Cigarette Smoking

Pregnancy

MotherToBaby says some chemicals in cigarette smoke cross the placenta and lower oxygen and food available to the fetus. CDC says smoking slows fetal growth before birth, and ACOG lists fetal growth restriction, low birth weight, preterm birth, placental problems, ectopic pregnancy, and perinatal mortality among reported risks. The Surgeon General finds sufficient evidence for causal relationships with fetal growth restriction and low birth weight, preterm delivery and shortened gestation, placenta previa, placental abruption, and early-pregnancy smoking with orofacial clefts. It says evidence for congenital malformations overall is inadequate and that miscarriage evidence is suggestive but not sufficient to infer causation. Miscarriage: Pineles et al. (2014) reviewed 98 studies. Any active smoking was associated with miscarriage RR 1.23 (95% CI 1.16-1.30; 50 studies); when exposure was measured during the pregnancy in which miscarriage risk was assessed, RR was 1.32 (1.21-1.44; 25 studies). The same review's pooled secondhand-smoke estimate was RR 1.11 (0.95-1.31; 17 studies), which was not statistically significant. Chen et al. (2026) included 21 studies of nonsmoking pregnant women and reported SHS associations with fetal loss OR 1.21 (1.11-1.31), spontaneous abortion OR 1.18 (1.02-1.37; 10 studies), and stillbirth OR 1.27 (1.10-1.46; 10 studies). These are observational pooled associations; the newer OR and older RR use different evidence sets and should not be read as directly comparable. Birth defects: The Surgeon General says evidence is sufficient for early-pregnancy smoking and orofacial clefts, but inadequate for congenital malformations overall. Yang et al. (2022) analyzed U.S. birth data for 12,144,972 mother-infant pairs and found associations with six subtypes: congenital diaphragmatic hernia, gastroschisis, limb reduction defects, cleft lip with or without cleft palate, cleft palate alone, and hypospadias. Bolin et al. (2022) studied 8,339 congenital heart-defect cases and 11,020 controls; modest associations appeared for selected defects, not all heart defects. These observational studies do not show that every defect is related to smoking. Placental and birth outcomes: Marufu et al. (2015) pooled 25 studies and reported smoking-associated stillbirth OR 1.47 (95% CI 1.37-1.57; I-squared 79%). The review searched evidence through 2012, so its estimate is not a current baseline risk. Karumai-Mori et al. (2025) studied 81,974 singleton pregnancies in Japan. Smoking at least 11 cigarettes/day in trimester 1 and/or 2 was associated with placental abruption aRR 2.54 (1.47-4.40); after simultaneous adjustment for SHS, aRR was 2.21 (1.21-4.06). Exposure was self-reported, and residual confounding remains possible. Chen et al. (2026) analyzed 17,381,709 U.S. singleton live births and reported fetal-growth restriction below the third percentile in 3.3% of nonsmokers and 10.0% of persistent smokers; adjusted RR for persistent smoking was 2.01 (1.99-2.04). Korhonen et al. (2026) studied 795,836 term singleton births in Finland; continued smoking was associated with NICU admission aOR 1.43, antibiotic treatment aOR 1.22, and stillbirth aOR 1.49. These large studies are observational and their population estimates are not individual predictions.

Breastfeeding

CDC says tobacco chemicals, including nicotine, can pass into breast milk and smoking decreases maternal milk supply, likely through nicotine's effect on prolactin. CDC and AAP say smoking is not a reason to stop breastfeeding; CDC says breast milk remains the recommended food if a parent cannot quit. Dahlstrom et al. (2008) studied 38 exclusively breastfed six-week-old infants and their mothers, including 18 smoking and 20 nonsmoking mothers. In the smoking group, milk nicotine was median 47 micrograms/L (range 8-192). Milk nicotine was correlated with lower heart-rate variability in boys (r=-0.74, p=0.02), but not girls. Differences in blood pressure between sleep states were also observed. The small study was observational, and it did not establish long-term health effects or isolate milk exposure from all household smoke exposure. Gillson et al. (2026) analyzed weighted PRAMS data for 203,802 U.S. postpartum respondents from 2016-2022. Self-reported smoking in the prior three months was associated with lower breastfeeding initiation and shorter duration across most racial and ethnic groups. The paper notes possible pregnancy/postpartum exposure overlap, self-report, unmeasured factors, and a cross-sectional design that cannot establish causality. Kunno et al. (2026) studied 95 pregnant participants and infants in Bangkok who had high household SHS exposure. Third-trimester urine nicotine metabolites were associated with birth weight and length and with post-delivery breast-milk nicotine concentration. This study does not show that milk exposure caused birth-size differences. Ilett et al. (2003) studied 15 lactating smokers who used nicotine patches at decreasing doses. Milk nicotine and cotinine were not lower on a 21 mg patch than while smoking, but were lower on 14 mg and 7 mg patches; the estimated infant nicotine-equivalent dose declined about 70% at 7 mg compared with smoking or 21 mg. This measured transfer, not long-term infant health. NRT is medicinal nicotine and is a distinct exposure from cigarette smoke.

What was tested

Human evidence includes systematic reviews and meta-analyses, national birth-record cohorts, case-control studies, a small prospective newborn study, and small milk-transfer and infant heart-rate studies. Most pregnancy outcome evidence is observational, so it can identify associations but cannot by itself establish that smoking caused an outcome in an individual. The 2008 breast-milk study included 38 mother-infant pairs, and the 2003 nicotine-patch study included 15 nursing mothers. No absolute risk for a general pregnant smoking population is used in the period summaries.