Vaccination of pregnant women against influenza: what is the optimal timing?

ABSTRACT

Pregnant women and young infants are at increased risk for severe influenza and its complications. Influenza vaccination during pregnancy is increasingly implemented as a strategy aiming to protect the pregnant woman, the fetus and the young infant. In clinical practice, the achievement of satisfactory protection for the pregnant woman without compromising the protection of her infant during the first months of life remains a challenge. Determinants that are implicated in the optimal timing of influenza vaccination in pregnancy include influenza season, trimester of pregnancy, maternal host factors and infant factors. This article addresses influenza vaccination in pregnancy and presents recent published evidence on issues that affect the optimization of the timing of maternal vaccination.

KEYWORDS:

• Influenza vaccination
• effectiveness
• pregnant women
• infants

The only reason for time is that everything doesn’t happen at once.

Albert Einstein (1879–1955)

Starting from the 1918 influenza pandemic, evidence consistently indicates that pregnant women are at a disproportionate risk for influenza-associated hospitalization, complications, and death.^1–5 Infants younger than six months of age with influenza are also at increased risk for severe illness, healthcare seeking, hospitalization, and death, accounting for most influenza-associated deaths among children.^6,7 Large studies conducted in various settings over the past two decades have shown that vaccinated pregnant women with the trivalent influenza vaccine (TIV) were less likely to seek healthcare, to be admitted to a hospital, to develop influenza-like illness (ILI) or influenza compared with unvaccinated women.^8–12 Beyond the benefits for the pregnant women per se, there is evidence that maternal vaccination also confers protection to offspring against fetal death,^10,13 prematurity, low birth weight and being small for gestational age.^8,14,15 Of equal importance, maternal vaccination also protects the young infants against healthcare seeking, ILI, influenza, and severe pneumonia up to the age of six months.^8,9,12,16,17 In our 2018–2019 study in a maternity hospital in Greece, the IVE of a quadrivalent influenza vaccine was 72% in pregnant women and 64.5% in their infants; beyond the protection of both women and infants against influenza and healthcare seeking, maternal vaccination was also associated with less frequent consumption of antibiotics in infants during the influenza season.¹⁸ Given the safety profile of influenza vaccination in pregnancy,^14,19,20 pregnant women constitute a high-priority group for influenza vaccination. Per public health perspective, vaccination of pregnant women is also cost effective.²¹ Yet, improving our understanding of the factors that impact IVE in pregnant women and young infants has practical implications for defining the optimal timing of maternal vaccination.

This article addresses the issue of optimal timing of influenza vaccination in pregnancy under the prism of recently published data on the role of each one of the following four players.

Influenza season

The unpredictable predominance and circulation of different influenza A subtypes and/or influenza B lineages in any influenza season means that the associated morbidity and mortality may vary substantially from season to season and from region to region.²² A recent New Zealand study over three influenza seasons found that influenza A/H1N1 virus infection posed a higher risk for hospitalization among pregnant women regardless of trimester.²³ In turn, IVE depends highly on antigenic matching between the vaccine and circulating influenza virus strains; in a United States study of in- and out-patients from 2015–2016 through 2017–2018 the IVE was higher against influenza A/H1N1 pdm09 (37% vs. 53%, respectively), lower against influenza A/H3N2 (19% vs. 23%, respectively), and moderate against influenza B viruses (46% for both settings).²⁴

Gradual waning of vaccine-induced antibody titers occurs over six months after vaccination, and may end to non-protective levels in case of low titers after vaccination.²⁵ Data from a European multicenter study over five influenza seasons found that IVE against influenza A/H3N2 reached 50.6% 38 days post-vaccination but waned to 0% from 111 days onwards, while IVE against influenza B waned from 70.7% 44 days post-vaccination to 21.4% at season end.²⁶ Similar decreasing IVE rates across all three types/subtypes of influenza vaccine were estimated with increasing time since vaccination among patients with respiratory illness seeking healthcare in the United States from 2011–2012 through 2014–2015.²⁷ In the latter study, maximum IVE was observed shortly after vaccination, followed by a monthly decline of approximately 7% for influenza A/H3N2 and influenza B and of 6–11% for influenza A/H1N1 pdm09.²⁷ Overall, IVE remained greater than zero for at least five to six months after vaccination.²⁷ Waning of vaccine-induced antibodies over time has practical implications for when influenza vaccine should be given.

Host comorbidities

There is no doubt that pregnancy constitutes a risk factor for serious influenza per se. Surveillance data from the United States showed that pregnant women accounted for up to one third of hospitalizations among females 15–44 years old per season.³ Similarly, a study from Spain estimated that pregnant women had almost 8 times increased risk of hospitalization than non-pregnant women.⁴

Over the last decade, substantial progress has been made on our understanding of the underlying immunologic mechanisms of severe influenza in pregnancy. In animal models, pregnant ferrets with influenza A/H1N1 pdm09 virus infection have higher levels of inflammatory cytokines in pulmonary tracts and significantly lower total CD8 T-cell counts and influenza A/H1N1 specific B-cell responses in blood than non-pregnant ferrets, which indicates a less effective immune response.²⁸ Pregnant murine models also show dysregulation of inflammatory antiviral responses in the context of influenza A virus infection, including decreased levels of IL-1α, IFN-β, and IFN-γ early in the course of infection and limited viral clearance and cytotoxic T-cell activity.²⁹ The ability of influenza A virus to infect and replicate in human placental cells and induce strong inflammatory responses, resulting in apoptosis, has been also documented.³⁰ Significant remodeling of placental architecture, pre-term labor, impairment of fetal growth, increased fetal mortality and maternal morbidity, indicating the potential of influenza to trigger a post-abortive mechanism with adverse fetal outcomes have been also shown.³¹ Murine models also found a protective effect of 17-β-estradiol on mouse morbidity and overall survival while progesterone-treated mouse had more prolonged morbidity and higher mortality rates.³²

Pregnant women express a unique immunological profile which is driven by sex hormones and consists of suppression of cell-mediated immunity and elevation of humoral responses.^31,33 Pregnant women elicit exaggerated proinflammatory immune response of monocytes and plasmacytoid dendritic cells to influenza A virus infection compared to non-pregnant women, as characterized by increase expression (at least 1.5-fold) of major histocompatibility complex class II, CD69 and IFN-γ-induced protein-10, and macrophage inflammatory protein 1β, which in turn may account for the pulmonary inflammation following influenza A virus infection.³⁴ In addition, higher IL-8 and TNF-α mRNA expression have been detected in pregnant women with influenza A/H1N1 pdm09 virus infection who died or survived compared with uninfected pregnant women, while IL-6 showed higher expression in infected pregnant women who died.³⁵ The elevated levels of pro-inflammatory cytokines in association with the detected decrease in the anti-inflammatory response of TGF-β mRNA and IFN-β, indicates an imbalance in their immune response and could also explain the increased severity of influenza virus infection and death of pregnant women.³⁵ In another study assessing the role of natural killer (NK)- and T-cell responses, pregnant women had a significantly increased chemokine response both before and after influenza vaccination compared to non-pregnant women, which in turn could drive pulmonary inflammation, explaining severe morbidity and mortality.³⁶ Lung structure and cardiopulmonary function are also stressed in pregnancy, and may increase the risk of hypoxemia and raise influenza severity.³³

The probability of severe influenza is even higher in pregnant women with comorbidities. In a study of 1992 women 15–44 years old (including 1748 pregnant women) hospitalized in Moscow with influenza during four consecutive influenza seasons, the risk of influenza increased by 2% with each year of age and was higher in women with underlying conditions, irrespective of the influenza season, circulating viruses or trimester.³⁷ In a South African study based on a decade-long surveillance data, the mean annual influenza-associated mortality was 74.9 per 100,000 HIV-infected pregnant women compared with 1.5 per 100,000 non-HIV-infected pregnant women.¹ Pregnant women with asthma elicit modified antiviral immune and inflammatory responses following influenza infection, which may persist for at least six months post-partum, and may relate to the increased susceptibility to influenza during pregnancy.³⁸ In turn, insufficient immune responses to vaccines may occur in persons with comorbidities.²⁵ Nonetheless, IVE trials in pregnancy almost exclusively enroll healthy pregnant women, underlying the need for more real-life research in pregnant women with comorbidities and high-risk pregnancies.

Trimester of pregnancy

Influenza is most severe in the third trimester of pregnancy. A New Zealand population-based study of women in reproductive age found that pregnant women in the third trimester experience higher influenza-associated hospitalization rates than those in the first or second trimester (relative risk: 2.5, 3.9, and 4.8 in first, second and third trimester, respectively, compared to non-pregnant women).²³

Vaccine-induced IgG antibodies start to increase approximately two weeks after vaccination. Going forward, Schlaudecker et al. studied the kinetics of maternal serological response to influenza vaccination in 71 healthy pregnant women compared to 67 non-pregnant controls during two influenza seasons.³⁹ They found that geometric mean titers (GMTs) significantly reduced in pregnant women following vaccination for all vaccine strains compared to non-pregnant women and progressively declined with vaccination later in pregnancy.³⁹ The most notable reduction concerned anti-H1N1 IgG1, where titers reduced by approximately 7% each week throughout pregnancy.³⁹

Issues related to fetus – infant

Neonates and young infants constitute a highly vulnerable group to infections given the functional immaturity of their immune system to mount efficient humoral immune responses. Therefore, maternal-derived IgG antibodies are crucial for their protection during the first months of life.

Influenza vaccination during pregnancy is expected to benefit the offspring via three routes: first, transplacental transfer of vaccine-derived IgG antibodies to the fetus; second, transfer of IgA antibodies to the infant through breast feeding; and third, post-partum protection (cocoon), since the mothers are the main sources of influenza virus infection for babies.

IgG is the only antibody class that efficiently crosses the human placenta and is released in fetal circulation. Transplacental transfer of antibodies is mediated by the FcRn receptor expressed on syncytiotrophoblast cells. Transfer is facilitated by high maternal IgG levels, though there is a threshold for FcRn saturation.^40,41 Gestational age at birth, birth weight, maternal disease, placental integrity, IgG subclass, nature of antigen also affect the transfer of maternal IgG antibodies, yet the relationships of these factors have not been elucidated so far.^40,41

IgG transfer begins as early as 13 weeks of gestation and subsequently progresses in an exponential manner, with the largest transfer accomplished in the third trimester.^40,41 It has been estimated that fetal IgG rises from approximately 10% of the maternal concentration at 17–22 weeks, to 50% at 28–32 weeks, reaching a fetal concentration between 29 and 41 weeks which is two times the concentration at 17–28 weeks.⁴⁰ Accordingly, preterm infants, and particularly those of <36 weeks of gestation receive a lower concentration of IgG antibodies.⁴⁰ This is even more critical, given that prematurity and low birth weight are associated with several maternal comorbidities, including gestational hypertension, diabetes and preeclampsia, which may also impact placental function and the antibody–FcRn interaction.⁴⁰

The rationale for conferring protection to the fetus and young infant through maternal vaccination relies on the rise of maternal IgG antibodies. Nonetheless, the time that elapses from vaccination is crucial for efficient transplacental antibody transfer.⁴¹ A recent meta-analysis found that pregnant women vaccinated in a later trimester had a greater fold increase in hemagglutination inhibition (HI) titers and higher HI titers in cord/newborn blood compared to women vaccinated earlier, which indicates that vaccination in late pregnancy increases the transfer of antibodies to the fetus.⁴² In addition, significantly higher H1N1 titers were detected in cord blood samples of infants whose mothers were vaccinated in either the second on the third trimester, while antibody titers were significantly lower when less than four weeks elapsed from vaccination to birth.⁴³ In contrast, preterm babies, especially those with a gestational age of <32 weeks, experience considerably lower transplacental transfer of IgG and thus have lower antibody concentrations compared with full-term babies.⁴¹ Moreover, the increasing survival of preterm babies, especially those with extremely low birth weight, may pose dilemmas in clinical practice regarding the optimal timing of influenza vaccination in pregnancy. In practice, vaccination in early pregnancy will confer protection for a greater pregnancy time, however protection may not last until delivery.⁴² On the other hand, vaccination in late pregnancy is expected to confer protection during early infancy. In a study in Mali over three influenza seasons, the overall IVE of TIV for infants was 37.3% against influenza; however, IVE was 70.2% in the first four months of age and diminished in the fifth and sixth months of age.⁴⁴ Similarly, in a trial conducted in 2011–2012 in South Africa, the IVE was estimated at 85.6% in infants ≤8 weeks of age, 25.5% in infants 8–16 weeks of age and 30.3% in infants 16–24 weeks of age.⁴⁵ In this latter study, the proportion of infants with HI titers ≥1:40 decreased from more than 56% in the first week of life to less than 40% at 16 weeks of age and to less than 10% at 24 weeks of age.⁴⁵ Therefore, the achievement of high maternal vaccine-derived IgG antibodies in the perfect timeframe is crucial in order to protect the young infant until it reaches the age-milestone for its own vaccination.

Chronic maternal infection, e.g. malaria and HIV, can injury the placenta and impair IgG transplacental transfer to the fetus.⁴⁰ Studies on pregnant women with preexisting type 2 diabetes mellitus indicate lower IgG levels and reduced leukocyte FcRn expression in maternal blood, cord blood and placental samples at delivery compared with normoglyxemic mothers, but increased FcRn expression in women with mild gestational hyperglycemia.⁴⁰ Notably, increased IgG transfer has been detected in women with pregnancy-induced hypertension.⁴⁰

Lastly, there is a gap in our knowledge regarding the role of maternal vaccination on breast feeding to confer immune protection during the first months of life. In a randomized trial, pregnant women vaccinated in the third trimester had significantly higher influenza-specific IgA levels in breast milk for at least six months postpartum compared to pneumococcal-vaccinated women.⁴⁶ Notably, breastfeeding significantly decreased the episodes of respiratory illness with fever in infants whose mothers were vaccinated against influenza but not in infants of pneumococcal-vaccinated mothers.⁴⁶ Further studies are needed to determine the underlying immune mechanisms of the protective effect of breastfeeding.

Conclusions

Influenza vaccination in pregnancy is increasingly implemented worldwide as a strategy to protect the mother and her offspring against severe influenza and its complications. Over the past decade, substantial progress has been made to understand the complexity of underlying mechanisms that are implicated in conferring protection through maternal vaccination (Table 1). Nonetheless, there are gaps in our knowledge, especially in pregnant women with comorbidities and high-risk pregnancies, as these factors may affect the transplacental transfer of maternal IgG antibodies.

Table 1. Issues to consider in vaccinating pregnant women against influenza

Two weeks are needed to elicit antibodies after vaccination Waning immunity after influenza vaccination Trimester of pregnancy Host comorbidities of pregnant woman High-risk pregnancy Preterm babies and extremely low birth weight babies Breast feeding to confer protection during the first months of life

In addition to host factors, the timing of vaccination is a key determinant of satisfactory protection both for the mother and her baby. In clinical practice, achieving a maximum of protection for the pregnant woman without compromising the protection of the infant during the early months of highest susceptibility is a challenge. Ideally, the vaccine should be given early in the third trimester and no later than two weeks before the delivery, in order to maximize the protection for the infant, yet the vaccine should be given even earlier in order to allow transplacental IgG transfer to preterm infants. However, given the risk that influenza imposes in pregnancy, pregnant women entering the influenza season, especially those in the second or third trimester, should be vaccinated against influenza, considering that two weeks are needed to elicit an immune response after vaccination and also the intra-season waning immunity. Real-life research is needed to address the gaps in knowledge about influenza and influenza vaccination in pregnant women with comorbidities and strengthen the impact of influenza vaccination strategies.

Disclosure of potential conflicts of interest

No conflict of interest to declare.

Acknowledgments

The opinions presented in this article are those of the authors, and not necessarily of their institutions.

Additional information

Funding

No funds were received for this study.