Clonazepam

Pregnancy

Miscarriage: Picot et al. (2026) reviewed ten observational studies with over 8,000 exposed pregnancies and reported an overall odds ratio of 1.68 (95% CI 1.48 to 1.90), with an odds ratio of 1.58 (1.39 to 1.80) after adjustment for publication bias. Its clonazepam subgroup estimate was 1.82 (1.54 to 2.16) from three studies and 468 exposed pregnancies. Ishikawa et al. (2026) studied 63,482 miscarriage cases and 190,383 live-birth controls in Japanese claims data; the adjusted odds ratio for benzodiazepines as a class was 1.241 (1.136 to 1.357), while sensitivity estimates ranged from 0.984 to 1.263. The Japanese study did not report clonazepam in its individual-drug results, and its article was corrected on 17 September 2026. These observational associations do not establish cause. EMA's PRAC requested further review of miscarriage, including confounding, absolute risk, dose, duration, and whether the signal applies to the whole class or specific medicines; no final EMA conclusion was identified by 14 October 2026. Birth defects: Animal data: The current H2 Pharma label reports that clonazepam given to pregnant rabbits during organ formation was linked to more fetal malformations at doses similar to or below therapeutic doses in patients. This is animal-only evidence, not a human pregnancy risk estimate; the underlying rabbit-study reports were not independently located. Human data: Fung et al. (2026) studied 2,478,434 Medicaid and 1,750,721 MarketScan pregnancies with live-born infants; the pooled adjusted relative risk for any major malformation was 0.99 (95% CI 0.96 to 1.03), and benzodiazepine-specific results were similar, but no clonazepam-specific estimate was reported. Noh et al. (2022) studied 3,094,227 live-birth pregnancies, including 40,846 with first-trimester benzodiazepine prescriptions, and reported class-level adjusted relative risks of 1.09 (1.05 to 1.13) for any major malformation and 1.15 (1.10 to 1.21) for heart defects. Both studies used live-birth populations and prescription records, which can miss early losses or actual ingestion. Longer-term child development: MotherToBaby says the evidence on behavior and learning is uncertain. Daugaard et al. (2020) studied 913,302 Danish children; 314 had prenatal clonazepam monotherapy prescription fills. The study reported an adjusted hazard ratio of 2.41 (95% CI 1.09 to 5.35) for intellectual disability and 2.99 (1.80 to 4.96) for the combined outcome of intellectual disability with delayed childhood milestones. Bjørk et al. (2022), in a five-country Nordic cohort, reported no consistently increased risk of neurodevelopmental disorders after clonazepam monotherapy; the study included 318 exposed children of mothers with epilepsy and 1,182 in the overall population. Sheehy et al. (2026) reported an association for prescription fills overlapping the 60 days before birth, but it studied a different exposure window and outcome. These observational findings differ and do not establish that clonazepam caused the outcomes. Fetal growth: Christensen et al. (2024) studied 4,494,918 live-born singleton children across five Nordic countries and found clonazepam prescription fills associated with small size for gestational age (adjusted odds ratio 1.27, 95% CI 1.10 to 1.48); among mothers with epilepsy, the adjusted odds ratio was 1.40 (1.03 to 1.89). MotherToBaby also reports two prescription-record studies associating clonazepam exposure with preterm birth or low birth weight. Late exposure: Sheehy et al. (2026) combined five population cohorts and found clonazepam fills overlapping the 60 days before birth associated with combined neurodevelopment diagnoses (adjusted hazard ratio 1.22, 95% CI 1.12 to 1.33; 585 exposed cases). This association does not establish cause, and the study did not assess short-term newborn adaptation. Exact Klonopin label warning: "Use of Klonopin late in pregnancy can result in sedation (respiratory depression, lethargy, hypotonia) and/or withdrawal symptoms (hyperreflexia, irritability, restlessness, tremors, inconsolable crying, and feeding difficulties) in the neonate (see PRECAUTIONS: Pregnancy). Monitor neonates exposed to Klonopin during pregnancy or labor for signs of sedation and monitor neonates exposed to Klonopin during pregnancy for signs of withdrawal; manage these neonates accordingly."

Breastfeeding

LactMed (revised 15 September 2024) says maternal clonazepam occasionally causes sedation in breastfed infants, especially with other central nervous system depressants, and recommends monitoring for drowsiness, weight gain, and developmental milestones. It notes the long half-life and possible accumulation. LactMed reports a maximum estimated infant dose of 2.5% of the mother's weight-adjusted dose in one mother taking 2 mg twice daily. Nishimura et al. (2021) studied 11 mothers taking multiple benzodiazepines, including clonazepam; milk-to-plasma ratios were below 1 and no infant abnormalities were found at one month, but the results are not clonazepam-only evidence. Kelly et al. (2012) followed 124 mothers using benzodiazepines; two infants had sedation reports, and the clonazepam subgroup included 22 infants with one report in a mother using clonazepam with flurazepam, bupropion, and risperidone. The small samples and mixed exposures limit conclusions about clonazepam alone. NHS SPS (published 15 February 2024) says clonazepam can be used with caution during breastfeeding, with monitoring for drowsiness, slowed breathing, poor feeding, and weight gain; it notes limited evidence, long half-life, and possible accumulation.

What was tested

Pregnancy evidence includes observational prescription-record cohorts, a systematic review and meta-analysis, and a descriptive study using pregnancy records from Spain and Norway. These designs can identify associations but cannot by themselves establish cause; records may not show whether a dispensed medicine was taken, and some studies exclude pregnancies not ending in live birth. The DARWIN EU P3-C3-009 report describes exposure and crude pregnancy-loss incidence, not a comparative causal safety analysis. In February 2026, EMA PRAC said the available data sources were unsuitable for causal inference as presented because key confounders were missing or incomplete and important biases were identified. Breastfeeding evidence includes small milk-level studies, infant case reports, and follow-up cohorts; several reports include other medicines or exposure during both pregnancy and breastfeeding.