COVID‐19 vaccination around the time of conception and risk of placenta‐mediated adverse pregnancy outcomes

Abstract

Introduction

Although numerous studies have documented no association between COVID‐19 vaccination during pregnancy and maternal and fetal health outcomes, fewer studies have evaluated fetal health effects after COVID‐19 vaccination around the time of conception and early pregnancy, a time when maternal exposures may affect early placentation and the subsequent risk of placenta‐mediated adverse pregnancy outcomes.

Material and Methods

We used province‐wide databases in Ontario to conduct a population‐based cohort study including all live and stillbirths ≥20 weeks' gestation with a last menstrual period (LMP) between April 1 and December 31, 2021. We deterministically linked birth registry data to the vaccine registry for all 80 253 eligible pregnancies; 31 209 (38.9%) received ≥1 dose of the COVID‐19 vaccine around the time of conception or first trimester. Using Cox regression, we estimated propensity score weighted hazard ratios (aHR) and 95% confidence intervals (CI) for associations between ≥1 dose of mRNA COVID‐19 vaccine during the periconceptional/first trimester exposure window (28 days before the LMP to the end of first trimester) and study outcomes: hypertensive disorders (gestational hypertension, preeclampsia, eclampsia), placental abruption, preterm birth (<37 weeks), small‐for‐gestational‐age (SGA) birth (<10th percentile), and stillbirth.

Results

COVID‐19 vaccination around the time of conception or first trimester was associated with a small increased risk of hypertensive disorders in pregnancy in exposed versus unexposed individuals (7.4% vs. 6.1%; aHR: 1.10, 95% CI: 1.03–1.17), mostly attributed to gestational hypertension (5.0% vs. 4.1%; aHR 1.13, 95% CI: 1.05–1.22). There was no increased risk of preeclampsia (1.8% vs. 1.5%; aHR 1.08, 95% CI: 0.95–1.22), eclampsia (0.1% vs. 0.1%; aHR: 1.12, 95% CI 0.65–1.95), placental abruption (0.8% vs. 1.0%; aHR: 0.77, 95%CI: 0.65–0.91), preterm birth (8.0% vs. 8.9%; aHR: 0.92, 95%CI: 0.87–0.97), SGA birth (8.9% vs. 9.3%; aHR: 1.00, 95%CI: 0.95–1.06), or stillbirth (0.4% vs. 0.6%; aHR: 0.66, 95%CI: 0.52–0.82).

Conclusions

This population‐based Canadian study provides additional evidence evaluating COVID‐19 vaccine administration around the start of pregnancy. While we identified no association with most placenta‐mediated outcomes, we report a slight increase in the rate of gestational hypertension. This could be a true association or attributed to residual confounding. Further research is needed to verify.

This population‐based cohort study of 80 253 pregnancies identified a small increase in the risk of hypertensive disorders of pregnancy associated with COVID‐19 vaccination around conception or the first trimester of pregnancy and no increased risk of other placenta‐mediated outcomes.

Key points.

This population‐based cohort study of 80 253 pregnancies identified a small increase in the risk of hypertensive disorders of pregnancy associated with COVID‐19 vaccination around conception or the first trimester of pregnancy and no increased risk of other placenta‐mediated outcomes.

1. INTRODUCTION

During the COVID‐19 pandemic, pregnant people were shown to be at higher risk of severe COVID‐19 illness ¹ , ² and many countries, including Canada, prioritized this population for COVID‐19 immunization when vaccines became available. ³ Although COVID‐19 vaccination during pregnancy was shown to be effective in preventing severe COVID‐19 illness in pregnant people ⁴ , ⁵ and their infants, ⁶ , ⁷ uptake among pregnant people has lagged behind that of non‐pregnant adults. ⁸ , ⁹ COVID‐19 vaccine hesitancy during pregnancy has been attributed in part to concerns regarding possible adverse effects on the fetus of COVID‐19 vaccines. ¹⁰

While there is now a substantial body of evidence supporting the safety of COVID‐19 vaccination during the second and third trimesters in terms of pregnancy and neonatal outcomes, ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ the evidence evaluating pregnancy outcomes after COVID‐19 vaccination around the time of conception and during the first trimester remains limited. Of the existing large, observational studies that have evaluated periconceptional and early pregnancy vaccination, the majority have focused on miscarriage or congenital anomalies, finding no associations. ¹⁶ , ¹⁷ , ¹⁸ Despite the critical importance of placental development early in pregnancy, ¹⁹ and evidence that SARS‐CoV‐2 infection in pregnancy can affect the placenta, ²⁰ , ²¹ few studies have evaluated the association between placenta‐mediated outcomes and COVID‐19 vaccination during the periconceptional period or early pregnancy.

Placenta‐mediated outcomes are a heterogenous but related set of clinically important conditions, including preeclampsia, placental abruption, intrauterine growth restriction leading to small‐for‐gestational‐age birth, and stillbirth, ²² affecting up to 20% of pregnancies and contributing to acute and long‐term morbidity. ²³ To support informed vaccine decision‐making for those planning pregnancy or early in pregnancy, additional evidence on the safety of COVID‐19 vaccination with respect to placenta‐mediated outcomes is needed. To address this, we carried out a population‐based cohort study of births in Ontario, Canada, to evaluate the association between receiving an mRNA vaccine (primary series or first booster dose) during the periconceptional period or first trimester of pregnancy and the risk of later placenta‐mediated pregnancy outcomes.

2. MATERIAL AND METHODS

We received ethical approval for this study from the Children's Hospital of Eastern Ontario Research Ethics Board and followed published guidance for conducting studies of COVID‐19 vaccination during pregnancy ²⁴ and reporting observational studies using routinely collected data sources. ²⁵

2.1. Study design, population, and data sources

This was a retrospective population‐based cohort study, conducted in the province of Ontario, which is Canada's most populous province. Pregnant people—as high‐risk individuals for severe COVID‐19—were prioritized in the province's COVID‐19 vaccination program in April 2021. ³ In April 2021, pregnant individuals were recommended to receive two doses of a mRNA COVID‐19 vaccine 8 weeks apart during any stage of pregnancy (i.e., any trimester). ³ In December 2021, pregnant individuals were also recommended to receive a COVID‐19 booster dose 3 months after the second dose of the primary series. ²⁶

Pregnant individuals with a last menstrual period (LMP) date between April 1, 2021 (i.e., initial vaccine recommendation for pregnant people) and December 31, 2021 (i.e., end of data availability) were eligible for inclusion in the study. We excluded records of individuals who gave birth at <20 weeks' gestational age and with a birth weight <500 g, or who had a pregnancy termination, as those events are not systematically collected in the province's birth registry. The follow‐up period began at 20 weeks' gestational age. A small number of individuals who received a non‐mRNA COVID‐19 vaccine were also excluded from the study.

To identify the study population and ascertain information on the study outcomes and other covariates (e.g., demographic and pregnancy characteristics, pre‐existing medical conditions, and lifestyle factors), we used data from the provincial birth registry (Better Outcomes Registry & Network [BORN] Ontario), ²⁷ which collects pregnancy, birth, and newborn information for approximately 140 000 births each year in Ontario. BORN systematically collects pregnancy and birth information for all live births and stillbirths ≥20 weeks' gestation or birth weight ≥500 g from hospitals, birth centers, and midwifery practice groups across the province. ²⁷ Data collection procedures for the BORN registry are described elsewhere ²⁷ ; briefly, pregnancy, birth, and other health information are collected from clinical encounters within the healthcare system. Agreement between clinical information in the BORN registry and patient charts and clinical administrative hospital databases is high. ²⁷ , ²⁸ We deterministically linked the study population to Ontario's COVID‐19 vaccination database (COVaxON), which contains information on all COVID‐19 immunization events in the province, including vaccine product, date of immunization, and number of doses received. We also linked the study population to Statistics Canada's 2016 Census and the Ontario Marginalization Index using maternal residential postal codes. ²⁹ The Census includes information on residential area class (i.e., rural vs. urban) and area‐based household income quintile, while ON‐Marg provides information on four area‐based measures of social and economic marginalization. Lastly, to identify laboratory‐confirmed SARS‐CoV‐2 infections before or during pregnancy, we also linked the study population to the Public Health Case and Contact Management Solution (CCM). ³⁰ Detailed summaries of all data sources are provided in Table S1.

2.2. Study measures

2.2.1. Periconceptional and first trimester COVID‐19 vaccination

We extracted information on receipt of COVID‐19 vaccine doses during the periconceptional period or the first trimester of pregnancy from the COVaxON database. Individuals who received any dose of an mRNA COVID‐19 vaccine (i.e., dose 1, 2, and/or first booster dose) during the periconceptional period (i.e., from 28 days prior to the LMP date up to 13 days after LMP date) or during the first trimester (i.e., from 14 days after the LMP date to 97 days of gestation) were considered exposed (see Figure 1 for exposure window of periconceptional/first trimester COVID‐19 vaccination). We excluded individuals who received any dose of mRNA COVID‐19 vaccine during a 30‐day ‘washout’ period before the start of the periconceptional period and those who received all doses prior to the ‘washout’ period as they were no longer eligible for vaccination during the exposure windows of interest (Figure 1). The unexposed comparison group was comprised of pregnant individuals who did not receive any doses of COVID‐19 vaccine during the periconceptional period or the first trimester, although they could have received doses prior to the 30‐day ‘washout’ period or during the second or third trimesters of pregnancy (see Figure 2 for illustration of vaccination timing relative to pregnancy in the study population).

FIGURE 1.

Exposure window for measuring periconceptional and first trimester COVID‐19 vaccination.

FIGURE 2.

Frequency of timing of first second, and/or first booster dose of COVID‐19 vaccine, relative to pregnancy. *The ‘Exposed’ group in the figure refers to those who received a first, second and/or first booster dose of COVID‐19 vaccine during the periconceptional period/first trimester (i.e., the exposure window). **The wash out period refers to a 30‐day timeframe before the start of the exposure window.

2.2.2. Placenta‐mediated adverse pregnancy outcomes

We studied the following placenta‐mediated pregnancy outcomes: hypertensive disorders of pregnancy (including gestational hypertension, preeclampsia, Hemolysis, Elevated liver enzymes, and Low Platelets [HELLP] syndrome, and eclampsia), placental abruption, preterm birth before 37 completed weeks of gestation, small‐for‐gestational‐age (SGA) at birth (i.e., <10th percentile of the sex‐specific birth weight for gestational‐age distribution based on a Canadian reference standard), ³¹ and stillbirth (i.e., intrauterine fetal death at ≥20 weeks of gestation). Detailed definitions of all study outcomes are provided in Table S2.

2.2.3. Covariates

To adjust for a range of variables potentially associated with receiving a COVID‐19 vaccine dose in the periconceptional period or first trimester and/or study outcomes, we used propensity score methods. The propensity score models included the following variables: maternal age at delivery (continuous); pre‐pregnancy body mass index ≥30 kg/m² (vs. <30); self‐reported smoking status (yes/no) or substance use during pregnancy (yes/no); public health unit region (seven regions); pre‐existing maternal health conditions (composite of: asthma, chronic hypertension, diabetes, heart disease, thyroid disease; yes/no); parity (nulliparous vs. non‐nulliparous); multi‐fetal pregnancy (yes/no); rural/urban residence; neighborhood income quintiles; neighborhood marginalization quintiles (four dimensions: households and dwellings, material resources, age and labour force, racialized and newcomer populations); calendar month of the LMP (categorical); and first prenatal care visit in the first trimester (yes/no). Additional details on covariate definitions can be found in Table S2.

2.3. Statistical analysis

We examined the distribution of baseline characteristics overall and by exposure group (received any COVID‐19 vaccine dose in the periconceptional/first trimester exposure window versus did not receive any dose in the periconceptional/first trimester exposure window). These were compared with weighted distributions (using stabilized inverse probability of treatment weights) using absolute standardized differences, where a value of <0.1 was considered indicative of a balanced distribution between the two groups. ³²

We used multiple imputation to address missing covariate values; five imputed datasets were generated using a fully conditional specification. Stabilized inverse probability of treatment weights were constructed from propensity scores representing the predicted probability of receiving one or more doses of a COVID‐19 vaccine in the periconceptional/first trimester exposure window, conditional on the covariates listed in Table S2. Additional information about stabilized inverse probability of treatment weight derivation can be found in Appendix S1.

We used Cox proportional hazards regression models with gestational age in days as the time scale to estimate hazard ratios (HR) with 95% confidence intervals (CI). Follow‐up for study outcomes started at 20 weeks of gestation and continued until the end of pregnancy for all outcomes except for preterm birth, for which follow‐up ended at the time of a preterm birth event or at 36 weeks + 6 days of gestation (on pregnancy day 258) for term births. Since the follow‐up period started after the periconceptional/first trimester exposure window, the exposure was treated as a time‐invariant variable in the analysis. We calculated unadjusted HRs for each outcome and computed adjusted HRs by using the stabilized inverse probability of treatment weights in the Cox models. ³² Each weighted outcome model was fitted using each of the five imputed datasets, and the results were combined using the “MIANALYZE” procedure in SAS Version 9.4 (SAS Institute, Cary NC).

In sensitivity analyses, we accounted for the potential impacts of any subsequent doses of COVID‐19 vaccine that were received during pregnancy but after the periconceptional/first trimester exposure window by introducing a time‐varying exposure for the additional doses using an extended Cox model; robust sandwich variance estimation was used to account for statistical dependence across repeated observations due to the time‐varying variable in this analysis.

We conducted two additional subgroup analyses to further explore the robustness of our exposure definition. First, we considered subgroups of the original exposed group based on the specific timing of the COVID‐19 vaccine receipt during the exposure window: (1) at least one dose of COVID‐19 vaccine received in the periconceptional period (with no COVID‐19 vaccine doses received in the first trimester); (2) at least one dose of COVID‐19 vaccine received in the first trimester of pregnancy (with no COVID‐19 vaccine doses received in the periconceptional period); and (3) doses of COVID‐19 vaccine received in both the periconceptional and the first trimester periods. Each of these exposure subgroups was compared with the original comparison group.

Second, we stratified the exposure group based on the number of COVID‐19 vaccine doses received within the original exposure window, resulting in two subgroups: (1) one dose of COVID‐19 vaccine received in the exposure window, and (2) two or more doses of COVID‐19 vaccine received in the exposure window. The same comparison group was used for each exposure subgroup.

Finally, to consider the potential influence of residual confounding in this observational study, we calculated E‐values using the “EValue” package in R (version 4.3.3) to quantify the strength of unmeasured confounding that would be necessary to explain away any observed findings. ³³ , ³⁴

3. RESULTS

We identified 81 610 live births and stillbirths—representing 80 253 pregnant individuals—who were eligible for periconceptional or first trimester vaccination and, therefore, included in the study. Of all live births and stillbirths, 31 701 (38.8%) occurred among individuals who received a COVID‐19 vaccine dose in the periconceptional period or first trimester (Figure 3).

FIGURE 3.

Flow diagram for selection of participants into analytic sample. The periconceptional period occurs from 28 days prior to the LMP date up to 13 days after LMP date. The first trimester occurs from 14 days after the LMP date to 97 days of gestation.

Characteristics of the study population overall and by COVID‐19 vaccination status during the periconceptional period or first trimester are presented in Table 1, with additional characteristics shown in Table S3. Overall, exposed pregnancies occurred earlier in calendar time than unexposed pregnancies. Compared with the unexposed group, individuals vaccinated in the exposure window of interest were more likely to be ≥30 years old, nulliparous, have a pre‐existing medical condition, have the first prenatal care visit in the first trimester, and reside in neighborhoods with higher median family income and lower material resources. Exposed individuals were less likely to report smoking and substance use during pregnancy and less likely to reside in rural areas (Table 1). Laboratory‐confirmed COVID‐19 infection before or during pregnancy was comparable between the two groups (Table S3). Standardized differences indicated good balance following propensity score weighting (Table 1; Figure S1).

TABLE 1.

Characteristics of the study population overall and by receipt status of COVID‐19 vaccination during the periconceptional period or the first trimester of pregnancy.

Characteristics	Unweighted							Stabilized inverse probability of treatment weighted a
	All live births and stillbirths (n = 81 610)		Received any COVID‐19 vaccine dose during the periconceptional period/first trimester (n = 31 701)		Did not receive a COVID‐19 vaccine dose during the periconceptional period/first trimester (n = 49 909)		Standardized difference b	Received any COVID‐19 vaccine dose during the periconceptional period/first trimester	Did not receive a COVID‐19 vaccine dose during the periconceptional period/first trimester	Standardized difference b
	n	% c	n	% c	n	% c		% c	% c
Maternal age at delivery (years)
<25	6957	8.5	1868	5.9	5089	10.2	0.16	8.2	9.0	0.03
25–29	19 556	24.0	6646	21.0	12 910	25.9	0.12	23.5	24.4	0.02
30–34	32 152	39.4	13 347	42.1	18 805	37.7	0.09	40.8	37.9	0.06
35–39	18 766	23.0	8254	26.0	10 512	21.1	0.12	23.3	22.6	0.02
≥40	4179	5.1	1586	5.0	2593	5.2	0.01	4.2	6.1	0.09
Mean (SD)	32.1 (5.0)		32.6 (4.6)		31.7 (5.2)			32.0 (4.9)	32.1 (5.2)
Date of last menstrual period (LMP)
April 2021	9441	11.6	4211	13.3	5230	10.5	0.09	11.3	11.5	0.01
May 2021	9964	12.2	5333	16.8	4631	9.3	0.23	11.9	12.1	0.01
June 2021	7791	9.5	4014	12.7	3777	7.6	0.17	9.4	9.5	0.00
July 2021	7037	8.6	3264	10.3	3773	7.6	0.10	8.5	8.7	0.00
August 2021	7117	8.7	2244	7.1	4873	9.8	0.10	9.3	8.9	0.01
September 2021	9260	11.3	2230	7.0	7030	14.1	0.23	12.3	11.4	0.03
October 2021	10 334	12.7	3465	10.9	6869	13.8	0.09	12.5	12.7	0.00
November 2021	10 252	12.6	3392	10.7	6860	13.7	0.09	12.2	12.5	0.01
December 2021	10 414	12.8	3548	11.2	6866	13.8	0.08	12.4	12.7	0.01
Parity
0 (nulliparous)	35 785	43.8	14 598	46.0	21 187	42.5	0.01	43.9	44.0	0.00
1 (primiparous)	28 018	34.3	11 460	36.2	16 558	33.2	0.07	34.0	34.3	0.01
≥2 (multiparous)	17 508	21.5	5517	17.4	11 991	24.0	0.06	22.1	21.7	0.01
Missing	299	0.4	126	0.4	173	0.3	0.16
Multiple birth
No	78 909	96.7	30 716	96.9	48 193	96.6	0.02	96.7	96.7	0.00
Yes	2701	3.3	985	3.1	1716	3.4	0.02	3.3	3.3	0.00
Pre‐existing medical condition d
No	72 459	88.8	27 756	87.6	44 703	89.6	0.06	88.9	88.8	0.00
Yes	9151	11.2	3945	12.4	5206	10.4	0.06	11.1	11.2	0.00
Smoked during pregnancy
No	74 953	91.8	29 609	93.4	45 344	90.9	0.09	93.0	93.6	0.02
Yes	4863	6.0	1324	4.2	3539	7.1	0.13	7.0	6.4	0.02
Missing	1794	2.2	768	2.4	1026	2.1	0.02
Substance use during pregnancy e
No	74 815	91.7	29 462	92.9	45 353	90.9	0.08	94.4	94.8	0.02
Yes	3833	4.7	1037	3.3	2796	5.6	0.11	5.6	5.2	0.02
Missing	2962	3.6	1202	3.8	1760	3.5	0.01
Maternal BMI (kg/m2)
<30.0	55 947	68.6	21 931	69.2	34 016	68.2	0.02	78.4	78.5	0.00
≥30.0	15 578	19.1	6173	19.5	9405	18.8	0.02	21.6	21.5	0.00
Missing	10 085	12.4	3597	11.3	6488	13.0	0.05
First prenatal care visit in the first trimester
Yes	73 447	90.0	29 060	91.7	44 387	88.9	0.09	93.3	93.5	0.01
No	5065	6.2	1416	4.5	3649	7.3	0.12	6.7	6.5	0.01
Missing	3098	3.8	1225	3.9	1873	3.8	0.01
Neighborhood median family income quintiles
Quintile 1—lowest	17 107	21.0	5846	18.4	11 261	22.6	0.10	21.7	21.3	0.01
Quintile 2	16 566	20.3	6286	19.8	10 280	20.6	0.02	20.4	20.5	0.00
Quintile 3	17 526	21.5	6952	21.9	10 574	21.2	0.02	21.5	21.7	0.00
Quintile 4	16 562	20.3	6913	21.8	9649	19.3	0.06	20.5	20.5	0.00
Quintile 5—highest	12 910	15.8	5577	17.6	7333	14.7	0.08	15.8	15.9	0.00
Missing	939	1.2	127	0.4	812	1.6	0.12
Rural residence
No	69 521	85.2	27 677	87.3	41 844	83.8	0.10	85.4	85.5	0.00
Yes	11 674	14.3	4022	12.7	7652	15.3	0.08	14.6	14.5	0.00
Missing	415	0.5	<6	0.0	413	0.8	0.13
Material deprivation quintile f
Quintile 1—least deprived	17 687	21.7	7845	24.7	9842	19.7	0.12	21.7	21.9	0.01
Quintile 2	16 014	19.6	6656	21.0	9358	18.8	0.06	19.8	19.9	0.00
Quintile 3	14 964	18.3	5854	18.5	9110	18.3	0.01	18.5	18.7	0.00
Quintile 4	14 982	18.4	5539	17.5	9443	18.9	0.04	18.8	18.7	0.00
Quintile 5—most deprived	16 184	19.8	5430	17.1	10 754	21.5	0.11	21.2	20.8	0.01
Missing	1779	2.2	377	1.2	1402	2.8	0.12
Residential instability quintile f
Quintile 1—least deprived	15 865	19.4	6067	19.1	9798	19.6	0.01	19.5	19.8	0.01
Quintile 2	14 796	18.1	5907	18.6	8889	17.8	0.02	18.5	18.5	0.00
Quintile 3	14 977	18.4	5975	18.8	9002	18.0	0.02	18.8	18.7	0.00
Quintile 4	14 924	18.3	6028	19.0	8896	17.8	0.03	18.7	18.7	0.00
Quintile 5—most deprived	19 269	23.6	7347	23.2	11 922	23.9	0.02	24.6	24.4	0.01
Missing	1779	2.2	377	1.2	1402	2.8	0.12
Dependency quintile f
Quintile 1—least dependent	26 181	32.1	10 492	33.1	15 689	31.4	0.04	32.4	32.6	0.00
Quintile 2	16 869	20.7	6643	21.0	10 226	20.5	0.01	21.0	21.0	0.00
Quintile 3	13 466	16.5	5285	16.7	8181	16.4	0.01	16.8	16.9	0.00
Quintile 4	12 205	15.0	4680	14.8	7525	15.1	0.01	15.4	15.3	0.00
Quintile 5—most dependent	11 110	13.6	4224	13.3	6886	13.8	0.01	14.3	14.2	0.00
Missing	1779	2.2	377	1.2	1402	2.8	0.12
Ethnic concentration quintile f
Quintile 1—least concentration	11 479	14.1	4334	13.7	7145	14.3	0.02	15.1	14.9	0.01
Quintile 2	12 932	15.8	5140	16.2	7792	15.6	0.02	16.3	16.2	0.00
Quintile 3	14 028	17.2	5854	18.5	8174	16.4	0.06	17.6	17.5	0.00
Quintile 4	17 414	21.3	7135	22.5	10 279	20.6	0.05	21.6	21.7	0.00
Quintile 5—most concentration	23 978	29.4	8861	28.0	15 117	30.3	0.05	29.5	29.7	0.00
Missing	1779	2.2	377	1.2	1402	2.8	0.12
PHU region of residence
South West	10 718	13.1	3796	12.0	6922	13.9	0.06	13.5	13.3	0.01
Central West	16 138	19.8	6314	19.9	9824	19.7	0.01	20.1	20.0	0.00
Central East	22 895	28.1	8483	26.8	14 412	28.9	0.05	28.3	28.4	0.00
Greater Toronto Region	15 581	19.1	6453	20.4	9128	18.3	0.05	18.8	19.2	0.01
Eastern	10 991	13.5	4914	15.5	6077	12.2	0.10	13.6	13.6	0.00
North West	1431	1.8	550	1.7	881	1.8	0.00	1.9	1.8	0.01
North East	2918	3.6	1068	3.4	1850	3.7	0.02	3.7	3.6	0.00
Missing	938	1.1	123	0.4	815	1.6	0.12

Abbreviations: BMI, body mass index; LMP, last menstrual period date; PHU, Public Health Unit; SD, standard deviation.

^a There are no missing values shown in the inverse probability‐of‐treatment‐weighted distributions of baseline characteristics, because multiple imputation was used to address missing values. Column percentages and weights for the weighted study population were based on imputation dataset 1.

^b Absolute standardized difference comparing those who received a dose of COVID‐19 vaccine in the periconceptional/first trimester exposure window and those who did not; standardized difference >0.10 indicates an imbalance in the distribution of the baseline characteristic between these two exposure groups.

^c Column percentages.

^d Composite of: asthma, chronic hypertension, diabetes, heart disease, thyroid disease. Sum of individual conditions does not equal the total number of individuals with any individual condition, as categories were not mutually exclusive (individual conditions shown in Table S3).

^e Self‐reported cannabis, opioid or alcohol use during pregnancy.

^f These are dimensions of the Ontario Marginalization Index. More information about this index is included in the Appendix.

Models applying stabilized inverse probability of treatment weighting that compared pregnancies vaccinated during the periconceptional period or during the first trimester to unvaccinated pregnancies showed no increase in risk of placental abruption (0.8% vs. 1.0%; aHR: 0.77, 95%CI: 0.65–0.91), preterm birth (8.0% vs. 8.9%; aHR: 0.92, 95%CI: 0.87–0.97), small‐for‐gestational age at birth (8.9% vs. 9.3%; aHR: 1.00, 95%CI: 0.95–1.06), or stillbirth (0.4% vs. 0.6%; aHR: 0.66, 95%CI: 0.52–0.82) (Table 2). We observed a slight increase in the risk of hypertensive disorders of pregnancy for individuals vaccinated during the periconceptional period or first trimester compared to unvaccinated individuals (7.4% vs. 6.1%; aHR: 1.10, 95% CI: 1.03–1.17). Within conditions contributing to hypertensive disorders of pregnancy, we observed no increase in risk of preeclampsia (aHR: 1.07, 95%CI 0.95–1.21), HELLP syndrome (aHR: 1.01, 95%CI 0.66–1.55), or eclampsia (aHR: 1.10, 95%CI 0.63–1.92) associated with periconceptional or first trimester vaccination; however, there was a 13% increased risk of gestational hypertension observed (aHR: 1.13, 95%CI 1.05–1.22). Results did not change substantively after further adjustment for additional doses received later in pregnancy (Table 2).

TABLE 2.

Association between any COVID‐19 vaccine dose received during the periconceptional period or first trimester of pregnancy and placenta‐mediated adverse outcomes.

Placenta‐Mediated Adverse Outcome	Received any COVID‐19 vaccine dose during the periconceptional period/first trimester (n = 31 701 infants) n (%) with outcome	Did not receive a COVID‐19 vaccine dose during the periconceptional period/first trimester (n = 49 909 infants) n (%) with outcome	Unadjusted hazard ratio a (95% CI)	Adjusted hazard ratio a , b (95% CI)	Adjusted hazard ratio a , b , c (95% CI), with further adjustment for additional dose(s) of COVID‐19 vaccine in pregnancy
Hypertensive disorders of pregnancy d , e	2301 (7.4)	2992 (6.1)	1.21 (1.14, 1.28)	1.10 (1.03, 1.17)	1.09 (1.02, 1.16)
Gestational hypertension d , f	1561 (5.0)	1991 (4.1)	1.23 (1.15, 1.32)	1.13 (1.05, 1.22)	1.12 (1.04, 1.21)
Preeclampsia d	566 (1.8)	726 (1.5)	1.22 (1.09, 1.37)	1.08 (0.95, 1.22)	1.07 (0.95, 1.21)
HELLP Syndrome d	39 (0.1)	55 (0.1)	1.11 (0.73, 1.67)	1.01 (0.66, 1.55)	1.01 (0.66, 1.55)
Eclampsia d	24 (0.1)	33 (0.1)	1.14 (0.67, 1.93)	1.12 (0.65, 1.95)	1.10 (0.63, 1.92)
Placental abruption d	247 (0.8)	494 (1.0)	0.79 (0.67, 0.92)	0.77 (0.65, 0.91)	0.77 (0.65, 0.91)
Preterm birth <37 weeks	2544 (8.0)	4463 (8.9)	0.89 (0.84, 0.94)	0.92 (0.87, 0.97)	0.92 (0.87, 0.98)
Small‐for‐gestational‐age infant	2807 (8.9)	4627 (9.3)	0.95 (0.91, 1.00)	1.00 (0.95, 1.06)	1.00 (0.95, 1.06)
Stillbirth	129 (0.4)	321 (0.6)	0.63 (0.51, 0.78)	0.66 (0.52, 0.82)	0.66 (0.53, 0.82)

^a Hazard ratios were estimated using an extended Cox model with a time‐fixed exposure variable (since periconceptional and first trimester vaccinations occur before the beginning of the follow‐up period, i.e., 20 weeks' gestation).

^b Modeling was performed on five multiple imputation datasets and adjusted using stabilized inverse probability of treatment weights derived from a propensity score model including the variables listed in Table S2.

^c Models included a time‐varying indicator variable signaling whether a dose of COVID‐19 vaccine was received after the periconceptional and first trimester exposure window. Since the comparison group contains individuals that may have been vaccinated later in pregnancy, but not in the exposure window, this variable applies to both the exposed and comparison group.

^d Percentages for hypertensive disorders of pregnancy and placental abruption are expressed as a percentage of all pregnancies in each exposure group (31 209 exposed and 49 044 unexposed individuals), and not as a percentage of all infants.

^e There were 449 records with hypertensive disorders in pregnancy, but the exact disorder was not specified. These were excluded from the analyses by individual disorder.

^f Records with preeclampsia were excluded from models related to gestational hypertension.

Sensitivity analyses evaluating COVID‐19 vaccination during the periconceptional period (Table S5), first trimester (Table S6), or both the periconceptional period and the first trimester of pregnancy (Table S7) similarly showed a slight increase in the risk of gestational hypertension. No association with other placenta‐mediated outcomes was identified—except for HELLP syndrome, which was only associated with COVID‐19 vaccine exposure when the vaccine was given during both the periconceptional period and the first trimester (aHR 1.95, 95%CI 1.09–3.47). We observed similar results for those who received one dose or two doses during the periconceptional period or first trimester (Table S8). Covariate balance was demonstrated for sensitivity analyses in Figures S2–S6, although some imbalance in LMP was noted in Figures S4 and S6.

E‐values indicated that a confounder or set of confounders would have to be associated with a 1.4‐fold increase in the risk of hypertensive disorders of pregnancy, a 1.9‐fold increase in the risk of placental abruption, a 1.4‐fold increase in the risk of preterm birth, and a 2.4‐fold increase in the risk of stillbirth to explain the observed effect estimates (Table S9).

4. DISCUSSION

Results from this large population‐based cohort study showed no increase in the risk of most adverse placenta‐mediated outcomes associated with periconceptional or first trimester exposure to the COVID‐19 vaccine. Although we identified no association with placental abruption, preterm birth, small‐for‐gestational age, or stillbirth, we identified a slight increase in the risk of hypertensive disorders of pregnancy associated with periconceptional vaccination, which was mostly attributed to gestational hypertension. These findings were confirmed in sensitivity analyses by the gestational window of vaccination and the number of doses. Our findings align with those of previous smaller cohort studies. ³⁵ , ³⁶ One U.S. study looked at a range of pregnancy outcomes, stratified by the trimester of exposure, finding a small (RR ≤ 1.10) but non‐significant increase in the rate of gestational hypertension and preeclampsia‐eclampsia‐HELLP syndrome associated with first trimester exposure to the COVID‐19 vaccine. ³⁵ A study in Israel showed a small, non‐significant increase (OR 1.30) in hypertensive disorders of pregnancy associated with COVID‐19 vaccination. ³⁶ However, it is equally worth noting that additional larger cohort studies of COVID‐19 vaccination during any stage of pregnancy have reported no association with preeclampsia or gestational hypertension. ³⁷ , ³⁸ , ³⁹

The pathophysiology of placenta‐mediated outcomes, including gestational hypertension, is not fully understood, likely involving multiple mechanisms, including impaired placentation, angiogenic imbalance, immune dysregulation, oxidative stress, mitochondrial dysfunction, and epigenetic modifications. ²² , ⁴⁰ Inflammatory and immune responses play a critical role in the development of hypertensive disorders in pregnancy. ⁴⁰ In hypertensive disordered pregnancies, the normal balance of the maternal immune system is disrupted, including increased activation of dendritic cells, natural killer cells, T helper cells, and B cells. ⁴¹ B cells additionally produce agnostic autoantibodies, which can stimulate vasoconstriction and exacerbate inflammation and oxidative stress. ⁴² Abnormal complement activation during pregnancy can amplify inflammation and neutrophil activation at the placenta. ⁴³ This imbalance in the maternal immune system can not only disrupt placental development but also contribute to systemic vascular dysfunction, leading to a proinflammatory shift and poor placental development. ⁴⁰

While it is possible that exposure to vaccination during early placentation could somehow trigger immunologic or other events leading to gestational hypertension, given the modest association observed and the observational study design, it is also possible that this finding could be the result of residual confounding. Given the low E‐values reported, this scenario is especially plausible. Furthermore, recent studies have shown that COVID‐19 illness during pregnancy may be associated with a two‐fold increase in hypertensive disorders, ⁴⁴ , ⁴⁵ resulting in a four‐fold increase in the risk of preterm birth. ⁴⁴ Given the evidence showing a reduced risk of COVID‐19 related complications associated with COVID‐19 vaccination during pregnancy, ¹¹ , ⁴⁶ the benefits of vaccination are likely to outweigh risks. However, a possible association with gestational hypertension cannot be ruled out, and further research on periconceptional COVID‐19 vaccination and gestational hypertension remains useful.

Several previous smaller studies of placental‐mediated outcomes have observed protective effects of COVID‐19 vaccination on placental health, including a lower risk of placentitis, ⁴⁷ focal perivillous fibrin deposits, and avascular fibrotic villi among vaccinated pregnancies. ⁴⁸ However, relatively few large‐scale studies of placenta‐mediated outcomes have been conducted. Smaller studies have shown no impact of COVID‐19 vaccination on the development of syncytiotrophoblast in the placenta ⁴⁹ or oxidative status of the placenta. ⁵⁰ A cohort study of 788 pregnant individuals prospectively monitored markers of placental functioning from the first trimester of pregnancy, showing no changes in placental growth factor or anti‐angiogenic factors, like soluble fms‐like tyrosine kinase 1. ³⁷

Recent national guidelines in the US and Canada recommend vaccination with an updated SARS‐CoV‐2 variant KP.2 formulated vaccine. ³ , ⁵¹ Vaccination within 6 months of the start of pregnancy or during pregnancy is likely needed to provide adequate protection against severe illness ⁴⁶ and is strongly recommended during pregnancy. ⁵² Because results from this study found no increased risk of most adverse placenta‐mediated outcomes, those who are planning pregnancy or in the early stages of pregnancy could receive an updated vaccine during this period.

This study used population‐based data from the provincial birth registry deterministically linked to the COVID‐19 vaccination database, which contains information on all COVID‐19 immunization events in the province, thereby including a large number of vaccinated pregnant individuals and minimizing potential selection bias and exposure misclassification. Moreover, these data were linked to Statistics Canada's 2016 Census and the Ontario Marginalization Index to obtain relevant sociodemographic information and to adjust for potential confounders.

However, this study also has some limitations. Uptake of COVID‐19 vaccination in the periconceptional period and first trimester was lower in younger pregnant individuals, those who reported smoking, and those who lived in lower‐income neighborhoods with higher material resources. These characteristics (and potentially other unmeasured factors) tend to be associated with lower vaccine uptake ⁸ , ⁹ and poorer pregnancy and birth outcomes; therefore, there may have been some residual confounding despite having adjusted for potential confounders using a propensity score‐based approach. Estimated E‐values indicated that modest (i.e., 2‐ to 3‐fold) to weak (i.e., <2‐fold) associations with an unmeasured confounder could explain the risk estimates observed in our study. In terms of COVID‐19 infection ascertainment, the number of cases might have been underestimated as individuals may not have sought COVID‐19 testing, especially later in the pandemic when testing became less common. In our study, the exposed group had earlier LMP dates on average than the comparison group, so the exposed group may have been more likely to seek testing and, therefore, have more confirmed COVID‐19 infection cases compared to the unexposed group. However, our adjustment for calendar time at LMP should have accounted for this. The Ontario birth registry does not systematically capture pregnancies ending prior to 20 weeks' gestation; therefore, the risk of pregnancy loss before 20 weeks' associated with COVID‐19 vaccination could not be evaluated; however, previous case–control studies have not found such association with COVID‐19 vaccination during the first trimester. ¹⁷ , ¹⁸ Lastly, our study only included mRNA vaccines as the use of other COVID‐19 vaccine types in Canada during the study period was uncommon.

5. CONCLUSION

This population‐based study provides additional evidence evaluating the safety of the COVID‐19 vaccine administered around the start of pregnancy, finding limited evidence in support of an increased risk of most adverse placenta‐mediated outcomes. While we identified a slight increase in the risk of gestational hypertension, the risk could be attributed to residual confounding, and the risk of hypertensive disorders is likely greater for COVID‐19 illness during pregnancy. These results, therefore, support continued provision of COVID‐19 mRNA vaccines to pregnant people to protect against COVID‐19‐associated complications.

AUTHOR CONTRIBUTIONS

Fell and Bruce had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Fell, Regan, Platt, Gravel, Bruce. Acquisition, analysis, or interpretation of data: Regan, Bruce, Lavin Venegas, Török, Platt, Gravel, Alton, Dimanlig‐Cruz, Shah, Barrett, Walker, El‐Chaâr, K Wilson, Sprague, MacDonald, Buchan, Kwong, SE Wilson, Håberg, Okun, Dhinsa, Dunn, Fell. Drafting of the manuscript: Regan, Lavin Venegas, Fell, Bruce. Critical revision of the manuscript for important intellectual content: Regan, Bruce, Lavin Venegas, Török, Platt, Gravel, Alton, Dimanlig‐Cruz, Shah, Barrett, Walker, El‐Chaâr, K Wilson, Sprague, MacDonald, Buchan, Kwong, SE Wilson, Håberg, Okun, Dhinsa, Dunn, Fell. Statistical analysis: Bruce, Regan, Fell, Gravel, Platt. Obtained funding: Fell, Sprague. Administrative, technical, or material support: Fell, Venegas, Török, Walker, Sprague, Dunn. Supervision: Fell, Regan, Platt. Other—Clinical input: El‐Chaar, Walker, Shah, Okun, Barrett, Kwong, Håberg.

FUNDING INFORMATION

This study was supported by funding from the Public Health Agency of Canada, through the Vaccine Surveillance Reference Group and the COVID‐19 Immunity Task Force. Dr. Håberg was partly funded by the Norwegian Research Council (project numbers 324312 and 262700), and Nordforsk project number 135876. Dr. Regan was partly funded by a National Institute of Allergy and Infectious Diseases Loan Repayment Program award (L40AI178819).

CONFLICT OF INTEREST STATEMENT

Dr. Regan reported receiving grants from the National Institutes of Health/National Institute of Allergy and Infectious Diseases and the EuroQol Research Foundation for research unrelated to the submitted work and membership on a Moderna Data Safety Monitoring Board. When this project was funded and initiated, Dr. Fell was employed by the University of Ottawa and had an academic appointment at the Children's Hospital of Eastern Ontario Research Institute; although she maintains adjunct appointments at both institutions, she is now employed by Pfizer and works on an unrelated topic. Dr. K. Wilson reported being CEO of CANImmunize Inc. and being on advisory boards for Medicago and Moderna. No other disclosures were reported.

ROLE OF THE FUNDER/SPONSOR

The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

DISCLAIMER

Parts of this article are based on data and information compiled and provided by BORN Ontario and the Ontario Ministry of Health; however, the analyses, conclusions, opinions, and statements expressed herein are solely those of the authors and do not reflect those of the funding or data sources; no endorsement is intended or should be inferred.

ETHICS STATEMENT

This study received ethical approval from the Children's Hospital of Eastern Ontario Research Ethics Board on April 20, 2021 (REB protocol number: 21/05PE). and followed published guidance for conducting studies of COVID‐19 vaccination during pregnancy and for reporting observational studies using routinely collected data sources.

Supporting information

Data S1

Regan AK, Bruce L, Lavin Venegas C, et al. COVID‐19 vaccination around the time of conception and risk of placenta‐mediated adverse pregnancy outcomes. Acta Obstet Gynecol Scand. 2025;104:1883‐1896. doi: 10.1111/aogs.70025

DATA AVAILABILITY STATEMENT

Restrictions apply to the availability of these data, which are not publicly available and cannot be shared publicly due to privacy restrictions. Data can be requested from the Better Outcomes Registry & Network. See https://www.bornontario.ca/en/index.aspx.