Flu vaccination of pregnant women and serious adverse events in their offspring

Flu vaccination of pregnant women and serious adverse events in their offspring

InformaMedici Project: This article was created with the intention of being a tool for citizens, capable of informing and updating their doctor on the latest news and / or publications. If the orders of doctors and scientific societies do not fulfill this task, the people will take care of it.

Download the PDF document and inform your doctor


Int J Environ Res Public Health. 2019 Nov 7; 16 (22).

Influence Vaccination of Pregnant Women and Serious Adverse Events in the Offspring

Alberto Donzelli - Executive Committee of the “Alineare Sanità e Salute” Foundation, 20122 Milan, Italy

Abstract

Pregnant women are increasingly considered a priority group to be vaccinated against the flu, but the evidence for this mainly refers to observational studies, subject to the "healthy vaccinated bias". Methods that use propensity scores, sometimes applied, reduce but cannot eliminate the residual confounding factors.
Meta-analyzes of observational studies show relative risks well beyond the thresholds that would confirm the efficacy of a universal vaccination for pregnant women without the need for randomized controlled trials (RCTs). Critical articles showed that in the four RCTs where the effects of this vaccination were investigated, there was a trend towards higher mortality in the offspring. There was a significant excess of in the larger RCT
presumed / serious neonatal infections, and also of total serious adverse events.
Many widely publicized observational results (regarding hormone replacement therapy in menopause, supplements with vitamin D, omega-3 fatty acids, etc.) have been refuted by RCTs. Therefore, as far as we know, international pressure to consider this vaccination as a "standard of care" is not justified. Furthermore, there is a risk of precluding subsequent independent RCTs for "ethical reasons", such as "not to deprive pregnant women of the group
to control the benefits of flu vaccination. " Instead, before promoting national campaigns for the universal vaccination of pregnant women, it is necessary to implement further large, independent and reassuring RCTs, also having the courage to challenge a current paradigm.
Until then, flu vaccination should be offered to pregnant women only after providing adequate information regarding uncertainties about its safety, to allow for truly informed choices, and in any case also promoting other protective behaviors.

Keyword: flu vaccination, pregnant women, healthy vaccinated bias, real world trials vs randomized trials, offspring death, increased serious adverse events, informed consent

Conflict of interest: the author declares that he has no conflict of interest.

finance: this study received no external funding


1. Introduction

WHO considers pregnant women to be a priority group for flu vaccination1 and global pressure is underway to make such vaccination universal. However, a Cochrane review of influenza vaccination in healthy adults2 reduced its effectiveness, highlighting that "the NNV (number of pregnant women who need to be vaccinated to avoid one getting flu) was 55 for mothers and 56 for newborns ". The review continued:

"The protective effect of vaccination for mothers and newborns was very modest ... more modest than that observed in other populations considered in this review ... we are not sure of the protection offered to pregnant women against ILI (influenza- like illness) and against the flu from the inactivated flu vaccine, or at least this protection was very limited ".

The first condition for a universal vaccination campaign should be the undisputed effectiveness of the procedure. In light of Cochrane's evidence-based conclusions, the WHO statement that pregnant women should be "the flu vaccination priority group" (statement appeared in 20121, before the publication of the three largest RCTs on this topic), while still coming from an authoritative source, it is not adequately supported by evidence.

The practical effectiveness of the vaccine varies based on correspondence with circulating strains. However, the NNV among pregnant women in the Cochrane review does not differ much from the figure calculated for healthy adults, which summarizes all the available evidence, including 52 clinical studies on more than 80.000 people, conducted during individual flu seasons, in the Americas and in Europe, over 40 years. However, it is also impossible to predict the effectiveness of the vaccine in a specific year, so that universal vaccination of pregnant women will always have the same universal denominator, independently of correspondence with a given strain, which can only be verified a posteriori.

The second condition for a universal vaccination should be its proven safety. Evidence on the efficacy and safety of influenza vaccination in pregnant women is still mainly based on observational studies.


2. Objectives, Materials and Methods

The goal of this perspective article is to present a different point of view on the flu vaccination of pregnant women. Both original data and personal opinions are presented, also analyzing new observational evidence, formulating possible alternative explanations for some favorable outcomes found in the offspring of pregnant women vaccinated, and quantifying the limited benefits in their offspring reported in recent meta-analyzes of observational studies ( with the addition of a maximum of two RCTs). The article adds some new evidence from RCTs available to date, raising new doubts about the safety of flu vaccination in pregnancy. The main articles analyzed are summarized in Table 1.

Finally, this perspective proposes some coherent implications in terms of health policies.


3. Results and their discussion

3.1. Observational studies and biases of healthy vaccinates

The observational study design is subject, among others, to the so-called "healthy adherent bias": individuals who adhere to preventive therapies are, at the same time, more likely to follow healthier lifestyles than patients who do not adhere. to such preventive strategies3,4. A healthy lifestyle includes diet, exercise, lower alcohol consumption, less risky behavior, and the search for better health care. These characteristics - difficult to recover in administrative databases - are associated with the outcomes of morbidity and mortality in observational studies.

Similarly, in the vaccination field, there is the "healthy vaccinated bias", which leads to an overestimation of the efficacy and safety of the vaccine5,6. Flu vaccination in the elderly has already provided a clear example of this type of bias7. Systematic studies of this bias and its opposite, the "confusion bias from indication"8 they showed that statistical adjustments can correct the indication confounding bias well enough, while they cannot adequately compensate for the healthy vaccinated bias.

This bias can be important in pregnant women: women with greater self-discipline and more educated generally have healthier behaviors and more adhering to the recommendations of doctors, obstetricians and health authorities9which greatly increases the likelihood of them being vaccinated9; unlike more disadvantaged women from a socio-economic and cultural point of view10,11. Other types of bias in observational studies were highlighted in 201612, and their correction removed statistical significance from further alleged benefits.
Previous publications have illustrated this bias, showing structural differences between cohorts of pregnant women who are adherent or not to flu vaccination5,6. A study by the CDC13 showed that unvaccinated women were to some extent less affected by "high risk diseases", which resulted in a slightly lower chance of being vaccinated: 46,3% versus 54% (absolute difference in points of 7,7 , XNUMX%); however, this feature was completely overwhelmed by the fact that the unvaccinated had serious disadvantages in terms of educational and socio-economic levels13.

This is not an isolated case. European studies have confirmed important structural differences between the cohorts of vaccinated and unvaccinated pregnant women. A prospective French cohort study14 found that cohorts of unvaccinated mothers had several disadvantages. The determinants associated with non-vaccination in a multivariate logistic regression included geographic origin: sub-Saharan African origin had an aOR (adjusted odds ratio) of 5,4 (95% CI 2,3, 12,7-XNUMX), the North African origin an aOR of
2,5 (1,3-4-7), and Asian origin an aOR of 2,1 (1,7-2,6), compared to women of French and European origin. In addition, if compared with managers and those who carried out intellectual professions, the categories of peasants, craftsmen and traders had an aOR of 2,3 (2,0-2,6), the intermediate professions an aOR of 1,3 (1,0 , 1,6-2,5), female workers and manual workers an aOR of 1,4 (4,4-XNUMX).

The probability of not having received the pandemic flu vaccine was lower among women who had stopped smoking before or in early pregnancy, with an aOR of 0,6 (0,4-0,8) compared to nonsmokers, while current smokers had an aOR of 1,2 (0,8-1,8).
It is quite likely that the worst socio-economic and behavioral conditions of unvaccinated women are the ones to explain the worst outcomes in their children, without having to question a missed vaccination. Having at least one associated co-morbidity acted in the opposite direction, but to a lesser extent and without achieving statistical significance: aOR 1,2 (0,9-1,5)14.

Even a significant obstetric history tended to increase the propensity of pregnant women to flu vaccination, but the difference did not reach statistical significance. Some of the characteristics mentioned above are not measured in pharmacological databases or are difficult to acquire even by applying a weighting with propensity score, because it is impossible to include unmeasured or unknown confounding factors in the propensity score15. Adherence to vaccination can also be associated with greater confidence in the proposed intervention, which in turn can lead to better outcomes6to some extent.
Furthermore, the propensity scores and their reliability strictly depend on how they are built, so the authors should always provide the independent auditors with the details of the methods used, so that the process can be replicated, and in any case the possibility that the results are affected by residual confounding factors.

A systematic review of the safety outcomes associated with flu vaccination in pregnancy16 it did not include randomized controlled trials (RCTs), although one of them was mentioned. Subsequently other systematic reviews were carried out, and we reported the results of the two most recent.

A systematic review on the safety of the inactivated flu vaccine in pregnancy regarding outcomes at birth17 it included 39 observational studies (25 retrospective and 9 prospective cohort studies, three case-control studies and 2 cross-sectional studies) and one RCT18. AORs were for: preterm birth (PTB) 0,87 (0,78-0,96), low birth weight (LBW) 0,82 (0,76-0,89), abnormalities
congenital 1,03 (0,99-1,07), small by gestational age (EMS) 0,99 (0,94-1,04), stillbirths 0,84 (0,65-1,08).

The latest systematic review19 carried out a Bayesian meta-analysis, reviewing as many articles of reasonable quality as possible. It found no significant decrease for any of the following adverse birth outcomes: PTB (OR 0,945, 95% credibility intervals (CrI) 0,736-1,345, P = 73,3%), LBW (OR 0,928, 95% CrI 0,432- 2,112, P = 76,6%), EMS (OR 0,971, CrI 95% 0,249-4,217, P = 63,3%), congenital malformations (OR 1,026, CrI 95% 0,687-1,600, P = 38,0%) , fetal death (OR 0,942, 95% CrI, 0,560-1,954, P = 61,6%).
The conclusion is: "The results show evidence of a null association between maternal flu vaccination and adverse birth outcomes."

After adjustments for the season at the time of vaccination and for the country's income level, the overall estimates comprising only cohort studies showed a significant reduction limited only to fetal deaths; however this is an implicit admission that fetal deaths tended to be higher in the two RCTs considered18,20, as it was in fact.

Lancet magazine recently reported an interesting debate about why "real world" research cannot replace RCTs to determine effectiveness.21. RCT may not be necessary to ascertain causality in rare situations where "a confounding factor is unlikely to account for associations with extreme relative risks, such as those of less than 0,25 or more than 4"21. Instead, "associations with relativ risks ranging from 0,5 to 2 are those most commonly reported in the analysis of real data, and those most susceptible to unrecognized confounding factors" 21. In these situations, RCTs are "irreplaceable". The conclusion is "In the absence of randomization, analyzes of most observational data from the" real world ", regardless of how sophisticated they are, can only generate assumptions"21. A letter of three
Cochrane reviewers reported even more stringent criteria: "Other studies have suggested different thresholds, with RR of 10 or higher, or of 5 or higher (or RR less than 0,2) in order to avoid an RCT" 22. Each of these thresholds would certainly have made other RCTs request in the face of the results of systematic reviews of observational studies such as the two mentioned above17, 19, when the two RCTs were excluded from the last.

A recently published retrospective cohort study23 found no association between 2009 women who received the pandemic H1N1 vaccine during pregnancy and most pediatric health indicators at 5 years of age. The study also cited the four RCTs commented on later, but avoided mentioning some articles5,6,24 who had questioned the reassuring conclusions of the authors of the aforementioned RCTs, not supported by their own data. This omission was made despite one of the main authors23 recently posted a comment25 to these criticisms, receiving a prompt reply letter24. An editorial from the retrospective study23 concludes: "Vaccination of pregnant women saves lives"26, but the RCTs tell a different story (Table 1).

Table 1 - Main articles reviewed and discussed in this perspective

flu reports 1

3.2. RCT: some new tests
Meta-analysis of observational studies (e.g. 17) show relative risks well beyond the thresholds that could support the effectiveness of this vaccination even without the need for RCT21,22.

However, four RCTs are currently available. After the first small RCT in Bangladesh27, the Bill & Melinda Gates Foundation has subsidized three large RCTs30 blinded, in part to overcome the problems of validity of the observational tests. One of them, the Matflu18, was implemented in a medium-high income country. It is considered "low risk of bias" by Cochrane reviewers2, with a number of subjects to be vaccinated (NNV) of 55 to avoid flu in mothers. The other two RCTs were in low-income countries20,28, where the expected benefits were greater.

Next to Matflu placebo controlled RCT18, included in the Cochrane2 review, the Nepalese RCT can also be considered20 placebo-controlled, where the NNV (unspecified) appears to be around 20; and the other two RCTs with active control: the small RCT in Bangladesh27, with NNV 17 (control group with 23-valent pneumococcal vaccine) and the much larger RCT in Mali28, with NNV 99 (control group with quadrivalent anti-meningococcal vaccine).

The overall NNV is not far from Cochrane estimates, close to the NNV for healthy adults2. In placebo-controlled RCTs, the best tests showed a "very modest" efficacy of vaccination at the population level2 and an excess of maternal local adverse events5,24. Additionally, offspring mortality tended to be higher among women vaccinated with flu shots than in control groups. The overall serious adverse events (SAEs) tended to be more numerous, as was shown in an infographic24. In the larger RCT, the excess of SAE among newborns was statistically significant, according to my personal calculations: the number of live births per vaccine group was 2064 for vaccinated women, 2041 in the group with active control; the total SAEs were 225 (10,90%) and 175 (8,57%) respectively28 S7 table. Therefore the RR was 1,27; 95% CI 1,05-1,53; NNH 42,98 (number of subjects that must be treated to obtain damage). A sensitivity analysis can exclude major congenital malformations, 6 (flu group) and 4 (control group), unrelated to vaccinations received at the 28th week of gestation from the SAE count. The analysis again shows a significant excess of SAE in the flu group: RR 1.27; 95% CI 1,05-1,53; NNH 44,80.

The abstract reports: "Presumptive neonatal infections were more common among infants in the group with inactivated trivalent flu vaccine than in the group with quadrivalent anti-meningococcal vaccine (n = 60 vs n = 37; p = 0.02)" 28, without however clearly specify that:

  • these presumed neonatal infections were SAE, not generic infections
  • Total SAEs (which included neonatal deaths: 52 vs. 37 in the control group) were also significantly higher in the offspring of flu-vaccinated mothers
  • the total of the SAE (not just the individual categories of SAE) constitutes a standard "hard" outcome.

An international debate that has emerged about the possible fetal adverse outcomes of this vaccination in pregnant women has led to a reanalysis of the RCT Matflu29. The authors concluded: "We have not found a beneficial effect of inactivated trivalent influenza vaccination in pregnancy against fetal adverse outcomes." But in reality the effectiveness of the vaccine tended to be worse for each indicator: fetal death (efficacy -21,2% [-150,8, 41,4]), LBW (-11,1% [-42,3 , 12,5]), EMS (-9,9% [-35,6, 11,0]), PTB (-21,3% [-60,5, 8,3]). In the analysis restricted to infants from mothers exposed to the flu season, the trend was equally unfavorable29.

The mechanism of hypothetical damage is unclear. It is possible that there is inflammatory stress related to this vaccination31 - 33. The inflammatory stress of a flu is certainly greater, but the ratio to consider is between 1 flu and 55 vaccinations. An interesting study34 covered 1.791.520 Swedish babies born over a long period of time, and found that fetal exposure to a maternal infection was associated with an increased long-term risk of neuro-psychiatric disease, even for mild infections of the urinary tract of the mother (UTI). In the discussion, the authors state: "we found convincing evidence that fetal exposure to infection (or inflammation) when the mother was hospitalized increased the risk to the baby ... during childhood or adulthood ... so regardless of whether the exposure was a serious maternal infection ... or an ICU during pregnancy ... the results were robust even after adjustment for an unknown confounding factor of moderate degree "34. In the absence of strong evidence of long-term benefit and safety, the precautionary principle should suggest to avoid any deliberate inflammatory stimulus during the vulnerable state of pregnancy, even if the authors34 - remaining within the current paradigm - they concluded by proposing primary prevention based on vaccinations, or on anti-inflammatory therapies.

There therefore appears to be a clear need for further broad, pragmatic RCTs conducted by independent institutions and researchers, with a long temporal extension35. A previous article suggested a way to fairly overcome "ethical problems" by recruiting only women who remained hesitant even after receiving balanced information on the pros and cons of vaccination during pregnancy, in the current state of knowledge35.

3.3. General comments
Given that observational studies on pregnant women (usually quite young and healthy) are subject to the bias of the healthy vaccinated, to promote pharmacological prevention intervention, particularly in the vulnerable period of pregnancy, public health authorities should not rely only or above all to observational tests5,6,24. This is even more true if the (insufficient) safety evidence from existing RCTs shows a trend in the opposite and alarming direction.

With the current state of knowledge, the international push and the proclaimed urgency to consider the flu vaccination of pregnant women "a standard of care" is not supported, and precludes new independent RCTs that could clarify the problem, with the topic of "considerations ethics, not to deny the benefits of vaccination to the placebo group ". This widespread attitude refers to what is described by epistemologists such as Kuhn36, when anomalies emerge that could call into question a dominant paradigm. At this point the majority of the scientific community can react with intolerance36, arriving even choosing to ignore the facts that could open a paradigm crisis, thus precluding the possibility of reaching more advanced syntheses. Instead, I agree with Popper's conclusion37 that the scientific attitude is a critical attitude, which does not seek (only) confirmations, but also seeks crucial controls potentially capable of arriving at falsifying a current theory, when anomalies may call it into question.

We should be very cautious about observational results not confirmed by correspondent RCTs. An example comes from hormone replacement therapy (HRT) to prevent chronic conditions in menopausal women. Estimated calculations of total damage versus benefit per 10.000 women / year associated with HRT are: estrogen + progestagen 971 damage vs 65 benefit; only estrogen 1329 damage vs 82 benefits38,39. Other examples are vitamin D supplements "for health"40, or omega-3 fatty acid supplements for cardiovascular disease41.


4. Conclusions

The RCTs available today raise safety signals about possible events that threaten the life of the offspring of mothers vaccinated against the flu during pregnancy. Before promoting universal paginvaccination in pregnancy, further extensive, independent and reassuring RCTs are needed, avoiding to hinder them in the name of "ethical reasons" without adequate foundation. The safety variables to be evaluated are all the adverse events solicited, with special attention for the SAE, within an extended time window and with many years of follow-up after the interruption of the double blind, to also detect subtle emerging differences in any direction in the long term.

In the meantime, vaccination could be offered to pregnant women, but informing them in a balanced way of the existing uncertainties (and also on the fact that most flu syndromes are not caused by the flu virus and are not preventable with vaccination), to allow truly an informed choice and consent. The claim that the offspring of unvaccinated mothers could suffer serious consequences should be avoided or balanced, since that the current RCTs show that the opposite may even be true. In addition, other previously detailed protective behaviors should also be promoted5.


Bibliography

  • 1. World Health Organization. Vaccines against influenza. WHO position paper. Geneve: Wkly Epidemiol. Rec. 2012, 87, 461–476. PMID: 23210147
  • 2. Demicheli, V .; Jefferson, T .; Ferroni, E .; Rivets, A .; Di Pietrantonj, C. Vaccines for preventing influenza in healthy adults. Cochrane Database Syst. Rev. 2018, CD001269. [PMID: 29388196]
  • 3. Simpson, SH; Eurich, DT; Majumdar, SR; Padwal, RS; Tsuyuki, RT; Varney, J. A metaanalysis of the association between adherence to drug therapy and mortality. BMJ 2006, 333, 15. [PMID: 16790458]
  • 4. Dormuth, CR; Patrick, AR; Shrank, WH; Wright, JM; Glynn, RJ; Sutherland, J .; Brookhart, MA Statin adherence and risk of accidents: A cautionary tale. Circulation 2009, 119, 2051–2057. [PMID: 19349320]
  • 5. Donzelli, A. Influenza vaccinations for all pregnant women? Better evidence is needed. Int. J. Environ. Res. Public Health 2018, 15, 2034; doi: 10.3390 / ijerph15092034. [PMID: 30231471]
  • 6. Donzelli, A. Influenza vaccinations for all pregnant women? So far the less biased evidence does not favor it. Human Vaccines & Immunother. 2019, doi: 10.1080 / 21645515.2019.1568161. [PMID: 30632885]
  • 7. Jackson, LA; Jackson, ML; Nelson, JC; Neuzil, KM; Weiss, NS Evidence of bias in estimates of influenza vaccine effectiveness in seniors. Int. J. Epidemiol. 2006; 35: 337-44. [PMID: 16368725]
  • 8. Remschmidt, C .; Wichmann, O .; Harder, T. Frequency and impact of confounding by indication and healthy vaccinee bias in observational studies assessing influenza vaccine effectiveness: A. systematic review. BMC Infect. Dis. 2015, 15, 429. [PMID: 26474974]
  • 9. Committee on Obstetric Practice and Immunization Expert Work Group; Centers for Disease Control and Prevention's Advisory Committee on Immunization, United States; American College of Obstetricians and Gynecologists. Committee opinion no. 608: Influenza vaccination during pregnancy. Obstet. Gynecol. 2014, 124, 648–651. [PMID: 25162283]
  • 10. Laenen, J .; Roelants, M .; Devlieger, R .; Vandermeulen, C. Influenza and pertussis vaccination coverage in pregnant women. Vaccines 2015, 33, 2125–2131. [CrossRef] [PMID: 25796339]
  • 11. Lindquist, A .; Kurinczuk, JJ; Redshaw, M .; Knight, M. Experiences, utilization and outcomes of maternity care in England among women from different socio-economic groups: Findings from the 2010 National Maternity Survey. BJOG 2015, 122, 1610–1617. [PMID: 25227878]
  • 12. Vazquez-Benitez, G .; Kharbanda, EO; Naleway, AL; Lipkind, H .; Sukumaran, L .; McCarthy, NL; Omer, SB; Qian, L .; Xu, S .; Jackson, ML; et al. Risk of Preterm or Small-for-Gestational-Age Birth After Influenza Vaccination During Pregnancy: Caveats When Conducting Retrospective Observational Studies. Am. J. Epidemiol. 2016; 184: 176-186. [PMID: 27449414]
  • 13. Kahn, KE; Black, CL; Ding, H .; Williams, WW; Lu, PJ; Fiebelkorn, AP; Havers, F .; D'Angelo, DV; Ball, S .; Fink, RV; et al. Influenza and Tdap vaccination coverage among pregnant women - United States, April 2018. MMWR Morb. Mortal. Wkly Rep. 2018, 67, 1055–1059. [PMID: 30260946]
  • 14. Freund, R .; Le Ray, C; Charlier, C .; Avenell, C .; Truster, V .; Tréluyer, JM .; Skalli, D .; Ville, Y .; Goffinet, F .; Launay, O .; Inserm COFLUPREG Study Group. Determinants of Non-Vaccination against Pandemic 2009 H1N1 Influenza in Pregnant Women: A Prospective Cohort Study. PLoS ONE 2011, 6, e20900. [PMID: 21695074]
  • 15. Brookhart, MA; Wyss, R .; Layton, JB; Stürmer, T. Propensity score methods for confounding control in nonexperimental research. Circ Cardiovasc Qual Outcomes 2013, 6, 604–611. [PMID: 24021692]
  • 16. McMillan, M .; Porritt, K .; Kralik, D .; Costs, L .; Marshall, H. Influenza vaccination during pregnancy: a systematic review of fetal death, spontaneous abortion, and congenital malformation safety outcomes. Vaccines 2015, 33, 2108–2117. [PMID: 25758932]
  • 17. Giles, ML; Krishnaswamy, S .; Macartney, K .; Cheng, A. The safety of inactivated influenza vaccines in pregnancy
    for birth outcomes: a systematic review. Hum Vaccin Immunother 2018, 15, 687–699. [PMID: 30380986]
  • 18. Madhi, SA; Cutland, CL; Kuwanda, L .; Weinberg, A .; Hugo, A .; Jones, S .; Adrian, PV; van Niekerk, N .; Treurnicht, F .; Ortiz, JR; et al. Flu vaccination of pregnant women and protection of their infants. N Engl J Med 2014, 371, 918–931. [PMID: 2549428]
  • 19. Jeong, S .; Jang, EJ; Jo, IJ; Jang, S. Effects of maternal influenza vaccination on adverse birth outcomes: A systematic review and Bayesian meta-analysis. PLoS ONE 2019, 14, e0220910. https://doi.org/10.1371/journal.pone.0220910. [PMID: 31412058]
  • 20. Steinhoff, MC; Katz, J .; Englund, JA; Khatry, SK; Shrestha, L .; Kuypers, J .; Stewart, L .; Mullany, LC; Chu, HY; LeClerq, SC; et al. Year-round influences immunization during pregnancy in Nepal: a phase 4, randomized, placebo-controlled trial. Lancet Infect Dis 2017, 17, 981–989. [PMID: 28522338]
  • 21. Gerstein, HC; McMurray, J .; Holman, RR Real-world studies no substitute for RCTs in establishing efficacy. Lancet 2019, 393, 210–211. [PMID: 30663582]
  • 22. Djulbegovic, B .; Glasziou, P .; Chalmers, I. The importance of randomized vs non-randomized trials. Lancet 2019, 394, 234–235. [PMID: 31448731]
  • 23. Walsh, LK; Donelle, J .; Dodds, L .; Hawken, S .; Wilson, K .; Benchimol, EI; Chakraborty, P .; Guttmann, A .; Kwong, JC; MacDonald, NE; et al. Health outcomes of young children born to mothers who received 2009 pandemic H1N1 influenza vaccination during pregnancy: retrospective cohort study. BMJ 2019, 366, l41–151. [PMID: 31292120]
  • 24. Donzelli, A. Reply Letter to: Fell, DB; Omer, SB; Edwards, KM Influence immunization during pregnancy: toward a balanced assessment of safety evidence. Hum Vaccin Immmunother 2019, 1–3. doi.org/10.1080/21645515.2019.1599679. [PMID: 31009301]
  • 25. Fell, DB; Omer, SB; Edwards, KM Influence immunization during pregnancy: toward a balanced assessment of safety evidence. Hum Vaccin Immmunother 2019, 1–3. doi.org/10.1080/21645515.2019.1599679. [PMID: 31009301]
  • 26. Håberg, SE; Wilcox, AJ Flu vaccination in pregnancy. BMJ 2019, 366, l4454. [Embase Identification Number (PUI) L628451203]
  • 27. Zaman, K .; Roy, E .; Arifeen, SE; Rahman, M .; Raqib R .; Wilson, E .; Omer, SB; Shahid, NS; Breiman, RF; Steinhoff, MC Effectiveness of maternal influenza immunization in mothers and infants. N Engl J Med 2008, 359, 1555–1564. [PMID: 18799552]
  • 28. Adegbola, R .; Nesin, M .; Wairagkar, N. Immunogenicity and efficacy of influenza immunization during pregnancy: recent and ongoing studies. Amer J Obstetr Gynecol 2012, 207, S28–32. [PMID: 22920055]
  • 29. Tapia, MD; Sow, SO; Tamboura, B .; Tégueté, I .; Pasetti, MF; Kodio, M .; Onwuchekwa, U .; Tennant, SM; Blackwelder, WC; Coulibaly, F .; et al. Maternal immunization with trivalent inactivated influenza vaccine for prevention of influenza in infants in Mali: a prospective, active-controlled, observer-blind, randomized phase 4 trial. Lancet Infect Dis 2016, 16, 1026–1035. [PMID: 2726106]
  • 30. Simões, EAF; Nunes, MC; Carosone-Link, P; Madimabe, R .; Ortiz, JR; Neuzil, KM; Klugman, KP; Cutland, CL; Madhi, SA Trivalent influenza vaccination randomized control trial of pregnant women and adverse fetal outcomes. Vaccines 2019, 37, 5397–5403. [PMID: 31331777]
  • 31. Christian, LM; Iams, JD; Porter, K .; Glaser, R. Inflammatory responses to trivalent influenza virus vaccine among pregnant women. Vaccines 2011, 29, 8982–8987. [PMID: 21945263]
  • 32. Christian, LM; Porter, K .; Karlsson, E .; Schultz-Cherry, S .; Jay, D .; Iams, JD Serum proinflammatory cytokine responses to influenza virus vaccine among women during pregnancy versus non-pregnancy. Am J Reprod Immunol 2013, 70, 45–53. [PMID: 23551710]
  • 33. Christian, LM; Porter, K .; Karlsson, E .; Schultz-Cherry, S. Proinflammatory cytokine responses correspond with subjective side effects after influenza virus vaccination. Vaccines 2015, 33, 3360–3366. [PMID: 26027906]
  • 34. al-Haddad, BJS; Jacobsson, B .; Chabra, S .; Modzelewska, D .; Olson, EM; Bernier, R .; Enquobahrie, DA; Hagberg, H .; Östling, S .; Rajagopal, L .; et al. Long-term Risk of Neuropsychiatric Disease After exposure to Infection in Utero. JAMA Psychiatry 2019, 76, 594–602. [PMID: 30840048]
  • 35. Donzelli, A .; Schivalocchi, A .; Giudicatti, G. Non-specific effects of vaccinations in high-income settings: how to address the issue? Hum Vaccin Immunother 2018, 14, 2904–2910. [PMID: 30019990]
  • 36. Kuhn, TS The Structure of Scientific Revolutions. Chicago University Press, Chicago, USA, 1962.
  • 37. Popper, KR The search has no end. An intellectual autobiography (1976). Armando Editore, Rome, Italy, 1997.
  • 38. US Preventive Services Task Force. Hormone therapy for the primary prevention of chronic conditions in postmenopausal women. Recommendation Statement. JAMA 2017, 318, 2224–2233. [PMID: 29234814]
  • 39. Gartlehner, G .; Patel, SV; Feltner, C .; Weber, RP; Long, R .; Mullican, K .; Boland, E .; Lux, L .; Viswanathan, M. Hormone therapy for the primary prevention of chronic conditions in postmenopausal women. Evidence report and systematic review for the US Preventive Services Task Force. JAMA 2017, 318, 2234–2249. [PMID: 29234813]
  • 40. Autier, P .; Boniol, M .; Pizot, C; Mullie, P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol 2014, 2, 76–89. [PMID: 24622671]
  • 41. Aung, T .; Halsey, J .; Kromhout, D .; Gerstein, HC; Marchioli, R .; Tavazzi, L .; Geleijnse, JM; Rauch, B; Ness, A; Galan, P; et al. Omega-3 Treatment Trialists' Collaboration. Associations of omega-3 fatty acid supplement use with cardiovascular disease risks: meta-analysis of 10 trials involving 77 917 individuals. JAMA Cardiol 2018, 3, 225–234. [PMID: 29387889]