Dayvigo (Lemborexant) During Pregnancy and Lactation: What the Evidence Shows

How Lemborexant Works: The Orexin Blockade

Lemborexant is a dual orexin receptor antagonist (DORA) that blocks both OX1R and OX2R, the two G-protein-coupled receptors activated by wake-promoting orexin-A and orexin-B neuropeptides [1]. By suppressing this arousal signal, the drug permits the brain's endogenous sleep circuitry to prevail without the broad CNS depression caused by benzodiazepines or Z-drugs.

The SUNRISE-1 trial (N=1,006) compared lemborexant 5 mg and 10 mg against placebo and zolpidem ER 6.25 mg in adults aged 55 and older with insomnia [2]. Both lemborexant doses significantly reduced latency to persistent sleep on polysomnography (LPS decreased by approximately 10 to 12 minutes vs. placebo at one month, P<0.001) while preserving next-morning postural stability better than zolpidem. These data supported FDA approval in December 2019 for insomnia characterized by difficulty with sleep onset and/or maintenance [3].

Why does mechanism matter for pregnancy? Orexin signaling is not limited to wakefulness. Preclinical research shows OX1R and OX2R expression in the placenta and uterine tissue of rodents, and orexin peptides may influence uterine contractility and placental blood flow [4]. Blocking these receptors during organogenesis raises theoretical concerns that extend beyond simple sedation.

What the FDA Label Says About Pregnancy

The Dayvigo prescribing information, revised under the Pregnancy and Lactation Labeling Rule (PLLR) that replaced the old letter-category system in 2015, includes a pregnancy subsection with the following key points [3]:

A pregnancy exposure registry does not currently exist for lemborexant. The label explicitly states: "There are no adequate data on the developmental risk associated with the use of DAYVIGO in pregnant women." It then directs clinicians to animal data showing adverse outcomes.

The label also notes that untreated insomnia during pregnancy carries its own risks, including associations with preeclampsia, gestational diabetes, and preterm birth [5]. This framing reflects FDA guidance encouraging a risk-benefit discussion rather than a blanket prohibition, but neither the manufacturer nor any major obstetric guideline recommends lemborexant during pregnancy given the absence of human safety data.

Animal Reproductive Toxicology: The Core Evidence

Because no human pregnancy data exist, the animal reproductive studies described in the Dayvigo label form the entire basis for fetal risk assessment [3].

Rat studies. Oral administration of lemborexant to pregnant rats during organogenesis (gestational days 6 through 17) at doses of 30, 100, and 300 mg/kg/day produced decreased fetal body weight and increased skeletal variations at the mid and high doses. The no-observed-adverse-effect level (NOAEL) for embryo-fetal development was 30 mg/kg/day, which generates a plasma exposure (AUC) roughly 50 times the human AUC at 10 mg. At first glance this margin appears reassuring. But the high-dose group showed increased post-implantation loss and visceral abnormalities, confirming the drug's teratogenic potential at sufficient exposure [3].

Rabbit studies. In pregnant rabbits dosed during organogenesis (gestational days 6 through 18), lemborexant at 25, 60, and 150 mg/kg/day caused embryo lethality (increased resorptions) at the highest dose. The developmental NOAEL was 60 mg/kg/day. Rabbits tend to be more pharmacokinetically sensitive to CNS agents than rats, so the margin between the NOAEL exposure and human therapeutic exposure narrows in this species [3].

Peri- and postnatal study (rats). When administered from gestational day 6 through lactation day 20, lemborexant at 300 mg/kg/day reduced pup survival and birth weight. Lower doses did not affect pup viability, but the finding demonstrates that late-gestational and early-postnatal exposure can compromise offspring outcomes [3].

These animal results do not prove that lemborexant causes birth defects in humans. Dose multiples between species are imprecise, and metabolism pathways differ. They do, however, establish a biological signal of harm that cannot be dismissed without countervailing human evidence, which does not exist.

Comparing the DORA Class: Suvorexant Pregnancy Data

Suvorexant (Belsomra), the first DORA approved in 2014, shares lemborexant's mechanism and has a longer post-marketing track record. Its prescribing information contains similar warnings: animal studies showed fetal skeletal abnormalities in rats and decreased fetal weight in rabbits, with no human controlled data [6].

A small case series published in 2022 described eight pregnancies with inadvertent first-trimester suvorexant exposure reported to the manufacturer's pharmacovigilance database [7]. No major malformations were identified, but the sample was far too small to draw safety conclusions. No comparable post-marketing case series has been published for lemborexant.

The practical takeaway: neither DORA has a human safety profile in pregnancy, and class-level animal signals of embryo-fetal toxicity apply to both agents.

Lactation: What We Know and What We Do Not

The Dayvigo label states that no data exist on lemborexant's presence in human milk, its effects on the breastfed infant, or its effects on milk production [3]. In a lactation study in rats, lemborexant and its metabolites were detected in milk at concentrations exceeding those in maternal plasma.

Several pharmacologic properties raise concern. Lemborexant is highly lipophilic (logP approximately 3.1), which favors partitioning into human milk. Its half-life of 17 to 19 hours means steady-state drug levels would persist across multiple feedings [3]. And because neonates have immature hepatic CYP3A4 activity (the primary enzyme metabolizing lemborexant), any drug transferred through milk would be cleared more slowly from the infant's circulation.

The American Academy of Pediatrics and the Academy of Breastfeeding Medicine have not issued specific guidance on DORAs during lactation. LactMed, the NIH database for drug and lactation information, notes the absence of human data for lemborexant and advises caution [8]. Some sleep medicine specialists suggest a "pump and hold" strategy if a single dose is taken, discarding milk for at least four half-lives (roughly 72 hours), but this approach is impractical for a medication intended for nightly use and has not been validated.

Risk-Benefit Framework for Insomnia in Pregnancy

Insomnia affects 46% to 78% of pregnant individuals, with prevalence peaking in the third trimester [9]. Untreated severe insomnia has been associated with adverse outcomes: a 2020 meta-analysis of 11 cohort studies (N=over 5.5 million pregnancies) found that insomnia diagnosis during pregnancy was associated with a 1.3- to 1.6-fold increased risk of preterm birth and a 1.5-fold increased risk of cesarean delivery [10]. These risks are real. They do not, however, automatically justify prescribing a medication with no human pregnancy data.

The American College of Obstetricians and Gynecologists (ACOG) recommends non-pharmacologic approaches first, specifically CBT-I delivered over four to six sessions [11]. A 2016 randomized trial (N=40) in pregnant women with insomnia showed CBT-I improved sleep-onset latency, wake after sleep onset, and self-reported sleep quality within six weeks, with benefits sustained postpartum [12]. Digital CBT-I programs have since demonstrated similar efficacy in larger pregnant cohorts.

When pharmacotherapy is necessary, ACOG practice guidance and UpToDate's obstetric drug reference suggest considering, in rough order of available data:

1. Diphenhydramine (25 to 50 mg): Extensive first-trimester exposure data from epidemiologic registries with no consistent teratogenic signal, though daytime sedation is a limitation [11].
2. Doxepin (3 to 6 mg, low-dose): Limited but existing human exposure data; no specific teratogenic signal identified in older tricyclic antidepressant registries, though data at the low insomnia dose are sparse [13].
3. Melatonin (0.5 to 3 mg): Endogenous hormone with physiologic roles in pregnancy; limited controlled trial data but no known teratogenic risk at low doses [14].

Benzodiazepines, Z-drugs (zolpidem, eszopiclone), and DORAs (suvorexant, lemborexant) are generally avoided unless the clinical situation is severe and the patient has failed first-line options, and even then, shared decision-making with maternal-fetal medicine consultation is standard practice.

Orexin Biology in Pregnancy: An Emerging Research Area

Orexin-A concentrations in maternal plasma rise during pregnancy, peaking in the third trimester, according to a 2017 cross-sectional study of 120 pregnant women published in the Journal of Clinical Endocrinology & Metabolism [15]. The authors proposed that elevated orexin levels may serve adaptive functions: supporting increased caloric intake, maintaining wakefulness during late pregnancy, and modulating uterine smooth muscle tone.

If this hypothesis is correct, pharmacologic blockade of orexin receptors during pregnancy could theoretically interfere with these adaptive responses. This remains speculative. No study has directly tested whether DORA exposure during human pregnancy disrupts energy balance, uterine contractility, or labor onset. But the preclinical signals (reduced pup survival in the rat peri-/postnatal study) are at least consistent with the possibility.

What to Do If You Took Dayvigo Before Knowing You Were Pregnant

Inadvertent early-pregnancy exposure to lemborexant is likely uncommon given the drug's niche prescribing, but it does happen. The prescribing information does not specify a threshold dose or duration associated with harm. Given that the animal NOAEL exposures substantially exceed the human therapeutic AUC, a few doses around the time of conception or during very early pregnancy would produce far lower cumulative fetal exposure than the doses that caused abnormalities in rats and rabbits.

Standard obstetric guidance for any inadvertent medication exposure applies: discontinue the drug, inform the prescribing clinician and obstetrician, and arrange for standard first-trimester screening (nuchal translucency ultrasound at 11 to 13 weeks, cell-free DNA screening if indicated) [11]. No additional monitoring specific to lemborexant exposure is currently recommended because the human risk profile is undefined.

Patients and clinicians can report pregnancy exposures to the FDA's MedWatch program (1-800-FDA-1088) and to Eisai's pharmacovigilance department. Accumulated case reports may eventually support a pregnancy exposure registry, which would improve future risk estimates.

Discontinuation Considerations Before Conception

Women planning pregnancy who take lemborexant nightly should discuss a taper and transition plan with their prescriber. Abrupt DORA discontinuation does not produce the physiologic withdrawal seen with benzodiazepines, but rebound insomnia (worsening sleep for one to three nights after stopping) is documented in the lemborexant clinical program [3].

A reasonable approach: begin CBT-I at least four to six weeks before planned discontinuation, reduce lemborexant from 10 mg to 5 mg for one to two weeks if applicable, then stop. The drug's elimination half-life of 17 to 19 hours means that greater than 99% clearance occurs within five days of the last dose. Conception attempts after five drug-free days would carry negligible residual exposure.

Paternal Exposure: A Brief Note

The Dayvigo label does not address male reproductive toxicity, and no published animal studies have evaluated spermatogenesis or male fertility endpoints with lemborexant specifically. Suvorexant's label similarly lacks male fertility data. For DORAs as a class, no signal of impaired male fertility has emerged from post-marketing surveillance [6]. Current evidence does not support discontinuing lemborexant in male partners before conception.