Lunesta vs Belsomra Side Effects: Eszopiclone vs Suvorexant Head-to-Head

How These Two Drugs Work Differently

Eszopiclone binds the alpha subunit of GABA-A receptors, amplifying inhibitory chloride currents to promote sedation. Suvorexant blocks orexin-1 and orexin-2 receptors, silencing the wake-promoting neuropeptide system rather than forcing sedation. This mechanistic split explains nearly every divergence in their adverse-event profiles.

GABA-A modulation produces dose-dependent CNS depression. The clinical consequence is a side-effect signature shared across Z-drugs and benzodiazepines: amnesia risk, ataxia, rebound insomnia on withdrawal, and complex sleep behaviors such as sleepwalking or sleep-driving [1]. Eszopiclone's 6-month efficacy trial by Krystal et al. (N=788) confirmed sustained sleep-onset and sleep-maintenance benefits, but also documented dysgeusia in 16.9% at 2 mg and 33.9% at 3 mg 1.

Orexin receptor antagonism works by a fundamentally different logic. Instead of amplifying inhibition, suvorexant reduces wakefulness drive. Herring et al. published the key phase III data in The Lancet Neurology (N=3,076 across two trials), showing that suvorexant improved subjective total sleep time by approximately 20 minutes over placebo at month 1 2. The authors noted that the orexin-blockade mechanism produced fewer of the "Z-drug-style complex sleep behaviors" seen with GABA-A agents [2]. That observation has since shaped prescribing decisions for patients with parasomnias or a history of sleepwalking.

Dysgeusia: Lunesta's Defining Side Effect

The single side effect that separates eszopiclone from every other prescription hypnotic is dysgeusia. That bitter, metallic aftertaste is not rare.

In the Krystal 2003 trial, 33.9% of patients receiving eszopiclone 3 mg reported unpleasant taste, compared to 3.1% on placebo 1. Even at the lower 2 mg dose, the rate was 16.9%. The FDA label lists dysgeusia as the most frequently reported adverse reaction at every approved dose 3. Many patients describe the taste as persisting into the following morning and affecting appetite. This is a pharmacokinetic artifact: eszopiclone's active metabolite (S-desmethylzopiclone) is secreted into saliva.

Suvorexant does not produce dysgeusia. Zero signal appeared across Herring et al.'s pooled phase III population or in subsequent post-marketing surveillance 2. For patients who have discontinued Lunesta specifically because of taste disturbance, this absence alone may justify a trial of Belsomra.

Next-Day Somnolence and Driving Impairment

Next-morning drowsiness is Belsomra's most reported side effect. In the Herring phase III trials, somnolence occurred in 7% of patients on suvorexant 20 mg vs. 3% on placebo 2. The FDA initially declined Merck's proposed 30 mg and 40 mg doses over concerns about residual morning sedation, approving only 10 mg and 20 mg 4.

Eszopiclone carries its own next-day burden. The FDA issued a 2014 dose-related warning recommending that the starting dose be limited to 1 mg because pharmacokinetic modeling showed that eszopiclone 3 mg could produce blood levels the following morning high enough to impair driving 3. A pharmacodynamic study published in the Journal of Clinical Psychopharmacology found that eszopiclone 3 mg impaired simulated driving performance 7.5 hours post-dose in a subset of adults, particularly women, who metabolize the drug more slowly 5.

The practical takeaway: both drugs can cause morning impairment, but the mechanism and dose-sensitivity differ. Eszopiclone's risk scales steeply from 2 mg to 3 mg. Suvorexant's risk is more linear and was the explicit reason the FDA capped the approved dose at 20 mg.

Complex Sleep Behaviors and the FDA Black-Box Warning

In 2019, the FDA mandated a boxed warning on all prescription insomnia drugs, covering eszopiclone, suvorexant, and others, for complex sleep behaviors including sleepwalking, sleep-driving, and engaging in activities while not fully awake 6. The agency reviewed 66 cases of serious injury or death associated with these behaviors across all hypnotics.

The Z-drug class (zolpidem, zaleplon, eszopiclone) accumulated the largest share of post-marketing reports. Eszopiclone's GABA-A mechanism produces a dissociative amnestic window during which patients may perform complex motor tasks with no subsequent recall 3. The 2019 FDA review noted that GABA-A modulating agents had a "disproportionately higher reporting rate" for these events compared to orexin receptor antagonists 6.

Suvorexant is not exempt from this risk. Post-marketing cases of complex sleep behaviors exist for Belsomra as well 4. The distinction is quantitative, not absolute. Herring et al. reported no sleep-driving events in the key trial population, and the rate of somnambulism-type events was numerically low 2. Patients with a documented history of parasomnias may carry less incremental risk with an orexin antagonist than with a Z-drug, though no randomized data directly test this hypothesis.

Abuse Potential and Dependence

Both eszopiclone and suvorexant are classified as Schedule IV controlled substances by the DEA. The scheduling is identical, but the pharmacology behind that classification is not.

Eszopiclone activates the same receptor complex as benzodiazepines. Cross-tolerance develops between Z-drugs and benzodiazepines, and abrupt discontinuation of eszopiclone after nightly use can trigger rebound insomnia for 1 to 2 nights 3. The Krystal 6-month study did document a mild rebound effect on the first discontinuation night, with sleep latency increasing by approximately 10 minutes above baseline before normalizing by night 2 1.

Suvorexant's human abuse-potential studies showed lower "drug-liking" scores than zolpidem at supratherapeutic doses 7. In a randomized crossover study of recreational sedative users, suvorexant 40 mg (twice the maximum approved dose) produced subjective effects that were rated lower than zolpidem 30 mg on visual analog scales for "take again" and "good effects" 7. Rebound insomnia following suvorexant discontinuation was not observed in the Herring phase III dataset through month 3 2.

For patients with a substance-use history or those requiring long-term hypnotic therapy, suvorexant's lower reinforcement profile may be clinically meaningful.

Head-to-Head Side-Effect Summary by Organ System

No published randomized controlled trial has compared eszopiclone and suvorexant directly. The comparison below synthesizes FDA prescribing information and the Krystal and Herring trials.

Central nervous system. Both drugs cause dizziness and somnolence. Eszopiclone adds amnesia and anxiety (reported in 1 to 3% of trial participants at 3 mg) 3. Suvorexant's CNS effects are predominantly somnolence and, at higher exploratory doses, sleep paralysis 4.

Gastrointestinal. Dysgeusia dominates eszopiclone's GI profile. Nausea occurred in 4 to 5% of eszopiclone patients at 3 mg 1. Suvorexant labels report dry mouth in a small percentage but no taste disturbance.

Psychiatric. The FDA label for Belsomra includes a warning about suicidal ideation, based on a signal in patients with pre-existing depression 4. Eszopiclone's label carries a similar psychiatric-symptoms warning, including worsening depression and hallucinations 3. Neither drug should be prescribed without screening for active mood disorders.

Infection/other. In the Krystal trial, eszopiclone-treated patients reported upper respiratory infection at rates similar to placebo (approximately 10%) 1. Suvorexant-treated patients in the Herring trial showed upper respiratory infection at comparable background rates 2. Neither drug appears to carry a unique infection risk.

Special Populations and Dose Adjustments

Elderly patients metabolize both drugs more slowly. The American Geriatrics Society Beers Criteria list all benzodiazepine-receptor agonists (including eszopiclone) as potentially inappropriate for adults aged 65 and older due to fall risk and cognitive impairment 8. Suvorexant is not excluded from Beers but is listed with a recommendation to use the lowest effective dose.

The FDA-approved geriatric starting dose for eszopiclone is 1 mg, the same as for younger adults but with a recommendation against exceeding 2 mg 3. For suvorexant, the starting dose remains 10 mg regardless of age, with caution advised in patients with hepatic impairment 4.

Women metabolize eszopiclone more slowly than men, which increases morning blood levels. The FDA's 2014 dose guidance specifically flagged female sex as a risk factor for next-day impairment at the 3 mg dose 3. Suvorexant's pharmacokinetics are also affected by body weight and sex, but the FDA did not issue sex-specific dosing recommendations.

Patients taking strong CYP3A4 inhibitors (ketoconazole, clarithromycin, ritonavir) require a dose reduction for suvorexant. The Belsomra label recommends not exceeding 10 mg when co-administered with moderate CYP3A4 inhibitors and contraindicates use with strong inhibitors 4. Eszopiclone is also metabolized by CYP3A4, and the label advises a starting dose of 1 mg with concurrent CYP3A4 inhibitors 3.

When to Choose One Over the Other

The choice between eszopiclone and suvorexant should be driven by three clinical variables: the patient's primary sleep complaint, their side-effect tolerance, and their risk profile.

Eszopiclone has stronger evidence for sleep-onset latency reduction. The Krystal trial showed a 14-minute advantage over placebo in median sleep-onset latency that persisted through 6 months 1. Suvorexant's primary efficacy outcome in the Herring trials was subjective total sleep time rather than onset latency 2. Patients whose chief complaint is difficulty falling asleep may respond more reliably to eszopiclone.

Suvorexant may be preferable for patients who cannot tolerate dysgeusia, have a history of complex sleep behaviors, have a substance-use history, or need a medication with lower rebound risk on discontinuation. The AASM 2017 Clinical Practice Guideline for pharmacologic treatment of chronic insomnia lists suvorexant as a recommended option and eszopiclone as a recommended option, with neither given preference over the other 9.

Cost may also factor in. Generic eszopiclone became available in 2014, and average retail price for a 30-day supply at 3 mg runs approximately $15 to $40 with a discount card. Belsomra remains brand-only through patent protection, with an average retail price exceeding $400 per month without insurance, though manufacturer coupons can lower the out-of-pocket cost.