Abstract
Objective:
To determine whether in utero exposure to maternal SARS-CoV-2 infection is associated with increased risk of adverse neurodevelopmental outcomes in children by 3 years of age.
Methods:
We conducted a retrospective cohort study of 18,124 live births to individuals who delivered between March 1, 2020 and May 31, 2021 within the Mass General Brigham health system. The exposure of interest was maternal SARS-CoV-2 infection, defined as a positive SARS-CoV-2 PCR test during pregnancy. The outcome of interest was presence of any neurodevelopmental diagnosis up to 36 months after birth, identified using ICD-10 diagnostic codes. To evaluate the association between SARS-CoV-2 exposure in pregnancy and these diagnoses, we used logistic regression models adjusting for maternal age, race and ethnicity, insurance type, hospital type, and preterm birth.
Results:
Among the 861 (4.8%) SARS-CoV-2–exposed pregnancies, 140 (16.3%) offspring received a neurodevelopmental diagnosis by 36 months after birth, compared with 1,680 of 17,263 (9.7%) unexposed offspring (unadjusted OR 1.80, [95% CI: 1.49–2.17], adjusted OR 1.29 [1.05–1.57], p=0.01). In sensitivity analyses, largest effects were observed in third trimester exposures overall (aOR 1.36 [1.07–1.72], p=0.01), and among male offspring (aOR 1.43 [1.05–1.91], p=0.02).
Conclusion:
Maternal SARS-CoV-2 infection in pregnancy was associated with increased risk of adverse neurodevelopmental diagnoses by age 3, with effects most pronounced following third-trimester exposure and in male offspring. These findings highlight the importance of long-term neurodevelopmental monitoring for SARS-CoV-2–exposed children.
Précis:
Exposure to maternal SARS-CoV-2 infection in utero was associated with greater odds of a neurodevelopmental diagnosis by age 3; risk was highest among male offspring and following third trimester infection.
Introduction
A substantial body of evidence links maternal infection during pregnancy to adverse neurodevelopmental outcomes in offspring. Large national registry studies have reported increased risks of neuropsychiatric disease, including autism spectrum disorder (ASD), cognitive delay, schizophrenia, and mood disorders, following in utero exposure to maternal infections.1–4 For example, a Swedish study of 2.3 million births found a 30% increase in risk for autism spectrum disorder among offspring of women hospitalized with an infection during pregnancy,2 and a subsequent analysis of 1.8 million births identified a 79% increase in risk for autism spectrum disorder and a 24% increase in risk of major depression among offspring exposed to any maternal infection in pregnancy, whether hospitalized or not.3 Relevant to SARS-CoV-2, which is only rarely vertically transmitted,5 adverse offspring neurodevelopmental effects from maternal viral infection can occur even without vertical transmission, as has been observed with influenza.6–8
The link between maternal infection and offspring neurodevelopmental disorders, even in the absence of vertical pathogen transmission to the fetus, has been demonstrated in animal models. Data from rodent and non-human primate models demonstrate that maternal immune activation disrupts normal fetal brain development through alterations in neuronal migration, synaptic function, and microglial activity.9–16 We therefore hypothesized that, like other viral and bacterial infections in pregnancy that are only rarely vertically transmitted, maternal infection with SARS-CoV-2 could be associated with increased rates of neurodevelopmental diagnoses (ND) in offspring.
In prior work, we identified an elevated risk of ND at 12 and 18 months of age in offspring exposed to maternal SARS-CoV-2 infection during pregnancy.17,18 These findings were consistent with smaller observational studies using standardized neurodevelopmental assessments during the first 1–3 years of life.19,20 However, not all studies have identified this elevated risk, particularly those that did not consider clinical diagnoses or employed shorter-term follow-up.21,22 We therefore sought to evaluate whether the previously-described 12- and 18-month risk for adverse ND outcomes persisted into early childhood by examining ND outcomes at three years of age in a large electronic health records-based cohort of linked maternal and offspring dyads. We hypothesized that SARS-CoV-2 infection would be associated with increased rates of adverse neurodevelopmental diagnoses in offspring at 3 years of age, with potential sex-specific and trimester-specific effects.
Methods
This retrospective cohort study included all live births of singleton or multiple gestations to individuals who delivered between March 1, 2020 and May 31, 2021 within the Mass General Brigham health system, which includes deliveries occurring at two academic medical centers and 6 community hospitals that share the same electronic data warehouse and governance. This time period was selected to ensure a well-phenotyped cohort of positive SARS-CoV-2 cases and true negative controls, as this coincided with universal SARS-CoV-2 polymerase chain reaction (PCR) testing on Labor and Delivery units across the hospital system, widespread testing for COVID-19 symptoms in pregnancy, and infection control procedures requiring confirmation of home antigen testing with laboratory-based testing, unlike later periods in the pandemic. The Mass General Brigham Institutional Review Board approved this study (Protocol #2021P000716) and allowed a waiver of informed consent because no patient contact was required, the study was considered to be minimal risk, and consent could not feasibly be obtained. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Eligible pregnancies had at least one SARS-CoV-2 PCR test recorded during pregnancy. Offspring were linked to the pregnant parent based on medical record number, time of birth, and offspring sex. We characterized maternal medical history using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) billing codes, problem lists, medications administered, and laboratory studies occurring between the date of the estimated last menstrual period and the discharge date for the delivery admission. SARS-CoV-2 vaccination status was obtained from documentation in the electronic data warehouse, which integrates vaccination orders, reports of vaccine status, and regional public health records. Electronic health records were also used to determine sociodemographic features, including maternal age, insurance type (public versus private payor), and self-reported race (Asian, Black, White, or other [American Indian or Alaska Native, Native Hawaiian or other Pacific Islander, more than 1 race]) and ethnicity (Hispanic or non-Hispanic). Gestational age at delivery was obtained from delivery records using best available obstetric dating.
The primary exposure of interest was maternal SARS-CoV-2 positivity during pregnancy, defined by the presence of a positive PCR result available in the electronic data warehouse and occurring between the date of the estimated last menstrual period and the discharge date for the delivery admission. The data warehouse integrates laboratory SARS-CoV-2 PCR results from in- and out-of-network hospital laboratories. Universal screening for SARS-CoV-2 by PCR on admission to Labor and Delivery departments was implemented across the Massachusetts General–Brigham health systems in April 2020 and continued through the study period. Individuals with positive results of home rapid antigen testing were advised to confirm and/or document a positive home test result with hospital-accessible PCR, per hospital infection control policy during the study period. Those individuals with only negative PCR results documented were thus considered to be truly negative. For pregnancies with multiple positive SARS-CoV-2 PCR tests, we used the first positive test date to determine trimester of exposure and classify timing of infection.
The primary outcome was defined as the presence of ≥1 ICD-10 code indicating a neurodevelopmental diagnosis (ND) within the first 3 years (36 months after birth) as included in the Healthcare Cost and Utilization Project level 2 developmental category (code 654), or diagnostic codes associated with an autistic disorder (ASD, F84.0). For all offspring, we extracted ICD-10-CM codes from billing codes and problem lists from all clinical encounters in the Mass General Brigham system through 36 months of age, including primary care visits, specialty consultations, emergency department visits, and hospitalizations. A complete list of diagnostic codes used to establish the ND outcome may be found in Appendix 1, available online at http://links.lww.com/xxx.
Demographic and clinical characteristics were compared between maternal SARS-CoV-2 exposed and unexposed groups using Wilcoxon rank-sum tests for continuous variables and Chi-square tests for categorical variables. Multivariable logistic regression was used to estimate odds ratios for ND associated with maternal SARS-CoV-2 exposure, adjusting for maternal age in years, maternal self-reported race and ethnicity, insurance type, hospital type, and preterm birth status, to yield adjusted estimates of effect and 95% confidence intervals (CIs). Covariates were selected a priori based on known associations with offspring neurodevelopmental outcomes and potential confounding relationships with SARS-CoV-2 exposure. Maternal age, race and ethnicity have previously been associated with offspring neurodevelopmental disorders.23–25 Insurance type serves as a proxy for socioeconomic status and social determinants of health.26 Hospital type was included due to differential referral patterns and diagnostic practices between academic and community hospitals. In these models, offspring exposed to maternal SARS-CoV-2 infection were compared with contemporaneous unexposed controls. We conducted three planned secondary analyses. First, to examine whether there were differences by trimester of exposure to maternal SARS-CoV-2 infection, we modeled exposure status (no exposure, 1st/2ndtrimester, 3rd trimester) as a 3-group categorical variable. We estimated marginal means for each exposure group and tested pairwise contrasts using the emmeans package (v1.11.2–8), applying Tukey’s adjustment for multiple comparisons. Second, we examined multivariable logistic regression models for male and female offspring separately in a pre-specified sex-stratified analysis. Third, to address potential differential healthcare utilization patterns, we restricted the analysis to offspring with documented follow-up visits within our healthcare system at 36 months of age. Nonindependence of multiple births was addressed by considering observations to be clustered within deliveries; the glm.cluster command in the R miceadds package (v3.11–6) was used to generate robust standard errors. All analyses used R, version 4.0.3 (R Project for Statistical Computing), with statistical significance defined as uncorrected 2-tailed P < 0.05.
Results
Among 18,124 live births included in the study, 861 offspring (4.8% of the cohort) were exposed to maternal SARS-CoV-2 infection (Table 1), as previously described (Appendix 2, available online at http://links.lww.com/xxx).18 Individuals with SARS-CoV-2 infection during pregnancy were more likely to deliver at academic medical centers (71.3% vs. 58.5%, p<0.001) and had a younger median maternal age (32.0 vs. 33.0 years, p<0.001). Individuals with SARS-CoV-2 infection during pregnancy were significantly more likely to identify as non-White (53.2% vs. 29.4%, p<0.001), Hispanic (38.3% vs. 13.0%, p<0.001), and have public insurance (48.0% vs. 17.8%, p<0.001). Individuals in this cohort were largely unvaccinated; 8% of individuals without SARS-CoV-2 infection during pregnancy had received at least 1 COVID-19 vaccine, whereas only 2% of individuals with SARS-CoV-2 infection had received at least 1 COVID-19 vaccine (p<0.001). Only 13 cases of SARS-CoV-2 infection were identified in individuals who had received at least one COVID-19 vaccine during pregnancy. Of the 861 pregnancies with SARS-CoV-2 infection, 533 (65.0%) of infections occurred in the third trimester, whereas 292 (35.0%) occurred in the first or second trimesters. SARS-CoV-2 infection during pregnancy was associated with a higher rate of preterm delivery (13.5% vs. 10.0%, p<0.001), and neonates in the exposure group had slightly lower birthweights (median 3255 g vs. 3335 g, p<0.001).
Table 1.
Clinical and sociodemographic characteristics of the cohort.
| Characteristic | Overall N = 18,124 | SARS-CoV-2 Unexposed N = 17,263 | SARS-CoV-2 Exposed N = 861 | p-value |
|---|---|---|---|---|
| Hospital Type | <0.001 | |||
| Academic medical center | 10,707 (59.1%) | 10,093 (58.5%) | 614 (71.3%) | |
| Community hospital | 7,417 (40.9%) | 7,170 (41.5%) | 247 (28.7%) | |
| Maternal age, years | 33.0 (30.0, 36.0) | 33.0 (30.0, 36.0) | 32.0 (28.0, 35.0) | <0.001 |
| Self-reported race | <0.001 | |||
| Asian | 1,801 (9.9%) | 1,754 (10.2%) | 47 (5.5%) | |
| Black or African American | 1,615 (8.9%) | 1,478 (8.6%) | 137 (15.9%) | |
| Unknown | 422 (2.3%) | 380 (2.2%) | 42 (4.9%) | |
| White | 12,586 (69.4%) | 12,183 (70.6%) | 403 (46.8%) | |
| None of the above | 1,700 (9.4%) | 1,468 (8.5%) | 232 (26.9%) | |
| Self-reported ethnicity | <0.001 | |||
| Hispanic | 2,580 (14.2%) | 2,250 (13.0%) | 330 (38.3%) | |
| Not Hispanic | 15,120 (83.4%) | 14,615 (84.7%) | 505 (58.7%) | |
| Unavailable | 424 (2.3%) | 398 (2.3%) | 26 (3.0%) | |
| Offspring sex | 0.72 | |||
| Female | 8,850 (48.8%) | 8,435 (48.9%) | 415 (48.2%) | |
| Male | 9,274 (51.2%) | 8,828 (51.1%) | 446 (51.8%) | |
| Trimester of Infection | ||||
| 1st | 66 (8.0%) | NA | 66 (8.0%) | |
| 2nd | 226 (27.0%) | NA | 226 (27.0%) | |
| 3rd | 533 (65.0%) | NA | 533 (65.0%) | |
| Unknown | 17,299 | 17,263 | 36 | |
| Public insurance | 3,491 (19.3%) | 3,078 (17.8%) | 413 (48.0%) | <0.001 |
| Mode of delivery | 0.42 | |||
| C-Section | 5,953 (32.8%) | 5,659 (32.8%) | 294 (34.1%) | |
| Vaginal | 12,171 (67.2%) | 11,604 (67.2%) | 567 (65.9%) | |
| Gestational age at delivery, weeks | 39.29 (38.14, 40.14) | 39.29 (38.14, 40.14) | 39.14 (37.71, 39.86) | <0.001 |
| Unknown | 10 | 8 | 2 | |
| Gestational age at delivery | <0.001 | |||
| Preterm | 1,840 (10.2%) | 1,724 (10.0%) | 116 (13.5%) | |
| Term | 16,284 (89.8%) | 15,539 (90.0%) | 745 (86.5%) | |
| Vaccination status during pregnancy | <0.001 | |||
| Fully vaccinated | 952 (5.3%) | 946 (5.5%) | 6 (0.7%) | |
| Partially vaccinated | 370 (2.0%) | 363 (2.1%) | 7 (0.8%) | |
| Unvaccinated | 16,802 (92.7%) | 15,954 (92.4%) | 848 (98.5%) | |
| Live offspring from a multiple gestation | 1,156 (6.4%) | 1,087 (6.3%) | 69 (8.0%) | 0.04 |
| Birthweight, grams | 3,330 (2,980, 3,655) | 3,335 (2,980, 3,657) | 3,255 (2,895, 3,590) | <0.001 |
| Unknown | 135 | 128 | 7 | |
| Birth length, inches | 19.75 (19.00, 20.50) | 19.75 (19.00, 20.50) | 19.50 (19.00, 20.08) | <0.001 |
| Unknown | 635 | 593 | 42 | |
Data presented as median (IQR) or N (%). Differences between groups assessed by Wilcoxon rank-sum test or Chi-square test. Preterm defined as delivery <37 weeks gestational age.
By age 3 (36 months), 140 (16.3%) of children exposed to maternal SARS-CoV-2 infection had received at least one ND, compared with 1,680 (9.7%) in the unexposed group. Of the 140 children exposed to maternal SARS-CoV-2 who had the primary outcome, 85 (60.7) received the first ND diagnosis between 18 and 36 months; among unexposed children, 975 (58.0%) received the first ND diagnosis between 18 and 36 months. The median days to first documentation of a ND was 524 days [interquartile range: 370–666] and 457 days [186–608] in exposed and unexposed groups, corresponding to 17 months [12–22] and 15 months [6–20] after birth, respectively. Individual ND diagnoses identified in the primary outcome are listed in Table 2. The most common ND identified included disorders of speech and language (F80.0, F80.1, F80.2, F80.81, F80.89, F80.9), developmental disorder of motor function (F82), autistic disorder (F84.0), and disorders of psychological development (F88, F89). Among the 1,820 children with the primary outcome, 1,261 (69.3%) had a ND documented in more than one clinical encounter.
Table 2.
Most common offspring neurodevelopmental diagnoses identified by exposure status.
| IC-10 Code | Description | Maternal SARS-CoV-2 Unexposed | Maternal SARS-CoV-2 Exposed |
|---|---|---|---|
| -- | Any neurodevelopmental diagnosis | 1,680 (9.7%) | 140 (16.3%) |
| F80.9 | Developmental disorder of speech and language, unspecified | 807 (4.7%) | 82 (9.5%) |
| F80.1 | Expressive language disorder | 641 (3.7%) | 46 (5.3%) |
| F82 | Specific developmental disorder of motor function | 283 (1.6%) | 23 (2.7%) |
| F84.0 | Autistic Disorder | 190 (1.1%) | 23 (2.7%) |
| F89 | Unspecified disorder of psychological development | 125 (0.7%) | <5 |
| F80.2 | Mixed receptive-expressive language disorder | 102 (0.6%) | 17 (2.0%) |
| F88 | Other disorders of psychological development | 96 (0.6%) | 15 (1.7%) |
| F80.0 | Phonological disorder | 60 (0.4%) | <5 |
| F80.89 | Other developmental disorders of speech and language | 13 (0.1%) | <5 |
| F80.81 | Childhood onset fluency disorder | 13 (0.1%) | <5 |
| F81.9 | Developmental disorder of scholastic skills, unspecified | 6 (0.03%) | <5 |
| F80.4 | Speech and language development delay due to hearing loss | <5 | <5 |
| F84.9 | Pervasive developmental disorder, unspecified | <5 | <5 |
| F80.82 | Social pragmatic communication disorder | <5 | <5 |
| F81.89 | Other developmental disorders of scholastic skills | <5 | <5 |
Number of diagnoses are not mutually exclusive. Counts with less than 5 are obscured to preserve data privacy per institutional policy.
After adjusting for maternal age, self-reported race and ethnicity, insurance type as a proxy for socioeconomic status, hospital type, offspring sex, preterm birth status, and vaccination status in a logistic regression model including all 18,124 offspring, we observed that maternal SARS-CoV-2 infection was independently associated with an increased risk of offspring ND by 3 years of age (adjusted OR 1.29 [95% CI: 1.05–1.57], p=0.01; Figure 1). Hispanic ethnicity, maternal public insurance, male sex, preterm birth, and delivery in an academic medical center versus community hospital were significantly associated with greater risk for ND diagnosis (p<0.001 in all cases).
Figure 1.
Association of maternal Sars-CoV-2 infection and covariates with neurodevelopmental diagnosis by age 3 years. Multivariable logistic regression model adjusting for maternal age, maternal race and ethnicity, maternal insurance type, delivery hospital type, offspring sex, preterm birth, and vaccination status in association with maternal SARS-CoV-2 positivity. Forest plot shows odds ratios and 95% CIs for each variable’s association with neurodevelopmental diagnosis, adjusted for all other variables shown. Maternal SARS-CoV-2 infection is the primary exposure of interest; other variables represent potential confounders.
In secondary analyses, we identified a significant association between maternal SARS-CoV-2 infection in the third trimester and the primary outcome (adjusted OR 1.36, 95% CI: 1.07–1.72, p=0.01) (Figure 2). Pairwise contrasts of the estimated marginal means indicated that offspring risk of pregnant individuals exposed during the 3rd trimester differed significantly from the unexposed group (p=0.03), while 1st/2nd trimester exposure was not significantly different from the unexposed group (p=0.66).
Figure 2.
Association of trimester of maternal Sars-CoV-2 infection with neurodevelopmental diagnosis by age 3 years. Multivariable logistic regression model evaluating the impact of trimester of maternal SARS-CoV-2 infection, adjusting for maternal age, maternal race and ethnicity, maternal insurance type, delivery hospital type, offspring sex, preterm birth, and vaccination status in association with maternal SARS-CoV-2 positivity. Forest plot shows odds ratios and 95% CIs for each variable’s association with neurodevelopmental diagnosis, adjusted for all other variables shown. Maternal SARS-CoV-2 infection is the primary exposure of interest, and here is divided by trimester of infection in pregnancy (first/second versus third); other variables represent potential confounders.
Among male offspring, third trimester maternal SARS-CoV-2 infection was significantly associated with an increased risk of ND at 3 years of age, with an adjusted OR of 1.43 [95% CI: 1.05–1.91], p=0.02 (Figure 3). The magnitude of risk in female offspring with third trimester exposure was more modest and not statistically significant (adjusted OR = 1.27, 95% CI: 0.85–1.85, p=0.23; Appendix 3, available online at http://links.lww.com/xxx).
Figure 3.
Association of trimester of maternal Sars-CoV-2 infection with neurodevelopmental diagnosis by age 3 years, male offspring only. Multivariable logistic regression model adjusting for maternal age, maternal race and ethnicity, maternal insurance type, delivery hospital type, and preterm birth in association with maternal SARS-CoV-2 positivity, grouped by trimester of infection. Only male offspring were included in this analysis (n=9,274). Forest plot shows odds ratios and 95% CIs for each variable’s association with neurodevelopmental diagnosis, adjusted for all other variables shown. Trimester of maternal SARS-CoV-2 infection is the primary exposure of interest; other variables represent potential confounders. Female offspring results depicted in Appendix 3 (available online at http://links.lww.com/AOG/E396).
Offspring with prenatal exposure to maternal SARS-CoV-2 were more likely to have at least one follow-up visit in the MGB system (43.3% vs. 38.3%, p=0.003). Of the 18,124 offspring included in the cohort, 5,855 children overall (32.3%) had at least one follow-up routine visit recorded in the EHR at or after 36 months of life, and 320 children with SARS-CoV-2 had one routine follow-up visit recorded (37.2%). Among offspring with a 36-month follow-up visit, 120 (37.5%) of 320 SARS-CoV-2 exposed and 1,637 (29.6%) of 5,535 unexposed children received an ND diagnosis. The effect size of maternal SARS-CoV-2 exposure remained similar to that observed in the overall cohort (adjusted OR 1.23, 95% CI: 0.95–1.59, p=0.11; Appendix 4, available online at http://links.lww.com/xxx), though the association in this smaller sample was no longer statistically significant.
Discussion
In this cohort study of 18,124 live births followed for 3 years, maternal SARS-CoV-2 infection during pregnancy was associated with a significantly increased risk of neurodevelopmental diagnoses. These findings were most pronounced in male offspring and following third-trimester infection. The sex-specific vulnerability aligns with previous human and animal data indicating greater susceptibility of the male placenta and fetal brain to maternal immune activation11,27–29, including in the setting of SARS-CoV-2.30–32 The trimester-specific findings highlight the third trimester as a critical window for brain development and enhanced susceptibility to maternal immune activation, as observed in other large cohorts of infection in pregnancy and in mouse models.33,34 These results extend our prior observations at 12 and 18 months17,18 to the early childhood period, suggesting persistent neurodevelopmental effects of in utero exposure to maternal SARS-CoV-2 infection.
These observations are particularly notable in light of their biological plausibility. We have previously shown that SARS-CoV-2 exposure affects the transcriptional programs and biological function of human placental Hofbauer cells,35 which may serve as a proxy cell type reflecting the effects of maternal exposures on fetal brain microglia, immune cells that play a key role in early neurodevelopment.36 Maternal SARS-CoV-2 infection has been demonstrated to be associated with sex-specific patterns of immune activation in the placenta,30 altered synaptic pruning behavior of induced microglialized placental cells,35 and more deleterious impacts on the male fetal brain in a humanized ACE2 knock-in mice.32 Maternal SARS-CoV-2 infection has also been shown to be associated with activation of feto-placental complement pathways, contributing to fetal inflammation independent of fetal virus exposure,37 as well as compromised blood vessel integrity and disrupted microglial organization associated with cortical hemorrhages in the human fetal brain exposed to maternal SARS-CoV-2 infection.38,39 Taken together, these studies provide a potential mechanism by which maternal SARS-CoV-2 infection may contribute risk to the developing fetal brain even in the absence of vertical transmission.
Strengths of our study include a large, diverse cohort, reliance on face-valid diagnostic codes, universal screening on Labor and Delivery units, and widespread reporting and confirmation of SARS-CoV-2 positivity among pregnant patients during the study period. Supporting the sensitivity of our approach, we observed associations between neurodevelopmental diagnoses and known clinical and demographic risk factors including male sex, preterm birth, Hispanic ethnicity, and public insurance status.25,26,40,41 Hispanic ethnicity and public insurance status may serve as proxies for racism and social determinants of health that could influence access to care, diagnosis patterns, or other systemic factors. The higher rates of ND observed at academic medical centers may reflect referral bias for high-risk pregnancies or more comprehensive diagnostic evaluation and follow-up postnatally. We ascertained SARS-CoV-2 infection status during pregnancy using hospital, clinic and state testing records, enabling us to reliably identify cases of infection throughout pregnancy as well as those occurring at the time of delivery. Due to infection control policies at the time pregnancies were ongoing, home antigen testing results were encouraged to be verified by PCR testing, improving the reliability of our group assignment.
By comparison, cohorts spanning longer time periods that extend later into the pandemic had increased potential for unreported and/or undetected cases of SARS-CoV-2 infections among their control populations. The observed association between maternal SARS-CoV-2 infection and preterm birth – an established and well-documented risk42,43 – serves as an internal validation of our exposure classifications and supports the reliability of our findings. This stands in contrast to other studies which have failed to identify the known association between maternal COVID-19 and preterm birth,21 again raising concerns for incorrect ascertainment of case-control status. A further strength of this cohort is that pregnant individuals were largely unvaccinated and immunologically naïve to SARS-CoV-2 infection during the study period, allowing us to isolate the association between SARS-CoV-2 infection and offspring neurodevelopment in an unvaccinated population.
We acknowledge limitations of our approach, including lack of systematic diagnostic assessment which may lead to misclassification, although this would likely lead to results biased toward the null hypothesis. In addition, children who were evaluated outside of the outpatient pediatric networks of these 8 hospitals could have received ND diagnoses not ascertained in our dataset. Although universal screening was practiced on Labor and Delivery units during the study period, asymptomatic infection during pregnancy may not be reliably detected. Here too, the misclassification of asymptomatic SARS-CoV-2 infections as negative cases would likely bias our findings toward the null hypothesis.
In conclusion, in this large cohort of over 18,000 live births, maternal SARS-CoV-2 infection was associated with significantly increased neurodevelopmental risk among 3-year-old children, with a stronger effect size in male compared to female offspring and greater effects noted after third-trimester exposure. Continued follow-up will be valuable in assessing the durability and clinical relevance of these early neurodevelopmental changes.
Supplementary Material
Funding Sources:
This study was supported by grants RF1MH132336 (Drs. Edlow and Perlis), R01 MH116270 (Dr. Perlis), and U54 MH118919 (Dr. Edlow) from the National Institute of Mental Health; grants R01 HD100022-02S2 (Dr. Edlow) and K12 HD103096-03 (Dr. Shook) from the NICHD; grant 870754 from SFARI (Dr. Edlow); and the Patricia and Scott Eston MGH Research Scholar Award from the MGH Executive Committee on Research (Dr. Edlow). Funders were not involved in the conduct of the study nor in the preparation of the manuscript.
Footnotes
The authors did not report any potential conflicts of interest.
Each author has confirmed compliance with the journal’s requirements for authorship.
Meeting Presentation: Presented at SMFM’s Annual Pregnancy Meeting in Denver, CO, January 27-February 1, 2025.
Financial Disclosure: Dr. Perlis reports payment for service on scientific advisory boards, outside the present work, from Alkermes, Circular Genomics, and Genomind. He holds equity in Circular Genomics. He is a paid editor at JAMA and JAMA Network Open. Dr. Edlow reports serving on the scientific advisory board for YourBio Health; as a consultant to Mirvie, Inc. and Merck & Co, Inc; and receiving research funding from Merck & Co., Inc, all outside the present work.
References:
- 1.Barr CE, Mednick SA, Munk-Jorgensen P. Exposure to influenza epidemics during gestation and adult schizophrenia. A 40-year study. Arch Gen Psychiatry. 1990;47(9):869–874. doi: 10.1001/archpsyc.1990.01810210077012 [DOI] [PubMed] [Google Scholar]
- 2.Lee BK, Magnusson C, Gardner RM, et al. Maternal hospitalization with infection during pregnancy and risk of autism spectrum disorders. Brain Behav Immun. 2015;44:100–105. doi: 10.1016/j.bbi.2014.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Al-Haddad BJS, Jacobsson B, Chabra S, et al. Long-term Risk of Neuropsychiatric Disease After Exposure to Infection In Utero. JAMA Psychiatry. 2019;76(6):594–602. doi: 10.1001/jamapsychiatry.2019.0029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jiang HY, Xu LL, Shao L, et al. Maternal infection during pregnancy and risk of autism spectrum disorders: A systematic review and meta-analysis. Brain Behav Immun. 2016;58:165–172. doi: 10.1016/j.bbi.2016.06.005 [DOI] [PubMed] [Google Scholar]
- 5.Movahed F, Haji Hosseini F, Heidari A, et al. COVID-19 vertical transmission from mothers to neonates: A systematic review and meta-analysis of 204 studies. J Infect Public Health. 2025;18(9):102825. doi: 10.1016/j.jiph.2025.102825 [DOI] [PubMed] [Google Scholar]
- 6.Mednick SA, Machon RA, Huttunen MO, Bonett D. Adult schizophrenia following prenatal exposure to an influenza epidemic. Arch Gen Psychiatry. 1988;45(2):189–192. doi: 10.1001/archpsyc.1988.01800260109013 [DOI] [PubMed] [Google Scholar]
- 7.Brown AS, Begg MD, Gravenstein S, et al. Serologic evidence of prenatal influenza in the etiology of schizophrenia. Arch Gen Psychiatry. 2004;61(8):774–780. doi: 10.1001/archpsyc.61.8.774 [DOI] [PubMed] [Google Scholar]
- 8.Brown AS. Prenatal infection as a risk factor for schizophrenia. Schizophr Bull. 2006;32(2):200–202. doi: 10.1093/schbul/sbj052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shook LL, Sullivan EL, Lo JO, Perlis RH, Edlow AG. COVID-19 in pregnancy: implications for fetal brain development. Trends Mol Med. 2022;28(4):319–330. doi: 10.1016/j.molmed.2022.02.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cordeiro CN, Tsimis M, Burd I. Infections and Brain Development. Obstet Gynecol Surv. 2015;70(10):644–655. doi: 10.1097/OGX.0000000000000236 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Han VX, Patel S, Jones HF, Dale RC. Maternal immune activation and neuroinflammation in human neurodevelopmental disorders. Nat Rev Neurol. 2021;17(9):564–579. doi: 10.1038/s41582-021-00530-8 [DOI] [PubMed] [Google Scholar]
- 12.Zhang J, Jing Y, Zhang H, Bilkey DK, Liu P. Maternal immune activation altered microglial immunoreactivity in the brain of postnatal day 2 rat offspring. Synapse. Published online September 26, 2018:e22072. doi: 10.1002/syn.22072 [DOI] [PubMed] [Google Scholar]
- 13.Koyama R, Ikegaya Y. Microglia in the pathogenesis of autism spectrum disorders. Neurosci Res. 2015;100:1–5. doi: 10.1016/j.neures.2015.06.005 [DOI] [PubMed] [Google Scholar]
- 14.Choi GB, Yim YS, Wong H, et al. The maternal interleukin-17a pathway in mice promotes autism-like phenotypes in offspring. Science. 2016;351(6276):933–939. doi: 10.1126/science.aad0314 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Smith SE, Li J, Garbett K, Mirnics K, Patterson PH. Maternal immune activation alters fetal brain development through interleukin-6. J Neurosci. 2007;27(40):10695–10702. doi: 10.1523/JNEUROSCI.2178-07.2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Racicot K, Mor G. Risks associated with viral infections during pregnancy. J Clin Invest. 2017;127(5):1591–1599. doi: 10.1172/JCI87490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Edlow AG, Castro VM, Shook LL, Kaimal AJ, Perlis RH. Neurodevelopmental Outcomes at 1 Year in Infants of Mothers Who Tested Positive for SARS-CoV-2 During Pregnancy. JAMA Netw Open. 2022;5(6):e2215787. doi: 10.1001/jamanetworkopen.2022.15787 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Edlow AG, Castro VM, Shook LL, Haneuse S, Kaimal AJ, Perlis RH. Sex-Specific Neurodevelopmental Outcomes Among Offspring of Mothers With SARS-CoV-2 Infection During Pregnancy. JAMA Netw Open. 2023;6(3):e234415. doi: 10.1001/jamanetworkopen.2023.4415 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shah AV, Howell HB, Kazmi SH, et al. Developmental screening of full-term infants at 16 to 18 months of age after in-utero exposure to maternal SARS-CoV-2 infection. J Perinatol. 2023;43(5):659–663. doi: 10.1038/s41372-023-01642-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Fajardo-Martinez V, Ferreira F, Fuller T, et al. Neurodevelopmental delay in children exposed to maternal SARS-CoV-2 in-utero. Sci Rep. 2024;14(1):11851. doi: 10.1038/s41598-024-61918-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jaswa EG, Huddleston HG, Lindquist KJ, et al. In utero exposure to maternal COVID-19 and offspring neurodevelopment through age 24 months. JAMA Netw Open. 2024;7(10):e2439792. doi: 10.1001/jamanetworkopen.2024.39792 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Favre G, Bromley RL, Bluett-Duncan M, et al. Neurodevelopmental outcomes of infants after in utero exposure to SARS-CoV-2 or mRNA-COVID-19 vaccine compared with unexposed infants: a COVI-PREG prospective cohort study. Clin Microbiol Infect. 2025;31(2):266–273. doi: 10.1016/j.cmi.2024.10.019 [DOI] [PubMed] [Google Scholar]
- 23.Sandin S, Schendel D, Magnusson P, et al. Autism risk associated with parental age and with increasing difference in age between the parents. Mol Psychiatry. 2016;21(5):693–700. doi: 10.1038/mp.2015.70 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Zablotsky B, Ng AE, Black LI, Blumberg SJ. Diagnosed developmental disabilities in children aged 3–17 years: United States, 2019–2021. NCHS Data Brief. 2023;(473):1–8. Accessed September 11, 2025. https://pubmed.ncbi.nlm.nih.gov/37440277/ [Google Scholar]
- 25.Gallegos A, Dudovitz R, Biely C, et al. Racial disparities in developmental delay diagnosis and services received in early childhood. Acad Pediatr. 2021;21(7):1230–1238. doi: 10.1016/j.acap.2021.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Straub L, Bateman BT, Hernandez-Diaz S, et al. Neurodevelopmental disorders among publicly or privately insured children in the United States. JAMA Psychiatry. 2022;79(3):232–242. doi: 10.1001/jamapsychiatry.2021.3815 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Braun AE, Carpentier PA, Babineau BA, et al. “Females Are Not Just ‘Protected’ Males”: Sex-Specific Vulnerabilities in Placenta and Brain after Prenatal Immune Disruption. eNeuro. 2019;6(6). doi: 10.1523/ENEURO.0358-19.2019 [DOI] [Google Scholar]
- 28.Kalish BT, Kim E, Finander B, et al. Maternal immune activation in mice disrupts proteostasis in the fetal brain. Nat Neurosci. 2021;24(2):204–213. doi: 10.1038/s41593-020-00762-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Schwarz JM, Sholar PW, Bilbo SD. Sex differences in microglial colonization of the developing rat brain: Sex differences in microglial colonization. J Neurochem. Published online January 2012:no-no. doi: 10.1111/j.1471-4159.2011.07630.x [DOI] [Google Scholar]
- 30.Bordt EA, Shook LL, Atyeo C, et al. Maternal SARS-CoV-2 infection elicits sexually dimorphic placental immune responses. Sci Transl Med. 2021;13(617):eabi7428. doi: 10.1126/scitranslmed.abi7428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Flemming A Foetal sex affects maternal and placental immune responses to SARS-CoV-2. Nat Rev Immunol. 2021;21(12):761. doi: 10.1038/s41577-021-00653-5 [DOI] [Google Scholar]
- 32.McMahon CL, Castro J, Silvas J, et al. Fetal brain vulnerability to SARS-CoV-2 infection. Brain Behav Immun. 2023;112:188–205. doi: 10.1016/j.bbi.2023.06.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kwok J, Hall HA, Murray AL, Lombardo MV, Auyeung B. Maternal infections during pregnancy and child cognitive outcomes. BMC Pregnancy Childbirth. 2022;22(1):848. doi: 10.1186/s12884-022-05188-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Creisher PS, Perry JL, Zhong W, et al. Adverse outcomes in SARS-CoV-2-infected pregnant mice are gestational age-dependent and resolve with antiviral treatment. J Clin Invest. 2023;133(20). doi: 10.1172/JCI170687 [DOI] [Google Scholar]
- 35.Shook LL, Batorsky RE, De Guzman RM, et al. Maternal SARS-CoV-2 impacts fetal placental macrophage programs and placenta-derived microglial models of neurodevelopment. J Neuroinflammation. 2024;21(1):163. doi: 10.1186/s12974-024-03157-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Batorsky R, Ceasrine AM, Shook LL, et al. Hofbauer cells and fetal brain microglia share transcriptional profiles and responses to maternal diet-induced obesity. Cell Rep. 2024;43(6):114326. doi: 10.1016/j.celrep.2024.114326 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Garcia-Flores V, Romero R, Xu Y, et al. Maternal-fetal immune responses in pregnant women infected with SARS-CoV-2. Nat Commun. 2022;13(1):320. doi: 10.1038/s41467-021-27745-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Massimo M, Barelli C, Moreno C, et al. Haemorrhage of human foetal cortex associated with SARS-CoV-2 infection. Brain. 2023;146(3):1175–1185. doi: 10.1093/brain/awac372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Perochon T, Krsnik Z, Massimo M, et al. Unraveling microglial spatial organization in the developing human brain with DeepCellMap, a deep learning approach coupled with spatial statistics. Nat Commun. 2025;16(1):1577. doi: 10.1038/s41467-025-56560-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Crump C, Sundquist J, Sundquist K. Preterm or early term birth and risk of autism. Pediatrics. 2021;148(3):e2020032300. doi: 10.1542/peds.2020-032300 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.May T, Adesina I, McGillivray J, Rinehart NJ. Sex differences in neurodevelopmental disorders. Curr Opin Neurol. 2019;32(4):622–626. doi: 10.1097/WCO.0000000000000714 [DOI] [PubMed] [Google Scholar]
- 42.Karasek D, Baer RJ, McLemore MR, et al. The association of COVID-19 infection in pregnancy with preterm birth: A retrospective cohort study in California. Lancet Reg Health Am. 2021;2(100027):100027. doi: 10.1016/j.lana.2021.100027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Smith LH, Dollinger CY, VanderWeele TJ, Wyszynski DF, Hernández-Díaz S. Timing and severity of COVID-19 during pregnancy and risk of preterm birth in the International Registry of Coronavirus Exposure in Pregnancy. BMC Pregnancy Childbirth. 2022;22(1):775. doi: 10.1186/s12884-022-05101-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.