Effects of new hypnotic drugs on cognition: A systematic review and network meta-analysis

Background Insomnia is a common disease, and the application of various types of sleeping pills for cognitive impairment is controversial, especially as different doses can lead to different effects. Therefore, it is necessary to evaluate the cognitive impairment caused by different sleeping pills to provide a theoretical basis for guiding clinicians in the selection of medication regimens. Objective To evaluate whether various different doses (low, medium and high) of anti-insomnia drugs, such as the dual-orexin receptor antagonist (DORA), zopiclone, eszopiclone and zolpidem, induce cognitive impairment. Methods The PubMed, Embase, Scopus, Cochrane Library, and Google Scholar databases were searched from inception to September 20th, 2022 for keywords in randomized controlled trials (RCTs) to evaluate the therapeutic effects of DORA, eszopiclone, zopiclone and zolpidem on sleep and cognitive function. The primary outcomes were indicators related to cognitive characteristics, including scores on the Digit Symbol Substitution Test (DSST) and daytime alertness. The secondary outcomes were the indicators associated with sleep and adverse events. Continuous variables were expressed as the standard mean difference (SMD). Data were obtained through GetData 2.26 and analyzed by Stata v.15.0. Results A total of 8702 subjects were included in 29 studies. Eszopiclonehigh significantly increased the daytime alertness score (SMD = 3.00, 95 % CI: 1.86 to 4.13) compared with the placebo, and eszopiclonehigh significantly increased the daytime alertness score (SMD = 4.21, 95 % CI: 1.65 to 6.77; SMD = 3.95, 95 % CI: 1.38 to 6.51; SMD = 3.26, 95 % CI: 0.38 to 6.15; and SMD = 3.23, 95 % CI: 0.34 to 6.11) compared with zolpidemlow, zolpidemhigh, DORAlow, and eszopiclonemid, respectively. Compared with the placebo, zopiclone, zolpidemmid, and eszopiclonehigh, DORA significantly increased the TST (SMD = 2.39, 95 % CI: 1.11 to 3.67; SMD = 6.00, 95 % CI: 2.73 to 9.27; SMD = 1.89, 95 % CI: 0.90 to 2.88; and SMD = 1.70, 95 % CI: 0.42 to 2.99, respectively). Conclusion We recommend DORA as the best intervention for insomnia because it was highly effective in inducing and maintaining sleep without impairing cognition. Although zolpidem had a more pronounced effect on sleep maintenance, this drug is better for short-term use. Eszopiclone and zopiclone improved sleep, but their cognitive effects have yet to be verified.


Introduction
Since the outbreak of coronavirus disease 2019 , more than 70 million people have developed depression, 90 million have developed anxiety, and hundreds of millions of people have developed insomnia disorders [1,2].A recent report by the World Health Organization showed that the global incidence rates of anxiety and depression significantly increased (by 25 %) in the first year of the pandemic [3].Large-scale epidemiological surveys in the USA and the UK also reported proportions of insomnia patients with depressive symptoms of 23 % and 21 %, respectively [4].Patients with severe anxiety disorders are more prone to depression and even suicidal behavior.Thus, improving sleep and reducing excessive arousal is urgent.
The first generation of hypnotic drugs were barbiturates, which act on the gamma-aminobutyric acid (GABA) system [5].GABA is an important inhibitory neurotransmitter in the central nervous system that mediates approximately 40 % of inhibitory nerve conduction and plays an important role in control of fear, anxiety, and convulsion-related neuronal hyperactivity [6].In the 1960s and 1970s, the second generation of sedative and hypnotic drugs-benzodiazepines-was developed.Compared with barbiturates, benzodiazepines had a wider safety range, high safety, and few adverse reactions.However, long-term use produced considerable adverse reactions, leading to drug dependence in nearly 30 % of patients [7,8].Third-generation nonbenzodiazepine sedatives and hypnotics selectively act on benzodiazepine receptors, exhibiting similar pharmacological characteristics to benzodiazepines [9].These drugs are characterized by fast sleep onset, increases in deep sleep and prolongation of sleep but do not affect the normal sleep architecture of healthy people and even improve the sleep architecture of patients [10].
At the end of the 20th century, researchers began to develop new sedative hypnotic drugs to reduce adverse reactions in patients.Zopiclone is a third-generation cyclopyrrolone hypnotic agent.This drug is a quick-acting sedative hypnotic agent mainly suitable for use in patients with difficulty sleeping and can shorten sleep latency because of its low likelihood of inducing dependence [11].Zopiclone is widely used to inhibit GABA receptors (i.e., a GABA receptor antagonist) and benzodiazepine drugs.Zopiclone is the r-isomer; in December 2004, eszopiclone (the s-isomer of this drug) was approved by the FDA for the treatment of insomnia.Eszopiclone is approximately 50 times stronger than zopiclone and has a significantly longer half-life.Compared to zopiclone, eszopiclone is better tolerated, less likely to lead to drug dependence, has greater efficacy, and induces fewer side effects [12].
In 1988, zolpidem was marketed in China as an effective drug for the short-term treatment of insomnia.Zolpidem is an imidazopyridine compound that selectively binds to the benzodiazepine ω1 receptor, which exerts inhibitory effects by partially activating GABA receptors (a ligand-gated ion channel) [13]; it exerts sedative and hypnotic effects but can easily lead to the development of drug tolerance and dependence.Due to its short peak latency and short half-life, it induces rapid sleep in insomnia patients and also reduces wakefulness after sleep onset (WASO), which is useful for the short-term treatment of insomnia [14].
However, individuals taking eszopiclone, zopiclone or zolpidem experience psychiatric distress, hangover-like symptoms, or even respiratory depression the next day in addition to the risks of developing drug tolerance and dependence [15].Effective sleep aids should maximize perceived sleep quality and avoid drug-related adverse effects without changing the underlying sleep architecture [16].Orexin is a neuropeptide produced by the ventrolateral hypothalamus (LH).Its isoforms (orexin A and orexin B) regulate the sleep-wake cycle [17].Thus, orexin antagonists have been used to treat insomnia; suppressing excessive wakefulness and inducing and maintaining sleep by inhibiting the orexin signaling system [18,19].The dual-receptor orexin antagonist (DORA) is a novel hypnotic drug; although animal experiments indicate slight cognitive facilitation of taking low doses of DORA, there have not been clinical studies or meta-analyses to determine whether DORA alters cognitive function in human subjects [20,21].
To date, no network meta-analysis comprehensively comparing the effects and applicability of commonly used hypnotics as well as hypnotic-induced cognitive impairments has been conducted.To provide a reference for the use of hypnotics, we categorized administered doses of four commonly used hypnotics as low, medium, or high doses and performed a comprehensive analysis of their efficacy and safety in terms of sleep, cognition, and adverse reactions.

Methods
This meta-analysis was guided by the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines [22].The review scheme was registered with the International Prospective System Review Register (PROSPERO) (unique identifier: CRD42022352911).

Inclusion and exclusion criteria
Inclusion criteria: (3) Research with data that cannot be extracted (e.g., conference presentations) or are missing; (4) Non-RCT publications, such as network meta-analyses, meta-analyses, systematic reviews, reviews, theoretical papers, methodological papers, animal experiments, case-control studies and cohort studies.

Literature screening and data extraction
After searching for relevant records, references were extracted according to the search strategy.Endnote X9 software was used to eliminate duplicate records consistent with the inclusion standards.Relevant full texts were downloaded to determine whether they were eligible (i.e., met the inclusion standards).Two authors independently screened, extracted and cross-checked the selected records according to these standards.Disagreements were resolved through discussion with a third investigator.The data extracted were as follows: first author, publication date, sample size, sex ratio of participants, mean age of participants, mean disease duration, number of cases, intervention measures in the treatment group, intervention measures in the control group, course of treatment, outcome measures, and adverse reactions.

Bias risk assessment
Using the Risk of Bias tool in Review Manager 5.3 software, two evaluators independently evaluated the quality of each study from seven perspectives.Disagreements were resolved by discussion with the third investigator.Review Manager 5.3 software was used to visualize the results of the bias risk assessment for the literature included.

Statistical methods
Stata v. 16.0 was applied.The standard mean difference (SMD) was generated as the effect size for continuous variables.Odds ratios (ORs) were produced for categorical variables.If only figures were provided, two researchers independently used GetData 2.26 to obtain the data and calculate the means.In cases where I 2 ≤50 % and p > 0.01, the fixed effect model was applied.Otherwise, the random effect model was used.If I 2 >75 %, Galbraith plots were drawn to exclude studies outside the outlines to eliminate heterogeneity.Publication bias was evaluated by funnel plots and Egger's test.A p value < 0.05 was considered statistically significant.

Study search and baseline characteristics and quality
In the preliminary phase, a total of 2085 articles were searched, after scrupulous screening, eventually, 27 RCTs and 2 case-control studies were eligible for analysis (Fig. 1) .The subjects came from America, Australia, British, and Germany, etc., their age mean ranged from 18 to 85, the general proportion of female was 47.3 %.

Headache
There were no significant differences in headache, as shown in Table 7.
In summary, DSST scores did not differ according to hypnotic, but there was a small-sample effect in the comparison of the placebo and zolpidem high .Daytime alertness did not consistently differ according to hypnotic, but there was a small-sample effect in the comparison of the placebo and eszopiclone low .The SE results were consistent.The WASO results significantly differed, but there was a small-sample effect in the comparison of eszopiclone low and eszopiclone mid .There were significant differences in the LPS.There were significant differences in the TST and a small-sample effect in the comparison of the placebo and zolpidem mid .There were no significant differences in headache or sleepiness.

Discussion
To the best of our knowledge, this is the first network meta-analysis of the effects of hypnotic agents on sleep and cognition.

Table 4
Matrix of pairwise comparisons of regimens on WASO (shown as standard mean difference and 95 % confidence intervals).The DSST is part of the revised Webster Adult Intelligence Scale and is often used to assess information processing, attention, and psychomotor performance [52,53].We found that none of the hypnotic agents, at any dose, significantly impacted DSST scores.
Despite numerous clinical studies on hypnotherapy for insomnia, however, there is no clear conclusion on which hypnotic is the best treatment.We believe that a traditional meta-analysis, restricted to 2by-2 comparisons, cannot provide valid methodological support for selecting the best intervention for hypnotherapy.In contrast, a network meta-analysis allows the comparison of multiple interventions.Therefore, this network meta-analysis provides the first comparison of the efficacy and safety of different hypnotic aids for insomnia using a frequency-based network meta-analysis framework, with the aims of comprehensively comparing direct and indirect treatments and providing more credible evidence for the clinical treatment of insomnia.
Over the years, DORA has been developed into a very successful hypnotic drug.DORA inhibits the hyperactive reticular activating system in insomnia by blocking orexin signal transduction [54].An ideal hypnotic would have the following effects: rapid induction of sleep, sleep maintenance throughout the night (providing sufficient sleep), and little residual effects (e.g., sleepiness) the next morning [55].DORA provides advantages such as improving sleep induction, enhancing metabolic waste removal, and improving circulation in the glial lymphoid system [56].In a follow-up survey, DORA was well tolerated without serious safety problems or rebound or withdrawal reactions [57].Regarding the effect of DORA on cognitive performance, orexinergic neurons directly project to the hippocampus and thereby influence learning, memory, and cognitive performance, indicating that DORA treatment may be a potential strategy to reverse the early cognitive impairment of insomnia patients [58].Nonbenzodiazepines, including zolpidem, zopiclone, and eszopiclone are less addictive, have fewer neuromuscular effects, and lead to less cognitive impairment than benzodiazepines [59].Zolpidem and dexzopiclone, in particular, only agonize hypnotic receptors and do not exert off-target effects on receptors involved in muscle relaxation, anxiety, or cognitive performance.Thus, nonbenzodiazepines are less disruptive to normal sleep architecture, are safer, and lead to less daytime sleepiness and other adverse effects than benzodiazepines [60].However, as a first-line therapeutic drug in clinical application, zolpidem does not have satisfactory cognitive results.In addition, there is no clear "gold standard" for measuring hypnotic-related cognitive impairment.Zolpidem leads to few sequelae and withdrawal symptoms, is less likely to lead to drug tolerance and dependence, and has a wide range of safety.However, when used with other central inhibitors, it can cause severe respiratory depression.Therefore, it is suitable for occasional and temporary insomnia.Short-term use of zolpidem as a hypnotic is known to have common adverse reactions such as hallucinations, excitement, nightmares, and depression.Zopiclone and eszopiclone are representatives of the third generation of sedative hypnotic drugs.They have high efficacy, few adverse reactions, rapid action and effectiveness up to 6 h, enabling patients to fall asleep quickly and maintain sufficient sleep depth.Long-term use does not lead to obvious drug resistance or rebound after drug withdrawal.The latest drug, dexzopiclone, is a dextro (r-) isomer of zopiclone and is 2 times more potent than the s-isomer (zopiclone) with reduced toxicity.In terms of the likelihood of addiction, benzodiazepines > zopiclone > zolpidem.The use of DORA as a potential preventive, therapeutic or neuroprotective drug that downregulates the orexinergic system [61] can not only treat sleep disruption in insomnia patients but also slow neurodegeneration process and cognitive impairment due to sleep loss.In short, DORA may treat a new aspect of (mainly mild to moderate) insomnia by improving sleep and enhancing cognition.

Limitations
1. Studies on each individual outcome were scarce; therefore, the sample sizes were small.Small sample sizes can result in higher variability, namely, wider 95 % confidence intervals, which indicates less stability of our results.2. Likewise, owing to the scarcity of studies for certain outcomes, we were unable to analyze an outcome evaluated by two articles even though I 2 ≥ 75 %; for these two articles, we describe the results only.3. We did not account for the duration of drug use.The efficacy of agents (such as zolpidem) in inducing sleep gradually diminishes over time; thus, there would be potential bias without adjusting for the treatment period.However, only a few studies reported this confounding factor.4. One of the articles included in our meta-analysis was not an RCT, potentially yielding bias.

Conclusion
In brief, DORA is a promising agent due to both its efficacy in treating insomnia and optimal safety, as indicated by a lack of cognitive impairment.Further head-to -head clinical studies are needed to confirm our findings.

Fig. 1 .
Fig. 1.Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of study search and selection for the meta-analysis.

Fig. 2 .
Fig. 2. Network relationship diagram.The larger the circle, the more treatment measures were used.Thicker lines between two points indicate more studies including both topics.

Table 1
Matrix of pairwise comparisons of regimens on DSST (shown as standard mean difference and 95 % confidence intervals).

Table 2
Matrix of pairwise comparisons of regimens on Daytime alertness (shown as standard mean difference and 95 % confidence intervals).

Table 5
Matrix of pairwise comparisons of regimens on LPS (shown as standard mean difference and 95 % confidence intervals).

Table 6
Matrix of pairwise comparisons of regimens on TST (shown as standard mean difference and 95 % confidence intervals).

Table 7
Matrix of pairwise comparisons of regimens on somnolence (shown as standard mean difference and 95 % confidence intervals).

Table 8
Matrix of pairwise comparisons of regimens on headache (shown as standard mean difference and 95 % confidence intervals).