INTRODUCTION

Infections by influenza viruses affect all age groups and can manifest as mild to severe disease. Pregnant women and young infants are particularly vulnerable to severe disease after influenza infection. Studies conducted during the 2009/H1N1 influenza pandemic showed that pregnant women with influenza had an increased risk of hospital admission compared to nonpregnant women.¹ Pregnant women might, however, be more likely to attend care and be preferentially hospitalized due to concerns regarding pregnancy complications. Reports evaluating seasonal influenza burden have, however, also shown that pregnant women experienced approximately three-fold higher rates of influenza-associated hospitalization than nonpregnant women.^2,3 A recent study among pregnant women with or without seasonal influenza infection suggested as well, that women with laboratory-confirmed influenza were more likely to be admitted to intensive care.⁴ Similarly, pregnant women experienced an approximately three-fold increased risk of death associated with influenza infection compared with nonpregnant women.⁵ In infants influenza infection is associated with increased rates of hospitalization and death.^6,7 A systematic review evaluating the incidence of laboratory-confirmed influenza hospitalizations among infants under 6 months of age reported rates ranging from 9.3 to 91.2 per 10,000 infants for seasonal influenza.⁸ Global mortality evaluations estimated that 3,339 influenza-associated in-hospital deaths occur in the first 3 months of life.⁹

It has been hypothesized that influenza infection during pregnancy may lead to inflammatory responses or immune dysregulation that can increase the risk of fetal death and preterm birth and interfere with the placental functions affecting the growth of the fetus.^10,11 Systematic reviews on the effect of influenza virus infection on pregnancy outcomes found that, influenza infection increased by approximately four-fold the risk of fetal death and was also associated with low birth weight in studies conducted during the 2009/H1N1 pandemic.^12,13 Fell et al. also identified that pregnant women with severe 2009/H1N1 illness, particularly those who required hospitalization, had an increased risk of preterm delivery.¹²

Generally, vaccination is the best strategy to protect the populations at highest risk of infection or severe disease. There is currently no influenza vaccine licensed in any country for use in infants younger than 6 months old. Nonetheless, infants can be protected from influenza illness via maternal influenza vaccination during pregnancy.^14,15,16,17 Randomized control trials (RCT) and observational studies have provided robust evidence that vaccination with inactivated influenza vaccines (IIV) during pregnancy can protect both the pregnant women and their infants from influenza disease. Live, attenuated influenza vaccines are contraindicated during pregnancy due to the hypothetical risk of transmission to the fetus and should be avoided.¹⁸ Most of the studies published to date on seasonal influenza vaccines have used trivalent formulations containing two influenza A viruses (A/H1N1 and A/H3N2) and a single influenza B virus. The first quadrivalent inactivated influenza vaccine was licensed in 2012, which in addition to the two influenza A viruses, also includes the two B viruses lineages, Victoria and Yamagata.¹⁹

PROTECTION AGAINST INFLUENZA DISEASE DURING PREGNANCY BY INFLUENZA VACCINATION

Four RCTs have investigated the effect of trivalent seasonal influenza vaccination during pregnancy in preventing illness in the women during pregnancy through approximately 6 months post-delivery and in their infants.^14,15,16,17 The trials included active surveillance for respiratory symptoms in the mothers and the infants, and polymerase chain reaction (PCR) testing was done to confirm influenza infection at the time of symptoms in three of the trials.^14,15,16,17 Two of these trials were conducted in South Asia, in Bangladesh and Nepal; and two in Africa: South Africa and Mali, in West-Africa. In the initial trial in Bangladesh during 2004–2005, 340 pregnant women in the third trimester were randomized to receive either pneumococcal polysaccharide vaccine or IIV. No laboratory confirmation was done for influenza infection, but women who received IIV had a reduced rate of respiratory illness with fever of 36% (95% CI: 4–57) compared with women in the pneumococcal vaccine group.¹⁷ In Nepal, where influenza viruses circulate year-round, two consecutive cohorts of pregnant women (17–34 weeks’ gestation) randomized to IIV or placebo were enrolled from April 2011 to April 2012 (cohort 1, N = 2090) and from April 2012 to September 2013 (cohort 2, N = 1603).¹⁵ The incidence of PCR-confirmed influenza was lower in vaccinated women than in placebo recipients in cohort 1 [vaccine efficacy: 45% (95% CI: 1–70)], but not in cohort 2 or the two cohorts combined [vaccine efficacy: 31% (95% CI: −10, 56)]. The lack of efficacy against PCR-confirmed influenza was probably because women were only tested when symptoms were reported in conjunction with fever. Nonetheless, in the two cohorts combined there were fewer influenza-like-illness (ILI) cases in the IIV-group compared with the placebo-group [vaccine efficacy: 19% (95% CI: 1–34)].¹⁵ The trial from South Africa included 2116 women without human immunodeficiency virus (HIV) infection in their second or third pregnancy trimester randomized to IIV or placebo in 2011 and 2012.¹⁴ During the study, 19 of the vaccinated women had a PCR-confirmed influenza infection compared with 38 women who received placebo, accounting for a vaccine efficacy of 50.4% (95% CI: 14.5–71.2).¹⁴ In Mali, 4193 third trimester pregnant women were randomly assigned to receive IIV or quadrivalent meningococcal vaccine.¹⁶ Study enrolment occurred from September 2011 to April 2013, and both Northern and Southern Hemisphere vaccine formulations were used. The overall vaccine efficacy point estimate against PCR-confirmed influenza was 70.3% (95% CI: 42.2, 85.8).¹⁶ A pooled analysis of the trials from Bangladesh, South Africa, and Mali estimated that from enrolment during pregnancy to the end of follow-up at 6 months post-partum, IIV was 50% (95% CI: 32–63) efficacious against PCR-confirmed influenza; with a 42% (95% CI: 12–61) efficacy during pregnancy and 60% (95% CI: 36–75) post-partum.²⁰

Observational studies from high-resource settings, using data collected during real-world situations, have also investigated the vaccine effectiveness of either seasonal or 2009/H1N1 pandemic influenza vaccines against influenza disease during pregnancy.^21,22,23 In Norway a nationwide registry-based retrospective cohort study among pregnant women, estimated that vaccination with monovalent 2009/H1N1 vaccines reduced the risk of a clinical diagnosis of influenza (with or without laboratory confirmation) by 70%.²¹ Furthermore, two case-control studies from the USA during the 2010 to 2012, and Denmark from 2010 to 2017, reported that the effectiveness of trivalent IIV during pregnancy in preventing PCR-confirmed influenza was 51–76% and 64%, respectively.^22,23

One study from Greece measuring the protection that the quadrivalent influenza vaccine formulation during the 2018–2019 influenza season, conferred to pregnant women reported a 72% effectiveness against PCR-confirmed influenza.²⁴

While it is difficult to measure the protective effect of vaccination against severe disease due to the small numbers of influenza-associated hospitalizations among pregnant women within each influenza season, observational studies have evaluated this outcome.^25,26 An Australian retrospective cohort study conducted during the 2012 and 2013 influenza seasons, reported that pregnant women vaccinated during pregnancy had an 81% (95% CI: 31–95) reduction in emergency department visits and 65% (95% CI: 3–87) reduction in inpatient hospital admissions. Laboratory-confirmed influenza hospitalizations were also lower in the vaccinated group compared to the unvaccinated group, but the difference was not significant (adjusted hazard ratio: 0.16, 95% CI: 0.01–1.76).²⁵ An international multi-site study using administrative data from healthcare systems with integrated laboratory, medical and vaccination records in four countries (Australia, Canada, Israel, and the USA), also reported an overall adjusted vaccine effectiveness of 40% (95% CI: 12–59) against PCR-confirmed influenza-associated hospitalization during pregnancy. In this international study, a sub-analysis by trimester at admission found a significant effect of 55% among women who were hospitalized in their first and second trimester.²⁶ Furthermore, the quadrivalent vaccine study in Greece showed that vaccination was associated with a decreased probability of ILI (61%), healthcare seeking (68%), and hospitalization (87%) among pregnant women.²⁴

EFFECT OF INFLUENZA VACCINATION DURING PREGNANCY ON INFLUENZA ILLNESS OR RESPIRATORY DISEASE IN THE INFANTS YOUNGER UNTIL 6 MONTHS OF AGE

The four RCTs previously mentioned and numerous observational studies have demonstrated the benefit of influenza vaccination during pregnancy in the offspring of the vaccinated mothers. In the trial conducted in Bangladesh, there was a 63% (95% CI: 5–85) reduction in laboratory-confirmed influenza and 29% (95% CI: 7–46) reduction in respiratory illnesses with fever, in the infants born to mothers who received IIV during pregnancy.¹⁷ In the placebo RCTs in South Africa and Nepal, influenza vaccination during pregnancy was associated with a reduction in PCR-confirmed influenza in the infants during the first 6 months of life of 49% (95% CI: 12–70) and 30% (95% CI: 5–48), respectively.^14,15 In Mali, during the entire study period the overall influenza vaccine efficacy in infants <6 months old was 33% (95% CI: 4–54) against PCR-confirm influenza.¹⁶ In a meta-analysis of the four trials, the calculated pooled efficacy of maternal influenza vaccination in preventing laboratory-confirmed influenza in infants <6 months old was 36% (95% CI: 22–48).²⁷ Moreover, observational studies from high-resource settings, also evaluated the effect of maternal IIV in preventing laboratory-confirmed influenza in infants <6 months old.^{22,28,29,30,31,32} The observational studies described significant reductions in the risk of laboratory-confirmed influenza among the infants born to vaccinated mothers that were generally higher than that reported in the clinical trials (71% to 41% vaccine effectiveness).²⁷ The only study evaluating the administration of the quadrivalent formulation during pregnancy reported a vaccine effectiveness against PCR-confirmed influenza of 65% among the infants.²⁴

Although the RCTs were not designed to assess the effect of maternal vaccination on infant hospitalizations and most of the influenza infections presented with mild symptoms; in a post-hoc analysis of the South African trial, infants born to vaccinated mothers had a 57% (95% CI: 7–81) risk reduction in all-cause acute lower respiratory tract infection during the first <3 months of life.³³ A similar observation of 31% (95% CI: 6–50) lower rate of lower respiratory tract infection was noted in the RCT from Nepal. A pooled analysis of these two RCTs and Mali, resulted in an overall vaccine efficacy of 20% (95% CI: 1–34) against all-cause lower respiratory tract infection.³⁴ Additionally, observational studies, also evaluated the impact of influenza vaccination during pregnancy on laboratory-confirmed influenza hospitalizations in infants.^{28,30,32,35,36} A meta-analysis of the earlier studies revealed a pooled vaccine effectiveness of 72% (95% CI: 39–87),²⁷ and more recently a population-based cohort study from Australia and a Spanish surveillance-based study also consistently found that infants aged <6 months born to mothers who were vaccinated during pregnancy compared with unvaccinated mothers had approximately 61% lower risk of influenza-associated hospitalization.^32,37

HUMORAL RESPONSES TO INFLUENZA VACCINATION DURING PREGNANCY AND TRANSPLACENTAL ANTIBODY TRANSFER

Hemagglutination-inhibition (HAI) assay is the laboratory method normally used to measure humoral responses to influenza vaccination, with a HAI titer of 1 : 40 or higher indicating seroprotection in healthy adults and seroconversion being defined as an increase in HAI titer of at least four-fold from pre- to post-vaccination.

In general, pregnant women mount good immune responses to influenza vaccination similar to non-pregnant adults.^{38,39,40,41,42,43,44} In the four RCTs, vaccinated women had significant increases in antibody titers to all vaccine strains, with HAI titer increases in Bangladesh and Nepal ranging from 2 to 18 times from before vaccination to delivery.^15,45 In a pooled analysis with South African and Mali data, one-month post-vaccination titers increased from baseline by a factor of 9.4, 6.0, and 9.6 for A/H1N1, A/H3N2 and B/Victoria, respectively.⁴⁶ The pooled analysis also found that 93%, 83% and 91% of the vaccinated women had seroprotective titers and that 77.0%, 64.8% and 84.5% seroconverted for A/H1N1, A/H3N2 and B/Victoria, respectively.⁴⁶ The South African trial also demonstrated that IIV induced functional influenza-specific neutralizing antibodies to the three vaccine strains, that were three-fold higher than the HAI titers, except against B/Victoria.⁴⁷

Maternal IgG antibodies are efficiently transferred from the mother to the fetus across the placenta; this transfer typically starts from 17 weeks of gestation, subsequently progresses in an exponential manner and peaks at 37–41 weeks, with enhanced neonatal Fc receptor expression.⁴⁸ After influenza vaccination during pregnancy vaccine induced antibodies are detectable in the newborn starting approximately 2 weeks after vaccination.^49,50 In full-term neonates the antibody concentrations can equal that of the mother and in some cases even exceed maternal levels. A wide range of infant-to-maternal antibody ratios have been reported, and studies evaluating the immunogenicity of monovalent 2009/H1N1 vaccines,^51,52 normally detected higher ratios (1.4 to 2.9) compared with the RCTs where seasonal trivalent vaccines was used (0.7 to 1.1).^14,45 In the pooled analysis of the trials in South Africa and Mali, infants born to IIV-recipients had higher titers at birth compared with those born to control mothers with 82%, 60%, and 81% of infants in the IIV group showing seroprotective titers for A/H1N1, A/H3N2, and B/Victoria, respectively, at birth.⁴⁶ However, by 6 months of age infants in both the IIV and control group had similar antibody levels. The level of maternal antibodies transferred to the fetus is affected by the time interval between vaccination and delivery.^49,51 In the two trials pooled analysis, longer time between maternal IIV receipt and birth was associated with higher transplacental antibody transfer, higher HAI titers, and likelihood of seroprotection at birth. Nonetheless, maternal vaccination at least 2–4 weeks before delivery appeared to be the minimum time needed to maximize infant seroprotection at birth.⁴⁶ A meta-analysis of studies that have evaluated the impact of when vaccination occurs during pregnancy on maternal and infant antibody levels found that compared with women vaccinated earlier in pregnancy, those vaccinated at a later trimester had a greater fold increase in HAI titers (1.33- to 1.96-fold) and higher HAI titers were also detected in their newborns blood (1.21- to 1.64-fold).⁵³

While we need to understand the full implications of vaccination timing for protection of the mother and the newborn. The programmatic distribution of influenza vaccines to pregnant women is challenging by trying to reconcile the time of vaccine availability with the best time to vaccinate during pregnancy to have good protection in the women and their infants. Since there is a cumulative transfer of antibodies pregnant women should not wait to late third trimester to get vaccinated, not only to protect themselves as early as possible in the influenza season, but also to afford some protection to neonates born prematurely since antibody titers are significantly lower when less than four weeks elapsed from vaccination to birth.⁵⁴ Preterm neonates, especially those born with gestational ages <32 weeks, have considerably lower transplacental transfer of IgG and thus have lower antibody concentrations compared with full-term babies.⁵⁵

DURATION OF PROTECTION AGAINST INFLUENZA ILLNESS OR RESPIRATORY DISEASE IN THE INFANTS

The duration of passive protection in the infants conferred by vaccination during pregnancy is subject to how quickly the maternal-acquired antibodies wane. In infants whose mothers received IIV during pregnancy, half-lives of 42–50 days have been described in South Africa and Bangladesh for the different influenza-specific hemagglutinin antibodies.^45,56 The impact of the antibody decay on infant protection was evaluated both in South African, where during the first 8 weeks of life vaccine efficacy against PCR-confirmed influenza was 86% instead of 49% in infants <6 months old, and in Mali where a 68% vaccine efficacy was reported among infants <4 months old compared with 33% in infants <6 months old.^16,57 The pooled analysis of the trials from Nepal, South Africa, and Mali found that the pooled vaccine efficacy in the infants was 56% (95% CI: 28–73) within the first 2 months of life, 39% (95% CI: 11–58) between 2 and 4 months, but the vaccine did not show efficacy after 4 months of age [19% (95% CI: −9–40) between 4 and 6 months].²⁰

Very few studies have evaluated the effects of influenza vaccination during pregnancy on acute respiratory infections in the infants beyond 6 months of age.^32,36,58 Foo et al., using linked administrative health records, evaluated the association of IIV during pregnancy with the risk of laboratory-confirmed influenza, hospitalization for influenza, or acute respiratory infection among children up to 5 years of age.³² Although, protective effects of maternal vaccination were identified in infants <6 months old, no consistent differences in overall risks were observed by maternal vaccination status for children aged 6 months to 5 years. Interestingly this study found an increase in the risk of laboratory-confirmed influenza among children aged 6 months to 2 years following maternal vaccination during the first trimester; these results should, however, be interpreted with caution, since as the authors mentioned, association could be due to residual confounding, differential exposure to influenza seasons, or the absence of vaccination information beyond 6 months of age.³²

SAFETY OF INFLUENZA VACCINATION DURING PREGNANCY

A major challenge for achieving high uptake of influenza vaccination during pregnancy relates to concerns of vaccine safety for the pregnant women and their fetus. Prior to the 2009/H1N1 pandemic, safety reports of influenza vaccination during pregnancy came mainly from post-marketing pharmacovigilance systems; these, however, have numerous limitations, and cannot causally attribute adverse events to vaccination.⁵⁹ After the accelerated licensure of the 2009/H1N1 vaccines and their use among large numbers of pregnant women, several large studies documented the safety of maternal influenza vaccination.^59,60,61,62 Robust assessments of the safety of influenza vaccination during pregnancy is crucial, and a number of systematic reviews and recent studies have reported on birth outcomes following vaccination.^{63,64,65,66,67,68,69} The outcomes more commonly evaluated were preterm birth, small for gestational age, fetal death (including stillbirth or spontaneous abortion), and low birth weight. Consistently, vaccination was not associated with an increase in any of these outcomes, and this was based on multiple studies involving more than 100,000 pregnant women.^{63,64,65,66,67,68} Evidence from two of these reviews suggest that in comparison to no influenza vaccination in pregnancy, vaccination was actually associated with an 8–18% reduction in incidence of low birth weight newborns.^64,67

The first pregnancy trimester is a vulnerable time for fetal organogenesis and development; therefore, it is essential to evaluate the safety of vaccination in relation to birth defects after vaccination in the first trimester. Systematic reviews and meta-analyses, case-control, and large cohort studies that have investigated the safety profile of influenza vaccination during pregnancy have consistently shown no association between seasonal or pandemic influenza vaccination and the prevalence of major birth defects in the offspring of mothers vaccinated early in pregnancy.^{63,69,70,71,72,73,74,75} This includes a recent population-based cohort study of 125,866 singleton births between 2012 and 2016 in Australia that reported that seasonal IIV exposure during the first trimester was not associated with diagnosis of any of the investigated structural birth defects diagnosed within 1 month (prevalence ratio 0.98, 95% CI: 0.77, 1.28) or 6 years (prevalence ratio 1.02, 95% CI: 0.78, 1.35) of life.⁷⁵

Despite most of the published studies evaluating safety of influenza vaccination during pregnancy focusing on possible adverse events during pregnancy, in the period immediately after birth, or in the first 6 months of life, an increasing number of reports are assessing in-utero exposure to influenza vaccines (either pandemic monovalent or seasonal) and its association with health outcomes beyond 6 months of life. Studies from Canada, the USA, Denmark, Sweden, and Australia reporting on the association between maternal influenza vaccination and several childhood conditions, including allergic or immune-related (e.g. asthma), nonimmune related (e.g., neoplasms, sensory disorders), neurologic, and behavioral (e.g., autism spectrum disorders) detected no increased risk for these early childhood morbidity endpoints and maternal vaccination among children 2–5 years old.^{76,77,78,79,80,81}

Several studies, including a narrative review, that evaluated the association between influenza vaccination and maternal health and obstetrical outcomes, including gestational diabetes, gestational hypertension, pre-eclampsia/eclampsia, chorioamnionitis, hyperemesis, proteinuria, urinary tract infection, venous complications, pulmonary embolus, and peripartum cardiomyopathy, found no significant increased risk of the outcomes evaluated among vaccinated women.^69,82,83,84

INFLUENZA VACCINATION OF PREGNANT WOMEN WITH UNDERLYING CONDITIONS

Pregnant women with comorbidities have an even higher probability of severe influenza and accordingly should be vaccinated.^5,85,86 Vaccine efficacy and immune responses to influenza vaccination were evaluated in pregnant women living with HIV.^14,56,87,88 In parallel to the South African trial described above, during 2011, a smaller immunogenicity trial was conducted among 194 pregnant women living with HIV.¹⁴ Before vaccination and 1-month post-vaccination, pregnant women living with HIV compared to those without HIV, had lower levels of HAI and neutralizing antibodies and fewer seroconverted.^47,56 Similar results were described in the USA and Malawi.^87,88 Correlation analyses of HAI titers and HIV viral load have not, in general, identified a correlation,^56,87,89 but improved immune responses were observed in women with CD4+-lymphocyte counts of >350 cells/μL.⁵⁶ The transplacental antibody transfer was similar in women living with and without HIV, but due to the lower antibody levels post-vaccination in the HIV group, their newborns had lower antibody titers at birth and were less likely to have seroprotective HAI titers.^56,88

To improve the vaccines’ immunogenicity in pregnant women living with HIV regimens using vaccines with double antigen concentration or two vaccine doses have been evaluated.^89,90 However, even though the immunogenicity was marginally increased, it was still lower than in pregnant women without HIV.

In the initial South African RCTs, despite women living with HIV having a reduced immune response to vaccination, the vaccine efficacy against PCR-confirmed influenza was similar in these women (58%; 95% CI: 0.2–82) and those without HIV (50%; 95% CI: 15–71).^14,56 The study was not powered to detect vaccine efficacy in the HIV-exposed infants; nevertheless, similar influenza attack rates were detected in infants born to vaccinated mothers (5.0%) or placebo recipients (6.8%).¹⁴ Based on these results more immunogenic vaccines are needed for pregnant women living with HIV to enhance transplacental transfer of vaccine-induced protective antibodies to their newborn infants.

PRACTICE RECOMMENDATIONS