Association of Maternal Smoking During Pregnancy With Childhood Blood Pressure and Hypertension in the ECHO Cohort.
Feb 16, 2026DOI: 10.1161/circulationaha.125.076520LetterAHA
Written by: Lyndsey E Shorey-KendrickChristine Ladd-AcostaHaozuo ZhaoJudy L AschnerCarrie V BretonCarlos A CamargoAndrea E Cassidy-BushrowElena ColicinoDana DabeleaAnne L DunlopShohreh F FarzanAssiamira FerraraJames E GernIrva Hertz-PicciottoMargaret R KaragasCatherine J KarrBarry LesterLeslie D LeveBrianna F MooreJenae M NeiderhiserEmily OkenT Michael O'SheaKeia SandersonJoseph B StanfordLeonardo TrasandeScott T WeissRosalind J WrightQi ZhaoYeyi ZhuCindy T McEvoyEliot R SpindelECHO Cohort Consortium
Elevated blood pressure (BP) and hypertension in childhood are strongly related to hypertension in adulthood, a major risk factor for cardiovascular morbidity and mortality.1 Previous studies suggest that maternal tobacco smoking during pregnancy (MSDP) is associated with increased childhood BP, but findings are inconsistent and mostly limited to systolic BP and self-reported MSDP.2 This study examined the association of MSDP with childhood systolic and diastolic BP and hypertension using data from the Environmental influences on Child Health Outcomes (ECHO) Cohort. The study protocol was approved by the single ECHO institutional review board. Written informed consent, or child assent with parent’s/guardian’s permission, was obtained for participation in the ECHO Cohort Data and Biospecimen Collection Protocol. Select de-identified data are available through NICHD’s Data and Specimen Hub (DASH).3We included children from singleton pregnancies with at least one BP measurement between 3–18 years and maternal self- or medical record report of any MSDP. We excluded cohorts with <10 children. We computed average systolic and diastolic BP Z-scores for each visit age adjusted for age, sex, and height.4 We utilized linear mixed models with random effects for individual participants nested within cohorts to test the association between MSDP and BP Z-scores, and Poisson regression models to test the association between MSDP and hypertension, defined as an average systolic and/or diastolic BP meeting or exceeding published thresholds [for children <13 y: ≥95th percentile, or ≥130/80 mm Hg (whichever is lower); for children ≥13y: ≥130/80] on ≥3 separate visits.1 We implemented unadjusted models and models adjusted for potential confounders identified previously (child race [parent report], maternal age at delivery, maternal pre-pregnancy BMI category, prenatal alcohol [self- or medical report], maternal education [highest level of school completed] as a proxy for socioeconomic status, and child sex [hypertension models only]). We also implemented models with additional adjustment for preterm birth (< 37 weeks). Sex- and age (< vs. ≥ 6 years)- stratified analyses were also performed. Models for MSDP and hypertension were fit with and without an interaction term between MSDP and child sex. Covariates with <20% missing data were imputed at the cohort type level using multivariate imputation by chained equations algorithm. We additionally examined associations with active smoking in a subset with maternal urine cotinine measured during the index pregnancy (no exposure: ≤1.6 ng/mL; active smoking: ≥75 ng/mL)5 and with secondhand smoke exposure based on the question “Did anyone (excluding the mother) ever smoke in the mother’s home(s) or dwelling(s) during the child’s pregnancy?” Only 20 subjects reported exposure to e-cigarettes in addition to traditional cigarettes.A total of 13120 children born between 1999–2020 from 52 cohorts met study criteria, with 9.3% prevalence of MSDP and nearly equal prevalence of each sex (52% male). Most reported race as White (58%), but others reported Black (18%), multiple (10%), Asian (5%), or other races (5%). The majority of children (7034) had BP data from one age; 3139 had measures at two ages, and 2947 had measures at ≥3 ages. The mean child age across measurement visits was 6.17 years. In the subset of children with BP measures at ≥3 visits (n=4158), 240 (5.8%; females, 82; males, 158) met criteria for hypertension.1 In a subset of participants (n=2763; 10 cohorts), maternal urine cotinine was measured during pregnancy (median of 25 weeks gestation; range = 7–41), and we observed high concordance (>95%) between self-reported MSDP and pregnancy urine cotinine–derived categories.Any MSDP was significantly associated with higher child diastolic BP (β=0.09 units; 95%CI: 0.04, 0.13; p<0.001) but not systolic BP (β=0.05 units; 95%CI: −0.01, 0.10; Figure 1). Any MSDP was also associated with a significant increase in the incidence rate ratio for hypertension (without interaction, 1.54; 95%CI: 1.04, 2.27). Cohort was removed from hypertension models to achieve convergence. Adding pre-term birth to models did not alter results (Figure 1). Results with sex-stratification or interaction suggest stronger effects among females compared to males (diastolic BP: females, 0.10; 95%CI: 0.03, 0.16; males, 0.08; 95%CI: 0.02, 0.15; hypertension interaction term, 0.31; 95%CI: 0.15, 0.66). Stratification at age 6 suggests weaker effect on diastolic BP in younger children (< 6 years, 0.07; 95%CI: 0.01, 0.13; ≥ 6years, 0.11; 95%CI: 0.06, 0.17), suggesting effects on BP may increase with age. Cotinine-determined active smoking exposure was associated with significantly higher diastolic and systolic BP Z-scores in crude models only and increased risk of hypertension before and after covariate adjustment. Secondhand smoke exposure only during pregnancy (1570 exposed vs 6017 unexposed) was not associated with increased BP Z-scores.Findings from this large multicohort US study support previous evidence that MSDP is a risk factor for increased childhood BP and hypertension. Our results extend previous studies of the relationship between MSDP and childhood BP to a large US population of children that included both males and females, a younger age group, and use of an objective biomarker of exposure.2 Further, our results suggest that decreasing MSDP may reduce prevalence of hypertension in children and therefore among adults, improving overall health.
AD
Dr. Anne Dunlop
Family Medicine · Women's HealthEmory Healthcareanne