Side Effects of COVID-19 Vaccines in Pregnant and Lactating Mexican Women and Breastfed Infants: A Survey-Based Study

COVID-19 vaccines’ safety has been extensively studied; however, further analysis is required in pregnant women, nursing mothers, and breastfed infants. Our aim was to compare the extension and severity of self-reported COVID-19 vaccine side effects in pregnant and breastfeeding women, and breastfed infants. In this cross-sectional study, COVID-19-vaccinated subjects were enrolled using an online survey in Mexico. Women were classified by pregnancy and breastfeeding status at the time of vaccination (n = 3167). After the first or only dose, there was a trend toward fewer systemic effects in pregnant women (p = 0.06). BNT162b2 (Pfizer–BioNTech) had a higher frequency of local symptoms in pregnancy. Lactating women experienced fewer local symptoms after the first or single dose (p = 0.04) and the opposite occurred after the second dose (p = 0.001). ChAdOx1 (AstraZeneca) increased the chances of developing both local and systemic symptoms after the first dose but decreased them after the second dose. The severity was similar across groups, although the result of lack of association in pregnancy requires studies with a larger sample size. Irritability was the most reported symptom in breastfed infants. This study contributes to the knowledge about the side effects in pregnant and lactating women, and breastfed babies.


Introduction
COVID-19 is a respiratory disease that paralyzed the world for several months by causing the death of more than six million people. The speed at which we travel the world in the 21st century caused a pandemic in just three months [1]. Furthermore, today's technology allows us to trace the virus from its first reports in a seafood market in Wuhan, China to the first case in every country. We now know the nucleotide sequence of the virus [2] and therefore vaccines were developed at unprecedented rates. In March 2021, a year after the first COVID-19 case in December 2019, vaccination started. In Mexico, from December to February 2021, vaccination was prioritized for healthcare workers, followed by the population aged 60 and over. From April to May 2021, schoolteachers and people aged 50 to 59 were vaccinated. From May to June 2021, people aged 40 to 49 were vaccinated, and finally from June 2021 to present, the remaining population was vaccinated [3]. To date, 13,337,416,815 doses have been administered worldwide, of which 222,580,922 doses have been administered in Mexico [1]. Pregnant women have an elevated risk of severe COVID-19 [4] and traces of mRNA vaccines have been detected in breast milk [5,6]; therefore, studying the safety of COVID-19 vaccines in these groups of the population is imperative, given they are already being exposed. Plenty of studies have reported the safety and efficacy of COVID-19 vaccines [7][8][9][10][11][12][13], but these are all clinical trials conducted on healthy participants with no participation of pregnant or breastfeeding women. However, observational studies had shown vaccines were safe in these populations, several of which were based on self-reported symptoms post-vaccination [8,. We have previously reported the side effects caused by seven COVID-19 vaccines in the Mexican population [3]. However, the extension and severity of side effects was not compared between pregnant women and non-pregnant/non-lactating women. Furthermore, in the present study we describe the extension and severity of side effects reported by breastfeeding women compared to non-pregnant/non-lactating women, which has not been shown before. Pregnant women's vaccination began in May 2021 in Mexico, and antibodies have been reported in the newborn [14] because they are transferred through the breast milk [19,22,43]. Additionally, local and systemic symptoms are expected in pregnant and lactating women [16], and breastfed infants [21]. Reports on the extension and severity of COVID-19 vaccine side effects in pregnant and lactating women are few. A recent literature review showed five studies describing post-vaccination side effects in lactating women [5,21,[43][44][45][46] and three in breastfed infants [5,21,46]. However, none of them corresponds to the Mexican population. Shimabukuro and colleagues (2021) have reported preliminary findings on mRNA COVID-19 vaccine safety in pregnant women [18], and several agencies only recommend the use of BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna) [47]. There are reports from several countries indicating pregnant women's attitudes toward COVID-19 vaccination during pregnancy, which demonstrate that fear of the side effects, lack of safety data, and mistrust in vaccines are the most reported reasons against vaccination [48][49][50][51][52]. Pregnant women not being included in the clinical trial might have built confusion, which led to mistrust in the vaccine even when updated evidence was being published. Social medial and political concerns have also been reported to lead to a negative perception towards vaccination.
In the present study we compared the extension and severity of self-reported COVID-19 vaccine side effects in pregnant and breastfeeding women. We also explored the presentation of side effects in breastfed infants.

Materials and Methods
This is a secondary analysis performed after a cross-sectional study conducted in Mexico from 12 August to 3 September 2021 (5 months after starting vaccination in the population other than healthcare workers, and 3 months after starting vaccination in pregnant women) focused on the female population. The database consisted of participants with a vaccine against COVID-19, mainly from Nuevo Leon, Mexico City, and the State of Mexico. The snowballing sampling technique was used for recruiting subjects. The survey was shared on social media platforms such as Facebook and WhatsApp; therefore, it was not possible to estimate the number of invited people. Word-of-mouth campaigns were also used with no proportional quotas by sociodemographic variables or by type of vaccine. Participants did not receive any kind of incentive. The self-applied electronic survey was designed in Spanish with the software tools QuestionPro (Survey Analytics LLC, San Francisco, CA, USA) and Google Forms (Google, Mountain View, CA, USA). The survey took an average of 5 min to complete [3]. For the present study, a minimum sample size of 385 was calculated based on a 50% estimated frequency of at least one local or systemic side effect assuming a 95% confidence level with a precision of 5% [53,54]. A total of 3204 women were eligible: 110 pregnant, 363 breastfeeding, and 2694 non-pregnant/non-breastfeeding; a total of 37 had to be excluded for simultaneously presenting pregnancy and lactation. No sample size was estimated for breastfed infants and the analysis of side effects in this population was exploratory. The protocol was approved by the Institutional Review Board (or Ethics Committee) of the Mexican Social Security Institute (R2021-1909-106). The study followed the Declaration of Helsinki's guidelines for research on human subjects, and all the participants signed their informed consent digitally before filling in the questionnaire. Participation was entirely voluntary, and withdrawal was allowed at any time without the need to justify the decision. There was no personal data collected that might enable the retrospective identification of the participant.

Study Variables
The variables included in the present study were vaccine data, type and number of doses, use of medication prior to vaccination to prevent symptoms (yes, no), and type of side effect (the participant chose from a list made with short-term side effects reported in the literature). The side effects were self-reported, similar to many observational studies [8,[20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39]. Dose 1 and 2 side effects were questioned in separate sections. The extension of the side effect was categorized as none, local (e.g., pain at the injection site), systemic (e.g., fever), or both. The severity of the side effect was classified as absent, mild, moderate, or severe, according to the need for taking medication to relieve symptoms, interruption of activities or missing work, having to visit a doctor, an emergency room, or required hospitalization [3] (Table S1). Information on severity was available only for the first or only dose.
Other variables were comorbidity (prediabetes, diabetes, hypertension, chronic renal failure, chronic obstructive lung disease, asthma, an immune disease, cancer, a cerebrovascular disease, other), history of allergies, COVID-19 infection (had symptoms consistent with COVID-19 disease and was positive with PCR or rapid nasal swab antigen test), sex, age, schooling, occupation, place of residence, and smoking. The nutritional status was assessed using a validated body mass index-body size pictorial method [55] that indicated low weight (shape A), normal weight (shapes B and C), overweight (shape D), and obesity (shapes E to J).

Statistical Analysis
Frequencies were obtained for the categorical variables, and means and standard deviations were obtained for the non-categorical variables. The chi-square test was used to compare the frequency distribution of sociodemographic, medical, and vaccination history characteristics among non-pregnant/non-lactating, pregnant, and lactating women. The Fisher's exact test or the chi-square test was used to compare the frequency distribution of the extension of vaccine side effects between non-pregnant/non-lactating and pregnant or lactating women. A multivariate ordinal logistic regression was used for estimating odds ratios (OR) and 95% confidence intervals (CI) between the type of vaccine (independent variable) and the extension of symptoms (dependent variable; 0 = absence of side effects, 3 = presence of local and systemic side effects) adjusting for potential confounders in each population of interest (pregnant and lactating), separately. Additionally, for severity of symptoms, 0 = absent and 3 = severe.

Pregnant vs. Breastfeeding vs. Non-Pregnant/Non-Breastfeeding Women's Side Effects
Pregnant and breastfeeding women differed from non-pregnant/non-breastfeeding women in sociodemographic, medical, and vaccination history characteristics. There were more pregnant and lactating women aged 30 to 39 years. There were fewer smokers among the pregnant and breastfeeding than non-pregnant/non-breastfeeding women. They also had fewer comorbidities and used less preventive medication than non-pregnant/nonbreastfeeding women. BNT162b2 and ChAdOx1 were the most common vaccines in pregnant and breastfeeding women (Table 1).

Extension of Side Effects in Pregnant Women
There was a trend toward fewer systemic effects after the first or only dose in pregnant compared to non-pregnant/non-breastfeeding women. They experienced fewer headaches, fevers, and hot flashes, and less muscle pain, fatigue or tiredness, malaise, dizziness, and chest pain. After the second dose, there was no difference in symptom presentation (Table S2). BNT162b2 had a higher frequency of local symptoms after the first and second dose. ChAdOx1 and Gam-COVID-Vac increased the chances of experiencing both local and systemic symptoms after the first dose compared to BNT162b2 (Table 2).

Extension of Side Effects in Breastfeeding Women
There were fewer local symptoms and more headaches, fatigue, bone or joint pain, nausea, and dizziness in lactating women after the first or only dose. In contrast, local symptomatology predominated after the second dose, especially arm and injection-site pain (Table S3). BNT162b2 had more local effects and ChAdOx1 increased the chances of experiencing both local and systemic effects after the first and second dose compared to BNT162b2 (Table 3). * p < 0.01, *** p < 0.001. a Adjusted for comorbidity, allergies, preventive use of medication to prevent symptoms before vaccination, and history of confirmed COVID-19 infection. b The frequency distributions of mRNA-1273 (n = 4) and Ad26.CoV2.S (n = 6) were not included due to the small sample sizes. c The frequency distributions of mRNA-1273 (n = 4), Ad26.CoV2.S (n = 1), and Gam-COVID-Vac (n = 1) were not included due to the small sample sizes.

Severity of Side Effects in Pregnant and Breastfeeding Women
The severity after the first or only dose was similar between pregnant, lactating, and non-pregnant/non-breastfeeding women (Table 4). After the first dose, ChAdOx1 and Gam-COVID-Vac slightly increased the chances of moderate/severe symptoms in pregnant and lactating women, (Table 5).   * p < 0.05, *** p < 0.001. a Adjusted for comorbidity, allergies, preventive use of medication to prevent symptoms before vaccination, and history of confirmed COVID-19 infection. b The frequency distribution of Ad5-nCoV (n = 5) was not included due to the small sample size, and was also excluded in pregnant women. c The frequency distributions of mRNA-1273 (n = 4) and Ad26.CoV2.S (n = 6) were not included due to the small sample sizes in pregnant and lactating women.

Exploratory Results on the Presentation of Side Effects in Breastfed Infants
More symptoms were reported in infants aged 4.1-6 months (28%) and in those who received the CoronaVac vaccine (23%) ( Table S4). Secondary symptoms were present in 11% of breastfed infants; irritability was the most mentioned effect (Table S5).

Discussion
Pregnant women are at a higher risk of severe COVID-19 [56][57][58]. Therefore, they must obtain a vaccination. One reason why pregnant women are hesitant to receive a COVID-19 vaccination is the lack of safety data from the clinical trials, where pregnant women were not included. Social media has been pointed out as a source of misinformation [50,59], and politics has also been involved in shaping women's perceptions about COVID-19 vaccination [60]. The present study provides evidence for the short-term safety of the COVID-19 vaccine in pregnant and lactating women.

Extension of Side Effects during Pregnancy
ChAdOx1 and Gam-COVID-Vac increased the chances of both local and systemic symptoms after the first dose. We listed a total of five types of local and 39 types of systemic symptoms in pregnancy. The most common symptoms were pain in the arm, injection-site pain, headaches, muscle pain, the desire to sleep, fatigue or tiredness, and a lack of energy. This was consistent with what has been reported in the literature [16,17,61,62]. The most cited study on self-reported side effects in pregnancy was published in 2021 [18], where injection-site pain, fatigue, headaches, and myalgia were the most commonly reported symptoms in 35,691 pregnant women who received mRNA-1273 and BNT162b2. Side effects were more frequently reported after the second dose.

Extension of Side Effects during Breastfeeding
Lactating women experienced fewer local symptoms after the first or only dose. However, the opposite occurred after the second dose, and arm and injection-site pain were common. Another study found that the frequency of reactions was higher after the second dose [46]. Pain, redness, and swelling at the injection site have been frequently reported after the second dose [45]. We found that ChAdOx1 registered higher chances for both local and systemic side effects after the first dose, but these chances were reduced after the second dose compared to BNT162b2. Golan and colleagues found a higher presentation of any injection-site symptoms with mRNA-1273 than BNT162b2, after the first and second dose [5].

Severity of Side Effects during Pregnancy and Breastfeeding
We found that breastfeeding status had no effect on severity. The statistically nonsignificant result could be due to the small sample size of pregnant women. Therefore, it should be considered as inconclusive. Studies with larger sample sizes are required to confirm that pregnancy does not make a difference in the severity of COVID-19 vaccine side effects.

Side Effects in Breastfed Infants
Children can present side effects after the vaccination of their mother. We found that irritability was the most frequent self-reported symptom, followed by fever and diarrhea, in agreement with the literature [21,63]. CoronaVac registered the highest frequency of effects. McLaurin and colleagues did not show differences between mRNA-1273 and BNT162b2 [63], while Bertrand and colleagues found more drowsiness in children whose mothers received mRNA-1273 than BNT162b2 [21]. Vaccination protects the infant by transferring IgA and IgG antibodies and immune cells through the breast milk [19,22,58,64].

Limitations
Unfortunately, self-reported side effects could not be validated by a physician. The technique used for recruiting participants could have biased inclusion of contacts whom they knew had similar experiences. Furthermore, people that suffer from more symptoms might have been more interested in answering the survey. The recall bias was another limitation; participants could have remembered symptoms that have affected their health or their daily activities more, and some results could have been overestimated. The BNT162b2 and ChAdOx1 vaccines predominated in the study population, which was consistent with the greater availability of these vaccines in Mexico or in the neighboring country, the USA. It is necessary to continue with studies that include a greater number of participants who received Gam-COVID-Vac, Ad26.CoV2.S, and mRNA-1273. Results for breastfed infants were exploratory and conclusions are non-definitive due to the nonspecific nature of symptoms and small sample size. Further studies are needed to answer the unknowns that are left after this study.

Conclusions
Pregnant women had a tendency toward fewer systemic symptoms than non-pregnant/ non-breastfeeding women after the first or only dose. BNT162b2 had a higher frequency of local symptoms in pregnancy. Lactating women experienced fewer local side effects after the first or single dose, but these were more frequent after the second dose. ChAdOx1 increased the chances of developing both local and systemic side effects after the first dose, but the opposite occurred after the second dose in lactating women. The severity was similar across groups, although the result of lack of association in pregnancy requires studies with a larger sample size. Irritability was the most reported symptom in breastfed infants. The present study contributes to the knowledge about the safety of COVID-19 vaccination during pregnancy and breastfeeding, which was one the main concerns against vaccination. This information can encourage women to obtain the vaccine, as these groups of the population are at higher risk for severe COVID-19.   Informed Consent Statement: All the participants had to give their informed consent digitally before filling in the questionnaire. Participation was entirely voluntary, and withdrawal was allowed at any time without the need to justify the decision. There was no personal data collected that might enable the retrospective identification of the participant.

Data Availability Statement:
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.