Several studies have reported that infants have the highest influenza infection and hospitalization rates in pediatrics, particularly for infants <6 months of age.1–6 One study reported hospital admission rates for laboratory-confirmed influenza of 4.5/1000 in children aged between 0 and 5 months compared with 0.9/1000 for children aged between 6 and 23 months.1 Another study conducted in Hong Kong reported rates of excess hospitalization for acute respiratory disease, which were attributable to influenza of around 284 per 10,000 children <1 year of age.2 Children also have an increased risk of mortality from influenza infection.1,7 Only one-third of all pediatric influenza deaths occur in children with underlying risk factors,7 most secondary to pneumonia, laryngotracheobronchitis or encephalopathy.8,9 Thus, young age alone is the most important risk factor for influenza-associated morbidity and mortality.
Pregnant women benefit themselves from influenza vaccination, being at increased risk of serious influenza-related complications, such as severe pneumonia, acute respiratory distress syndrome and death (in addition to the increased risk of stillbirths, neonatal death and preterm delivery). These complications are facilitated by physiological changes during pregnancy including reduced lung capacity and attenuation of immune responses.10–12 In the United States and Canada, the immunization of pregnant women with inactivated trivalent influenza vaccines has thus been recommended since 1997 in the second and third trimester, and since 2004 at any time during pregnancy.13 In contrast, most European countries have only recommended such vaccination after the H1N1/09 pandemic. The World Health Organization and few European countries (including the United Kingdom and France) currently recommend influenza immunization at any time during pregnancy.14 In Switzerland, vaccination is currently only recommended after the first trimester of gestation and only for healthy women to reduce the occurrence of random temporal associations with spontaneous miscarriage (http://www.bag.admin.ch/themen/medizin/00682/00684/02535/index.html?lang=fr).
Maternal immunization against influenza during pregnancy may protect newborns and infants for the first months after birth through the passive transfer of maternal antibody in cord blood.14–16 It is well-known that IgG antibodies cross the placenta via active transport from the mother to the fetus.15 Several studies have reported an effective transmission of maternal antibodies to newborns, following seasonal or pandemic influenza immunization during pregnancy.16–21 However, we were not aware of any multivariate analysis identifying independent factors influencing neonatal humoral immunity, including past immunization and/or past influenza-like illness (ILI). The 2010/2011 winter season provided a unique opportunity to address these questions, as the seasonal vaccine included the pandemic A/California/7/2009 (H1N1) strain (influence of past exposure/past immunization), a novel A/Perth/16/2009 (H3N2) strain (“primary” responses) and the same B/Brisbane/60/2008 strain as in the 2009/2010 vaccine.
In the present study, we assessed influenza antibody responses to 3 vaccine strains in the cord blood of infants born between December 2010 and June 2011. We compared geometric mean titers (GMTs) and seroprotection rates in infants born to vaccinated and nonvaccinated mothers and identified independent determinants of neonatal humoral immunity.
This was a single center, cross-sectional study enrolling pregnant women in their last trimester to assess the presence of antibodies against the 3 influenza strains included in the 2010–2011 influenza vaccine [A/California/7/2009 (H1N1), A/Perth/16/2009 (H3N2) and B/Brisbane/60/2008] in cord blood samples. To detect a difference of at least 20% in the rate of protective antibody titers between the 2 groups (vaccinated and nonvaccinated mothers) with a power of 90% at a level of significance of 0.05 (2 sided), a sample size of 92 participants in each group was required. Seroprotection was defined as a hemagglutination inhibition (HAI) assay titer ≥1:40 for the H3N2 strain—a new and moderately immunogenic strain. Allowing for a drop out of 10%, we intended to recruit 100 future mothers in each group. Recruitment occurred between December 2010 and June 2011.
The primary endpoint was the antibody level against each influenza vaccine strain measured at birth in cord blood by HAI, allowing the calculation of a GMT for each strain in each group. The secondary endpoint was the seroprotection level as defined as HAI titer ≥1:40 in each group.
Participants eligible for the study were pregnant women delivering after 36 weeks of gestation, that is, the time required for efficient maternal antibody transfer and accepting to provide a cord blood sample just after delivery. Exclusion criteria were as follows: known or suspected impairment of maternal immune function resulting from congenital immune deficiency, immunosuppressive therapy [including chemotherapy, radiation therapy and long-term systemic corticosteroid therapy (prednisone or equivalent at >20 mg/d for >2 consecutive weeks) within the past 3 months] or HIV infection. ILI was identified through medical history (high fever with rhinorrhea, sore throat, asthenia and/or myalgia), without laboratory confirmation.
The study was approved by the Ethics Committee of Gynecology, Obstetrics and Pediatrics of the University Hospital of Geneva. Recruitment (information, signing of consent forms and filling in of questionnaires) occurred at the outpatient clinic of the maternity ward of the University Hospital of Geneva during prenatal checkups and at times of vaccination. Women not contacted prior to labor were also eligible and could be recruited while in the labor suite, before delivery. All women were at a normal risk setting.
Sampling Procedures and Laboratory Assays
A sample of umbilical cord blood (3 mL) was taken just after birth and transferred to the Laboratory of Vaccinology of the University Hospital of Geneva. Samples were centrifuged and aliquoted, and sera were frozen at 20°C until assessed. All samples were analyzed by HAI using A/California/7/2009 (H1N1), A/Perth/16/2009 (H3N2)-like virus and B/Brisbane/60/2008-like virus strains. HAI assays were performed according to routine procedures, essentially as described.22,23
Descriptive statistics were performed; the presence of influenza-specific antibodies in serum from cord blood was assessed. The proportion of samples with protective antibody titer (HAI seropositive:titer ≥ 1:40) and the HAI GMT with corresponding 95% confidence intervals (CIs) were calculated for each group according to the maternal vaccine status. For vaccinated mothers, the influence of the timing of immunization during pregnancy (number of weeks elapsed between vaccination and delivery) was assessed. Comparisons of GMTs and seroprotection rates between children of vaccinated and nonvaccinated mothers were compared using t and χ2 tests. A univariate regression analysis was performed to assess the effect of different factors on the likelihood of increased antibody titer or seroprotection rates. Linear regressions were performed to assess effects on antibody titers, a positive rate ratio (RR) suggesting an increased likelihood of a higher antibody titer. Logistic regressions were performed to assess the effect on the probability of seroprotection, a positive odds ratio (OR) suggesting that there was an increased probability of seroprotection. Multivariate analysis was performed with adjustments for maternal age, previous pandemic influenza vaccination or history of previous ILI and the interval between time of vaccination and delivery using a linear model for GMT and a logistic model for seroprotection. Goodness of fit of the regression models was checked (distribution of the residuals for the linear model and Hosmer-Lemeshow test for the logistic model). P <0.05 was considered as statistically significant. Stata (version 9.1; Stata Corp, College Station, TX) was used for all statistical analyses.
A total of 188 pregnant women were enrolled between December 2010 and June 2011. Among these, 101 mothers had been vaccinated against influenza during their pregnancy and 87 had not. The mean maternal age was 32 ± 5.6 years, the mean gestational age was 40 ± 1.5 weeks and 90 of the 188 women (48%) were primiparous. Population characteristics of the 2 groups (immunized and nonimmunized mothers) were similar except that significantly more vaccinated women had also been vaccinated against H1N1 (P = 0.01) and seasonal influenza (P = 0.02) in 2009. Mean gestational age at immunization was 29 ± 8 weeks. The mean time interval between vaccination and delivery was 77 ± 54 days (Table 1). All women had received nonadjuvanted, inactivated 2010/2011 trivalent influenza vaccines, mostly Mutagrip (Sanofi Pasteur MSD, Lyon, France), a split vaccine including mostly the neuraminidase and hemagglutinin surface antigens. This study was not designed to analyze vaccine safety data, but no safety concerns were reported by participating vaccinated women.
HAI Titers Against Influenza A (H1N1), A (H3N2) and B in Cord Serum Samples
Figure 1 presents the GMTs with corresponding 95% CI of anti-influenza A (H1N1), A (H3N2) and B HAI antibodies in cord serum samples. The GMTs were significantly higher in the cord serum samples of infants born to vaccinated compared with nonvaccinated women (P < 0.001). Similarly, significantly more babies (85%) born to vaccinated mothers reached seroprotection (HAI ≥1:40) for all 3 influenza strains, whereas less than one-third of neonates of nonimmunized mothers had seroprotective titers (Fig. 1).
Influence of the Interval Between Maternal Vaccination and Delivery on Cord Blood GMTs and Seroprotection Rates
Figure 2 and Table 2 describe the influence of the interval between vaccination and delivery on the GMTs and seroprotection rates in the cord serum samples of infants born to women vaccinated between <15 days and >121 days before delivery. Immunization of pregnant women <15 days before delivery did not significantly increase cord GMTs and seroprotection rates compared with nonvaccinated women. The regression analysis indicated a nonsignificant increase of GMTs (1–5.2-fold) and seroprotection rates (1–3.4-fold).
In contrast, vaccinating pregnant women at any time during pregnancy (but >15 days before delivery) conferred similar HAI-GMTs seroprotection rates (80–94%), which was confirmed by regression analysis (see below).
Determinants of Cord Antibodies to Specific Influenza Antigens Using HAI
Univariate linear regression and logistic analyses were performed to identify the determinants of HAI-GMTs and seroprotection rates, respectively (Table 3). GMTs were 7.0–8.8 times higher and seroprotection rates were 11.3–14.9 times higher in cord serum samples of vaccinated compared with nonvaccinated women, for all vaccine strains. Maternal age (increase per 10 years) significantly reduced influenza antibodies against the H1N1 and H3N2 strains. Gestational age at delivery, parity or prior immunization with 2009/2010 seasonal influenza vaccine exerted no significant influence. Prior 2009, immunization against pandemic influenza H1N1 was associated with markedly higher GMTs (RR = 4.5, P < 0.001) and seroprotection rates (OR = 5.9, P = 0.002) against the homologous H1N1 strain included in the seasonal 2010/2011 vaccine. In univariate analysis, previous pandemic immunization was also associated with higher seroprotection rates against the H3N2 and B strains. In contrast, the influence of a 2009 seasonal immunization was not significant, despite the inclusion of the same B/Brisbane/60/2008 strain in 2010/11. A history of ILI after August 2009 increased the H1N1 GMTs (RR = 1.7, P = 0.02), but not the likelihood of seroprotection (OR = 1.5, P = 0.1) and did not influence antibody levels to the H3N2 and B viral strains.
The interval between immunization and delivery showed that vaccinating pregnant women <15 days before delivery did not allow for sufficient time to produce an increase in HAI antibody titers or the rate of seroprotection compared with nonvaccinated women. In contrast, vaccination at any time during pregnancy prior to 15 days before delivery consistently increased HAI antibody titers and the probability of seroprotection (RR between 5.2 and 14.4 and OR between 6.4 and 37.9, depending on the length of the interval and the type of strain).
Factors identified as significant in univariate regression analysis were added to the multivariate model (increasing age per 10 years, H1N1 vaccination in 2009, a previous history of ILI after August 2009 and the interval between vaccination and delivery). The effect of vaccination during pregnancy in comparison with no vaccination was analyzed by subcategorizing the interval of time between vaccination and delivery. This was done to gain a better understanding of the dynamics of serological response as a function of time. The variable “vaccination during pregnancy” was not included in the multivariate analysis as the different intervals between vaccination and delivery were all subcategorized from the variable vaccination during pregnancy.
Vaccination >15 days before delivery was the strongest independent predictor of a cord blood GMT increase (RR = 5–17) and of an increased likelihood of seroprotection (OR = 5.8–34.4), depending on the vaccine strain and the interval between vaccination and delivery. The effect of the intervals showed that vaccinating mothers <15 days before immunization did not increase significantly the GMT and the probability to be protected for newborns of vaccinated mothers compared with newborns of nonvaccinated mothers. Second, pandemic H1N1 immunization in 2009 was also associated with higher GMTs (RR = 3.1) and an increased likelihood of seroprotection (OR = 5.9), however, only for the H1N1 strain. There was no residual impact of pandemic immunization on the other vaccine strains, indicating that univariate analysis was influenced by confounders. The attenuating effect of increasing age on vaccine responses remained significant in multivariate analysis, although limited to H1N1 (GMTs and seroprotection rates) and H3N2 (GMTs only). The goodness of fit of the multivariate model was adequate. The distribution of the residuals for the linear regression was found to be approximately normal, and the goodness of fit for the logistic regression was not rejected by the Hosmer-Lemeshow test.
This study shows that vaccination of pregnant women facilitates significant transplacental antibody transfer to the babies for all strains of influenza virus contained in the 2010/2011 influenza vaccine. Less than one-third of neonates from nonimmunized mothers had seroprotective titers compared with 85% among vaccinated women. Thus, past maternal exposure to influenza is not sufficient to confer neonatal immunity, even during a season immediately following a pandemic outbreak of a strain with a high attack rate.24 Previous studies have also reported a significant increase in antibody levels to influenza virus in the cord blood of infants born to vaccinated mothers compared with nonvaccinated mothers.16–18 However, there were no data on the intensity of this effect and the impact of confounding factors. In this study, we report for the first time the impact of variables as history of H1N1 vaccination, past ILI and age on blood cord GMTs and seroprotection rates after trivalent influenza vaccination (Table 4; multivariate regression analysis). We observed a 5–17 fold increase in antibody titer and a 6–34 fold increase in seroprotection rates for women immunized earlier than 15 days before delivery. These are important findings reinforcing that vaccination of pregnant women is indeed a very efficient way to protect newborns against influenza illness during the first weeks or months of life.
One-third of unvaccinated mothers had cord HAI antibodies above the seroprotective threshold. As influenza immunization in Switzerland has been recommended only since the preceding year for pregnant women and not at all for healthy nonpregnant women, this is best explained by cross-reactive antibodies elicited by past influenza infection. Around 16% of women reported a previous ILI since August 2009. Similar prevalence rates of seroprotective H1N1 antibody titers in unvaccinated women have been reported before.17,25
An important and new finding of the present study is that previous vaccination with H1N1 in 2009 (but not previous ILI since 2009) increased the H1N1 GMT and seroprotection rates for H1N1. This suggests that previous vaccination may increase the antibody response to subsequent vaccination during pregnancy, probably through reactivation of immune memory.
One remaining question concerns the ideal time during pregnancy a woman should be vaccinated, to facilitate optimal production and transfer of antibodies and therefore maximal protection of the neonate. We found that vaccinating pregnant women during the second and third trimester (as recommended in Switzerland) but at least 15 days before delivery induced a 5–17 fold increase in the GMTs and a 6–34 fold increase in seroprotection rates compared with nonvaccinated women. Immunizing mothers <15 days before delivery was not sufficient to significantly increase antibody titers in the newborn. Vaccinating close to delivery can therefore probably be considered as not sufficient in conveying immunity to the newborn, although a certain indirect effect may still be beneficial through the protection of the mother and therefore a reduced risk of transmission to the baby during the first months of life. Similarly, protection levels seemed to diminish if the flu jab was given >120 days before delivery (Fig. 2). However, numbers were too small to allow enough power for valid conclusions. Little data is available comparing the impact of different time intervals between influenza vaccination and delivery on cord blood antibody levels: we only identified 1 study comparing titers after vaccination in the second and third trimester; no difference was reported.18 Hence, it can probably be said that the second and third trimester remains a relatively extended ideal interval for vaccination during pregnancy, allowing for flexibility in administration.
The present study has several limitations. First, for practical reasons, it was not possible to conduct a randomized-controlled trial. We carried out a cross-sectional study, increasing the potential impact of confounding factors. Among the group of women vaccinated during pregnancy, for example, a higher percentage may have also been open to immunization in the past and would therefore have higher baseline titers. Although multivariate analysis aimed to control at least partly for this factor by including the potential confounder of H1N1 vaccination in 2009, a possible impact cannot be completely excluded. Second, the study was conducted between December 2010 and June 2011, that is, partly overlapping with the influenza season. It is thus possible that women with subclinical influenza infection were enrolled, which may have increased their antibody titers independently of immunization. However, this should equally affect both study groups, and any difference should therefore be random. Third, for practical reasons, we did not collect paired maternal and umbilical cord blood samples. Cord titers may therefore not reflect maternal immune responses 100%. However, a previous study by Steinhoff et al. assessing HAI antibody titers in maternal and cord blood samples reported comparable levels.21 Another study reported no direct immune response to influenza antigens in infants.26 For this study, it was therefore considered reasonable to assume that cord blood antibody levels reflect maternal and not fetal antibody responses. Finally, it was not the scope of this study to follow-up infants after delivery to monitor antibody levels and correlate them to possible influenza infections, which would have allowed for conclusions on true vaccine efficacy. From past studies, it can be expected that protection of the newborn by maternal antibody may last for up to 6 months.18–21,27–29 A randomized-controlled trial in Bangladesh demonstrated that women vaccinated with inactivated influenza vaccine during pregnancy presented significantly higher influenza antibody titers at delivery than a control group vaccinated with pneumococcal polysaccharide vaccine. Their infants also presented reduced rates of influenza infection during the first 6 months of life (63% reduction).20,21 Two retrospective studies neither report any reduction in ILI outpatient visits for vaccinated women nor were hospitalization rates due to pneumonia or influenza infection lower for infants born to vaccinated mothers. However, maternal immunization rates (0.7–20%) were very low in these 2 studies.30,31 Overall literature thus seems to imply the efficacy of maternal influenza vaccination in the prevention of neonatal morbidity.
This study shows that influenza vaccination at any time during the second or third trimester of pregnancy, but at least 15 days before delivery, is an effective strategy to increase cord blood antibody titers and seroprotection rates in neonates. It is thus likely to reduce the risks of influenza infection during the first months of life.
The authors thank the women who participated in the study and all the midwives who conducted the recruitment and blood sampling.
1. Poehling KA, Edwards KM, Weinberg GA, et al.New Vaccine Surveillance Network. The underrecognized burden of influenza in young children. N Engl J Med. 2006;355:31–40
2. Chiu SS, Lau YL, Chan KH, et al. Influenza-related hospitalizations among children in Hong Kong. N Engl J Med. 2002;347:2097–2103
3. Izurieta HS, Thompson WW, Kramarz P, et al. Influenza and the rates of hospitalization for respiratory disease among infants and young children. N Engl J Med. 2000;342:232–239
4. Neuzil KM, Mellen BG, Wright PF, et al. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. N Engl J Med. 2000;342:225–231
5. Glezen WP, Taber LH, Frank AL, et al. Influenza virus infections in infants. Pediatr Infect Dis J. 1997;16:1065–1068
6. Neuzil KM, Zhu Y, Griffin MR, et al. Burden of interpandemic influenza in children younger than 5 years: a 25-year prospective study. J Infect Dis. 2002;185:147–152
7. Bhat N, Wright JG, Broder KR, et al.Influenza Special Investigations Team. Influenza-associated deaths among children in the United States, 2003-2004. N Engl J Med. 2005;353:2559–2567
8. Morishima T, Togashi T, Yokota S, et al.Collaborative Study Group on Influenza-Associated Encephalopathy in Japan. Encephalitis and encephalopathy associated with an influenza epidemic in Japan. Clin Infect Dis. 2002;35:512–517
9. Kappagoda C, Isaacs D, Mellis C, et al. Critical influenza virus infection. J Paediatr Child Health. 2000;36:318–321
10. Gaunt G, Ramin K. Immunological tolerance of the human fetus. Am J Perinatol. 2001;18:299–312
11. Dodds L, McNeil SA, Fell DB, et al. Impact of influenza exposure on rates of hospital admissions and physician visits because of respiratory illness among pregnant women. CMAJ. 2007;176:463–468
12. Steinhoff MC, Omer SB. A review of fetal and infant protection associated with antenatal influenza immunization. Am J Obstet Gynecol. 2012;207(3 suppl):S21–S27
13. Harper SA, Fukuda K, Uyeki TM, et al.Centers for Disease Control and Prevention (CDC) Advisory Committee on Immunization Practices (ACIP). Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep. 2004;53(RR-6):1–40
14. . World Health Organization (WHO). Vaccines against influenza WHO position paper. November 2012. Wkly Epidemiol Rec. 2012;87:461–476
15. Siegrist CA. Mechanisms by which maternal antibodies influence infant vaccine responses: review of hypotheses and definition of main determinants. Vaccine. 2003;21:3406–3412
16. Englund JA. Maternal immunization with inactivated influenza vaccine
: rationale and experience. Vaccine. 2003;21:3460–3464
17. Puleston RL, Bugg G, Hoschler K, et al. Observational study to investigate vertically acquired passive immunity in babies of mothers vaccinated against H1N1v during pregnancy. Health Technol Assess. 2010;14:1–82
18. Eick AA, Uyeki TM, Klimov A, et al. Maternal influenza vaccination and effect on influenza virus infection in young infants. Arch Pediatr Adolesc Med. 2011;165:104–111
19. Zuccotti G, Pogliani L, Pariani E, et al. Transplacental antibody transfer following maternal immunization with a pandemic 2009 influenza A(H1N1) MF59-adjuvanted vaccine. JAMA. 2010;304:2360–2361
20. Zaman K, Roy E, Arifeen SE, et al. Effectiveness of maternal influenza immunization in mothers and infants. N Engl J Med. 2008;359:1555–1564
21. Steinhoff MC, Omer SB, Roy E, et al. Influenza immunization in pregnancy—antibody responses in mothers and infants. N Engl J Med. 2010;362:1644–1646
22. Siegrist CA, van Delden C, Bel M, et al.H1N1 Study Group; Swiss HIV Cohort Study (SHCS). Higher memory responses in HIV-infected and kidney transplanted patients than in healthy subjects following priming with the pandemic vaccine. PLoS One. 2012;7:e40428
23. Gabay C, Bel M, Combescure C, et al.H1N1 Study Group. Impact of synthetic and biologic disease-modifying antirheumatic drugs on antibody responses to the AS03-adjuvanted pandemic influenza vaccine
: a prospective, open-label, parallel-cohort, single-center study. Arthritis Rheum. 2011;63:1486–1496
24. Glatman-Freedman A, Portelli I, Jacobs SK, et al. Attack rates assessment of the 2009 pandemic H1N1 influenza A in children and their contacts: a systematic review and meta-analysis. PLoS One. 2012;7:e50228
25. Tsatsaris V, Capitant C, Schmitz T, et al.Inserm C09-33 PREFLUVAC (Immunogenicity and Safety of an Inactivated Nonadjuvanted A[H1N1v] Influenza Vaccine
in Pregnant Women) Study Group. Maternal immune response and neonatal seroprotection from a single dose of a monovalent nonadjuvanted 2009 influenza A(H1N1) vaccine: a single-group trial. Ann Intern Med. 2011;155:733–741
26. Englund JA, Mbawuike IN, Hammill H, et al. Maternal immunization with influenza or tetanus toxoid vaccine for passive antibody protection in young infants. J Infect Dis. 1993;168:647–656
27. Håberg SE, Trogstad L, Gunnes N, et al. Risk of fetal death after pandemic influenza virus infection or vaccination. N Engl J Med. 2013;368:333–340
28. Esposito S, Bosis S, Morlacchi L, et al. Can infants be protected by means of maternal vaccination? Clin Microbiol Infect. 2012;18(suppl 5):85–92
29. Blanchard-Rohner G, Siegrist CA. Vaccination during pregnancy to protect infants against influenza: why and why not? Vaccine. 2011;29:7542–7550
30. Black SB, Shinefield HR, France EK, et al.Vaccine Safety Datalink Workgroup. Effectiveness of influenza vaccine
during pregnancy in preventing hospitalizations and outpatient visits for respiratory illness in pregnant women and their infants. Am J Perinatol. 2004;21:333–339
31. France EK, Smith-Ray R, McClure D, et al. Impact of maternal influenza vaccination during pregnancy on the incidence of acute respiratory illness visits among infants. Arch Pediatr Adolesc Med. 2006;160:1277–1283