Association of COVID-19 Vaccination During Early Pregnancy With Risk of Congenital Fetal Anomalies

Pregnant individuals with SARS-CoV-2 infection experience increased maternal and neonatal morbidity.^1-3 Although effective COVID-19 vaccines became available in December 2020, pregnant people were excluded from initial trials. Whereas data suggest that COVID-19 vaccines are safe and effective during pregnancy, there is concern about whether the vaccines are associated with risks to the fetus.⁴ We evaluated the association between COVID-19 vaccination during early pregnancy and risk of major fetal structural anomalies identified on ultrasonography.

This cohort study included pregnant people receiving care at a quaternary medical center in Chicago, Illinois, who completed a fetal anatomic survey between March and November 2021 and had COVID-19 vaccination records. This study followed the STROBE reporting guideline. The Northwestern University institutional review board provided approval, with a waiver of informed consent because data were collected retrospectively and each participant could not be contacted. Most pregnant people who delivered neonates at the center received outpatient care in community practices. Patient characteristics were abstracted from electronic medical records (EMRs). Race and ethnicity were self-reported and were included to assess whether vaccination uptake varied among racial and ethnic groups. First vaccination date was obtained from EMRs and the Illinois Comprehensive Automated Immunization Registry Exchange and included both messenger RNA and adenovirus vector vaccines. Thirty days before conception until 14 weeks’ gestation was considered the teratogenic window. Participants were considered unvaccinated if there was EMR documentation of declination of vaccination. Fetal congenital anomalies were defined as structural anomalies identifiable in the second trimester (eg, 18-24 weeks’ gestation) that may affect a neonate’s life expectancy, health, or functioning and were categorized according to the Brighton Collaboration Congenital Anomalies Working Group recommendations.⁵ Functional defects (eg, galactosemia) cannot be assessed using ultrasonography and thus were excluded. Sonographic interpretation was performed by perinatologists or obstetrics and gynecology specialists with additional training in obstetric ultrasonography. Vaccination status was not routinely available to clinicians at the time of ultrasonography.

Primary analyses compared unvaccinated individuals and those vaccinated outside the teratogenic window (ie, individuals without a potential teratogenic exposure) with those vaccinated within the teratogenic window. Analyses were conducted using Stata, version 15.0. Sensitivity analyses used a narrower teratogenic window to categorize exposure (2-10 weeks’ gestation). Two-sided P < .05 was considered significant.

Of 3156 patients (100% female; mean [SD] age, 33.4 [4.6] years) who met the inclusion criteria, 2622 (83.1%) received at least 1 vaccine dose and 1149 (43.8%) were vaccinated within the teratogenic window (Table 1). An anomaly was identified in 27 of 534 unvaccinated people (5.1%) and 109 of 2622 people who received at least 1 dose of vaccine (4.2%) (P = .35). Similar findings were seen when the teratogenic window was narrowed (Table 2). After controlling for potential confounders (age at delivery, nulliparity, chronic hypertension, and hemoglobin A_1c level during the first trimester), vaccination within the teratogenic window was not associated with presence of a congenital anomaly identified on ultrasonography (adjusted odds ratio, 1.05; 95% CI, 0.72-1.54).

In 3% to 5% of births in the US, neonates are born with structural defects, which are associated with increased infant morbidity, mortality, and billions of dollars in cost.^5,6 Our findings suggest that COVID-19 vaccination during early pregnancy is not associated with an increased risk of fetal structural anomalies identified with ultrasonography.

The findings are limited by the retrospective, single-center origin of the data and by limitations of EMRs. Not all confounding variables could be reasonably ascertained (eg, folic acid intake). Furthermore, ultrasonography markers are surrogate outcomes, and many pregnancies in the data set are ongoing; thus, neonatal outcomes were not uniformly available. Given the urgent need for safety data on COVID-19 vaccines, these preliminary findings may be useful when considering vaccination during early pregnancy. The adjustment for preexisting risk factors (eg, hyperglycemia) allowed for a better understanding of associations between vaccination and anomalies. Clinicians may use this evidence in counseling their patients on the safety of vaccination.

Accepted for Publication: January 10, 2022.

Published Online: April 4, 2022. doi:10.1001/jamapediatrics.2022.0164

Corresponding Author: Rachel S. Ruderman, MD, MPH, Department of Obstetrics and Gynecology, Northwestern University Feinberg School of Medicine, 250 E Superior St, 5th Floor, Chicago, IL 60611 (rachel.ruderman@northwestern.edu).

Author Contributions: Drs Ruderman and Miller had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Ruderman, Perry, Miller.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Ruderman, Trawick, Perry, Millan, Miller.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Trawick, Perry, Millan, Miller.

Administrative, technical, or material support: Ruderman, Trawick, Perry, Allen, Miller.

Supervision: Ruderman, Miller.

Conflict of Interest Disclosures: Dr Miller reported serving as the site principal investigator for a Pfizer phase 2/3 randomized clinical trial of the COVID-19 vaccine during pregnancy outside the submitted work. No other disclosures were reported.