., as little as 2-3 hours before waking) for nocturnal awakenings to restore sleep.

In clinical studies, zaleplon has been shown to shorten sleep onset and to cause minimal or no next-day impairment of psychomotor and memory performance in patients with insomnia, even when administered in the middle of the night. This benefit is in contrast to other short-acting hypnotics such as zolpidem, zopiclone, and triazolam, which can cause detrimental residual effects on psychomotor and cognitive functions anywhere from five to ten hours after dose administration. That said, these agents are also more effective than zaleplon at increasing total sleep time and improving overall sleep quality. However, zaleplon has the advantage of causing no rebound insomnia or withdrawal symptoms after discontinuation following short-term (i.e., up to four weeks) use; even the short-acting non-benzodiazepine hypnotics zolpidem and zopiclone have been shown to cause rebound insomnia in some patients. The most common side effects associated with zaleplon are headache, dizziness, nausea, and daytime drowsiness (incidence is similar to placebo).

In one of the largest published randomized, double-blind, placebo-controlled studies of zaleplon, 615 adult patients with Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Re vised (DSM-III-R)-defined insomnia received 7 nights of placebo followed by 28 nights of treatment with one of five possible treatments (zaleplon 5mg, 10 mg, or 20 mg; zolpidem 10 mg; or placebo) followed by 3 nights of placebo washout. (Note: The manufacturer’s recommended dose of zaleplon is 10 mg in nonelderly adults and 5mg in elderly individuals.) The primary efficacy variable was the patient’s assessment of sleep latency. Secondary efficacy variables included the patient’s assessment of sleep duration, number of awakenings, and sleep quality. Sleep variables obtained from daily postsleep questionnaires were averaged for each week of the double-blind treatment period. In addition, rebound insomnia and withdrawal effects were assessed from sleep variable data derived from postsleep questionnaires completed after each night of the placebo run-out phase.

At study completion, investigators noted a significant dose-response trend with increasing doses of zaleplon for all four weeks of treatment. Median sleep latency was significantly reduced with zaleplon 5mg (p < 0.02), 10 mg (p < 0.001) and 20 mg (p < 0.001), as well as with zolpidem 10 mg (p < 0.05), compared with placebo during week 1; this significant decrease in sleep latency persisted through week 4 with zaleplon 10 and 20 mg, and through week 3 with zaleplon 5mg and zolpidem 10 mg. Sleep maintenance (which was based on patients’ subjective assessments of sleep duration and number of awakenings) was not significantly different between the zaleplon groups and placebo (except for an improvement with zaleplon 20 mg in week 1), and no evidence of rebound insomnia or withdrawal symptoms after discontinuation of zaleplon was observed. In contrast, zolpidem 10 mg increased sleep duration and improved sleep quality at most time points compared with placebo, but the drug also caused a higher incidence of withdrawal symptoms and rebound insomnia on the first night after treatment discontinuation when compared with placebo. No significant differences in the frequency of treatment-emergent adverse events between any of the active treatment groups and the placebo group were observed. (The most common treatment-emergent AE was headache).

Comparable safety and efficacy findings were reported from another large trial (n = 595) with a similar design by a separate group of investigators. These trials, as well as several other short-term trials in nonelderly and elderly individuals with chronic insomnia, support zaleplon’s safety and efficacy in treating insomniacs who have difficulty falling asleep.