Systematic review and meta-analysis: The associations of prenatal exposure to SARS-CoV-2 infection and COVID-19 vaccination with child neurodevelopment

Abstract

Objective:

We aimed to systematically review and meta-analyze evidence on the associations between prenatal exposure to SARS-CoV-2 infection or COVID-19 vaccination and child neurodevelopmental outcomes.

Method:

We searched MEDLINE, EMBASE, PsycINFO, Web of Science, Scopus, and Cochrane CENTRAL for original research on neurodevelopmental outcomes in children prenatally exposed to SARS-CoV-2 infection and COVID-19 vaccination published in any language before November 8, 2024. We performed meta-analyses on any outcome reported in ≥3 controlled studies.

Results:

Seventy studies were identified on neurological (n = 7), neuroimaging (n = 12), motor (n = 3), audiological (n = 29) and neurodevelopmental (n = 35) assessments, and neurodevelopmental disorders (n = 2) with median sample sizes of N = 117 (IQR: 44–340) and follow-up $\le$ 36 months. Meta-analyses of neonatal auditory screenings (n = 10), Ages and Stages Questionnaire (ASQ-3) (n = 9), and ASQ Social-Emotional (ASQ-SE) (n = 3) data suggested a higher risk of transient hearing impairment [RR = 2.01, 95% CI, 1.39–2.91] and delays in fine motor [RR = 1.55, 95% CI, 1.14–2.10] and problem-solving [RR = 1.32, 95% CI, 1.01–1.74] skills in children prenatally exposed to SARS-CoV-2 compared to unexposed children.

Conclusion:

Prenatal SARS-CoV-2 exposure was associated with early impairments in hearing, fine motor, and problem-solving skills, which appear to resolve with time. No associations were identified with atypical neurological or neuroimaging outcomes. No adverse neurodevelopmental effects were reported in the two studies that examined prenatal COVID-19 vaccination. Study quality was generally moderate, with small sample sizes, inappropriate control groups, and unmeasured confounding. Taken together, the current body of research does not support a causal relationship between prenatal exposure to SARS-CoV-2 infection or COVID-19 vaccination and adverse neurodevelopmental outcomes.

INTRODUCTION

As of July 2025, the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) has infected over 778 million people worldwide,¹ including many pregnant individuals. According to the Developmental Origins of Health and Disease (DOHaD) framework, prenatal exposures can shape long-term health trajectories, even in the absence of acute symptoms at birth.² Therefore, the widespread prenatal exposure to SARS-CoV-2 raises concerns about potential long-term sequelae for the developing child.

Animal research has consistently linked maternal infection during pregnancy to functional brain changes and impaired neurodevelopment in the offspring, including autism spectrum disorder (ASD)-like behaviors,³ motor delays,⁴ and deviations in cognitive development⁵. Several biological mechanisms have been proposed to account for these effects, including maternal immune activation, placental dysfunction and activation of the fetal immune system.^6–10 While our previous research found no association between prenatal SARS-CoV-2 infection and adverse birth outcomes,¹¹ other studies have suggested that severe maternal infection with SARS-CoV-2 may increase the risk of preterm birth and low birthweight.¹² In contrast, prenatal exposure to COVID-19 vaccination has not been shown to increase the risk of adverse maternal, delivery, or neonatal outcomes.¹³ However, the longer-term effects of prenatal exposure to SARS-CoV-2 infection and COVID-19 vaccination on auditory, motor, cognitive, socioemotional, and neurobiological development remain to be elucidated.

Despite the reassuring safety data on COVID-19 vaccination during pregnancy, public concern and vaccine hesitancy remain widespread, often fueled by misinformation and fears about long-term effects on child health and development.¹⁴ Notably, to mount effective protection, vaccines must elicit an immune response which, although typically brief, may range from mild to severe with fever and elevated cytokine levels.^15,16 Thus, both SARS-CoV-2 infection and COVID-19 vaccination during pregnancy trigger a maternal immune response, a shared biological mechanism previously linked to adverse neurodevelopment.

Several studies have explored the impact of prenatal SARS-CoV-2 exposure on child development, yielding inconsistent results. While some studies reported significant associations between prenatal exposure to SARS-CoV-2 and delays in motor¹⁷ and auditory^18,19 domains, others did not.^20–22 Similarly, a number of review articles have reported mixed findings regarding the link between prenatal SARS-CoV-2 exposure and offspring neurodevelopment with some reviews indicating higher rates of early motor development delays in exposed children,²³ while others reported no differences between the exposed and unexposed groups.²⁴ However, these reviews predominantly examined short-term clinical characteristics²³, focused on a narrow range of developmental outcomes,^23,24 and searched a limited number of databases, resulting in few included studies.²⁴

A clearer understanding of whether and how prenatal SARS-CoV-2 and COVID-19 vaccination exposures affect neurodevelopment has important clinical and public health implications, including informing infection prevention strategies, vaccine recommendations, and prenatal counseling. Robust evidence showing no negative impact on neurodevelopment would offer critical reassurance to pregnant individuals and their families and help address ongoing misinformation. We aimed to systematically review the literature on the associations of prenatal exposure to SARS-CoV-2 and COVID-19 vaccines with neurodevelopmental outcomes in the child at any age, including neurological, neuroimaging, auditory, cognitive, motor, psychosocial, and behavioral domains. Meta-analyses were conducted for any outcomes reported in three or more studies with controls.

METHOD

This study was preregistered (PROSPERO, CRD42023451206; for protocol deviations see Supplement 1, available online) and followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA)²⁵ (for PRISMA checklist see Supplement 2, available online) and Synthesis Without Meta-analysis (SWiM)²⁶ guidelines. To investigate the associations of prenatal exposures to SARS-CoV-2 and COVID-19 vaccination with adverse neurodevelopment in the child, we systematically reviewed the literature on any available measure of child development, including neurological, auditory, motor, cognitive, behavioral, and psychosocial domains, as well as the neuroimaging literature. The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Eligibility criteria

The inclusion criteria were: 1) any original research, including prospective or retrospective studies, cohort, case-control, and cross-sectional designs; 2) participants of any age prenatally exposed to SARS-CoV-2 (e.g., established via PCR test) and/or COVID-19 vaccination (e.g., extracted from the electronic medical record [EMR]); 3) developmental (e.g., behavior) or neuroimaging (e.g., MRI) outcomes in the child and 4) written in any language. Studies were excluded if: 1) the full text was unavailable despite attempts to contact the authors for access; and 2) only postmortem tissues or animal models were investigated.

Literature search and selection process

The initial literature search was conducted in MEDLINE, EMBASE, PsycINFO, Web of Science, Scopus, and Cochrane CENTRAL on August 28, 2023, including studies from 2019 onwards. The search was rerun on November 8, 2024. Unpublished and preprinted studies were searched in bioRxiv and medRxiv. The search strategies included all appropriate controlled vocabulary and keywords related to the type of exposure (e.g., ‘SARS-CoV-2’, ‘COVID-19 vaccination’), the target population (e.g., ‘prenatal’, ‘child’), and the outcomes (e.g., ‘neurodevelopment*’, ‘neuroimaging’). Full search strategies are available in Supplement 3, available online. Studies written in languages other than English (n = 2)^27,28 were translated using DeepL (https://www.deepl.com/en/translator) and verified through multiple iterations of forward and backward translation.

Study selection and quality assessment

Please see Figure 1 for the PRISMA flowchart. Three blinded reviewers (MR, RT, and CK) screened titles and abstracts, assessed full texts for eligibility, and rated the quality of the selected articles using the Newcastle-Ottawa Scale (NOS).²⁹ Each manuscript was randomly assigned to two of the three reviewers for independent assessment, ensuring complete coverage while avoiding fixed reviewer pairs. For the detailed NOS scores per study, see Supplement 4, available online. Disagreement between the reviewers was resolved in discussions with the last author (ASR).

Figure 1.

Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 Flow Diagram

Data synthesis

Meta-analyses were conducted for outcomes reported in three or more studies with unexposed control groups. The meta-analyzed outcomes included neonatal auditory screenings, combining Auditory Brainstem Response (ABR) and Otoacoustic Emissions (OAE) tests, as well as the Ages and Stages Questionnaire (ASQ-3) and the Ages and Stages Questionnaire: Social-Emotional (ASQ:SE-2). ABR and OAE are widely used dichotomous (pass/fail) screening tools that assess different aspects of auditory function (neural conduction to the brainstem and outer hair cell function in the cochlea, respectively) with high clinical concordance. While combining them may marginally reduce specificity, this approach enhances feasibility and statistical power without compromising overall validity. Detailed descriptions of these outcomes are provided in, available online 5. We also incorporated unpublished ASQ-3 data from our prospective pregnancy cohort, which is described in Supplement 6, available online (for sample characteristics see Table S1, available online; for study results see Tables S2-S5, available online) and in further detail elsewhere.¹¹

For outcomes not amenable to meta-analysis, we used a structured narrative synthesis, in accordance with SWiM guidelines.²⁶ Studies were grouped by outcome domain (e.g., neurological, motor, neuroimaging), and within each domain, we reported the study design, measurement tools, and sample size, the presence or absence of unexposed control groups, the direction and consistency of findings, whether statistically significant group differences were observed, and the quality of evidence (low, moderate, or high) based on NOS scores. This narrative synthesis is reflected in the main text and Table 1, which summarizes methodologies, key findings, and covariate adjustments. For studies reporting outcomes across multiple timepoints or subgroups (e.g., by symptom severity or trimester of exposure), we explicitly noted these details and reported whether findings were consistent. For outcomes that we were unable to meta-analyze, we defined consistency as ≥70% of studies within a domain reporting effects in the same direction. Discrepant findings were examined in relation to methodological differences. As with all narrative syntheses, our approach may be subject to interpretive bias. The absence of pooled estimates for studies not included in the meta-analysis limits our ability to quantify heterogeneity.

Table 1.

Methodologies, main findings and quality assessments of the included studies by outcome category

Author	Groups	N	Child age	Assessment	Reporting method	Main findings	Sociodemographic factors reported	Adjusted covariates	NOS score
Neurological assessment
Hill et al., 202476	Exposed Unexposed	16 14	0–3 mo	HNNE HINE	Health professional	– No differences at birth and 3 mo follow-up	• Maternal age • Education • Household income • Infection severity • Vaccination status • GA at delivery • Parity • Delivery Mode • Pregnancy complications	None	7
Yan et al., 202184	Exposed Unexposed	5 15	NA	HNNE	Not stated	↓ Lower total HNNE score in exposed children	• Maternal age • Delivery mode • Pregnancy complications • Delivery complications • Infection severity • Child sex • GA at delivery • Birthweight • SGA • NICU admission	None	5
Hill et al., 202343	Exposed Unexposed	4 4	3 mo	HINE	Health professional	– No differences between the groups	• Maternal age • Education • Employment • Marital status • Immigrant status (25.0% yes, 75.0% no) • Infection severity • Maternal postnatal attachment • Child sex • GA at delivery • Birthweight • Birth length • Multiple pregnancy • Delivery mode • Pregnancy complications	None	7
Carrasco Colom et al., 202375	Exposed No control group	95 -	<24 h	NA	Health professional	• 100% Typical	• Maternal age • GA at infection • Infection severity • ICU admission • Child sex • GA at delivery • Birthweight • Birth length • Delivery mode • Preterm birth • SGA • NICU admission • Breastfeeding	None	4
aÖzdemir and Yilmaz, 202350	Exposed No control group	31 -	At birth	NA	Health professional	• 87.1% Typical (n = 27) • Hypotonia and personal-social delay (n = 1) • Microcephalia (n = 2) • Macrocephalia (n = 1)	• Maternal age • Infection severity • Vaccination status • GA at infection • Trimester of infection • Parity • Multiple pregnancy • Child sex • Child age at assessment • Delivery mode • Neonatal infection • Neonatal infection severity • Preterm birth • Neonatal medical conditions	None	4
Kurokawa et al., 202345	Exposed total Symptomatic Asymptomatic No control group	90 34 56 -	At birth	NA	Health professional	↑ Higher risk in neurological complications in neonates born to symptomatic mothers (20.6%) compared to asymptomatic mothers (3.6%)	• Maternal age • Trimester of infection • Maternal medical history • Pregnancy complications • Child sex • Preterm birth • Birthweight • Delivery complications • Neonatal complications • NICU admission	None	6
bMulkey et al., 202277	Exposed total Symptomatic Asymptomatic Infected at birth No control group	34c 16 12 6 -	NA	NA	EMR	17.6% neurological abnormalities (no details) in children born to symptomatic mothers(n = 6)	• Maternal age • Maternal race/ethn (47.0% Hispanic, 53.0% non-Hispanic) • Infection severity • Pregnancy complications • Child sex • GA at delivery • GA at infection • Delivery mode • Preterm birth • Neonatal infection • NICU admission	None	3
Neuroimaging
Andescavage et al., 202479	Exposed Unexposed Prepandemic	47 55 108	6 wks	MRI	Health professional	↑ Increased gray matter volume in exposed children ↑ Higher gyrification indices and deeper sulcal depth in post-pandemic group	• Maternal age • Child sex • Child race/ethn (4.3% Asian or Pacific Islander, 19.5% Black, 26.4% Other or Unknown; 46.7% White; • 20.0% Hispanic, 80.0% non-Hispanic) • GA at delivery • Birthweight • Birth length	• Education • Employment • Child race/ethn • Child sex	6
Alves de Araujo et al., 202369	Exposed Unexposed	201 18	6–7 mo	cUS	Health professional	↑ Higher risk of abnormalities in deep brain white matterand basal ganglia in exposed children	• Maternal medical history • Child sex • GA at delivery • Birthweight • Preterm birth • Neonatal medical conditions	• GA at delivery • Birthweight	6
Sugak et al., 202127	Exposed Prepandemic	159 159	1–5 days	cUS	Not stated	• 94.3% Typical in exposed children (n = 150) • 98.1% Typical in unexposed children (n = 156) ↑ Higher risk of choroid plexus cysts and striatal vasculopathy in exposed children	• Child sex • GA at delivery • Preterm birth • Birthweight • Child age at assessment	None	4
cZeng et al., 202185	Exposed Unexposed	3 15	44 wks corrected gestational age	MRI	Health professional	– No differences between the groups	• Maternal age • Infection severity • Pregnancy complications • Child sex • GA at delivery • Birthweight • Delivery mode • SGA • NICU admission • Neonatal infection • Neonatal complications	None	4
Carrasco Colom et al., 202375	Exposed No control group	85 -	3 mo	cUS	Health professional	• 95% Typical (n = 81) • Intraventricular hemorrhage (n = 1) • Increase in extra-axial space (n = 1) • Lenticulostriated vasculopathy (n = 1) • White matter oedema (n = 1)	• Maternal age • GA at infection • Infection severity • ICU admission • Child sex • GA at delivery • Birthweight • Birth length • Delivery mode • Preterm birth • SGA • NICU admission • Breastfeeding	None	4
Santos et al., 202453	Exposed No control group	60 -	<12 wks	cUS	Health professional	• 90% Typical (n = 50) • Mild ventricular dilatation (n = 4) • Moderate ventricular dilatation (n = 1) • Periventricular cysts (n = 1)	• Maternal age • Maternal race/ethn (1.5% Asian, 11.7% Black, 70.1% Multiracial, 16.1% White) • Comorbidities • Parity • Maternal postpartum depression • Child sex • Delivery mode • Preterm birth • Birthweight • Breastfeeding	None	6
aÖzdemir and Yilmaz et al., 202350	Exposed No control group	31 -	9 mo	cUS n = 18 MRI n = 8 EEG n = 4	Not stated	• 100% Typical cUS and EEG (n = 31) • 96.8% Typical MRI (n = 30) (one child had Joubert syndrome)	• Maternal age • Infection severity • Vaccination status • GA at infection • Trimester of infection • Delivery mode • Parity • Child sex • Child age at assessment • Multiple pregnancy • Neonatal infection • Neonatal infection severity • Preterm birth • Neonatal medical conditions	None	4
bMulkey et al., 202277	Exposed No control group	14 -	4 days	MRIn = 6 cUSn = 8	EMR	• 71.4% Typical (n = 10) • Unspecified abnormal imaging results (n = 3) • Lenticulostriate vasculopathy on cUS (n = 1)	• Maternal age • Maternal race/ethn (47.0% Hispanic, 53.0% non-Hispanic) • GA at infection • Infection severity • Pregnancy complications • Child sex • GA at delivery • Delivery mode • Preterm birth • NICU admission • Neonatal infection	None	3
Martenot et al., 202383	Exposed No control group	1 -	7 mo	EEG	Health professional	• 100% Typical (n = 1)	• GA at delivery • Child sex • Birthweight • Preterm birth • NICU admission • Neonatal complications	None	2
Yan et al., 202184	Exposed and infected at birth Unexposed	5 15	1 wk	MRI	Health professional	↑ Higher risk of white matter insults in the basal ganglia region in exposed children • Decreased gray matter in the thalamus, caudate nucleus, and parahippocampal gyrus in exposed children	• Maternal age • Infection severity • Pregnancy complications • Child sex • GA at delivery • Birthweight • Delivery mode • SGA • Delivery complications • NICU admission	None	5
Kurokawa et al., 202345	Exposedand neurological abnormalities No control group	9 -	<1 mo	MRI n = 6 EEG n = 6 cUS n = 3	Health professional	• Intracranial hemorrhage (n = 5) • Hypoxic brain injury (n = 3) • Frontal lobe hematoma (n = 1)	• Maternal age • Trimester of infection • Maternal medical history • Pregnancy complications • Child sex • Birthweight • Preterm birth • Delivery complications • Neonatal complications • NICU admission	None	6
McKissic et al., 202446	Exposed and neurological abnormalities No control group	7 -	0–2 mo	cUS MRI	EMR	• Diffuse/patchy white matter echogenicity (n = 4) • White matter cystic changes (n = 4) • White matter restricted diffusions (n = 3)	• Infection severity • Placental pathologies • Child sex • GA at delivery • Birthweight • Preterm birth • Neonatal infection	None	4
Motor development
Fajardo Martinez et al., 202317	Exposed Prepandemic	124 115	3–8 mo	Prechtl MOS-R	Health professional	↓ Lower MOS-R scores in exposed children	• Maternal age • Maternal comorbidities • Infection severity • Trimester of infection • Multiple pregnancy • Child sex • GA at delivery • Birthweight • Delivery mode • Preterm birth • Neonatal conditions	None	6
Aldrete-Cortez et al., 202231	Exposed Unexposed	28 28	3–5 mo	Prechtl MOS-R	EMR	↓ Lower MOS-R scores in exposed children	• Maternal age • Education • Marital status • Pregnancy complications • Child sex • GA at delivery • Delivery mode • Birthweight • Birth length	• Length of hospitality stay	8
Arguelho et al., 202470	Exposed No control group	16 -	3–6 mo	Prechtl GMA MOS-R HINE	Health professional or remote video visit	• 20% aberrant monotonous fidgety movements (n = 3) • 85.8% suboptimal global HINE at 3 mo (n = 14) • No motor delays at 6 mo follow-up	• Maternal age • Education • Household income • Marital status • Trimester of infection • Infection severity • Child sex • GA at delivery • Birthweight	None	4
Neurodevelopmental disorders
Edlow et al., 202241	Exposed Unexposed	222 7772	0–12 mo	ICD-10	EMR	↑ Higher risk of neurodevelopmental diagnosis in exposed children, particularly after 3rd trimester exposure	• Maternal age • Maternal race/ethn (9.9% Asian, 8.4% Black, 9.4% Other, 3.2% Unknown, 69.0% White; 14.6% Hispanic, 82.1% non-Hispanic, 3.3% Unavailable) • Insurance status • Trimester of infection • Maternal comorbidities • Child sex • GA at delivery • Birthweight • Birth length • Delivery mode • Multiple pregnancy • Preterm birth	• Maternal age at delivery • Maternal race/ethn • Insurance status • Trimester of infection • Child sex • Preterm birth	8
Edlow et al., 202340	Exposed Unexposed Prepandemic (born 2018) Prepandemic (born 2019)	883 17,432 13,952 15,386	0–12 mo 18 mo nexposed = 555 nctrl = 13,452	ICD-10	EMR	↑ Higher risk of neurodevelopmental diagnoses in exposed male participants at age 12 mo – No differences between the exposed and unexposed groups at age 18 mo ↑ Higher risk of neurodevelopmental diagnoses in pandemic group at age 12 mo compared to 2019 cohort – No differences between the pandemic and 2018 cohorts at age 12 mo	• Maternal age • Maternal race/ethn (9.9% Asian, 8.9% Black, 9.3% Other, 2.6% Unknown, 69.3% White; • 14.3% Hispanic, 82.9% non-Hispanic, 2.8% Unavailable) • Insurance status • Trimester of infection • Maternal comorbidities • Child sex • GA at delivery • Birthweight • Birth length • Delivery mode • Multiple pregnancy • Preterm birth	• Maternal age at delivery • Maternal race/ethn • Insurance status • Hospital type • Preterm birth	8
Audiological assessment
Tanyeri Toker et al., 202319	Exposed Unexposed	570 570	0–1 mo	ABR	Health professional	↑ Higher risk of hearing impairment at first test in exposed children ↑ Higher risk of hearing impairment at second test in exposed children – No differences between groups on the third test – No effects of trimester of exposure detected	• Maternal age • Child sex • GA at delivery • Birthweight • Delivery mode	None	8
Alan and Alan, 202130	Exposed Unexposed	118 118	0–1 mo	ABR	Health professional	↑ Higher risk of hearing impairment at first test in exposed children – No differences at second test	• Maternal age • Child sex • GA at delivery • Birthweight • Delivery mode	None	8
Mostafa et al., 202264	Exposed Unexposed	34 921	0–15 days	TEOAE ABR	Health professional	↑ Higher risk of failing the first test in exposed children – No differences on the second test	• Parity • Child sex	None	3
Swaminathan et al., 202458	Exposed Unexposed	50 50	3–18 mo	ABR OAE	Health professional	↑ OAE: Higher risk hearing impairments on the first test in exposed children – OAE: Typical hearing results in all children at follow-up ↑ ABR: Higher risk of hearing impairments in exposed children	None	None	7
Moghadasi-Boroujeni 202328	Exposed Unexposed	35 35	0–15 days	TEOAE ABR	Health professional	↑ Higher risk of failing the first hearing test in exposed children – No differences on the second test	• Maternal age • Child sex • GA at delivery • Birthweight • Delivery mode	None	5
Yilmaz et al., 202220	Exposed Unexposed	39 39	0–30 days	ABR	Health professional	↑ Higher risk of hearing impairment at first test in exposed group – No differences on the second and third test	• Maternal age • Child sex • GA at delivery • Birthweight	None	8
Celik et al., 202180	Exposed Unexposed	37 36	NA	TEOAE CLS OAE DPOAE	Health professional	↓ Worse cochlear function in exposed children	• Maternal age • Child sex • Birthweight • Delivery mode	None	3
Rajanna et al., 202318	Exposed Unexposed	942 942	1 mo 3 mo 6 mo	OAE1mo AABR	Health professional	– OAE: No differences at 1 mo – AABR: No differences at any follow-ups	• Maternal age • Trimester of infection • Child sex • GA at delivery • Birthweight	None	6
Ahmed et al., 202467	Exposed Unexposed	311 1251	0–3 yrs	ABR OAE	Health professional	– No differences between the groups	• Maternal age • Maternal race/ethn (2.9% Asian, 16.1% Black, 0.3% Native American, 50.7% Other or Mixed, 11.2% Unknown, 18.8% White) • Insurance status • Child sex • GA at delivery • Preterm birth • Delivery mode • Child age at assessment	• GA at delivery • Delivery mode • Child age at assessment	7
Santos et al., 202453	Exposed Unexposed	69 68	1 mo 2 mo 4 mo 12 mo	OAE	Not stated	– No differences between the groups	• Maternal age • Maternal race/ethn (1.5% Asian, 11.7% Black, 70.1% Multiracial, 16.1% White) • Comorbidities • Parity • Maternal postpartum depression • Child sex • Preterm • Birthweight • Delivery mode • Breastfeeding	• Maternal age at delivery • Maternal race/ethn • Parity • Comorbidities (undefined)	6
Cianfrone et al., 202439	Exposed Unexposed	16 141	0–6 mo	TEOAE ABR	Health professional	– No difference between the groups	• Maternal age • Infection severity • Trimester of infection • Child sex • Birthweight Breastfeeding • Delivery mode	None	6
Kirbac et al., 202344	Exposed Unexposed	36 36	0–6 months	ABR	Health professional	– No differences between groups at any follow-up	• Maternal age • Child sex • GA at delivery • Birthweight • Delivery mode • NICU admission	None	7
Veeranna et al., 202259	Exposed Unexposed	15 40	43 wks	DPOAE ABR	Health professional	– DPOAE: No differences between groups – ABR: No differences between groups	• Child sex • GA at delivery	None	4
Verdaguer et al., 202460	Exposed Unexposed	15 19	0–12 mo	ABR	Not stated	– No differences between the groups	• Maternal age • Infection severity • Child sex • Birthweight • Delivery mode • Preterm birth	None	6
Oskovi-Kaplan et al., 202149	Exposed Prepandemic	458 339	0–15 days	ABR TEOAE	EMR	– No differences on the first test – No differences on the second test	• Maternal age • Child sex • GA at delivery • Birthweight • Delivery mode	None	7
Senthil et al., 202456	Exposed No control group	1910 -	0–2 wks	ABR OAE	Not stated	• OAE: 19.9% hearing impaired at birth (n = 380) • OAE: 84.2% of these children had hearing impairments at 2 wk follow-up (n = 320) • ABR: 3.1% hearing impairments at 2 wk follow-up (n = 10)	• Infection severity • Child sex	None	5
Yildiz et al., 202163	Exposed No control group	199 -	0–15 days	ABR	Health professional	• 10.5% hearing impairment at first test (n = 21) • 100% Typical at second test	• Maternal age • Maternal comorbidities • Trimester of infection • GA at infection • Child sex • Birthweight • Delivery mode • Neonatal infection • NICU admission	None	5
Buonsenso et al., 202237	Exposed No control group	143 -	1–6 mo	TEOAE ABR (n = 34)	Health professional	• TEOAE: 19% hearing impairment at birth (n = 27) • TEOAE: 100% Typical at 1 mo follow-up • ABR: 100% Typical between 3–6 mo	• Maternal age • Trimester of infection • Infection severity • Birthweight • Delivery mode	None	5
Apa et al., 202332	Exposed No control group	119 -	0 mo 3 mo 12 mo	TEOAE ABR PTA	Health professional	• TEAOE: 100% Typical at birth • ABR: 4.2% hearing impairments at birth (n = 5) • ABR: 1.6% hearing impairments at 3 mo (n = 2) • PTA: 100% Typicalat 12 mo follow-up	• Maternal age • Infection severity • Parity • Child sex • GA at delivery • Birthweight • Preterm birth • Neonatal infection	None	5
Apa et al., 202433	Exposed No control group	115 -	2mo 12mo 18 mo	OAE PTA	Health professional	• OAE: 0.9% hearing impairment at 2 mo (n = 1) • PTA: 4.5% hearing impairment at 12 mo (n = 5)	• Trimester of infection • Child sex • Child age at assessment	None	5
Goulioumis et al., 202381	Exposed No control group	111 -	0.5–5.5 mo	TEOAE ABR	Health professional	• 0.9% hearing impairment at first test (n = 1) • 100% Typical at 1 mo follow-up	• Trimester of infection	None	4
Carrasco Colom et al., 202375	Exposed No control group	95 -	1 mo 2 mo 6 mo	ABR	Health professional	• 6% hearing impairment (n = 6) • 100% Typical at 6 mo follow-up test	• Maternal age • GA at infection • Infection severity • ICU admission • Child sex • GA at delivery • Birthweight • Birth length • Preterm birth • SGA • Delivery mode • Breastfeeding • NICU admission	None	4
Ghiselli et al., 202221	Exposed No control group	63 -	0–4 mo	TEOAE ABR	Health professional	• 6.3% hearing impairment at birth (n = 4) • 1.6% hearing impairment at 1 mo follow-up (n = 1)	• Maternal age • Pregnancy complications • Child sex • GA at delivery • Birthweight • Delivery mode • Breastfeeding	None	6
Kosmidou et al., 202282	Exposed No control group	32 -	3 mo 4 mo 9 mo	TEOAE	Health professional	• 22% hearing impairment at first test (n = 7) • 100% Typical at both follow-ups	• Infection severity • Child sex • GA at delivery • Birthweight	None	4
Schuh et al., 202255	Exposed No control group	30 -	At birth 6 mo	ABR	Not stated	• 10% hearing impairment at first test (n = 3) • 3.3% hearing impairment at 6 mo (n = 1)	• Maternal age • Maternal race/ethn (100.0% Latina) • Insurance status • Trimester of infection • Infection severity • Maternal ICU admission • Child sex • GA at delivery • Delivery mode • Preterm birth • NICU admission • Neonatal infection	None	3
Sanz Lopez et al., 202454	Exposed No control group	90 -	2.5 mo 10 mo	ABR ASSR	Not stated	• 100% Typical at both follow-ups	• Child sex • Neonatal infection	None	6
Orioli et al., 202474	Exposed No control group	31 -	At birth	BAEP	Not stated	• 100% Typical	• Housing conditions • Trimester of infection • Child sex • GA at delivery • Birthweight • Delivery mode • Neonatal infection	None	4
aÖzdemir and Yilmaz, 202450	Exposed No control group	31 -	At birth	ABR	Not stated	• 100% Typical	• Maternal age • Infection severity • Vaccination status • GA at infection • Trimester of infection • Parity • Multiple pregnancy • Child sex • Delivery mode • Preterm birth • Child age at assessment • Neonatal infection • Neonatal infection severity • Neonatal medical conditions	None	4
bMulkey et al., 202277	Exposedtotal Symptomatic Asymptomatic Infected at birth No control group	34 16 12 6 -	At birth	NA	EMR	• 8.8% hearing impairment (n = 3)	• Maternal age • Maternal race/ethn (47.0% Hispanic, 53.0% non-Hispanic) • Infection severity • GA at infection • Pregnancy complications • Child sex • GA at delivery • Delivery mode • Preterm birth • NICU admission • Neonatal infection	None	3
Neurodevelopmental questionnaires
Ages and Stages Questionnaire – Third Edition (ASQ-3) and Social Emotional – Second Edition (ASQ:SE-2)
Garrido-Torres et al., 202473	Exposed (mild) Exposed (severe) Unexposed	346 53 244	6mo 12 mo	ASQ-3	Parental report	– No difference between the groups at 6 mo ↑ Higher risk of personal-social delay in children born mothers with severe infection compared to mild infection and unexposed at 12 mo	• Maternal age • Maternal race/ethn (8.9 Non-White, 89.3% White, 1.8% missing) • Education • Employment • Perceived stress • Substance use during pregnancy • Child sex	• Maternal age • Trimester of infection • Symptoms severity • Perceived stress • Substance use during pregnancy • Child sex	8
Jaswa et al., 2024b91	Exposed Unexposed	217 1786	12 mo 18 mo 24 mo	ASQ-3	Parental report	– No differences between the groups at any age	• Maternal age • Maternal race/ethn (4.5% Asian, 1.8% Black, 8.4% Hispanic, 3.7% Multiracial/Other, 87.8% White) • Education • Income • Trimester of infection • Infection-related fever • Maternal anxiety symptoms • Maternal depression symptoms • Child sex • Preterm birth	• Maternal age • Maternal race/ethn • Education • Income • Trimester of infection • Infection-related fever • Maternal anxiety symptoms • Maternal depression symptoms • Child sex • Preterm birth	7
Namakin et al., 202348	Exposed Unexposed	161 181	6 mo	ASQ-3	Health professional	– No differences between the groups – No differences by timing or severity of exposure	• Maternal age • Maternal citizenship (98.1% Iranian, 1.9% Non-Iranian) • Education • Employment • Underlying disease history • Child sex	None	8
Shuffrey et al., 202268	Exposed Unexposed Prepandemic	114 141 62	6 mo	ASQ-3	Parental report	– No differences between exposed and unexposed groups ↓ Lower gross motor, fine motor, and personal-social scores in the pandemic group	• Maternal age • Maternal race/ethn (5.9% Asian, 12.5% Black, 0.8% Native American, 1.2% Native Hawaiian, 27.1% Multiracial/Other, 13.7% Unknown, 38.8% White; 50.2 % Hispanic) • Education • Parity • Child sex • GA at delivery • Delivery mode • Child age at assessment	• Maternal age • Maternal race/ethn • Education • Parity • GA at delivery • Child sex • Delivery mode • Child age at assessment	8
Vrantsidis et al., 202461	Exposed Unexposed	96 800	12 mo 24 mo	ASQ-3 ASQ:SE-2	Parental report	– No differences between the groups	• Maternal age • Maternal race/ethn (1.1% Black, 2.8% East Asian, 2.3% Hispanic, 2.6% Indigenous, 0.6% 5.1% Multiracial/Other, Southeast Asian, 2.3% South Asian, 0.7% West Asian, 81.7% White, 0.8% Missing) • Education • Household income • Food security • Prepregnancy medical conditions • Child sex • GA at birth • Birthweight • Delivery mode • Child age at assessment	• Household SES (education, income, food security) • Prepregnancy medical conditions • Trimester of infection • Symptoms severity • Child sex • Child age at assessment	9
Rommel et al., unpublished, Sept 2025	Exposed Unexposed	81 272	6 mo	ASQ-3	Parental report	– No differences between the groups	• Maternal age • Maternal race/ethn (9.3% Asian, 13.0% Black, 27.8% Hispanic, 5.1% Other, 44.8% White) • Education • Income • Pre-pregnancy BMI • Trimester of infection • Infection severity • Vaccination status • Maternal postpartum depression • Child sex • GA at delivery • Birthweight	• Maternal race/ethn • Education • Maternal postpartum depression • GA at delivery • Child sex	8
Wu et al., 202162	Exposed Unexposed	57 78	0–3 mo	ASQ-3 ASQ:SE-2	Parental report	– No differences between the groups	• Maternal age • Education • Employment • Pregnancy complications • Child sex • GA at infection • Birthweight • Preterm birth • NICU admission • Birth complications • Length of mother-baby separation • Breastfeeding	• Child sex • Birthweight • delivery • Preterm birth • NICU admission • Length of mother-baby separation • Length of breastfeeding	6
Cheng et al., 202138	Exposed Unexposed	9 9	8–10 mo	ASQ-3	Parental report	↓ Lower fine motor scores in exposed children	• Maternal age • Maternal height • Maternal weight • Maternal BMI • Metabolic disorders of pregnancy • Parity • Child sex • GA at delivery • Birthweight • Delivery mode • Preterm birth • Delivery complications	None	5
Hill et al., 202476	Exposed Unexposed	16 14	12 mo	ASQ-3	Parental reported	↓ Lower communication, problem-solving, and personal-social scores in exposed children	• Maternal age • Education • Household income • Infection severity • Vaccination status • Parity • GA at delivery • Delivery Mode • Pregnancy complications	None	7
Favre et al., 202489	Exposed Vaccine exposed Unexposed	76 153 101	12 mo	ASQ-3	Parental reported	– No difference between the groups	• Maternal age • Maternal race/ethn • (1.8% Asian or Pacific Islander, 1.8% Black, 4.8% Hispanic, 3.6% Other, 7.9% Unknown, 79.7% White) • Parental education • Marital status • Maternal BMI • Maternal addiction • Maternal medical history • Parity • Pregnancy complications • GA at birth • Birthweight • Delivery mode • NICU admission	• Maternal age • Maternal race/ethn • Education • Marital status • Maternal BMI • Maternal addiction • Parity • Pregnancy complications • Birthweight • NICU admission	7
Jaswa et al., 2024a88	Vaccine exposed Unexposed	1692 795	12 mo 18 mo	ASQ-3	Parental report	– No difference between the groups at 12 mo and 18 mo	• Maternal age • Maternal race/ethn (4.6% Asian, 2.1% Black, 8.5% Hispanic, 3.9% Multiracial/Other, 89.3% White) • Education • Income • Trimester of infection • Infection-related fever • Maternal anxiety symptoms • Maternal depression symptoms • Child sex • Preterm birth	• Maternal age • Maternal race/ethn • Education • Income • Maternal depression symptoms • Maternal anxiety symptoms • SARS-CoV-2 infection • Child sex • Preterm birth	8
Ayed et al., 202234	Exposed No control group	298 -	10–12 mo	ASQ-3	Parental report	• 10.1% neurodevelopmental delays (n = 30)	• Maternal age • Maternal nationality (37.6% Kuwaiti, 62.4% Non-Kuwaiti) • Education • Paternal education • GA at infection • Infection severity • Delivery mode • Parity • Pregnancy complications • Child sex • GA at delivery • Birthweight • Preterm birth • Multiple pregnancy • NICU admission • Neonatal infection • Breastfeeding	None	5
Fajardo-Martinez et al., 202442	Exposed No control group	80 -	4–28 mo	ASQ-3	Health professional EMR	• 6%-9% neurodevelopmental delays, particularly in communication and fine motor domains (n = 5–7)	• Maternal age • Maternal race/ethn (25.3% Asian or other, 2.2% Black or Multiracial, 44% Hispanic, 28.6% White) • Education • Insurance status • Maternal medical history • Maternal medical conditions • GA at infection • Infection severity • Vaccination status • Child sex • GA at delivery • Birthweight • Delivery mode • Preterm birth • SGA • NICU admission • Neonatal complications	None	5
Shah et al., 202357	Exposed No control group	51 -	16–18 mo	ASQ-3	Telephone interview	• 51% development delay on at least one domain (n = 12)	• Maternal age • Insurance status • Delivery mode • Infection severity • Child sex • GA at delivery • Birthweight • SGA • NICU admission • Child age at assessment	None	5
Konduri and Joshi, 202486	Exposed No control group	45 -	9–15	ASQ-3	Telephone interview	• 28.8% developmental delays, particularly gross motor and problem-solving skills (n = 13) • Higher risk of delay in male (9/20) than female (4/25) participants	• Maternal age • Maternal comorbidities • Child sex • GA at delivery • Birthweight • Length of mother-baby separation	None	4
Santos et al., 202453	Exposed No control group	43 -	4 mo (n = 42) 6 mo (n = 43) 12 mo (n = 28)	ASQ-3	Health professional	• 36% developmental delays at 4mo, particularly fine motor and problem-solving (n = 15) • 7% developmental delays at 6mo, particularly gross motor (n = 3) • 32% developmental delays at 12mo, particularly fine motor and personal-social (n = 9)	• Maternal age • Maternal race/ethn (1.5% Asian, 11.7% Black, 70.1% Multiracial, 16.1% White) • Parity • Comorbidities • Maternal postpartum depression • Child sex • Preterm birth • Birthweight • Delivery mode • Breastfeeding	None	6
Martenot et al., 202383	Exposed No control group	26 -	10 mo	ASQ-2	Telephone interview	• 4% gross motor delay (n = 1)	• GA at delivery • Child sex • Birthweight • Preterm birth • NICU admission • Neonatal complications	None	2
Schuh et al., 202255	Exposed No control group	15 -	0–6 mo	ASQ-3	EMR	• 100% Typical	• Maternal age • Maternal race/ethn (100.0% Latina) • Insurance status • Trimester of infection • Infection severity • Maternal ICU admission • Delivery mode • Child sex • GA at delivery • Preterm birth • Neonatal infection • NICU admission	None	3
Yan et al., 202184	Exposed No control group	5 -	9 mo	ASQ-3	Telephone evaluation	• 100% Typical	• Maternal age • Infection severity • Pregnancy complications • Child sex • GA at delivery • Birthweight • Delivery mode • SGA • Delivery complications • NICU admission	None	5
dJackson et al., 202490	Exposed and infected at birth Unexposed	96 243	21–32 mo	ASQ-3 ASQ:SE-2	Parental reported and phone interview	↑ ASQ-3: Higher risk of personal-social skills delay in exposed children ↑ ASQ:SE-2: Higher risk of social-emotional delay in exposed children	• Maternal age • Maternal race/ethn (11.8% Asian, 5.3% Black, Multiracial or Other, 82.9% White) • Paternal race/ethn (11.2% Asian, 6.2% Black, Multiracial or Other, 79.4% White) • Parental education • Index of deprivation • Trimester of infection • Infection severity • Child sex • GA at delivery • Birthweight • Breastfeeding	• Maternal race/ethn • Parental education • Index of Multiple Deprivation • Child sex • Child age at assessment	8
bMulkey et al., 202277	Exposedtotal Symptomatic Asymptomatic Infected at birth No control group	34 16 12 6 -	3 mo (n = 34) 10 mo (n = 18) 14 mo (n = 4)	ASQ-3	EMR	↑ Higher risk of fine motor and personal-social skills delay in children born to symptomatic mothers than asymptomatic mothers	• Maternal age • Maternal race/ethn (47.0% Hispanic, 53.0% non-Hispanic) • Infection severity • GA at infection • Pregnancy complications • Child sex • GA at delivery • Delivery mode • Preterm birth • Neonatal infection • NICU admission	None	3
Other neurodevelopmental questionnaires
Firestein et al., 202371	Exposed Unexposed	112 256	5–11 mo	DAYC-2	Trained research staff via telehealth	– No differences between the groups	• Maternal age • Maternal race/ethn (3.2% Asian, 11.2% Black, 1.5% Native American, 0.7% Native Hawaiian, 11.2% Other, 12.7% Unknown, 59.6% White) • Parity • Insurance status • Child sex • GA at delivery • Delivery mode • Preterm birth • Child age at assessment	• Maternal race/ethn • Maternal age • Insurance status • Parity • Delivery mode • GA at delivery • Child sex	9
Liu et al., 202222	Exposed Unexposed	33 67	0–13 mo	DDST	EMR	– No differences between the groups	• Maternal age • Parity • Maternal BMI • Maternal comorbidities • Child sex • GA at delivery • Birthweight • Birth length • Preterm birth	• GA at delivery • Child sex	6
Firestein et al., 2024 (Cohort 1)72	Exposed Unexposed	130 997	16–30 mo	M-CHAT-R	EMR	↓ Lower risk of positive screen for parent-reported autism spectrum disorder in exposed children	• Maternal age • Insurance status • Maternal race/ethn (3.9% Asian, 16.0% Black, 10.8% Declined, 0.5% Native American, 36.5% Other, 8.2% Unknown, 24.0% White; 59.6% Hispanic, 23.9% non-Hispanic, 5.7% Declined, 10.8% Unknown) • Child sex • GA at delivery • Preterm birth • Child age at assessment	• Maternal age • Maternal race/ethn • Insurance status • Child sex • GA at delivery • Delivery mode • Child age at assessment	9
Firestein et al., 2024 (Cohort 2)72	Exposed Unexposed	101 201	16–30 mo	M-CHAT-R	Parental report	– No difference between the groups	• Maternal age • Maternal race/ethn (, 3.4% Asian, 10.1% Black, 16.4% Declined, 0.8% Native American, 0.3% Native Hawaiian, 28.3% Other, 40.8% White; 43.6% Hispanic, 42.3% non-Hispanic, 13.8% Declined, 0.3% Unknown) • Insurance status • Child sex • GA at delivery • Preterm birth • Child age at assessment	• Maternal age at delivery • Maternal race/ethn • Insurance status • Child sex • GA at delivery • Delivery mode • Child age at assessment	9
Vrantsidis et al., 202461	Exposed Unexposed	60 640	6–24 mo	IBQ-R-VSF (6mo) ECBQ (24mo)	Parental report	↑ IBQ-R-VSF: Greater regulatory behavior in exposed children – ECBQ: no differences between the groups	• Maternal age • Maternal race/ethn (1.1% Black, 2.8% East Asian, 2.3% Hispanic, 2.6% Indigenous, 0.6% 5.1% Multiracial/Other, Southeast Asian, 2.3% South Asian, 0.7% West Asian, 81.7% White, 0.8% Missing) • Education • Household income • Food security • Prepregnancy medical conditions • Child sex • GA at birth • Birthweight • Delivery mode • Child age at assessment	• Household SES • Prepregnancy maternal medical conditions • Trimester of infection • Symptoms severity • Child sex • Child age at assessment	9
Pinheiro et al., 202451	Exposed Unexposed	224 225	6–12 mo	SWYC	Phone interview	– No differences between the groups	• Education • Maternal SES (Brazilian Socioeconomic Classification Criteria) • Family conflict • Family substance use • Trimester of infection • Maternal ICU • GA at delivery • Birthweight • Delivery mode • Neonatal infection	• Food insecurity • Family conflict • Family substance use • Maternal depression • Delivery mode	8
Silva et al., 202378	Exposed Unexposed	27 27	1–12 mo	SWYC	Phone interview Home visit	↑ Higher risk of motor developmental delay and socioemotional alterations in the exposed group	• Food insecurity • Parental depression • Family violence • Parental substance abuse • Child sex • GA at delivery • Preterm birth	• GA at delivery	3
Werchan et al., 202465	Exposed Unexposed	50 117	6–12 mo	BITSEA	Parental report	– No differences between the groups	• Education • Household income • Maternal history of mental illnesses • Maternal prenatal stress • Child sex	• SES • (education, income) • Prior maternal mood/anxiety disorder • Postpartum depression	7
Ayesa-Arriola et al., 202335	Exposed Unexposed	21 21	6 wks	NBAS	Health professional	– No differences between the groups	• Maternal age • Education • Trimester of infection • Infection severity • Child sex • Birthweight • Birth length • Delivery mode • Preterm birth	• Maternal age • Child sex • GA at delivery • Child age at assessment	7
Hill et al., 202476	Exposed Unexposed	16 14	12 mo	SP-2 VABS-3	Parental report	– No differences between the groups	• Maternal age • Education • Household income • Infection severity • Vaccination status • Parity • Pregnancy complications • GA at delivery • Delivery Mode	None	7
Purpura et al., 202466	Exposed Unexposed	11 11	0–3 mo	NBAS SP-2 IBQ-R	Health professional and parental report	↓ NBAS: Lower scores in autonomic stability, number of smiles, and reflexes in exposed children – SP-2: No differences between the groups – IBQ-R: No differences between the groups	• Maternal age • Education • Maternal employment • Child sex • GA at delivery • Birthweight • Delivery mode	None	6
Bianco et al., 202336	Exposed Unexposed	63 110	6 mo	IBQ-R	Parental report	– No differences between the groups ↓Lower temperament with maternal post-natal self-reported stress	• Maternal age • Maternal race/ethn (6.4% Asian, 13.9% Black, 0.5% Native American, 1.2%, 5.8% Multiracial, Native Hawaiian, 22.5% Other, 11.0% Unknown, 38.7% White; 49.7% Hispanic, 44.5% non-Hispanic, 5.8% Unknown) • Education • Household income • Child sex • GA at delivery	• Maternal age • Education • Child sex • Child age at assessment	8
Fajardo-Martinez et al., 202442	Exposed Prepandemic	128 128	12–36 mo	Bayley-III	Health professional EMR	• 91.6% Typical ↑Higher risk of developmental delay in pandemic group	• Maternal age • Maternal race/ethn (9.0% Asian, 21.9% Black or Multiracial, 15.6% White; 15.6% Hispanic) • Education • Insurance status • Maternal medical history • Maternal medical conditions • GA at infection • Infection severity • Vaccination status • Child sex • GA at delivery • Birthweight • SGA • Preterm birth • Delivery mode • NICU admission • Neonatal complications	• Maternal age • Trimester of infection • Symptoms severity • Maternal fever during pregnancy • Maternal mental illness • Birthweight • Child sex • Delivery mode • Preterm birth	5
Munian, Hazra, and Ray, 202147	Exposed No control group	127 -	0–6 mo	DP3	EMR	• 100% Typical	• Child sex • GA at delivery • Birthweight • Delivery mode • Neonatal complications	None	5
Mand et al., 202487	Exposed No control group	110 -	12–31 mo	ePsychomotor development	Parental report	• 100% Typical	None	None	3
Carrasco Colom et al., 202375	Exposed No control group	55 -	12 mo	Bayley-III	Not specified	• 96% Typical (n = 53) • 2% Mild development deficit (n = 1) • 2% Unspecified developmental disorder (n = 1)	• Maternal age • GA at infection • Infection severity • ICU admission • Child sex • GA at delivery • Birthweight • Birth length • Preterm birth • SGA • Delivery mode • NICU admission • Breastfeeding	None	4
Orioli et al., 202474	Exposed No control group	35 -	6 mo	Bayley-III	Health professional	• Developmental delay prevalence:   - 26% Cognitive  - 63% Receptive language  - 31% Expressive language  - 43% Fine motor  - 49% Gross motor	• Housing conditions • Trimester of infection • Child sex • GA at delivery • Birthweight • Delivery mode • Neonatal infection	None	4
Rood et al., 202352	Exposed No control group	13 -	3 mo	fVan Wiechen scheme	Health professional	• 85% Typical (n = 11) • 15% unspecified neurodevelopmental delays (n = 2, one born preterm)	• Maternal age • Maternal comorbidities • Infection severity • Child sex • GA at delivery • Delivery mode	None	4
Apa et al., 202433	Exposed No control group	56 -	8–19 mo	CDI-WG	Parental report	• Language delay:  - 8.9% for sentence comprehension  - 12.5% for word comprehension  - 5.4% for words production  - 3.6% for gestures	• Trimester of infection • Child sex • Child age at assessment	None	5

Note: Unless otherwise stated, Exposed refers to prenatal exposure to SARS-CoV-2 infection, Pandemic group refers to both exposed and unexposed children born after March 2020, Education refers to maternal education and Income refers to household income.

↓ = significantly lower; ↑ = significantly higher; – = no significant difference

ABR= auditory brainstem response; ASQ= Ages and Stages Questionnaire; ASQ-SE= ASQ-Social-Emotional; ASSR= auditory steady-state response; BAEP= brainstem auditory evoked potentials; BITSEA= Brief Infant-Toddler Social and Emotional Assessment; CDI-WG= MacArthur Bates Communication Development Inventory-Words and Gestures; CLS= contralateral suppression; cUS= cranial ultrasound; DAYC-2= Developmental Assessment of Young Children-2; DDST= Denver Developmental Screening Test; DP3= Developmental Profile-3; DPOAE= Distortion Product Otoacoustic Emissions; ECBQ= Early Childhood Behavior Questionnaire; EEG= electroencephalography; EMR= electronic medical records; GA= gestational age; GMA= General Motor Assessment; HINE= Hammersmith Infant Neurological Examination; HNNE= Hammersmith Neonatal Neurological Examination; IBQ-R= Infant Behavior Questionnaire-Revised; IBQ-R-VSF= IBQ-R Very Short Form; ICD-10= International Classification of Diseases-10^th Revision; M-CHAT-R= Modified Checklist for Autism in Toddlers-Revised; mo= months; MOS-R= Motor Optimality Score-Revised; MRI= magnetic resonance imaging; NA= not available; NBAS= Neonatal Behavioral Assessment Scale; n_ctrl = number of controls; NICU= Neonatal Intensive Care Unit admission; NOS= Newcastle-Ottawa scale; OAE= Otoacoustic emissions; Prechtl= video-based method test for functional assessment of early movements repertoire; PTA= Pure Tone Average; Race/ethn= race and ethnicity; SP-2= Sensory Profile-2; SES= socioeconomic status; SGA= Small for Gestational Age; SWYC= Survey of Well-being of Young Children; TEOAE= Transitory Evoked Otoacoustic Emissions; VABS-3= Vineland Behavior Scale-3; wks= weeks; yrs= years.

^aThe exposed group included pregnant individuals who tested positive for SARS-CoV-2 during pregnancy or at the time of delivery and/or children who tested positive for SARS-CoV-2 as neonates, but results were reported as one group

^b28/34 participants were exposed to SARS-CoV-2 during pregnancy and 6/34 after birth, but results were reported as one group

^c5 neonates were not considered as they were infected with SARS-CoV-2 at birth

^d84/96 participants were exposed to SARS-CoV-2 during pregnancy and 12/96 after birth only, but results were reported as one group

^eHellbrüge and Pechstein developmental assessment

^fDutch equivalent of the Bayley-III assessment

Statistical analysis: meta-analysis

We conducted random-effects meta-analyses using Restricted Maximum Likelihood (REML) and inverse-variance weighting (metafor R package, version 4.0.4). Statistical significance was determined at α = 0.05. For dichotomous outcomes (i.e., neonatal auditory screening, categorical ASQ-3 and ASQ:SE-2), we extracted the number of individuals screening positive versus negative for auditory or developmental concerns to calculate risk ratios (RRs). For continuous outcomes (i.e., ASQ-3 scores), we synthesized standardized mean differences (SMDs) based on adjusted effect size estimates reported in the original publications. Neonatal auditory assessments (ABR and OAE) are typically performed within the first weeks of life, facilitating pooling across studies. Because ASQ-3 uses age-standardized, domain-specific cut-offs, we pooled studies across age groups. We assessed heterogeneity using the I² statistic, with values >50% indicating substantial heterogeneity, and evaluated potential publication bias through visual inspection of funnel plots and Egger’s tests.

RESULTS 

Systematic review: article selection and overview 

Our search yielded 11,925 articles of which 4,797 were duplicates. We screened 7,128 titles and abstracts, excluding 6,753 articles because they did not (i) investigate the outcomes of interest, or (ii) examine the exposure of interest. Full-text screening was performed on 375 articles. Seven additional articles were retrieved via cross-referencing. We identified 70 studies (Figure 1), published over 4 years (2021–2024). For an overview of the study characteristics, see Supplement 7, available online. For information on the covariates by exposure group, see Supplement 8, available online. For a list of the excluded studies at full-text screening stage, see Supplement 9, available online. We categorized studies by outcome into neurological (n = 7), neuroimaging (n = 12), motor development (n = 3), neurodevelopmental disorders (n = 2), audiological assessments (n = 29), and neurodevelopment (n = 35). Some studies reported more than one outcome. The median sample size was N = 117 (interquartile range: 44–340) with ages ranging 0–36 months.

Prenatal exposure to SARS-CoV-2 was based on nasal-swab polymerase chain reaction tests (PCR) (n = 37)^{17,19,20,30–63}, self-report (n = 2)^64,65 or rapid antigen test (n = 1).⁶⁶ Several studies employed multiple detection methods, including PCR and at least one other: serology^{21,22,27,67–75}, rapid antigen tests^76–78, self-report^67,79, or suspected infection based on clinical criteria⁷³ (see Supplement 7). Ten studies did not specify detection methods.^{18,28,80–87}

Only two studies examined prenatal COVID-19 vaccination as exposure^88,89, relying on self-report⁸⁸ and an unspecified detection method.⁸⁹ Most other studies did not report COVID-19 vaccination status, possibly because the vaccine was not available or recommended for pregnant individuals during their data collection periods.

For an overview of the studies’ methodologies and main findings, see Table 1.

Neurological assessment 

We identified seven studies on neurological outcomes (ages 0–3 months) (Table 1).^{43,45,50,75–77,84}

Of the three studies with control groups (total exposed n = 25; total unexposed n = 33)^43,76,84, one reported significantly lower scores on the Hammersmith Infant Neurological Examination (HINE) in the exposed group (exposed n = 5; unexposed n = 15), with differences in the reflex, orientation, and behavior subdomains⁸⁴, while the other two reported no differences between the groups. Similarly, two studies without control groups (total n = 126)^50,75 reported that neurological development was generally typical in infants prenatally exposed to SARS-CoV-2. Based on our predefined threshold, findings in the neurological domain were consistent, with 4 out of 5 studies (80%) reporting no adverse neurological effects of prenatal SARS-CoV-2 exposure.

Two additional studies investigating the effect of infection severity on neurological outcomes found a higher likelihood of abnormal neurological outcomes in children born to mothers with symptomatic SARS-CoV-2 infections (total n = 50) compared to children born to asymptomatic mothers (total n = 68).^45,77 However, these studies did not include a comparison group of unexposed children, and in one of the studies 67% of neonates were born preterm and 56% under emergency conditions.⁴⁵ Thus, the effect of infection severity remains inconclusive.

The quality of the evidence ranged from low (1 study) and moderate (4 studies) to high (2 studies) (Table 1).

Neuroimaging 

Twelve studies examined neuroimaging outcomes in children aged 0–9 months.^{27,45,46,50,53,69,75,77,79,83–85} Most studies reported structural findings utilizing cranial ultrasound (cUS) (n = 4)^27,53,69,75 or magnetic resonance imaging (MRI) (n = 3).^79,84,85 One study employed electroencephalography (EEG) to investigate suspected seizures.⁸³ Four studies combined techniques, utilizing cUS^45,46,50,77, MRI^45,46,50,77, and EEG^45,50 (Table 1).

One MRI study (exposed n = 47; unexposed n = 55) found increased gray matter volume in SARS-CoV-2 exposed infants (aged 6 weeks) compared to unexposed infants, particularly following third trimester exposure.⁷⁹ Moreover, a cohort study using cUS reported a higher risk of cerebral deep white matter abnormalities in prenatally exposed infants (8.9%; aged 6 months, n = 201) compared to unexposed infants (n = 18) but did not provide incidence data for the unexposed controls.⁶⁹ Sugak et al. (2021)²⁷ reported higher incidences of choroid plexus cysts and striatal vasculopathy using cUS in neonates (aged 1–5 days) prenatally exposed to SARS-CoV-2 (5.7%) compared to neonates born pre-pandemic (1.9%), but concluded that despite the significant difference, the majority (94.3%) of infants prenatally exposed to SARS-CoV-2 did not exhibit any atypical clinical manifestations.²⁷ These findings are in line with six small studies without control groups (total exposed n = 194)^{50,53,75,77,83,85}, which documented neuroimaging abnormalities using cUS, MRI, and/or EEG but found that neuroimaging results were normal in 70–100% of the SARS-CoV-2 exposed children. Based on our predefined threshold, neuroimaging findings were consistent, with 6 out of 8 studies (75%) reporting no evidence of atypical brain morphology or functioning in children prenatally exposed to SARS-CoV-2.

Three small MRI studies with and without control groups (total exposed n = 21; total unexposed n = 15) examined prenatally SARS-CoV-2 exposed infants who presented with neurological abnormalities^45,46 or were infected with SARS-CoV-2 at birth.⁸⁴ These studies reported differences in white matter, as well as reduced gray matter volume. It remains unclear whether these differences are attributable to intrauterine viral exposure, the infection at birth or other confounding morbidities.

The quality of the evidence ranged from low (2 studies) to moderate (10 studies) (Table 1).

Motor development

Three studies investigated motor development using the Motor Optimality Scale-Revised (MOS-R)^17,31,70 or HINE⁷⁰ (total exposed n = 168; total unexposed n = 143; aged 3–8 months) (Table 1). All three identified motor abnormalities in SARS-CoV-2 exposed infants, particularly in fidgety movements^31,70 and overall motor development.¹⁷ However, one of these studies included 6-month follow-up data and found no motor delays at that time.⁷⁰

Motor outcomes were also evaluated as subscales within questionnaires assessing global neurodevelopment across ten studies.^{22,35,42,47,52,71,74,75,78,87} Four small studies (total exposed n = 116, total unexposed n = 115, age 0–12 months)^22,35,74,78 reported motor abnormalities in children prenatally exposed to SARS-CoV-2. However, these abnormalities were primarily observed in preterm-born children^22,78, and no differences were observed at follow-ups at 7 and 13 months.²² The remaining six studies (total exposed n = 545; total unexposed n = 384; age 0–36 months)^{42,47,52,71,75,87} found no significant associations between prenatal exposure to SARS-CoV-2 and motor development delays. Based on our predefined threshold, findings in this domain were inconsistent (7/13 studies, 54%), suggesting that while prenatal SARS-CoV-2 exposure may be associated with early motor abnormalities, these effects appear to be transient and may be influenced by prematurity or methodological differences across studies.

The quality of the evidence ranged from low (2 studies) and moderate (8 studies) to high (3 studies) (Table 1).

Neurodevelopmental disorders 

Two US-based studies from the same group^40,41 investigated the association between prenatal exposure to SARS-CoV-2 and ICD-10 diagnoses of neurodevelopmental disorders (Table 1). After adjusting for race and ethnicity, insurance status, child sex, maternal age, and preterm status, exposed children were more likely (6.3%; n = 14/222) to receive a motor (e.g., specific developmental disorder of gross motor coordination) or language (e.g., receptive language disorder) disorder diagnosis than unexposed children (3.0%; n = 227/7550) in the first 12 months of life (Table 1).⁴¹ However, adjusting for length of hospital stay nullified the associations.⁴¹ The subsequent study comparing children prenatally exposed to SARS-CoV-2 (n = 883) to unexposed controls (n = 17,432), found a higher risk of neurodevelopmental diagnoses at age 12 months among exposed boys but not girls.⁴⁰ At age 18 months, this risk was no longer statistically significant.⁴⁰ Importantly, neither study controlled for family history of mental illness, a salient risk factor for adverse neurodevelopment. Based on our predefined threshold, findings in this domain were consistent, showing no long-term associations between prenatal SARS-CoV-2 exposure and diagnosed neurodevelopmental disorders.

The quality of the evidence was high (Table 1).

Audiological assessment 

We identified 29 studies reporting on audiological outcomes in children aged 0–36 months^{18–21,28,30,32,33,37,39,44,49,50,53–56,58–60,63,64,67,74,75,77,80–82} (Table 1). Audiological assessments were conducted by healthcare professionals (e.g., physician)^{18–21,28,30,32,33,37,39,44,58,59,63,64,67,75,80,81}, or retrieved from the electronic medical record (EMR).^49,77 Eight studies lacked data acquisition details.^{50,53–56,60,74,82}

Of the 15 studies with a control group^{18–20,28,30,39,44,49,53,58–60,64,67,80}, seven reported a significantly higher risk of failing the first auditory test in exposed compared to unexposed children (total exposed n = 883, total unexposed n = 1,769).^{19,20,28,30,58,64,80} The remaining eight studies (total exposed n = 1,862, total unexposed n = 2,836) found no significant differences between the groups.^{18,39,44,49,53,59,60,67} Follow-up results reported in 12 studies showed no differences in hearing abilities between the exposed and unexposed groups at the second^{18,20,28,30,44,49,53,58,59,64,67} or third¹⁹ screening, with most children showing typical hearing.

Of 14 studies without control groups^{21,32,33,37,50,54–56,63,74,75,77,81,82}, three reported auditory responses within typical ranges (total exposed n = 152).^50,54,74 In the remaining eleven studies (total exposed n = 2,851)^{21,32,33,37,55,56,63,75,77,81,82}, the risk of failing the first auditory test ranged from 1% to 22%. However, in at least three of these eleven studies, infants presented risk factors for hearing impairment (e.g., middle ear effusion, preterm birth).^21,81,82 The eleven studies that provided follow-up data reported no increased risk of hearing impairment in exposed children at follow-up.^{21,32,33,37,54–56,58,63,81,82}

Based on our predefined threshold, findings were consistent with a pattern of transient auditory impairment following prenatal SARS-CoV-2 exposure, with 23 out of 29 studies (79%) reporting either no initial differences or early abnormalities that resolved on follow-up.

Eleven studies conducted sensitivity analyses examining the effects of infection timing.^{19–21,28,30,32,33,39,49,50,81} Three large case-control studies (total exposed n = 615, total unexposed n = 496) found an increased risk of hearing impairments in children exposed during the second^20,30, or second and third trimesters.⁴⁹ However, the trimester-based sample sizes were small^30,49 or not provided.²⁰ The remaining eight studies found no significant differences based on exposure timing.^{19,21,28,32,33,39,50,81}

The quality of the evidence ranged from low (4 studies) and moderate (18 studies) to high (7 studies) (Table 1).

Neurodevelopmental outcomes

Thirty-five studies investigated neurodevelopmental outcomes in children aged 0–36 months, with sample sizes ranging from N = 5 to N = 2,487^{22,33–36,38,42,47,48,51–53,55,57,61,62,65,66,68,71–78,83,84,86–91} (Table 1). Neurodevelopmental assessments were parent-reported (n = 23)^{33,34,36,38,51,57,61,62,65,66,68,72,73,76,78,83,84,86–91}, conducted by trained personnel in clinical settings (n = 7)^{35,42,48,52,66,71,74}, retrieved from the EMR (n = 5)^{22,47,55,72,77} or unspecified (n = 1).⁷⁵

ASQ-3 and ASQ:SE-2

The Ages and Stages Questionnaire (ASQ-3) was the most commonly used tool, employed in 19 studies (total N = 5,435, total exposed n = 1,838).^{34,38,42,48,53,55,57,61,62,68,73,76,77,83,84,86,89–91}

Seventeen studies examined ASQ-3 categorically (i.e., typical development/developmental delay)^{34,42,48,53,55,57,61,62,68,73,77,83,84,86,89–91}, eight of which included unexposed controls.^{48,61,62,68,73,89–91} Six case-control studies (total exposed n = 721; total unexposed n = 3,087; aged 3–24 months) found no differences in the risk of developmental delays between the exposed and unexposed groups.^{48,61,62,68,89,91} The other two studies reported significantly higher risk of personal-social delays in exposed children (total exposed n = 495; total unexposed n = 487; aged 12–32 months).^73,90 However, one of these two studies linked developmental delays to severe SARS-CoV-2 infections only⁷³, and the other study combined prenatally exposed (n = 84/96) and postnatally infected (n = 12/96) children⁹⁰, obscuring the specific effects of prenatal SARS-CoV-2 exposure.

The nine other categorical ASQ-3 studies (total exposed n = 597; aged 3–36 months) did not include a control group and found prevalences of developmental delay in exposed children ranging from 0% to 36%.^{34,42,53,55,57,77,83,84,86} Developmental delays were most frequently observed in the fine motor^{34,42,53,57,77}, problem-solving^{34,53,57,77,86}, and personal-social^34,53 subdomains. Ten further studies reported mean ASQ-3 scores and standard deviations (total exposed n = 603; total unexposed n = 1,324; aged 3–12 months).^{38,48,53,61,62,68,76,83,84,89} Of those, only two reported significantly lower scores in exposed (total n = 25) compared to unexposed (total n = 23) children^38,76, while the other eight did not. Interestingly, while Shuffrey and colleagues⁶⁸ found no differences in ASQ-3 scores between exposed and unexposed children born in the pandemic, they reported lower mean gross motor, fine motor and personal-social scores in pandemic-born children (n = 255) compared to pre-pandemic controls (n = 62).

Three studies investigated social-emotional development using the ASQ:SE-2^61,62,90 (total exposed n = 249 total unexposed n = 1,121; aged 12–32 months). Two of these studies found no differences between the exposed and unexposed groups, while one identified a significantly higher risk of social-emotional delays in the exposed (n = 96) compared to the unexposed (n = 243) group.⁹⁰ However, 12.5% of the exposed children in this study were infected at birth, and it is unclear to what extent the results are specific to prenatal SARS-CoV-2 exposure.⁹⁰

Based on our predefined threshold, categorical ASQ-3 findings were consistent, with 6 out of 8 controlled studies (75%) reporting no evidence of developmental delays following prenatal SARS-CoV-2 exposure. Continuous ASQ-3 findings were also consistent, with 8 out of 10 studies (80%) reporting no significant differences in mean developmental scores between exposed and unexposed children. ASQ:SE-2 findings were largely consistent, with 2 out of 3 studies (67%) reporting no social-emotional delays; the one study reporting delays included a subset of children infected postnatally, limiting interpretability.

The quality of the evidence ranged from low (3 studies) and moderate (8 studies) to high (8 studies) (Table 1).

Other neurodevelopmental outcomes

Neurodevelopment, including behavioral, motor, cognitive, and social-emotional development, was further assessed using other measures (Table 1).^{22,33,35,36,42,47,51,52,61,65,66,71,72,74–76,78,87}

Five studies on behavioral and temperament outcomes (total exposed n = 171; total unexposed n = 796; aged 1–24 months)^{35,36,61,66,76} found no differences between exposed and unexposed children.

Eight studies investigating developmental delays in children aged 0–36 months using the Bayley-III or translated equivalent (exposed n = 231)^42,52,74,75; the Developmental Assessment of Young Children (DAYC) (exposed n = 112; unexposed n = 256)⁷¹; the Hellbrüge and Pechstein developmental assessment (exposed n = 110)⁸⁷; the Developmental Profile-3 (DP3) (exposed n = 127)⁴⁷; or the Denver Developmental Screening Test (DDST) (exposed n = 33, unexposed n = 67)²² further reported largely typical development following prenatal SARS-CoV-2 exposure.

Social-emotional development was assessed in three studies, two of which used the Survey of Well-being of Young Children (SWYC) (total exposed n = 251, total unexposed n = 252; aged 1–12 months)^51,78 and one used the Brief Infant-Toddler Social and Emotional Assessment (BITSEA) (exposed n = 50, unexposed n = 117; aged 6–12 months).⁶⁵ One study reported delays in exposed children,⁷⁸ while the other two found no group differences.^51,65

One study using the parent-reported Modified Checklist for Autism in Toddlers (M-CHAT) in two large cohorts (Cohort 1 Exposed n = 130 and Cohort 1 Unexposed n = 997; Cohort 2 Exposed n = 101 and Cohort 2 Unexposed n = 201; aged 16–30 months) reported no association between prenatal SARS-CoV-2 exposure and higher rates of positive autism screens.⁷²

Lastly, two studies without control groups (total n = 91; aged 6–19 months) identified language delays in receptive language⁷⁴ and word and sentence comprehension³³ in children prenatally exposed to SARS-CoV-2.

Based on our predefined threshold, findings across behavioral, cognitive, and social-emotional domains were largely consistent with no persistent effects of prenatal SARS-CoV-2 exposure. The quality of the evidence ranged from low (2 studies) and moderate (8 studies) to high (8 studies) (Table 1).

COVID-19 vaccination and neurodevelopment

Two studies investigating the impact of prenatal COVID-19 vaccination on neurodevelopment used the ASQ-3 (total exposed n = 1,845; total unexposed n = 896; aged 12–18 months) (Table 1) and found no differences between the exposed and unexposed groups.^88,89

Based on our predefined threshold, both studies consistently reported no neurodevelopmental differences following prenatal COVID-19 vaccination. The quality of the evidence was high (Table 1).

Meta-analysis

Audiological assessment

The audiological results are presented in Figure 2. Ten studies reported at least two subsequent auditory evaluations in exposed and unexposed children.^{18–20,28,30,44,49,58,64,67} Sample sizes ranged from n = 34 to n = 942 in the exposed group and from n = 35 to n = 955 in the unexposed group. We found a higher risk of failing the first auditory screening in exposed compared to unexposed children [RR = 2.01, 95% CI, 1.39–2.91] (Figure 2a). Heterogeneity was high (I² = 84.1%). No significant differences between the groups were found for the second auditory screening [RR = 1.40, 95% CI, 0.75–2.63] (Figure 2b). Heterogeneity was low (I² = 30.1%). Publication bias is unlikely (Supplement 10).

Figure 2.

Forest plots for the first (a) and second (b) auditory screening (i.e., ‘pass’ vs ‘referral’) with 95% confidence intervals (CIs).

Note: In the study by Ahmed et al. (2024), 3/43 subjects (7%) did not attend a second screening. Since the exposure of the missing subjects is unknown, we reported the full sample for the second screening (i.e., 43 instead of 40)

ASQ-3

The categorical ASQ-3 meta-analysis results, based on eight studies^{48,61,62,68,73,89–91} and our unpublished data (Table S2), are presented in Figure 3a-e. Sample sizes ranged from n = 57 to n = 357 in the exposed group and from n = 79 to n = 1,559 in the unexposed group. We found higher risk of developmental delays in fine motor [RR = 1.55, 95% CI, 1.14–2.10] and problem-solving [RR = 1.32, 95% CI, 1.01–1.74] skills in exposed compared to unexposed children. There were no significant differences between the groups for any other ASQ-3 subscale, with risk ratios ranging from 0.84 (95% CI, 0.49–1.40) to 1.39 (95% CI, 0.99–1.95); all confidence intervals overlapped 1.0. Heterogeneity was low (I²<27%). Publication bias is unlikely for all subdomains except communication (Supplement 11).

Figure 3.

Forest plots for the categorical Ages and Stages Questionnaire (ASQ)-3 (a-e) and ASQ:SE-2 (f) outcomes with 95% confidence intervals (CIs).

Note: Plots are shown for the (a) communication, (b) gross motor, (c) fine motor, (d) problem-solving, and (e) personal-social subdomains of the ASQ-3.

The ASQ-3 mean score meta-analysis results (i.e., means and SDs), based on six studies^{38,48,61,62,68,89} and our unpublished data, are reported in Supplement 12. Sample sizes ranged from n = 9 to n = 161 in the exposed group and from n = 9 to n = 800 in the unexposed group. We found no significant differences between the exposed and unexposed groups on any subdomain, with standardized mean differences ranging from 0.91 (95% CI, 0.78–1.07) to 1.06 (95% CI, 0.95–1.17); all confidence intervals overlapped 1.0.

ASQ:SE-2

The categorical ASQ:SE-2 meta-analysis results, based on three studies^61,62,90, are presented in Figure 3f. Sample sizes ranged from n = 57 to n = 92 in the exposed group and from n = 78 to n = 707 in the unexposed group. We found no significant differences between the exposed and unexposed groups [RR = 1.26, 95% CI, 0.44–3.60]. The heterogeneity was high (I² = 79.8%). Publication bias is likely (Supplement 10).

DISCUSSION

The evidence on the association between prenatal SARS-CoV-2 exposure and child neurodevelopment is inconsistent, and the risk of residual confounding is high, particularly due to socioeconomic and psychosocial factors as well as parental psychopathology. The literature is further limited by small sample sizes and short follow-up periods. Across all identified studies, the quality of evidence ranged from low to high, with most studies being of moderate quality. Below, we discuss the most consistent and well-supported findings and their clinical implications, as well as potential mechanisms underlying the associations.

Auditory impairment

In our meta-analysis of 10 studies examining hearing impairment within the first 18 months of life, we found that prenatal SARS-CoV-2 exposure was associated with an increased risk of failing the initial newborn auditory screening. Reassuringly, the differences in auditory impairment between the exposed and unexposed groups were no longer present at subsequent hearing assessments. While false-positive results on initial hearing tests are common,⁹² it is possible that our findings point to transient sensorineural hearing loss (SNHL), a condition previously linked to other prenatal infections (e.g., TORCH infections).⁹³ Viruses can damage inner ear hair cells, as well as auditory organs, pathways, and centers, either directly or via the immune reactions they trigger.⁹³ Some studies have also attributed initial hearing impairments to middle ear effusion,^21,81 which may result from fluid accumulation in the middle ear, a condition that has been linked to infections and that typically resolves within one month from birth.⁹⁴ While the mechanisms underlying the association between prenatal exposure to SARS-CoV-2 and early hearing impairment remain to be elucidated, our findings suggest no long-term adverse effects of prenatal SARS-CoV-2 exposure on auditory functions.

Neuroimaging

Most neuroimaging studies on prenatal SARS-CoV-2 exposure included small samples, and some only included prenatally exposed children who also presented with neurological abnormalities and/or acute SARS-CoV-2 infection at birth. These studies are further limited by short follow-up periods (≤9 months) and the lack of control groups. Nonetheless, the research reassuringly demonstrates that, while some prenatally SARS-CoV-2 exposed children show white matter and other structural brain changes, most children prenatally exposed to SARS-CoV-2 do not present with atypical clinical manifestations in brain morphology or functioning.

Developmental delay

Our meta-analysis of nine studies investigating developmental delay at ages 3–32 months indicated that children prenatally exposed to SARS-CoV-2 were at higher risk of delays in fine motor and problem-solving skills than their unexposed peers. Fine motor skills, such as grasping, object manipulation, or drawing, involve the use of smaller muscles and depend on well-coordinated and highly efficient brain networks that quickly relay real-time sensory input from various sensory cortices to the motor cortex.⁹⁵ Prenatal SARS-CoV-2 exposure and the resulting immune activation may impact the development of such intricate networks. However, as outlined above, more neuroimaging research is required to better understand these potential effects. While the results of our meta-analysis indicate that prenatal SARS-CoV-2 exposure is linked to fine motor and problem-solving skills in the first year of life, two longitudinal studies suggest that these delays may be transient.^22,91 This idea is in line with work by Edlow and colleagues who reported that children prenatally exposed to SARS-CoV-2 had a higher risk of neurodevelopmental disorders in the first year of life but not at age 18 months.^40,41 Nevertheless, since fine motor skills have been linked to language development and autism spectrum disorder,⁹⁶ and problem-solving skills have been implicated in the risk of neurodevelopmental and mood disorders,⁹⁷ our findings highlight the importance of continued follow-up of children prenatally exposed to SARS-CoV-2 to assess the scope of these developmental differences.

Confounding

In studying the effects of prenatal SARS-CoV-2 exposure on neurodevelopment, it is crucial to consider factors that may influence the likelihood of contracting SARS-CoV-2 and that also increase the risk of adverse neurodevelopment. Unfortunately, few studies were able to adequately adjust for confounding (see Table 1).

SARS-CoV-2 disproportionately affected economically and socially marginalized individuals⁹⁸, who are more likely to experience adverse social determinants of health (e.g., economic hardship), which in turn are linked to adverse child development.⁹⁹ Yet, fewer than half of the reviewed studies accounted for race/ethnicity or socioeconomic status. Furthermore, maternal psychopathology, which is one of the most salient risk factors of adverse neurodevelopment and a factor linked to SARS-CoV-2 infection risk,¹⁰⁰ was only considered by a quarter of the studies. Additionally, individuals infected with SARS-CoV-2 during pregnancy, particularly those with severe COVID-19, may have other pre-existing chronic illnesses (e.g., diabetes) or lifestyle factors that contribute to neurodevelopmental risk in the child.¹⁰¹ However, fewer than 10% of studies controlled for maternal comorbidities, and only two took lifestyle factors (e.g., prenatal drug use) into account.^73,89 Lastly, genetic vulnerability, which may underlie the predisposition to SARS-CoV-2 and the risk of psychopathology in the child thereby contributing to the observed associations,¹⁰² remains unexamined. Thus, our ability to draw causal inferences about the association between prenatal exposure to SARS-CoV-2 infection and adverse neurodevelopment is greatly diminished given the general lack and overall variability of confounding factors in the included studies.

Clinical implications

The evidence for a causal relationship between prenatal SARS-CoV-2 exposure and adverse neurodevelopment is currently limited. While we observed associations with early hearing impairments, fine motor and problem-solving skills, these effects appear to be transient, resolving over time. It is important to recognize that the absence of statistically significant associations does not confirm the absence of an effect,¹⁰³ particularly given the limitations of the current literature (e.g., short follow-up durations, heterogeneous measurement tools, and variability in confounder adjustment). Nonetheless, the results may offer some reassurance to parents, clinicians, and policymakers.

Many neurodevelopmental conditions are not typically diagnosed until school age. As such, the relatively short follow-up periods in the existing studies may be insufficient to fully capture the long-term impact of prenatal SARS-CoV-2 exposure. Inflammatory cascades triggered by maternal immune activation could, in theory, exert subtle and enduring effects on synaptic development, myelination, or cortical organization, particularly when compounded by psychosocial stressors or underlying genetic vulnerability. While the current findings do not indicate sustained developmental impairments, long-term consequences cannot yet be precluded. Our findings point to the importance of rigorous longitudinal studies with extended follow-up to clarify the extent, nature, persistence, and clinical relevance of early developmental findings.

Beyond direct viral exposure, the pandemic profoundly disrupted everyday life impacting social interactions, employment, care-as-usual (e.g., mother-infant separation at birth) and mental health. Notably, several studies suggested that children born during the pandemic display poorer neurodevelopmental outcomes than pre-pandemic cohorts, regardless of SARS-CoV-2 exposure status.^40,68,79 Thus, it will be crucial to examine and address the unique developmental challenges faced by the pandemic generation.

This study represents the first systematic review and meta-analysis to examine the associations between prenatal exposure to SARS-CoV-2 infection and COVID-19 vaccination and a broad range of neurodevelopmental outcomes. It has several methodological strengths. A comprehensive literature search was conducted by an experienced medical librarian. To enhance inclusivity, we included studies published in any language and verified their accuracy through multiple iterations of forward and backward translation. Study selection, data extraction, quality and risk of bias assessments were performed rigorously to ensure validity and accuracy. We included a wide range of study designs (e.g., prospective, retrospective, and cross-sectional) and conducted meta-analyses for auditory and behavioral outcomes. Formal assessments indicated no publication bias for audiological measures and most ASQ-3 domains, but publication bias cannot be ruled out for other neurodevelopmental domains. We incorporated all relevant outcomes from our own unpublished cohort regardless of statistical significance to minimize selective reporting bias.

However, the clinical implications of the review’s findings are constrained by the quality of the included studies. Many studies had small sample sizes, limited follow-up durations, and inadequate confounder selection. The predominance of cross-sectional and retrospective designs limits causal inference, given the unclear temporality, recall bias, and missingness. Most studies did not account for broader contextual factors related to the pandemic (e.g., maternal stress, health and childcare disruptions, financial hardship), making it difficult to disentangle direct effects of viral exposure from broader pandemic-related influences.

There was substantial heterogeneity across all domains stemming from differences in exposure definition (e.g., timing and severity of maternal SARS-CoV-2 infection), child age at assessment, measurement tools, study design, population characteristics, and approaches to confounder adjustment. Formal investigation of heterogeneity (e.g., stratification, subgroup or meta-regression analyses) was largely precluded by the small number of studies per outcome measure, underscoring the need for harmonized protocols, standardized assessments, and transparent reporting in future research.

Further limitations included the absence of native speaker validation for two non-English studies, which may have introduced minor misinterpretation risk. Follow-up durations were limited, with children assessed up to a maximum age of 36 months, precluding detection of later-onset neurodevelopmental conditions such as ADHD. Additionally, the effect of infection timing, despite its potential importance, could not be adequately explored due to the scarcity of available data. Finally, our systematic literature search concluded in November 2024. To preserve methodological integrity, studies published thereafter were not incorporated into the discussion or interpretation of the results.

Collectively, these limitations highlight the need for large, longitudinal studies with extended follow-up, careful confounder adjustment, and standardized outcome measurements to clarify the long-term neurodevelopmental consequences of prenatal SARS-CoV-2 infection and COVID-19 vaccination exposure.

We observed transient auditory impairments and delays in fine motor and problem-solving skills following prenatal SARS-CoV-2 exposure. However, due to the heterogeneity and quality of the evidence, as well as the potential for unmeasured confounding, these associations are unlikely to be causal. We found no link between prenatal SARS-CoV-2 exposure and neurological or neuroimaging outcomes. Additionally, no adverse neurodevelopmental effects were reported following prenatal exposure to COVID-19 vaccination. Comprehensive, longitudinal studies across diverse populations will be crucial to fully understand the long-term consequences of both prenatal SARS-CoV-2 exposure and the broader impacts of the COVID-19 pandemic.

Data sharing

This meta-analysis did not involve the collection of new data and instead analyzed data from previously published studies. Data included in the meta-analysis and scripts used for the analysis can be found at https://github.com/asrommel/SARS-CoV-2_review