Safety of Influenza Vaccines

Currently available IIVs are generally well-tolerated by children. A large post-licensure population-based study assessed IIV3 safety in 251,600 children aged <18 years (including 8,476 vaccinations in children aged 6–23 months) enrolled in one of five health care organizations within the Vaccine Safety Datalink (VSD) during 1993–1999 (390). This study noted no increase in clinically important medically attended events during the 2 weeks after IIV administration compared with control periods 2–4 weeks before and after vaccination. In a retrospective cohort study of VSD data from 45,356 children aged 6–23 months during 1991–2003, IIV3 was not associated with statistically significant increases in any clinically important medically attended events other than gastritis/duodenitis during the 2 weeks after vaccination compared with control time periods before and after vaccination (391). Most vaccinated children with a diagnosis of gastritis/duodenitis had self-limited vomiting or diarrhea. Several diagnoses, including acute upper respiratory illness, otitis media and asthma, were significantly less common during the 2 weeks after influenza vaccination. Although there was a temporal relationship with vaccination, the vaccine did not necessarily cause or prevent these conditions. A subsequent VSD study of 66,283 children aged 24–59 months noted diagnoses of fever, gastrointestinal tract symptoms, and gastrointestinal disorders to be significantly associated with IIV3 (392). Upon medical record review, none of the events were determined to be serious, and none was associated with complications.

Fever, malaise, myalgia, and other systemic symptoms that can occur after vaccination with IIV3 most often affect persons who have had no previous exposure to the influenza virus antigens in the vaccine (e.g., young children) (393). These reactions are generally self-limited and subside after 1–2 days. Studies suggest the frequency of fever after IIV in children may vary in different influenza seasons and settings. In a study of 791 healthy children aged 1 through 15 years, post-vaccination fever was noted among 12% of those aged 1 through 5 years, 5% among those aged 6 through 10 years, and 5% among those aged 11 through 15 years (112). An observational study assessed post-vaccination fever frequency in 314 children aged 24-59 months receiving IIV during the 2013-14 influenza season. On the vaccination day to 2 days after vaccination (risk window 0-2 days), 7.1% and 6.0% of children had fever after IIV4 and IIV3, respectively (394). A clinical trial assessed fever in 142 children aged 6-47 months randomized to receive acetaminophen, oral placebo, or ibuprofen immediately following IIV and for 24 hours after vaccination. In this study, post-vaccination fever was observed in only two children (one in the acetaminophen group and one in the ibuprofen group) (395).

Since the 2013–14 season, several IIV4 formulations have been licensed. Over the subsequent seasons, fewer IIV3s have been marketed, while available IIV4s have increased. For the 2019-20 season, it is anticipated that all IIVs licensed for children will be IIV4s. IIV4s include products licensed for children as young as age 6 months. In pre-licensure studies of IIV4s, overall frequencies of most solicited adverse events (AEs) were similar to the corresponding comparator IIV3s. Most injection site and systemic AEs are temporary and mild to moderate in severity. Among children, the most common safety complaint was a modest increase in injection site pain (127, 128, 130, 396). The first post-licensure review of VAERS reports covering the 2013–14 and 2014–15 seasons noted that the most common AEs reported following receipt of IIV4 among children aged 6 months through 17 years were injection site reactions and fever. No specific safety concerns were identified; the safety profile was similar to that of IIV3 (397).

Febrile Seizures: Febrile seizures are not uncommon in young children. At least one febrile seizure is experienced by 2%–5% of children aged 6–60 months. Nearly all children who have a febrile seizure recover quickly and are healthy afterward (398). Febrile seizures may occur in the context of febrile illnesses, including influenza. In an observational study of 143 children aged 6 months through 5 years who presented with febrile seizures to an an emergency department in Australia between March 2012 and October 2013, influenza was isolated from 19 (13%) (399). Seizures occurred within 14 days of administration of a vaccine in 16 (11%) children, none of whom had received an influenza vaccine within this period.

Prior to the 2010–11 influenza season, an increased risk for febrile seizures following receipt of IIV3 had not been observed in the United States (391, 400). During the 2010–11 influenza season, CDC and FDA conducted enhanced monitoring for febrile seizures (primarily among children under 5 years of age) and febrile reactions following receipt of influenza vaccines. This heightened surveillance followed reports of an increased risk for fever and febrile seizures (up to 9 febrile seizures per 1,000 vaccine doses) in young children in Australia associated with a 2010 Southern Hemisphere IIV3 produced by CSL Biotherapies (now Seqirus) (401).

Following these events in Australia, from July 2010 through the 2016-17 season, the ACIP did not recommend use of the U.S.-licensed CSL IIV3, Afluria, for children aged <9 years (344, 402). Subsequent laboratory investigation by CSL into the potential etiology of these reactions concluded that the 2010 Southern Hemisphere formulation induced a stronger inflammatory cytokine response than that associated with previous formulations of the vaccine, or with other IIVs. This was hypothesized to be related to the introduction of the viruses B/Brisbane/60/2006 and A/California/7/2009 to the vaccine, and believed to be mediated by higher concentrations of residual lipid and RNA remaining in the vaccine following splitting of the B, and to a lesser extent, the H1N1 components (403). At the time, lower concentrations of the detergent splitting agent taurodeoxycholate (TDOC) were used for the influenza A(H1N1) and influenza B viruses (0.9% and 0.5%, respectively) than for the influenza A(H3N2) component (1.5%). Increasing the concentration of TDOC to 1.5% for all three viruses resulted in attenuation of the cytokine response in an in vitro model (404, 405). In a study comparing fever rates among 402 children aged 5 through 9 years, 302 of whom received a trivalent Afluria produced using 1.5% TDOC for the B viruses and 100 of whom who received a licensed comparator (non-CSL) IIV4, prevalence of fever was similar in both groups (8.2% for Afluria IIV3 versus 9.2% for the comparator IIV4) (406). In a randomized trial of 5- through 17-year-olds comparing Afluria IIV4 (manufactured using 1.5% TDOC for all four viruses) with a licensed comparator IIV4, higher prevalence of fever was observed with Afluria IIV4 (4.5% versus 3.6% for 5- through 8-year-olds and 2.1% versus 0.8% for 9- through 17-year-olds); these differences were not statistically significant (407). In a more recent clinical trial comparing Afluria IIV4 with a licensed comparator IIV4 among children aged 6 through 59 months, no febrile seizures occurred in either group within 7 days post vaccination; two occurred in the Afluria IIV4 arm, but were late after vaccination (at 43 and 104 days following vaccination) and were judged to be unrelated to receipt of vaccine. The proportion of children who experienced fever was similar between the two groups (408, 409).

Subsequent to the events in Australia during 2010, surveillance among children receiving U.S.-licensed influenza vaccines in two different surveillance systems (VAERS and VSD) during the 2010–11 influenza season detected safety concerns for febrile seizures in young children following receipt of IIV3 (410, 411). Further assessment of this signal through the VSD determined that risk for febrile seizures was increased in children aged 6 months–4 years from the day of vaccination until the day after (risk window: day 0–1). The risk was higher when children received concomitant PCV13 (i.e., when the two vaccines were administered at the same health care visit) and peaked at approximately age 16 months (411), but the effect of other concomitant vaccines was not evaluated. The magnitude of the increased risk for febrile seizures in children aged 6–23 months in the United States observed in this study (<1 per 1,000 children vaccinated) was substantially lower than the risk observed in Australia in 2010 (401). After evaluating the data on febrile seizures from the 2010–11 season and taking into consideration benefits and risks of vaccination, ACIP recommended no policy change for use of IIV (412, 413).

A follow-up VSD study assessed the risk for febrile seizure on days 0-1 with the concomitant administration of IIV3 and all other routine childhood vaccines in children aged 6-23 months during 5 influenza seasons (2006-07 through 2010-11) (414). This study found that there was an increased risk for febrile seizure when IIV3 was administered simultaneously with either PCV or DTaP-containing vaccines, but no increased risk when IIV3 was administered alone. The increased risk with these vaccine combinations was observed to have been present in seasons prior to 2010-11. Another U.S. study performed in follow-up the 2010–11 season findings analyzed data from the separate FDA-sponsored PRISM (Post-licensure Rapid Immunization Safety Monitoring) system population. This analysis found no association between receipt of IIV3 (adjusted for concomitant 13-valent pneumococcal conjugate vaccine [PCV13] or diphtheria, tetanus, and acellular pertussis vaccine [DTaP]) and febrile seizures among children 6-59 months of age during 2010-11 (IRR adjusted for age and seasonality=1.36; 95%CI 0.78, 2.39) (415). Same-day IIV3 and PCV13 vaccination was not associated with more febrile seizures compared with separate-day vaccination (1.08 fewer febrile seizures per 100,000 with same day administration; 95%CI -5.68, 6.09).

However, surveillance findings in some subsequent seasons for febrile seizures in young children following influenza vaccine have been consistent with the original findings of an increased risk in 2010-11. During the 2011–12 season (for which the influenza vaccine composition was the same as that of the 2010–11 season), an observational clinical study showed that risk for fever in the 0–1 days after vaccination was higher when children 6 to 23 months old received IIV3 and PCV13 concomitantly versus receipt of IIV3 or PCV13 without the other product (416). The viral composition of U.S. influenza vaccines was changed for the 2013-14 season, and this same composition was used for the 2014-15 season. VSD surveillance for the 2013–14 and 2014–15 seasons found an elevated risk for febrile seizures among 6- through 23 month-olds 0–1 days after concomitant receipt of IIV3 and PCV13 (RR=5.30; 95%CI 1.87, 14.75). There was no significant increased risk following administration of IIV3 without PCV13 (417). Similarly, analysis of 2013-14 data from the PRISM system revealed no increased risk for seizure following either IIV3 or LAIV when an individual-level, self-controlled risk interval comparison method was used, but did reveal increased risk for IIV3 and PCV13 administered concomitantly (but not alone) when using a method comparing current and historical rates (418). Surveillance for febrile seizures following receipt of IIVs is ongoing through the Vaccine Adverse Event Reporting System (VAERS; https://vaers.hhs.gov/index), and VSD conducts near real-time sequential monitoring for seizures following receipt of IIV during the influenza season. Inactivated influenza vaccines and other childhood vaccines (including PCV13) may be given concomitantly.

Safety of Full-Dose IIV4 for children aged 6 through 35 months: The dose of IIV given to persons aged ≥3 years is 0.5 mL. During several seasons prior to November 2016, the only influenza vaccines licensed for children 6 through 35 months of age were Fluzone (IIV3) and Fluzone Quadrivalent (IIV4, Sanofi Pasteur, Swiftwater, Pennsylvania), given in a 0.25mL dose (half the dose given to persons aged ≥3 years). The rationale for this reduced dose was greater frequency of fever and other reactogenicity events noted in studies conducted during the 1970s among children in this age group, primarily with older, whole-virus vaccines (419-423). Whole-virus IIVs are no longer available in the United States, having been replaced with split-virus and subunit IIVs.

As a group, the currently available IIVs are generally less reactogenic than the previous whole-virus products (424). Evaluations of several currently available IIVs have been conducted and have reported favorable safety profiles when administered at a 0.5mL dose for children in this age group (144, 425-427). Recent comparative studies of 0.5mL doses of IIV4s among children aged 6 through 35 months have included FluLaval Quadrivalent (426), and Fluzone Quadrivalent (427), each compared with 0.25mL Fluzone Quadrivalent. In each instance, safety and reactogenicity were comparable between the two groups. In a randomized trial of Fluarix Quadrivalent vs. non-influenza control vaccines that was conducted among 11,795 children aged 6 through 35 months, frequencies of injection site and systemic reactions were similar between Fluarix Quadrivalent and the control vaccines (144).

In placebo-controlled studies of IIV3 among older adults, the most frequent adverse reaction of vaccination was soreness at the vaccination site (affecting 10%–64% of recipients) that lasted <2 days (428, 429). These injection site reactions typically were mild and not commonly associated with interference with recipients’ ability to conduct usual daily activities. Placebo-controlled trials demonstrate that among older persons and healthy younger adults, administration of IIV3 is not associated with higher proportions of systemic symptoms (e.g., fever, malaise, myalgia, and headache) when compared with placebo injections (428-430). In a VAERS analysis of 18,245 reports from 1990 through 2005, in the most common AEs among adults aged ≥18 years included injection site reactions, pain, fever, myalgia, and headache (431). This VAERS review identified no new safety concerns. Fourteen percent of the IIV3 VAERS reports in adults were classified as SAEs; (defined as those involving death, life-threatening illness, hospitalization or prolongation of hospitalization, or permanent disability (432)), similar to proportions seen in VAERS for other adult vaccines. The most common SAE reported after IIV3 in VAERS in adults was Guillain-Barré syndrome (GBS) (see Guillain-Barré Syndrome and IIV). However, VAERS cannot assess whether a vaccine caused an event to occur. During the 2013-14 and 2014-2015 influenza seasons, VSD found no increased risk after IIVs for 6 outcomes in populations that included adults (acute disseminated encephalomyelitis [ADEM], anaphylaxis, Bell’s palsy, GBS, encephalitis, and transverse myelitis) (417). Another VSD study found that overall there was no increased risk for venous thromboembolism (VTE) after IIV in adults in adults aged ≥50 years (433).

Injection site and systemic AEs were more frequent after vaccination with high-dose IIV3 (HD-IIV3; Fluzone High-Dose; Sanofi Pasteur, Swiftwater, Pennsylvania), which contains 180 µg of HA antigen (60 per vaccine virus) than following standard dose IIV3 (15 µg per virus; Fluzone; Sanofi Pasteur, Swiftwater, Pennsylvania), but were typically mild and transient. In one study, 915 (36%) of 2,572 persons who received HD-IIV3 reported injection site pain, compared with 306 (24%) of 1,262 who received SD-IIV3 (288). Among Fluzone High-Dose recipients, 1.1% reported moderate to severe fever; this was significantly higher than the 0.3% of Fluzone recipients who reported this systemic AE (RR=3.6; 95%CI 1.3, 10.1). A randomized study of HD-IIV3 versus SD-IIV3 including 9,172 participants found no difference in occurrence of SAEs or several specific AEs of interest (including GBS, Bell’s Palsy, encephalitis/myelitis, optic neuritis, Stevens-Johnson syndrome, and toxic epidermal necrolysis) (434). Safety monitoring of HD-IIV3 in VAERS during the first year after licensure indicated a higher-than-expected number of gastrointestinal events compared with standard-dose vaccine, but otherwise no new safety concerns were identified. Most of the reported gastrointestinal events were nonserious (435). A survey of adults aged ≥65 years in the Minneapolis Veterans Affairs Health Care System who received influenza vaccines (547 high-dose and 541 standard dose) during October 2015 found that injection site and systemic adverse reactions were more common after HD-IIV3 than after SD-IIV3 during the week after vaccination (with 37% of HD-IIV3 recipients and 22% of SD-IIV3 recipients reporting at least one such symptom). There was no significant difference in prevalence of severe adverse reactions or healthcare visits between groups (436).

A trivalent MF59-adjuvanted IIV3 (aIIV3), Fluad (Seqirus, Summit, New Jersey) was approved in November 2015 for use in persons aged ≥65 years. In clinical trials among persons in this age group, some injection site and systemic AEs were observed to occur more frequently following aIIV3 compared with unadjuvanted SD-IIV3; most were mild in severity. Proportions of persons experiencing SAEs was similar between the two groups (299, 300). In addition, rates of immune-mediated diseases after aIIV3 and SD-IIV3 were similar. In a post-marketing review of 630 VAERS reports of adverse events among aIIV3 recipients submitted between July 2016 and June 2018, no new patterns of events were noted, though a relatively high proportion of reports involved administration of the vaccine to persons under 65 years of age (a population for which the vaccine is not currently licensed). Proportions of the most commonly reported adverse events, injection site pain and erythema, was similar to that observed with HD-IIV3 and SD-IIVs (437).

Fewer post-marketing safety data have thus far accumulated for IIV4s, which first became available during the 2013–14 season, compared with IIV3. Among adults the most common safety complaints were injection site pain and systemic reactions, such as myalgia, headaches, and fatigue (126, 129, 131, 132, 438). The first post-licensure safety assessment of VAERS reports covering the 2013–14 and 2014–15 seasons noted a safety profile similar to that of IIV3. The most common AE reported following receipt of IIV4 among adults aged 18 through 64 years was injection-site pain. No specific safety concerns were identified (397).

Cell culture-based IIV3 (ccIIV3), was licensed by FDA in 2013; a quadrivalent formulation was licensed in 2016. ccIIV3 appeared to have a similar safety profile to other, previously licensed IIVs. A review of 629 VAERS reports related to ccIIV3 during the 2013–14 and 2014–15 seasons noted that injection site and systemic symptoms were the most commonly reported AEs; no concerning pattern of AEs was identified (439). In pre-licensure studies, the safety profile of ccIIV4 was similar to that of ccIIV3 (440).

Overall, fewer safety data pertaining to persons with specific underlying medical conditions are available relative to data from healthy populations. Most studies in these populations are small, limiting the extent to which uncommon or rare AEs may be captured. Few studies directly compare outcomes among persons with high risk conditions with those observed in healthier populations.

A study of 52 children aged 6 months through 4 years with chronic lung disease or congenital heart disease reported fever among 27% and irritability and insomnia among 25% (441). Another of 33 children aged 6–18 months with bronchopulmonary dysplasia or congenital heart disease reported that one child had irritability and one had a fever and seizure after vaccination (424). No placebo comparison group was used in these studies. One prospective cohort study found that the rate of AEs was similar among hospitalized persons who were either ≥65 years or 18–64 years of age and who had one or more chronic medical conditions compared with outpatients; injection site soreness was the most common complaint (442).

Several randomized clinical trials comparing IIV to placebo among persons with chronic obstructive pulmonary disease (COPD) and asthma have reported safety outcomes. A study of 125 COPD patients at a Thai hospital clinic reported that significantly more patients in the vaccine group had injection site reactions (27% versus 6% placebo; p = 0.002) (443). The most common injection site reactions among vaccinated patients were swelling, itching and pain when touched. However, these symptoms were generally rated as mild and lasted <48 hours. There were no significant differences between the two groups in systemic reactions, such as headache, myalgia, fever, skin rash, nor in lung function, dyspneic symptoms, and exercise capacity at 1 and 4 weeks.

Evidence indicates that IIVs are well tolerated in children (444) and adults (200) with asthma. A multicenter, randomized, double-blind, placebo-controlled crossover trial involving 2,032 asthmatic subjects aged 3–64 years found a similar frequency of asthma exacerbations during the 2 weeks following either vaccination or placebo injection (28.8% versus 27.7%). Only myalgia was reported more frequently following IIV3 (25% versus 21% placebo; p<0.001) (199). A randomized study of IIV3 versus placebo among 262 adults with asthma noted that vaccination was associated with a decline in peak expiratory flow; however, this effect was no longer significant when adjusted for the presence of concomitant symptomatic colds (445). A randomized crossover design study of IIV3 versus saline placebo showed no significant difference in the occurrence of asthma exacerbations during the 14 days post-vaccination (446).

A non-randomized study compared AEs following receipt of IIV among 105 adults with type 2 diabetes with those occurring among 108 in nondiabetics. Local reactions such as tenderness, pain, redness, and swelling occurred less frequently in the diabetic group. Differences in systemic reactions such as myalgia, tiredness, headache, malaise, chills, and arthralgia were not statistically significant (447).

Transient increases in replication of HIV-1 in the plasma or peripheral blood mononuclear cells of HIV-infected persons after vaccine administration have been observed in some, but not all, studies (448-453). However, IIV does not appear to have a clinically important impact on HIV infection or immunocompetence in HIV-infected persons. CD4+ T lymphocyte cell counts or progression of HIV disease have not been demonstrated to change substantially after influenza vaccination among HIV-infected persons compared with unvaccinated HIV-infected persons (452). Limited information is available about the effect of antiretroviral therapy on increases in HIV RNA levels after either influenza virus infection or influenza vaccination (454, 455).

IIV generally has been shown to be well-tolerated in both adult and pediatric solid organ transplant recipients (265). In small studies, IIV vaccination did not affect allograft function or cause acute rejection episodes in recipients of kidney (267, 270, 456, 457), heart (458), lung (456) or liver transplants (274, 459, 460). A literature review concluded that there is no convincing epidemiologic link between vaccination and allograft dysfunction (265). Guillain-Barré syndrome in a liver transplant recipient (461) and rhabdomyolysis leading to acute renal allograft dysfunction (462) after IIV vaccination have been reported. Several case reports of corneal graft rejection have been reported following receipt of IIV (463-466). However, no studies specifically designed to evaluate whether IIV is associated with increased risk for corneal graft rejection have been conducted.

Some studies have evaluated safety of high-dose or adjuvanted IIVs in immunocompromised populations. A randomized trial compared HD-IIV3 and SD-IIV3 in 190 persons with HIV infection; injection site pain was reported more frequently among the high-dose recipients, but overall the authors reported that no significant differences were observed in injection site or systemic reactions between the two groups (260). In a randomized trial of HD- versus SD-IIV3 among 161 adult solid organ transplant recipients, frequencies of injection site reactions and fever were similar between the two groups. Participants who received HD-IIV3 had a higher frequency of systemic symptoms, particularly gastrointestinal symptoms and arthralgia, but this difference was not statistically significant (277). In a randomized trial of adjuvanted versus unadjuvanted IIV3 which enrolled 73 adult allogeneic hematopoietic stem cell transplant recipients (≥12 weeks post-transplant), no significant differences were reported in the prevalence of injection site or systemic reactions between the two groups. Fever occurred more frequently among those who received the adjuvanted vaccine (254).

Immune checkpoint inhibitors (including drugs such as nivolumab, pembrolizumab, ipilimumab, and atezolizumab) are medications used to treat cancer which block pathways that inhibit activity of T-cells, enhancing anti-tumor activity of these cells (467). Because they inhibit processes which down-regulate immune response, there is concern for immune-related adverse events. In a 2018 report which described 23 lung cancer patients receiving checkpoint inhibitors who received IIV3, 52% experienced an adverse event which was considered immune-related. Six of these events were considered severe or life threatening (including 2 cases of colitis, 2 of encephalitis, and one each of pneumonitis and neuropathy) (468). Subsequent reports have noted fewer safety concerns for influenza vaccination of persons receiving these agents (469-471). In a cohort study of 127 persons with lung cancer receiving nivolumab, 42 of whom received IIV3 and 85 of whom were unvaccinated, risk of immune-related adverse events did not differ significantly between the two groups (rate ratio=1.20; 95%CI 0.51, 2.65 for all events; rate ratio=2.07; 95%CI 0.28, 15.43 for serious events) (469). In a case-control study comparing 101 persons receiving checkpoint inhibitors who had been diagnosed with myocarditis with 201 patients also receiving these agents who had not been diagnosed with myocarditis, receipt of influenza vaccine was more common among those persons who had not been diagnosed with myocarditis (40%) than those who had received this diagnosis (25%) (470). In a retrospective review including 370 persons who received IIV within 65 days of checkpoint inhibitor therapy, overall 75 (20%) experienced an immune-related adverse event; the majority of these were judged mild to moderate in severity (471).

Guillain-Barré Syndrome (GBS) is an autoimmune demyelinating disease of the peripheral nervous system which most commonly presents with rapid-onset muscle weakness (472). The annual incidence of GBS is 10–20 cases per 1 million adults (472). Multiple bacterial and viral infectious illnesses, most notably Campylobacter jejuni gastrointestinal infections and upper respiratory tract infections, have been implicated as triggers of the autoimmune processes associated with GBS (473-476). Influenza infection has also been associated with the onset of GBS. An analysis of 405 patients admitted to a single facility identified an association between serologically confirmed influenza virus infection and GBS, with time from onset of influenza illness to GBS of 3–30 days (477).

An association between GBS and receipt of IIVs has been noted during some influenza seasons. In particular, the 1976 swine influenza vaccine was associated with an increased frequency of GBS, estimated at one additional case of GBS per 100,000 vaccinated persons (478). Since that time, evidence for an association between IIVs and GBS has been variable and inconsistent across influenza seasons, but in general an association of similar magnitude to that noted during the 1976-77 season has not been demonstrated (479, 480). During three of four influenza seasons studied during 1977–1991, the overall relative risk estimates for GBS after influenza vaccination were not statistically significant (481-483). However, in a study of the 1992–93 and 1993–94 seasons, the overall relative risk for GBS was 1.7 (95%CI 1.0, 2.8; p = 0.04) during the 6 weeks after vaccination, representing approximately one additional case of GBS per 1 million persons vaccinated. Cases occurred most frequently in the second week after vaccination (480). Similarly, data from the systems monitoring influenza A(H1N1) 2009 monovalent vaccines suggest that the increased risk for GBS is approximately one or two additional cases per 1 million persons vaccinated, which is similar to that observed in some years for seasonal IIV (484-490). An analysis of chart-confirmed GBS cases from the Medicare population during the 2009-10 season estimated that vaccination was associated with an attributable risk of 2.84 per 1 million doses (491). A subsequent four-season study in this population (2010-11 through 2013-14) obtained a similar estimate of excess risk for the 2010-11 season, but not for the three subsequent seasons (during which time the vaccine continued to contain an H1N1pdm09-like virus) (492). Of note, some studies have observed a higher risk for GBS following influenza infection than that following influenza vaccination (477, 493).

Because GBS is more likely to occur in persons who have experienced it previously, (472), the likelihood of coincidentally experiencing GBS after influenza vaccination is expected to be greater among persons who have had a prior episode than those who have not. Whether influenza vaccination specifically might increase the risk for recurrence of GBS is unknown. In a Kaiser Permanente Northern California database study among >3 million members conducted over an 11-year period, no cases of recurrent GBS were identified after influenza vaccination in 107 persons with a documented prior diagnosis of GBS, two of whom had previously developed GBS within 6 weeks of influenza vaccination (494).

Thimerosal is an ethyl mercury-containing antimicrobial compound. It is primarily used in multi-dose vial preparations of IIVs as a preservative to inhibit microbial growth. For these preparations, the mercury content from thimerosal (as reported in package inserts) is ≤25 µg of mercury per 0.5 mL dose.

Although the evidence is reassuring regarding health risks associated with exposure to vaccines containing thimerosal (495-507), the U.S. Public Health Service and other organizations have recommended that efforts be made to eliminate or reduce the thimerosal content in vaccines as part of a strategy to reduce mercury exposures from all sources (495, 499). Single-dose vial and syringe preparations of IIVs, as well as LAIV4 and RIV4, which are expected to be available during the 2019-20 season do not contain thimerosal.

Adjuvants are compounds added to vaccines to improve immune response. Only one adjuvanted influenza vaccine, Fluad (adjuvanted inactivated influenza vaccine, trivalent [aIIV3], Seqirus) is currently licensed in the United States (299). Fluad contains the squalene-based oil-in-water adjuvant, MF59. Currently, it is indicated for persons aged 65 years and older. In a randomized controlled study comparing aIIV3 with IIV3 among 7,000 persons aged ≥65 years, prevalence of some injection site and systemic reactions within the first 7 days after vaccination were higher in the aIIV3 group (300). Those who received aIIV3 were more likely to experience pain or tenderness, or myalgia; however, most of these reactions were mild in severity. Prevalence of SAEs was similar between the two groups. Though not licensed for children in the U.S., aIIV3 has been used in the pediatric population in Europe and Canada. In studies conducted among children, MF59-adjuvanted vaccines were associated with greater likelihood of injection site redness and pain, fever, irritability, and loss of appetite than non-adjuvated control vaccines (508).

During the 2009 pandemic, three monovalent H1N1pdm09 vaccines containing squalene-based oil in water adjuvants were used globally (in addition to unadjuvanted inactivated and live attenuated H1N1pdm09 vaccines). The AS03-adjuvanted vaccine, Pandemrix (GSK), was used widely in some European nations, particularly in Scandinavian countries; the AS03-adjuvanted vaccine, Arepanrix (GSK), was used in Canada; and the MF59-adjuvanted vaccine, Focetria (Novartis), was widely used globally.   Epidemiological studies conducted in European countries after the pandemic have consistently found an increased risk of narcolepsy associated with Pandemrix, especially in children (509); although like all retrospective observational studies, they are subject to limitations, including possible awareness and detection bias (510-513). No similar risk has been detected with Arepanrix or Focetria (514, 515). The reasons for the Pandemrix finding, as well as the lack of association with Arepanrix and Focetria, are still the subject of scientific investigation. Adjuvanted monovalent H1N1pdm09 were not licensed or used in the United States during the 2009 pandemic and no AS03-adjuvanted seasonal influenza vaccines are currently licensed in the United States. An MF59-adjuvanted trivalent seasonal influenza vaccine is approved for individuals aged 65 years and older.

Oculorespiratory syndrome (ORS), an acute, self-limited reaction to IIV, was first described during the 2000–01 influenza season in Canada (516, 517). ORS was initially noted to be associated with one vaccine preparation (Fluviral S/F, Shire Biologics, Quebec, Canada; not available in the United States) during the 2000–01 influenza season (517). After changes in the manufacturing process of the vaccine preparation associated with ORS during the 2000–01 season, the incidence of ORS in Canada diminished greatly (518). The cause of ORS has not been established; however, studies suggest that the reaction is not IgE-mediated (519). When assessing whether a patient who experienced ocular and respiratory symptoms should be revaccinated, providers should determine if concerning signs and symptoms of IgE-mediated immediate hypersensitivity are present (see Immediate Hypersensitivity Reactions After Receipt of Influenza Vaccines). Health care providers who are unsure whether symptoms reported or observed after receipt of IIV represent an IgE-mediated hypersensitivity immune response should seek advice from an allergist/immunologist.

Shedding of the live attenuated vaccine virus is common after receipt of LAIV. In general, shedding is more common among younger recipients, among whom it may also be of longer duration. Among 345 LAIV3 recipients aged 5–49 years for whom shedding was assessed by viral culture of nasal swabs, 29% had detectable virus in nasal secretions. Prevalence of shedding was inversely related to age, and maximal shedding occurred within 2 days of vaccination. The symptoms most frequently reported after vaccination (runny nose, headache, and sore throat) did not correlate with the presence of shedding (522). In a study of 200 children aged 6 through 59 months, shedding of at least one vaccine virus was detected in 79% of children overall, and was more common among younger children (89% of 6- through 23-month-olds as compared with 69% of 24- through 59-month-olds) (523). Shedding had stopped in most cases by 11 days post vaccination. Vaccine virus was detected from nasal secretions in one (2%) of 57 HIV-infected adults who received LAIV3 compared with none of 54 HIV-negative participants (524), and in three (13%) of 24 HIV-infected children compared with seven (28%) of 25 children who were not HIV-infected (525).

Transmission of shed LAIV vaccine viruses from vaccine recipients to unvaccinated persons has been documented, but has not been reported to be associated with serious illness. One study of 197 children aged 9–36 months (98 of whom received LAIV3 and 99 of whom received placebo) in a child care center assessed the potential for transmission of LAIV3 vaccine viruses. Among vaccine recipients, 80% shed one or more vaccine virus (mean duration: 7.6 days). One influenza B vaccine virus strain isolate was recovered from a placebo recipient, and was confirmed to be vaccine-type virus. This transmitted virus isolate retained the cold-adapted, temperature-sensitive, attenuated characteristics. The placebo recipient from whom the influenza B vaccine virus strain was isolated had symptoms of a mild upper respiratory illness. The estimated probability of transmission of vaccine virus within a contact group with a single LAIV recipient in this population was 0.58% (95%CI = 0, 1.7) (526).

In a study of genotypic and phenotypic stability of LAIV vaccine viruses, nasal and throat swab specimens were collected from 17 study participants for 2 weeks after vaccine receipt. Virus isolates were analyzed by multiple genetic techniques. All isolates retained the cold-adapted and temperature-sensitive phenotypes (527). In a separate experimental study, serial passage of the LAIV H1N1pdm09 monovalent vaccine virus in Madin-Darby canine kidney (MDCK) cells at increasing temperatures resulted in a variant that reproduced at higher temperatures and produced severe disease in mice (528).

Among healthy children aged 60–71 months enrolled in one clinical trial, some signs and symptoms were reported more often after the first dose among LAIV3 recipients (n = 214) than among placebo recipients (n = 95), including runny nose (48% vs. 44%), headache (18% vs. 12%), vomiting (5% vs. 3%), and myalgia (6% vs. 4%, respectively). However, these differences were not statistically significant (529). In other trials, signs and symptoms reported after LAIV3 administration have included runny nose or nasal congestion (18%–82%), headache (3%–46%), fever (0–32%), vomiting (3%–17%), abdominal pain (2%), and myalgia (0–21%) (308, 309, 315, 530-534). In general, these symptoms were associated more often with the first dose and were self-limited. In a placebo-controlled trial in 9,689 children aged 1–17 years which assessed pre-specified medically attended outcomes during the 42 days after vaccination, LAIV3 was associated with increased risk for asthma, upper respiratory infection, musculoskeletal pain, otitis media with effusion, and adenitis/adenopathy. In this study, the proportion of SAEs was 0.2% in LAIV3 and placebo recipients; none of the SAEs was judged to be related to the vaccine by the study investigators (530).

LAIV3 has been associated with wheezing among younger children in some studies. In a comparison of LAIV3 and IIV3 children in aged 6– 59 months which excluded children with recent medically diagnosed or treated wheezing or a history of severe asthma (328), the proportion of children who experienced medically significant wheezing following vaccination did not differ between the two vaccines among children 24 through 59 months of age. Wheezing was observed more frequently following the first dose among previously unvaccinated younger LAIV3 recipients, primarily those aged <12 months (an age group for which LAIV is not licensed). In a randomized placebo-controlled safety trial among children without a history of asthma, an increased risk for asthma events (RR=4.1; 95%CI 1.3, 17.9) was documented among the 728 children aged 18–35 months who received LAIV3. Of the 16 children with asthma-related events in this study, seven had a history of asthma on the basis of subsequent medical record review. None required hospitalization, and increased risk for asthma events was not observed in other age groups (530). A 14-year follow-up study conducted among children who had enrolled in this trial at <3 years of age reported no findings indicating that the children who received LAIV had an increased risk of subsequent asthma diagnosis (535).

In an open-label trial conducted between 1990 and 2002, 18,780 doses of LAIV3 were administered to 11,096 children aged 18 months through 18 years. Among those aged 18 months through 4 years, no increase was reported in asthma visits 0–14 days after vaccination compared with the pre-vaccination period. A significant increase in asthma events was noted 15–42 days after vaccination, but only in the first year of the study (536). Among the 2,196 children in this study who had a history of intermittent wheezing but were otherwise healthy, no increased risk was observed for MAARI, including acute asthma exacerbation either 0–14 or 0–42 days after receipt of LAIV3 compared with the pre- and post-vaccination reference periods (537).

A review of 460 VAERS reports (including persons aged 2 through 70 years) following distribution of approximately 2.5 million doses of LAIV3 during the 2003–04 and 2004–05 influenza seasons did not indicate any new safety concerns (538) (note, however, that LAIV4 is licensed for persons aged 2 through 49 years). Respiratory events (such as influenza-like illness, rhinitis, pharyngitis, sinusitis, and asthma) were the most common conditions reported. Few (9%) of the reports described events considered serious. During 2005–2012, VAERS received 2,619 reports in children aged 2 through 18 years after receipt of LAIV3 (539). Consistent with the earlier VAERS study, few (7.5%) of these reports were serious and no new AE patterns were identified. A VSD self-controlled case series analysis of 396,173 children who received LAIV3 from September 2003 through March 2013 revealed a significant association of anaphylaxis and syncope with receipt of LAIV (540). Among the 5 syncope cases, 4 had also received injectable vaccines concurrently. Both of these AEs were rare, with an estimated rate of 1.7 events per 1 million doses for anaphylaxis and 8.5 events per 1 million doses for syncope.

LAIV4 has been available in the U.S. since the 2013-14 season, and was not recommended for use in the U.S. during the 2016-17 and 2017-18 seasons. Fewer post-marketing surveillance data specific to this formulation have accumulated than are available for LAIV3. For the 2013–2014 influenza season, in which approximately 12.7 million doses of LAIV4 were distributed, VAERS received 779 reports of events occurring following LAIV4 (599 of which occurred in children aged 2 through 17 years). The safety profile of LAIV4 was consistent with pre-licensure clinical trials and data from post-licensure assessment of LAIV3 (541). In an analysis of health maintenance organization data for the 2013-14 season including persons aged 2 through 49 years, risk of wheezing after LAIV4 was not higher than after IIV or unvaccinated children when the analysis was conducted in the total population. There was a slightly higher risk of wheezing events among children aged 2 through 4 years who received LAIV4 relative to unvaccinated controls (hazard ratio=1.50; 95%CI 1.03, 2.20). Of the 66 LAIV4 recipients who experienced these events, 4 were assessed in an emergency department, but not hospitalized (542). An open label safety study of LAIV4 among 100 children aged 2 through 6 years conducted in Japan during the 2014-15 season reported a safety profile consistent with that of previous studies. The most frequently reported solicited symptoms were runny/stuffy nose (51%), cough (34%), fever (10%), and sore throat (7%). The only AE that occurred among greater than 5% of children was nasopharyngitis, in 13% (543). All AEs were characterized as mild in severity. In a randomized placebo-controlled trial of LAIV conducted among 1,301 children aged 7 through 18 years in Japan, the prevalence of AEs was similar among the LAIV4 and placebo groups (24.3% and 25.9%, respectively). Only 2.9% of these events were characterized as moderate in severity; none were rated as severe. The most commonly reported AE was nasopharyngitis, occurring in 8.1% of LAIV4 recipients and 8.3% of placebo recipients. SAEs were uncommon, occurring in 0.3% of LAIV4 recipients and 0.7% of placebo recipients; those in the LAIV4 recipients occurred > 100 days post-vaccination and were judged unrelated to receipt of study vaccine (543).

Published data on the use of LAIV for children with diagnosed asthma and other medical conditions conferring higher risk for severe influenza illness are relatively limited compared with available data for children without such conditions. A prospective nonrandomized population-based study conducted in the United Kingdom which compared rates of hospitalization and selected adverse events through 6 days and 42 days post-vaccination among 11,463 children and adolescents during the 2013-14 and 2014-15 seasons found no difference in risk of all-cause hospitalization or hospitalizations for lower respiratory events between LAIV recipients and unvaccinated children at either time interval. Risk of hospitalization for any cause and for lower respiratory conditions were significantly lower among LAIV recipients compared to IIV recipients, at both 6 days and 42 days. While vaccine recipients and unvaccinated controls had been matched on various factors in order to control for severity of underlying illness, the investigators noted that the different vaccine groups could have differed in inherent risk for hospitalization, and that residual confounding could have contributed to the differences observed between the two vaccine groups (544).

Among healthy adults aged 18–49 years in one clinical trial, signs and symptoms reported significantly more often (p<0.05; Fisher exact test) among LAIV3 recipients (n = 2,548) than placebo recipients (n = 1,290) within 7 days after each dose included cough (14% versus 10%), runny nose (44% versus 27%), sore throat (27% versus 16%), chills (8% versus 6%), and tiredness/weakness (25% versus 21%) (529). A review of 460 reports (involving persons aged 2 through 70 years) to VAERS after administration of approximately 2.5 million doses of LAIV3 during the 2003–04 and 2004–05 influenza seasons did not indicate any new safety concerns. Respiratory symptoms were the most common types of events reported. Serious adverse events were uncommon (538).

Limited data assessing the safety of LAIV use for certain groups at higher risk for influenza-related complications are available. In a study of 57 HIV-infected persons aged 18–58 years with CD4+ counts >200 cells/µL who received LAIV3, no SAEs judged to be related to vaccine were reported during a 1-month follow-up period (524). No significant difference was noted in the frequency of AEs or viral shedding among 24 HIV-infected children aged 1–8 years on effective antiretroviral therapy who were administered LAIV3 compared with 25 HIV-uninfected children receiving LAIV3 (525). In a study comparing immunogenicity and shedding of LAIV4 among 46 HIV-infected (CD4+ counts >200 cells/µL) and 56 uninfected persons aged 2 through 25 years, prevalence of most AEs were similar between the two groups. Shedding of vaccine virus was somewhat more prevalent among the HIV-infected participants, 67% of whom shed any vaccine virus up to 14–21 days post-vaccination, compared with 50% of uninfected participants (p = 0.14) (545).

Data on the relative safety of LAIV and IIV are limited for children and adults with chronic medical conditions conferring a higher risk for influenza complications. Among 1,940 children aged 2 through 5 years with asthma or prior wheezing enrolled in multinational randomized trials of LAIV3 and IIV3, risk of lower respiratory illness, and hospitalization were not significantly increased among children receiving LAIV3 relative to IIV3; however, increased prevalence of rhinorrhea (8.1% LAIV versus 3.1% IIV3; p = 0.002) and irritability (2.0% versus 0.3%, p = 0.04) were observed among LAIV3 recipients (546). A study of LAIV and IIV3 among children aged 6 through 17 years with asthma noted no significant difference in wheezing events after receipt of LAIV3 (329). In a study conducted between 2007 and 2014 among children aged ≥2 years with a history of asthma who received either LAIV or IIV, LAIV was associated with a lower likelihood of asthma exacerbation than IIV (547). A VSD study of the safety of LAIV in persons with asthma conducted during 3 influenza seasons (2008-2009 through 2010-2011) found no association between receipt of LAIV and increased risk of medically attended respiratory AEs (548). Available data are insufficient to determine the level of severity of asthma for which administration of LAIV would be inadvisable.

A two-season Canadian retrospective cohort study was conducted among 198 persons aged 2 through 19 years with cystic fibrosis who received LAIV3 compared rates of hospitalization, antibiotic prescriptions and diary-recorded symptoms during days 0 through 6 post-vaccination (defined as the at-risk period) with those occurring during days 7 through 55 (the control period). Hospitalization and antibiotic prescription events were not greater during the at-risk period. Self-reported symptoms occurred more commonly during the at-risk period; the pulmonary symptoms with the greatest magnitude of risk were chest congestion/increased sputum, and wheezing (549).

Under the previous FDA labeling regulations, influenza vaccines were classified as either Pregnancy Category B or Category C on the basis of risk of reproductive and developmental adverse effects and on the basis of such risk weighed against potential benefit. In 2014, new regulations updated the format and content requirements of labeling for human prescription drugs and biological products, including vaccines. Under these new regulations, the previous pregnancy risk categories are replaced with a narrative summary of risk based on human and animal data for the specific product. In accordance with a defined implementation plan, most influenza vaccines are now labeled using the new format.

In general, there are no pre-licensure studies of influenza vaccines among pregnant women. The majority of available data come from post-licensure studies. However, influenza vaccines have been administered to pregnant women for more than five decades, and overall have a reassuring safety profile. The vast majority of published data and clinical experience involve use of IIVs, which have been available for the longest period of time, and which have been recommended for use for some populations of pregnant women since the early 1960s. Data are more limited for RIV (which has only been available since 2013), and for LAIV (which has not been recommended for use during pregnancy because it is a live vaccine).

In a retrospective cohort analysis of healthcare organization data covering nine influenza years (2008-2016) and including 247,036 pregnant women, 53% were vaccinated. The vast majority of women received IIV; only 156 received LAIV (550). Vaccination was distributed relatively evenly by trimester (32% during the first, 31% during the second, and 30% during the third). Vaccination was significantly associated with lower risk of several outcomes, including maternal influenza illness (OR=0.49; 95%CI 0.39, 0.62), maternal fever (OR=0.40; 95%CI 0.35, 0.45), preeclampsia (OR=0.93; 95%CI 0.90, 0.96), placental abruption (OR=0.89; 95%CI 0.82, 0.96), stillbirth (OR=0.88; 95%CI 0.78, 0.99), and infant admission to a neonatal intensive care unit (OR=0.89; 95%CI 0.87, 0.92).

IIVs

Overall experience with the use of IIVs during pregnancy has been reassuring. Substantial data have accumulated which do not indicate fetal harm associated with IIVs administered during pregnancy. However, data specifically concerning administration of these vaccines during the first trimester are limited (551). A 2015 review of studies of maternal influenza vaccination and pregnancy outcomes noted that women vaccinated in the first trimester were underrepresented in these studies, contributing to imprecision in estimates for risk of outcomes such as fetal death, spontaneous abortion and congenital malformations (552). Most available data are from observational studies rather than controlled trials. Differences in methodology (for example, clinical definitions for outcomes of interest) complicates pooling of data and comparisons among estimates.

Background rates of spontaneous abortion (miscarriage) vary from 10.4% in women aged <25 years to 22.4% in women aged >34 years (553). Considering the number of pregnant women vaccinated, spontaneous abortion following (but not attributable to) influenza vaccination would be expected to occur due to chance. Most studies (554-562) and systematic reviews (552, 563, 564) that have evaluated risk of spontaneous abortion following receipt of IIVs have not found an association. However, data on the use of influenza vaccines are more limited during the early first trimester, when spontaneous abortions are more likely to occur. As discussed above, small sample sizes limit ability to obtain a precise estimate of risk. A systematic review and meta-analysis of seven published observational studies (four involving unadjuvanted A[H1N1]pdm09 monovalent vaccine, two involving adjuvanted A[H1N1]pdm09 monovalent vaccine, and one involving A/New Jersey/8/76 monovalent vaccine) found no significant difference in risk for spontaneous abortion between vaccinated and unvaccinated women (RR=0.91; 95%CI 0.68, 1.22) (563). Some reviews of studies involving seasonal and 2009(H1N1) monovalent IIV in pregnancy have concluded that no evidence exists to suggest harm to the fetus from maternal vaccination (565, 566). A cohort study from the Vaccines and Medications in Pregnancy Surveillance System (VAMPSS) of vaccine exposure during the 2010-11 through 2013-14 seasons found no significant association of spontaneous abortion with influenza vaccine exposure in the first trimester or within the first 20 weeks of gestation (554).

Several case-control studies evaluating receipt of IIVs and subsequent spontaneous abortion have been conducted within the VSD. The first of these, conducted during the 2005–06 and 2006–07 seasons, found no significant increase in the risk for pregnancy loss in the 4 weeks following seasonal influenza vaccination (557). A subsequent study using similar methods reported an increased risk for spontaneous abortion during days 1 to 28 days after receiving IIV3 during either the 2010–11 or the 2011– 12 seasons; the increased risk was seen among women who had also received a H1N1pdm09-containing vaccine in the previous season (567). A larger follow-up study conducted over three seasons (2012-13 through 2014-15) found no association between IIV receipt and spontaneous abortion in any of the studied seasons, regardless of vaccination status in the previous season (568).

Multiple studies have found no increased risk for stillbirth among women who received IIV during pregnancy (185, 552, 555, 563, 569-574). A review of health registry data in Norway noted an increased risk for fetal death associated with clinically diagnosed (not laboratory-confirmed) influenza A(H1N1) pdm09 infection, but no increased risk for fetal mortality associated with vaccination (63). A systematic review and meta-analysis of seven published observational studies (four involving unadjuvanted A[H1N1]pdm09 monovalent vaccine, two involving adjuvanted A[H1N1]pdm09 monovalent vaccine, and one involving A/New Jersey/8/76 monovalent vaccine) reported decreased risk for stillbirth among women who were vaccinated (RR=0.73; 95%CI 0.55, 0.96 for all studies; RR=0.69; 95%CI 0.52, 0.90 for studies of influenza A(H1N1)pdm09 vaccines) (563).

A matched case-control study of 225 pregnant women who received IIV3 within the 6 months before delivery determined that no SAEs occurred after vaccination and that no difference in pregnancy outcomes was identified among these pregnant women compared with 826 pregnant women who were not vaccinated (575). Reviews of VAERS reports during 1990–2009 (576) and 2010–2016 (577), concerning pregnant women after receipt of IIV3 did not find any new or unexpected pattern of adverse pregnancy events or fetal outcomes.

Data are reassuring with regard to the risk of congenital malformations following maternal influenza vaccination, with a large number of studies noting no association (552, 554, 573, 578-582). A systematic review and meta-analysis of studies of congenital anomalies after vaccination including data from 15 studies (14 cohort studies and one case-control study), eight of which reported data on first-trimester immunization, found that risk for congenital malformations was similar for vaccinated and unvaccinated mothers. In the cohort studies, events per vaccinated versus unvaccinated were 2.6% versus 3.1% (5.4% versus 3.3% for the subanalysis involving first-trimester vaccination). In the case-control study, there was no association between congenital defects and influenza vaccination in any trimester (OR=0.96; 95%CI 0.86, 1.07) or specifically in the first trimester (OR=1.03; 95%CI 0.91, 1.18). With respect to major malformations, there was no increased risk after immunization in any trimester (OR=0.99; 95%CI 0.88, 1.11) or in the first trimester (OR=0.98; 95%CI 0.83. 1.16) (581). A case-control analysis from VAMPSS of data from the 2011-12 through 2013-14 seasons noted an elevated OR for omphalocele (OR=5.19; 95%CI 1.44, 18.7) during the 2011-12 season, no other significant associations were found (579). A study examining safety of exposure to pandemic H1N1pdm09-containing vaccine during the 2009-2010 and 2010-11 seasons noted an increased risk for anophthalmia/microphthalmia (OR=8.67; 95%CI 1.10, 68.5); the authors noted the possibility that this association could be due to chance, in the context of multiple comparisons (582). A VSD analysis comparing occurrence of selected major birth defects among women with and without first trimester IIV exposure found no increased risk of the selected defects (578). A retrospective cohort study including 15,510 infants born to active duty military mothers who received IIVs while pregnant during 2008-09 (seasonal vaccine) and 2009-2010 (pandemic vaccine) found no increased risk of birth defects (as reported in medical records through the first year after birth) following receipt of pandemic vaccines as compared with seasonal vaccines (583).

Assessments of association between influenza vaccination and preterm birth and small for gestational age infants have yielded inconsistent results, with most studies reporting no association or a protective effect against these outcomes (574, 584-590). Some authors have noted that protective effects observed in some studies may be attributable to biases (589). A VSD study of 46,549 pregnancies during the 2009-2010 season found a strong protective effect against preterm birth of monovalent H1N1pdm09 vaccination which was no longer present with adjustment for potential biases such as differences in vaccine availability, timing of vaccination, and likelihood of vaccination associated with baseline characteristics of the study populations (591).

Few studies have assessed infant health outcomes outside the neonatal period, among infants born to mothers receiving IIV during pregnancy. A retrospective cohort study of electronic medical record data including nearly 197,000 women noted no association between receipt of IIV in any trimester and diagnosis of an autism spectrum disorder (ASD) in the child. When data were analyzed by trimester, an increased risk was noted following vaccination during the first trimester (adjusted HR=1.20; 95%CI 1.04, 1.39) (592). This association was no longer statistically significant after adjusting for multiple comparisons. A VSD matched case-control study of 413,034 infants born between January 2004 through June 2014 found no association between maternal receipt of influenza vaccine and infant hospitalization during the first 6 months of life (aOR=1.00; 95%CI 0.96, 1.04) or death (aOR=0.96; 95%CI 0.54, 1.69) (593).

RIVs

Experience with the use of RIVs in pregnancy is limited compared to that with IIVs, as these vaccines have been available only since the 2013-14 influenza season. In two pre-licensure studies, 23 pregnancies occurred among participants who received RIV3. Complete follow-up was available for 18, for which outcomes included 11 uneventful, normal, term births; 2 in which the recipients experienced pregnancy-related AEs but delivered healthy infants; 4 elective terminations; and one spontaneous abortion (594). VAERS has received 3 RIV3 reports and 7 RIV4 reports involving pregnant women (CDC, unpublished data). A pregnancy registry has been established for RIVs (124).

LAIVs

As a live virus vaccine, LAIV has not been recommended for use during pregnancy. However, occasional reports of its use for pregnant women are reported to VAERS. Among 27 reports to VAERS involving inadvertent administration of LAIV3 to pregnant women during 1990– 2009, no unusual patterns of maternal or fetal outcomes were noted (576). Of 127 reports of administration of LAIV3/4 to pregnant women submitted to VAERS from July 2010 through May 2016, no AE was reported in 112 instances. The remaining 15 included two reports each of spontaneous abortion, elective termination, and nasal congestion and one report each of transverse myelitis, abdominal pain, preterm delivery, chest pain with dyspnea secondary to trauma, pure cell aplasia, headache, common cold, pulmonary hypertension in a newborn infant, and one unspecified pregnancy complication. Only the instance of pulmonary hypertension in the infant was reported as a serious event (577). Among 138 instances of administration of LAIV3 to pregnant women noted in a health insurance claims database, reported outcomes occurred at similar rates to those reported in literature among unvaccinated women (595).

Vaccine components can occasionally cause allergic reactions, also called immediate hypersensitivity reactions. Immediate hypersensitivity reactions are mediated by preformed immunoglobulin E (IgE) antibodies against a vaccine component, and usually occur within minutes to hours of exposure (596). Symptoms of immediate hypersensitivity range from urticaria (hives) to angioedema and anaphylaxis. Anaphylaxis is a severe life-threatening reaction that involves multiple organ systems and can progress rapidly. Symptoms and signs of anaphylaxis can include (but are not limited to) generalized urticaria, wheezing, swelling of the mouth, tongue, and throat, difficulty breathing, vomiting, hypotension, decreased level of consciousness, and shock. Minor symptoms such as red eyes or hoarse voice also might be present (596, 597).

Vaccines contain multiple components that may potentially cause allergic reactions. These include the vaccine antigen, residual animal proteins, antimicrobial agents, preservatives, stabilizers, or other vaccine components. Manufacturers use a variety of compounds to inactivate influenza viruses and may add antibiotics to prevent bacterial growth. Package inserts for specific vaccines should be consulted for additional information. The ACIP has recommended that all vaccine providers be familiar with their office emergency plan and be certified in cardiopulmonary resuscitation (598). The Clinical Immunization Safety Assessment (CISA) Project (www.cdc.gov/vaccinesafety/ensuringsafety/monitoring/cisa), a collaboration between CDC and medical research centers with expertise in vaccinology and vaccine safety, has developed an algorithm to guide evaluation and revaccination decisions for persons with suspected immediate hypersensitivity after vaccination (596).

Anaphylaxis after receipt of influenza vaccines is rare. In a review of VAERS reports from 2010 through 2016, an estimated median annual rate of 0.2 per million administered doses was calculated (599). A VSD study conducted during 2009–2011 observed that the incidence of anaphylaxis in the 0–2 days after any vaccine was 1.31 (95%CI 0.90, 1.84) cases per million vaccine doses in all ages. The incidence of anaphylaxis in the 0–2 days after IIV3 (without other vaccines) was 1.35 (95%CI 0.65, 2.47) per million doses administered in all ages (600). Anaphylaxis occurring after receipt of influenza vaccines is occasionally reported to VAERS (431, 538, 601, 602). A VSD study of children aged <18 years in four health maintenance organizations during 1991–1997 estimated the overall risk for post-vaccination anaphylaxis after any type of childhood vaccine to be approximately 1.5 cases per million doses administered. In this study, no cases were identified among IIV3 recipients (603).

Most currently available influenza vaccines (with the exceptions of RIVs and ccIIV4) are prepared by propagation of influenza viruses in embryonated eggs, and therefore may contain egg proteins such as ovalbumin. Among influenza vaccines for which ovalbumin content was disclosed during the 2011–12 through 2016–17 seasons, reported maximum amounts were ≤1 µg/0.5 mL dose for IIVs and <0.24 µg/0.2 mL dose for LAIV4. Of the three vaccines produced using nonegg based technologies, currently only RIV3 and RIV4 (Flublok and Flublok Quadrivalent; Protein Sciences, Meriden, Connecticut) are considered egg-free. For ccIIV4 (Flucelvax Quadrivalent; Seqirus, Holly Springs, North Carolina), ovalbumin is not directly measured. Influenza viruses for ccIIV4 are propagated in mammalian cells rather than in eggs; however, egg proteins are potentially introduced at the start of manufacture, because some of the original viruses received from the WHO are egg-derived. From that point forward, no eggs are used, and dilutions at various steps during the manufacturing process result in a theoretical maximum of 5×10^-8 μg/0.5 mL dose of total egg protein (Seqirus, data on file, 2016).

Reviews of studies of experience with use of IIV, and more recently LAIV, indicate that severe allergic reactions to the currently available egg-based influenza vaccines in persons with egg allergy are unlikely. In a 2012 review of published data, including 4,172 egg-allergic patients (513 reporting a history of severe allergic reaction) there were no noted occurrences of anaphylaxis following administration of IIV3, though some milder reactions did occur (513). Subsequently, several evaluations of LAIV use in persons with egg allergy have been published. In a prospective cohort study of children aged 2 through 16 years (68 with egg allergy and 55 without), all of whom received LAIV, none of the egg-allergic subjects developed signs or symptoms of an allergic reaction during the one hour of postvaccination observation, and none reported adverse reactions that were suggestive of allergic reaction or that required medical attention after 24 hours (514). In a larger study of 282 egg-allergic children aged 2 through 17 years (115 of whom had experienced anaphylactic reactions to egg previously), no systemic allergic reactions were observed after LAIV administration (515). Eight children experienced milder, self-limited symptoms that might have been caused by an IgE-mediated reaction. In another study of 779 egg-allergic children aged 2 through 18 years (270 of whom had previous anaphylactic reactions to egg), no systemic allergic reactions occurred. Nine children (1.2%) experienced milder symptoms, possibly allergic in nature within 30 minutes of vaccination (four rhinitis, four localized/contact urticaria, and one oropharyngeal itching) (516). A study that compared adverse reactions in eight egg-allergic and five nonegg-allergic children when given increasing doses of egg protein (517) showed only mild symptoms of rhinitis after exposure to 10–100 µg. This is substantially more than the concentration of ovalbumin reported on the LAIV package insert (<0.24 µg per 0.2 mL dose). All eight egg-allergic children tolerated LAIV doses without any allergic symptoms. These data indicate that LAIV4 may be administered safely to persons with a history of egg allergy. However, ACIP recommends that LAIV4 not be used in any population during the 2017–18 season because of concerns regarding effectiveness against influenza A(H1N1)pdm09.

Occasional cases of anaphylaxis in egg-allergic persons have been reported to the Vaccine Adverse Event Reporting System (VAERS) after administration of influenza vaccines (508, 509). ACIP will continue to review available data regarding anaphylaxis cases following influenza vaccines.