Safety of in fl uenza vaccination on adverse birth outcomes among pregnant women: A prospective cohort study in Japan
Satoko Ohfuji
a,b,*, Masaaki Deguchi
c, Daisuke Tachibana
d, Masayasu Koyama
d, Tetsu Takagi
e, Takayuki Yoshioka
f, Akinori Urae
g, Kazuya Ito
a,h, Tetsuo Kase
a,b, Akiko Maeda
a, Kyoko Kondo
i, Wakaba Fukushima
a,b, Yoshio Hirota
a,h,j, for the Osaka Pregnant Women In fl uenza Study Group
1aDepartmentofPublicHealth,OsakaCityUniversityGraduateSchoolofMedicine,1-4-3,Asahi-machi,Abeno-ku,Osaka-shi,Osaka545-8585,Japan
bResearchCenterforInfectiousDiseaseSciences,OsakaCityUniversityGraduateSchoolofMedicine,1-4-3,Asahi-machi,Abeno-ku,Osaka-shi, Osaka545-8585,Japan
cDepartmentofObstetricsandGynecology,KishiwadaCityHospital,1001,Gakuhara-cho,Kishiwada-shi,Osaka596-8501,Japan
dDepartmentofObstetricsandGynecology,OsakaCityUniversityGraduateSchoolofMedicine,1-4-3,Asahi-machi,Abeno-ku,Osaka-shi, Osaka545-8585,Japan
eTakagiLadiesClinic,1-13-44,Kamihigashi,Hirano-ku,Osaka-shi,Osaka547-0002,Japan
fOsakaBranch,MedisciencePlanningInc.,3-6-1,Hiranomachi,Chuo-ku,Osaka-shi,Osaka541-0052,Japan
gHeadOffice,MedisciencePlanningInc.,1-11-44,Akasaka,Minato-ku,Tokyo107-0052,Japan
hCollegeofHealthcareManagement,960-4,Takayanagi,Setaka-machi,Miyama-shi,Fukuoka,835-0018,Japan
iOsakaCityUniversityHospital,1-4-3,Asahi-machi,Abeno-ku,Osaka-shi,Osaka545-8585,Japan
jClinicalEpidemiologyResearchCenter,SOUSEIKAI,3-5-1,Kashii-Teriha,Higashi-ku,Fukuoka-shi,Fukuoka813-0017,Japan
ARTICLE INFO Articlehistory:
Received29November2019
Receivedinrevisedform9January2020 Accepted19January2020
Keywords:
Adversebirthoutcomes Influenzavaccine Pregnantwomen Prospectivecohortstudy Vaccinesafety
ABSTRACT
Background:Pregnantwomenareinthehighestprioritygroupfor receivinginfluenzavaccination.
However,theymaybereluctant toreceivethevaccination duetoconcernsaboutthe influenceof vaccinationonthefetuses.
Methods:Thisprospectivecohortstudyof10330pregnantwomenexaminedthesafetyofinfluenza vaccinationintermsofadversebirthoutcomes.Influenzavaccinationduringpregnancywasdetermined fromquestionnairesbeforeandafterthe2013/2014influenzaseason.Allsubjectswerefolloweduntilthe endoftheirpregnancy.Adversebirthoutcomes,includingmiscarriage,stillbirth,pretermbirth,lowbirth weight,andmalformation,wereassessedbyobstetricianreports.
Results:Adversebirthoutcomeswerereportedfor641(10%)ofthe6387unvaccinatedpregnantwomen and356(9%)ofthe3943vaccinatedpregnantwomen.Evenafteradjustingforpotentialconfounders, vaccinationduringpregnancyshowednoassociationwiththeriskofadversebirthoutcomes(oddsratio 0.90,95%confidenceinterval0.76–1.07).Vaccinationduringthefirstorsecondtrimesterdisplayedno associationwithadversebirthoutcomes,whereasvaccinationduringthethirdtrimesterwasassociated withadecreasedriskofadversebirthoutcomes(oddsratio0.70,95%confidenceinterval0.51–0.98).
Conclusions:Influenzavaccinationduringpregnancydidnotincreasetheriskofadversebirthoutcomes, regardlessof thetrimesterinwhich vaccinationwasperformed,whencompared tounvaccinated pregnantwomen.
©2020TheAuthor(s).PublishedbyElsevierLtdonbehalfofInternationalSocietyforInfectiousDiseases.
ThisisanopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/licenses/by-nc- nd/4.0/).
Introduction
InNovember2012,theWorldHealthOrganizationpresenteda position paperplacing pregnant womeninthe highestpriority grouptoreceiveinfluenzavaccination,duetotheexpectationsof vaccineeffectivenessinpreventinginfluenzaamongmothersand theirinfants(WorldHealthOrganization,2012). Indeed,several epidemiological studieshave indicated thatmaternal influenza
*Corresponding authorat: Corresponding authorat:Department ofPublic Health,OsakaCityUniversityGraduateSchoolofMedicine,1-4-3,Asahi-machi, Abeno-ku,Osaka-shi,Osaka545-8585,Japan.
E-mailaddress:[email protected](S.Ohfuji).
1OthermembersofthestudygrouparelistedintheAppendixA.
https://doi.org/10.1016/j.ijid.2020.01.033
1201-9712/©2020TheAuthor(s).PublishedbyElsevierLtdonbehalfofInternationalSocietyforInfectiousDiseases.ThisisanopenaccessarticleundertheCCBY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).
ContentslistsavailableatScienceDirect
International Journal of Infectious Diseases
j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / i j i d
vaccinationprovideseffectiveprotectionagainstinfantinfluenza (Benowitz et al., 2010; Black et al., 2004; Ohfuji et al., 2018;
Steinhoffetal.,2012;Zamanetal.,2008). Ingeneral, however, pregnant women tend to be concerned about the influence of vaccinesonthe fetuses,whichmay leadto somereluctanceto undergovaccination.Infact,apreviousstudyidentifiedconcerns aboutvaccinesafetyasthemostsignificantreasonforpregnant women not undergoing influenza vaccination (Prospero et al., 2019). Particularly in Japan,influenza vaccinationfor pregnant womenisperformedas‘voluntaryvaccination’.Inthissituation, positivevaccinationbehaviorsamongpregnantwomenarelikely toremainsuboptimaluntilthesafetyconcernsregardingeffectson fetusescanbeaddressed.
A review of previous reports on the safety of influenza vaccinationamongpregnantwomenrevealednostudiesexamin- ing the influence of influenza vaccination among Japanese pregnantwomenonadversebirthoutcomes.SincetheJapanese populationtendstoshowgreaterconcern aboutvaccine safety thanotherpopulations(Hanleyetal.,2015;Nakayama,2019),this lackofevidenceamongtherelevantpopulationmightpresenta barriertoachievingadequatecoveragewithinfluenzavaccination for pregnant women. Additionally, the proportions of preterm delivery, lowbirth weightinfants, and malformedinfants vary betweencountries(Källén,2012;Morisakietal.,2017;Sepkowitz, 1995).
Wethereforeconductedaprospectivecohortstudytoexamine vaccinesafetyincomparisonwiththeincidenceofadversebirth outcomes (including miscarriage, stillbirth, preterm birth, low birthweight,andcongenitalmalformation)betweenvaccinated andunvaccinatedpregnantwomeninJapan.Ingeneral,pregnant womenwhoreceiveinfluenzavaccinationarelikelytobeolderor tohave anunderlying illness suchashypertension ordiabetes, representingconditionsthatmaybringaboutahigherincidenceof adverse birth outcomes. In this study, the safety of influenza vaccinationinpregnantwomenwasevaluatedwithconsideration oftheeffectofdifferencesinsuchbackgroundcharacteristics.
Methods Studysubjects
This study was conducted with the cooperation of 117 maternity hospitals and clinics affiliated with the Obstetrical GynecologicalSocietyofOsaka,Japan.Studysubjectscomprised Japanese pregnant women (regardless of gestational week) attending the collaborating hospitals and clinics before the beginningofthe2013/14influenzaseason(i.e.,betweenOctober andDecember2013).InJapan,pregnantwomentypicallyundergo influenzavaccinationatamaternityclinicorprimarycareclinic betweenOctoberandDecember,asa voluntaryvaccination. All studysubjects receivedanexplanationof thestudy fromtheir obstetrician and verbally provided informed consent prior to participation.
ThestudyprotocolwasapprovedbytheEthicsCommitteeat theOsakaCityUniversityGraduateSchoolofMedicine,andwas performedinaccordancewiththeDeclarationofHelsinki.
Informationcollection
At thetimeofrecruitment,studysubjectscompleteda self- administeredquestionnairetoprovidethefollowinginformation:
dateofrecruitment,age,gestationalageatrecruitment,expected dateofbirth,heightandweightbeforepregnancy,smokingand alcoholdrinkinghabits,underlyingillnesses,influenzavaccination status for the 2013/14 season, and month of vaccination for vaccinatedsubjects.Theaccuracyofgestationalageatrecruitment
wasconfirmedbyreferringtotheexpecteddateofbirth.Tocollect informationonreceiptofinfluenzavaccinationafterrespondingto thequestionnaireatrecruitment,thestudysubjectsweresenta secondquestionnaireaftertheendofthe2013/14influenzaseason (May2014).Inthispost-seasonquestionnaire,besidesvaccination statusforthe2013/14seasonandmonthofvaccination,wealso askedthefollowingquestionsaboutpregnancyoutcomesandtheir babies:dateofdeliveryandbirthweightandheightoftheirbabies.
Toconfirmtheseself-reportedpregnancyoutcomesandneonatal characteristics, the obstetrician-in-charge was contacted and asked to provide the following information from the medical recordsofeachsubject:pregnancyoutcome(live birth,miscar- riage,orstillbirth),andifalivebirthwasdelivered,thedateof delivery, gestationalweek atdelivery, birth weightandheight, Apgarscoresat1minand5min,andpresenceandnameofany congenitalmalformations.Inaddition,informationonpregnancy- induced complications (i.e., multiple pregnancy, pregnancy- induced hypertension, gestational diabetes, hospitalization due tothreatenedmiscarriage,placentaprevia,fetalgrowthrestriction, abruptioplacentae,andintrauterineinfection)wasalsocollected bytheirobstetricians.
Statisticalanalysis
The primary exposure was influenza vaccination during pregnancy, determined from information on the month of vaccination and month of delivery. Subjects who received vaccination in the same month as the delivery, or for whom information onthe monthofvaccinationwasunavailable were excludedfromtheanalysis.
The study outcome was adverse birth outcomes including miscarriage (terminationof pregnancy before gestational week 22),stillbirth(deadatbirthoraftergestationalweek22),preterm birth(livebirthatlessthangestationalweek37),and/orlowbirth weight(birthweight<2500g)forallstudysubjects.Miscarriage andstillbirthincludedtherapeuticabortions.Informationonlow birth weight was primarily based on information from the obstetrician.Ifinformationwasunavailablefromtheobstetrician, complementary datawere obtainedfrom the self-administered questionnaire.Inaddition,Apgarscoresat1minand5minwere alsoassessedusingthreecategories:0–3,verylow;4–6,low;7–10, healthy. Also, for women in the first trimester, congenital malformationwasassessed asanother study outcome.Genetic andchromosomalabnormalitieswerenotincludedincongenital malformation, because these occur at conception and are uninfluenced by vaccination. For detailed analyses, congenital malformationswereclassifiedinto10categoriesbyorgansystem (i.e., central nervous system; ophthalmological, otological, or orofacial; cardiac; respiratory; cleft lip and/or cleft palate;
gastrointestinal;genitourinaryorrenal;muscularorlimbdefects;
orother),accordingtoInternationalClassificationofDiseases10th revision(ICD-10)codes,andwerecomparedbetweenunvaccinat- edandvaccinatedwomen.
Withregardtoexplanatoryvariables,agewascategorizedinto
<30, 30–34, and >34 years old. Body mass index (BMI) was
calculatedasweightdividedbyheightsquared(kg/m2),andthen classifiedintothreecategoriesaccordingtoconventionalcut-off values.Gestationalagewasdefinedasgestationalweekatthetime ofvaccinationforvaccinatedwomenoratthetimeofrecruitment forunvaccinatedwomen,andwascategorizedintofirsttrimester (<16weeks),secondtrimester(16–27weeks),andthirdtrimester (>27weeks).Gestationalageatvaccinationwascalculatedusing theinformationonthemonthofvaccination,gestationalageat recruitment,anddateatrecruitment,andconsideringthedateof vaccination as the 15th day (median) of the month. Calendar monthatthestartofpregnancywascalculatedbyinformationon
thedateofrecruitmentandgestationalageatrecruitment,andwas classifiedintofourseasons.Thefollowinginfluenza-relatedhigh- risk conditions were included according to a previous report:
chronicrespiratorydisorders(includingasthma),cardiovascular disorders (excluding isolated hypertension), kidney disease, liverdisease, neurologicaldisorders,blood disorders,metabolic disorders (including diabetes), immunocompromised states (suchasmalignanttumors,connectivetissue disorders,inflam- matory bowel disease, and chronic rheumatism), and obesity (BMI25.0kg/m2)(CentersforDiseaseControlandPrevention, 2013). Underlying obstetric and gynecological illnesses were includedasinfertility,myomauteri,ovariandiseases,endometri- osis,diseasesintheneckoftheuterusincludingseveredysplasiaor cancer,endometrial polyp, adenomyosisuteri, habitual miscar- riage,etc.
Alogisticregressionmodelwasusedtocalculatetheoddsratio (OR)and 95% confidence interval(CI)for associationsbetween influenza vaccination duringpregnancy and adversebirth out- comes. Themultivariate modelincludedall variablesrelatedto vaccination status (i.e., exposure variables) or adverse birth outcomes (i.e., outcome index) showing values of p< 0.05 in theunivariateanalyses.TheChi-squaretestandWilcoxonrank- sumtestwereusedwhereappropriate.
In addition, in order toseparately evaluatethe influenceof influenza vaccination on adverse birth outcomes according to gestationalweek,stratifiedanalysesbytrimesterwereconducted.
All analysesweretwo-tailed andwereconductedusing SAS version9.3software(SASInstitute,Cary,NC,USA).
Results
Of the 20 420 pregnant Japanesewomen recruited,12 838 respondedto thepost-season questionnaire. Amongthese, 301 vaccinatedwomenwereexcluded,becausevaccinationhadbeen performedinthemonthofdelivery;whetherthevaccinationhad beenperformedbeforeorafterdeliverywasthusunclear.Another 233 vaccinated women were excluded because of a lack of information on themonth ofvaccination. Informationon birth outcomes was then obtained for 10 330 women from their obstetricians,andthesewomenthereforecomprisedthesubjects foranalysis(Figure1). Table1 showsthecharacteristics ofthe studysubjects.Atotal of3943women(38%)receivedinfluenza vaccinationduringpregnancy.Amongthese,aboutone-thirdhad receivedthevaccinationduringeachofthefirst,second,andthird trimesters.Vaccinatedwomenwereolderandmorelikelytohave underlyingobstetricandgynecologicalillnesses,whereasunvac- cinatedwomenappearedtoshowhigherfrequenciesofobesity, hypertension, or fetalgrowth restriction as pregnancy-induced
complications,andofhavingsmokingoralcoholdrinkinghabits duringpregnancy.
Table2showsbirthoutcomesforthestudysubjects.Miscar- riage orstillbirth was reported for0.1% ofsubjects, each with similar proportions in unvaccinated and vaccinated women.
Preterm birth occurred in 4.1% of subjects, again with similar proportionsinthetwogroups.Ontheotherhand,lowbirthweight wassignificantlymorefrequentamongunvaccinatedwomenthan amongwomenvaccinatedduringpregnancy(8%vs.7%).
Atotalof 997subjects (10%)reportedmiscarriage, stillbirth, pretermbirth,and/orlowbirthweightasadversebirthoutcomes (Table3).Womenwhohadreceivedinfluenzavaccinationduring pregnancy reported slightly fewer adverse birth outcomes compared with unvaccinated women, although the difference wasnotsignificant(9%vs.10%,respectively;p=0.09).Inaddition, pregnantwomen30yearsold,withBMI<18.5kg/m2,underlying obstetricandgynecologicalillnesses,pregnancy-inducedcompli- cations, orasmokinghabitduringpregnancyweresignificantly more likely to present with adverse birth outcomes. After considering the effects of these potential confounders in the multivariateanalysis,vaccinationduringpregnancydidnotshow any significant associationwith adverse birth outcomes when comparedto unvaccinatedwomen(OR0.90,95% CI0.76–1.07).
However,age30years,lowerBMIbeforepregnancy,andsome pregnancy-induced complications were significantly associated withadversebirthoutcomes.
Adverse birth outcomes were examined separately in sub- groups accordingto thetrimesteratvaccinationforvaccinated womenoratrecruitmentforunvaccinatedwomen(Table4).Inthe firsttrimester,althoughcongenitalmalformationwasregardedas one of the adversebirth outcomes,no significantdifference in theseadversebirthoutcomeswasseenbetweenunvaccinatedand vaccinated women (13% each). In the second trimester, the proportionofadversebirthoutcomeswasbroadlysimilaramong unvaccinated and vaccinated women. In the third trimester, however, vaccinated women had significantly fewer reports of adversebirthoutcomes(6%vs.9%,respectively),especiallyforlow birthweight(6%vs.8%,respectively),thanunvaccinatedwomen.
Eveninthemultivariateanalysiswithconsiderationoftheeffectof potentialconfounders,womenwhoreceivedvaccinationduring the firstorsecondtrimestershowed nosignificantelevation in adversebirthoutcomescomparedwithunvaccinatedwomen(first trimester:OR1.07,95%CI0.81–1.40;secondtrimester:OR0.87, 95% CI 0.65–1.16). On the other hand, women who received vaccination during the third trimester showed a significantly decreasedORforadversebirthoutcomes whencomparedwith unvaccinatedwomen(OR0.70,95%CI0.51–0.98).
Discussion
The study findings demonstrated that influenza vaccination duringpregnancywasnotassociatedwithanyincreaseinadverse effectsonthefetus.Thisresultisconsistentwithpreviousstudies fromothercountries.Todate,severalrandomizedcontrolledtrials ofpregnantwomenhaveshownthattheincidencesofmiscarriage, stillbirth, preterm birth, low birth weight, and congenital malformationsamonginfluenzavaccinationgroupsweresimilar to thoseinplacebogroups (Michikawaetal.,2018;OsakaCity, 2019;Steinhoffetal.,2012).Mostcohortstudieshavealsoshown thatvaccinatedandunvaccinatedpregnantwomendisplaysimilar incidences of miscarriage, stillbirth, preterm birth, low birth weight,orcongenitalmalformationintheirbabies(Baumetal., 2015;Blacketal.,2004;Chambersetal.,2013;Chambersetal., 2016;Clearyetal.,2014;deVriesetal.,2014;Fabianietal.,2015;
Felletal.,2012;Felletal.,2017;Kharbandaetal.,2017;Maetal., 2014;Madhietal.,2014;McHughetal.,2017;Nordinetal.,2014a;
Figure1.Theenrollmentprocessforthestudy.
Olsen etal.,2016; Omonet al.,2011; Oppermann et al.,2012;
OskoviKaplanandOzgu-Erdinc,2018;Pasternaketal.,2012;Regan etal.,2016;Steinhoffetal.,2017;Sugiura-Ogasawaraetal.,2019;
Vazquez-Benitezetal.,2016).Inapreviouscase–controlstudy,no associationwasidentifiedbetweeninfluenzavaccinationduring pregnancyandmiscarriage(Ludvigssonetal.,2013).
Ingeneral,concernshavebeenraisedregarding theeffectof maternalmedications,includingvaccination,onthefetuswithin thefirsttrimester,sincethefirsttrimesterisacrucialperiodfor embryogenesisofthemajororgans.However,thepresentstudy showed that even pregnant women who received influenza vaccinationduringthefirsttrimestershowedsimilarincidences of miscarriage, stillbirth, preterm birth, low birth weight, and congenital malformations when compared with unvaccinated
women.Theresultssuggestnoadverseinfluencesonthefetus, evenwhenprovidinginfluenzavaccinationtopregnantwomenin thefirsttrimester.
Besides,pregnantwomenwhoreceivedinfluenzavaccination duringthethirdtrimesterwerelesslikelytohavebabieswithlow birth weight. This was an unexpected finding. One possible interpretation is that recent advancesin medical checkups for pregnancyhaveenabledbetterdiagnosisoffetalgrowthrestriction duringpregnancy.Pregnantwomendiagnosedwithfetalgrowth restriction during the third trimester might thus have been reluctant to receive influenza vaccination. Such a difference in vaccinationbehaviormightresultinapparentincreasesinbabies with low birth weight among unvaccinated pregnant women.
However,thepresentstudydidnotcollectinformationaboutthe Table1
Characteristicsofthepregnantwomen(N=10330).a
Characteristics Total Unvaccinated Vaccinated p-Value
n(%) n(%) n(%)
Total 10330(100) 6387(62) 3943(38)
Gestationalageatrecruitmentorvaccination(weeks) <16(firsttrimester) 2826(27) 1705(27) 1121(28) <0.01 16–27(secondtrimester) 3328(32) 1738(27) 1590(40)
>27(thirdtrimester) 4176(40) 2944(46) 1232(31)
Calendarmonthatpregnancystart March–May(Spring) 3506(34) 1997(31) 1509(38) <0.01
June–August(Summer) 2908(28) 1434(22) 1474(37)
September–November(Autumn) 1997(19) 1215(19) 782(20) December–February(Winter) 1919(19) 1741(27) 178(5)
Age(years) Median(range) 32(15–51) 32(15–51) 33(16–47) <0.01
<30 3273(32) 2273(36) 1000(25) <0.01
30–34 3673(36) 2154(34) 1519(39)
>34 3384(33) 1960(31) 1424(36)
Bodymassindexbeforepregnancy(kg/m2) <18.5 1636(16) 1007(16) 629(16) 0.04
18.5–24.9 7585(75) 4640(74) 2945(76)
>24.9 960(9) 640(10) 320(8)
Influenza-relatedhigh-riskconditions Present 2321(22) 1456(23) 865(22) 0.31
Underlyingobstetricandgynecologicalillness Present 1916(19) 1075(17) 841(21) <0.01
Pregnancy-inducedcomplications n/N(%) n/N(%) n/N(%)
Multiplepregnancy Present 149/10328(1) 91/6386(1) 58/3942(1) 0.85
Pregnancy-inducedhypertension Present 338/10302(3) 234/6367(4) 104/3935(3) <0.01
Gestationaldiabetes Present 276/10322(3) 185/6383(3) 91/3939(2) 0.07
Hospitalizationduetothreatenedmiscarriage Present 526/10318(5) 324/6379(5) 202/3939(5) 0.91
Placentaprevia Present 41/10325(0.4) 28/6384(0.4) 13/3941(0.3) 0.39
Fetalgrowthrestriction Present 271/10315(3) 190/6377(3) 81/3938(2) <0.01
Abruptioplacentae Present 36/10324(0.4) 25/6384(0.4) 11/3940(0.3) 0.35
Intrauterineinfection Present 81/10313(1) 54/6376(1) 27/3937(1) 0.37
Smokinghabit Presentduringpregnancy 306/9645(3) 261/5983(4) 45/3662(1) <0.01
Alcoholdrinkinghabit Presentduringpregnancy 66/9661(0.7) 54/5996(0.9) 12/3665(0.3) <0.01
aDataareexpressedasthenumber(%)unlessindicatedotherwise.
Table2
Birthoutcomesofthestudysubjects.a
Birthoutcomes Total Unvaccinated Vaccinated p-Value
n(%) n(%) n(%)
Pregnancyoutcomes Livebirth 10305(99.8) 6370(99.7) 3935(99.8) 0.38
Miscarriage 11(0.1) 6(0.1) 5(0.1)
Stillbirth 14(0.1) 11(0.2) 3(0.1)
Gestationalageatdelivery(weeks) 22–36(pretermbirth) 421(4.1) 258(4.1) 163(4.2) 0.81
37–41 9839(95.5) 6084(95.5) 3755(95.4)
42+ 45(0.4) 28(0.4) 17(0.4)
Birthweight(g) Median(range) 3030(428–4670) 3032(484–4670) 3030(428–4615) 0.39
<2500(lowbirthweight) 812(7.9) 531(8.3) 281(7.1) 0.03
2500 9493(92.1) 5839(91.7) 3654(92.9)
Apgarscoreat1min 0–3 58(0.6) 43(0.7) 15(0.4) 0.01
4–6 155(1.5) 105(1.7) 50(1.3)
7–10 10078(97.9) 6212(97.7) 3866(98.3)
Apgarscoreat5min 0–3 10(0.1) 8(0.1) 2(0.1) 0.16
4–6 32(0.3) 22(0.3) 10(0.3)
7–10 10233(99.6) 6323(99.5) 3910(99.7)
aDataareexpressedasthenumber(%)unlessindicatedotherwise.
Table3
Associationbetweenbackgroundcharacteristicsincludinginfluenzavaccinationandadversebirthoutcomes.a
Characteristics Outcomes Univariate Multivariateb
n/N(%) OR(95%CI)
p-Value
OR(95%CI) p-Value
Total 997/10330(10)
Influenzavaccinationduringpregnancy Unvaccinated 641/6387(10) 1.00 1.00
Vaccinated 356/3943(9) 0.89(0.78–1.02)
0.09
0.90(0.76–1.07) 0.24
Gestationalageatrecruitmentorvaccination(weeks) <16(firsttrimester) 291/2826(10) 1.00 1.00 16–27(secondtrimester) 354/3328(11) 1.04(0.88–1.22)
0.06
1.25(0.95–1.65) 0.12
>27(thirdtrimester) 352/4476(8) 0.80(0.68–0.94)
<0.01
1.05(0.73–1.51) 0.79
(Trendp<0.01) (Trendp=0.89)
Calendarmonthatpregnancystart March–May(Spring) 358/3506(10) 1.00 1.00
June–August(Summer) 292/2908(10) 0.98(0.83–1.16) 0.82
1.01(0.78–1.31) 0.93
September–November(Autumn) 213/1997(11) 1.05(0.88–1.26) 0.59
1.23(0.86–1.76) 0.25
December–February(Winter) 134/1919(7) 0.66(0.54–0.81)
<0.01
0.79(0.60–1.03) 0.08
Age(years) <30 264/3273(8) 1.00 1.00
30–34 380/3673(10) 1.32(1.12–1.55)
<0.01
1.34(1.09–1.64)
<0.01
>34 353/3384(10) 1.33(1.12–1.57)
<0.01
1.32(1.06–1.63) 0.01
(Trendp<0.01) (Trendp=0.01)
Bodymassindexbeforepregnancy(kg/m2) <18.5 206/1636(13) 1.44(1.22–1.70)
<0.01
1.45(1.18–1.78)
<0.01
18.5–24.9 689/7585(9) 1.00 1.00
>24.9 83/960(9) 0.95(0.75–1.20)
0.66
0.78(0.55–1.09) 0.14
(Trendp<0.01) (Trendp<0.01)
Influenza-relatedhigh-riskconditions Absent 768/8009(10) 1.00 1.00
Present 229/2321(10) 1.03(0.88–1.21)
0.69
1.19(0.95–1.50) 0.13
Underlyingillnessesinobstetricsandgynecology Absent 787/8414(9) 1.00 1.00
Present 210/1916(11) 1.19(1.02–1.40)
0.03
0.97(0.79–1.19) 0.76
Pregnancy-inducedcomplications
Multiplepregnancy Absent 894/10179(9) 1.00 1.00
Present 102/149(68) 22.5(15.9–32.1)
<0.01
14.8(9.73–22.4)
<0.01
Pregnancy-inducedhypertension Absent 894/9964(9) 1.00 1.00
Present 99/338(29) 4.21(3.29–5.37)
<0.01
3.81(2.79–5.21)
<0.01
Gestationaldiabetes Absent 957/10046(10) 1.00 1.00
Present 37/276(13) 1.47(1.03–2.09)
0.03
1.41(0.92–2.16) 0.12
Hospitalizationduetothreatenedmiscarriage Absent 821/9792(8) 1.00 1.00
Present 172/526(33) 5.31(4.37–6.46)
<0.01
4.45(3.49–5.68)
<0.01
Placentaprevia Absent 971/10284(9) 1.00 1.00
Present 24/41(59) 13.5(7.25–25.3)
<0.01
17.2(8.38–35.1)
<0.01
Fetalgrowthrestriction Absent 768/10044(8) 1.00 1.00
Present 224/271(83) 57.6(41.7–79.5)
<0.01
66.1(46.4–94.0)
<0.01
Abruptioplacentae Absent 982/10288(10) 1.00 1.00
Present 13/36(36) 5.36(2.71–10.6)
<0.01
6.22(2.55–15.2)
<0.01
Intrauterineinfection Absent 977/10232(10) 1.00 1.00
Present 15/81(19) 2.15(1.22–3.79)
<0.01
2.13(1.11–4.07) 0.02
Smokinghabit Absent 887/9339(9) 1.00 1.00
Presentduringpregnancy 43/306(14) 1.56(1.12–2.17)
<0.01
1.45(0.96–2.19) 0.08
Alcoholdrinkinghabit Absent 926/9595(10) 1.00 1.00
Presentduringpregnancy 6/66(9) 0.94(0.40–2.18) 0.88
1.15(0.44–2.95) 0.78
OR,oddsratio;CI,confidenceinterval.
aMiscarriage,stillbirth,pretermbirth,orlowbirthweightwereincluded.
bModelincludedvariablesinthistable.
timingofdiagnosesoffetalgrowthrestriction.Itisthusdifficultto determinehowsuchdiagnosesaffectedthevaccinationbehaviors ofpregnantwomen.
Variouslimitations needtobeconsideredwheninterpreting theresultsofthisstudy.First,toincreasetheresponserate,we decidedtocollect informationonvaccinationmonth insteadof vaccination date, resulting in the exclusion of 301 vaccinated womenwhohadreceivedvaccinationinthesamemonthasthe delivery. Besides, the trimester at vaccination for vaccinated women might have been misclassified into the neighboring categoryin somesubjects, since calculations weremade using informationonthemonthofvaccination,dateofrecruitment,and gestationalweekatrecruitment,andthedateofvaccinationwas regarded asthe 15thof eachmonth. Since welackedaccurate informationonthedateofvaccinationfromtheclinicatwhich patientsreceivedvaccination,thisrepresentsthemostimportant limitationofthepresentstudy.
Second,sinceinformationonvaccinationstatusandexplanatory variableswasbasedonself-reportsfrompregnantwomen,some datasuchasbodyweightbeforepregnancy,smoking,andalcohol drinking status might have been underreported. However, the presentdesignusingaprospectivecohortstudyislesssusceptibleto misclassification dueto recall errors than a case–control study design.Besides,toconfirmtheaccuracyofself-reporteddata,the dateofdeliveryandbirthweight,whichwereobtainedusingtwo methods(self-reportandobstetricianreport),wereexaminedby comparinginformationfrombothsources.Amongthesubjectsfor whomthedateofdeliverywasavailablefrombothself-reportand obstetricianreport(n=8227),thecorrelationcoefficientbetween self-reportandobstetrician reportwas0.988(p<0.01).Among subjectsforwhombirthweightwasavailablefrombothreports (n=8273),thecorrelationcoefficientwas1.000(p<0.01).Basedon these confirmations, the self-reported information used in the presentstudywasexpectedtoberelativelyreliable.
Third,the subjects analyzedcomprised 10330women who answeredthepost-seasonquestionnaireandhadbirthoutcomes providedbytheirobstetricians,fromamongthe20420women recruitedbeforetheseason.Thisfollow-upproportionmighthave affectedthestudyresults.Forexample,ifwomenwhoexperienced miscarriageorstillbirthasthepregnancyoutcometendedtobe
lesslikelytoanswerthepost-seasonquestionnaire,aselectionbias forstudysubjectswouldhavebeenpresent.Actually,considering thenumberofstillbirthsandlivebirthsinOsakaof1621and69968 in 2014(Håberget al.,2013), the proportion ofmiscarriage or stillbirthamongthepresentstudysubjects(0.1%)appearedlower thanamongthegeneralpopulation(2%).Ontheotherhand,the proportions of preterm birth, low birth weight, or congenital malformation in Japanwere reported as 5.1%, 8.3%, and 3–5%, respectively,in2013(Nordinetal.,2014b),representingpropor- tions broadly comparable to those in the present study. The possibilityofselectionbiasthusappearslowintheassessmentof pretermbirth,lowbirthweight,orcongenitalmalformations,but the possibility of selection bias due to study dropout in the assessmentsofmiscarriageorstillbirthcannotberuledout.
Fourth,sincethestudysubjectswerepregnantwomenunder clinicalfollow-upatobstetricfacilitiesinOsakaPrefecturebefore the beginningof the 2013/14 influenza season, some concerns remain about the generalizabilityof the results.Further inves- tigationsofdifferentseasonsandregionsisdesirabletoconfirm thevalidityofthepresentstudyfindings.
This study has the following strengths. First, with the cooperationoftheObstetricalGynecologicalSocietyofOsaka,it was possible to investigate the safety of influenza vaccination amongpregnantwomeninalargecohortexceeding10000study subjects,covering15%ofpregnantwomeninthestudyarea.This also enabled the examination of the effects of the timing of influenzavaccinationonadverse birth outcomes.Second,since informationonpregnancyoutcomeswasbasedonreportsfrom theobstetriciansofthestudysubjects,theaccuracyofinformation wasconsideredhigh.Infact,theproportionsofpretermbirth,low birth weight,congenitalmalformations,andpregnancy-induced complicationsinthepresentstudywerecomparabletothoseof thegeneralpopulationinJapan(Munozetal.,2005;Nordinetal., 2014b;Sheffieldetal.,2012).Additionally,maternalage,BMI,and the proportion of smokers during pregnancy were similar in another study in Japan (Munoz et al., 2005). In addition, the present study detected known risk factors for adverse birth outcomes, such asmaternal age, pregnancy-inducedcomplica- tions,andsmokingduringpregnancy(Irvingetal.,2013).These findingssuggestthereliabilityofthestudyresults.
Table4
Birthoutcomesofstudysubjectsaccordingtotrimester.
Birthoutcomes Firsttrimester p-Value Secondtrimester p-Value Thirdtrimester p-Value
Unvaccinated Vaccinated Unvaccinated Vaccinated Unvaccinated Vaccinated
n(%) n(%) n(%) n(%) n(%) n(%)
Adversebirth outcomesa
Present 229(13) 142(13) 0.56 119(11) 155(10) 0.11 265(9) 87(6) 0.04
Pregnancy outcomes
Livebirth 1697(99.5) 1116(99.6) 0.61 1733(99.7) 1588(99.9) 0.46 2940(99.9) 1231(99.9) 1.00
Miscarriage 6(0.4) 5(0.4) 0(0) 0(0) 0(0) 0(0)
Stillbirth 2(0.1) 0(0) 5(0.3) 2(0.1) 4(0.1) 1(0.1)
Pretermbirth Present 64(4) 55(5) 0.12 97(6) 79(5) 0.74 97(3) 29(3) 0.07
Lowbirthweight Present 148(9) 86(8) 0.34 155(9) 125(8) 0.27 228(8) 70(6) 0.02
Congenital malformation
Present 55(3.2) 33(3.0) 0.67 – – – –
Categoriesbyorgan system
Centralnervoussystem 0(0) 1(0.1) 0.25 Ophthalmological,otological
ororofacial
1(0.1) 3(0.3)
Cardiac 7(0.4) 3(0.3)
Respiratory 0(0) 1(0.1)
Cleftlipand/orcleftplate 1(0.1) 1(0.1)
Gastrointestinal 0(0) 1(0.1)
Genitourinaryorrenal 5(0.3) 2(0.2) Muscularorlimbdefects 7(0.4) 1(0.1)
Others 0(0) 1(0.1)
Unknown 34(2.0) 19(1.7)
aMiscarriage,stillbirth,pretermbirth,lowbirthweight,orcongenitalmalformationwereincludedforwomeninthefirsttrimester.Forwomeninthesecondorthird trimester,miscarriage,stillbirth,pretermbirth,orlowbirthweightwereincluded.
In conclusion, this cohort study indicates that influenza vaccinationof pregnant women hadno adverse effects on the fetus regardless ofthe trimester inwhich the vaccinationwas performed.Thesafety ofinfluenzavaccinationamongpregnant womeninJapanwasalsosuggested.
Funding
ThisworkwassupportedbyaresearchgrantforResearchon EmergingandRe-emergingInfectiousDiseases,HealthandLabor SciencesResearchGrantsfromtheMinistryofHealth,Laborand Welfare,Japan(H23-SHINKO-IPPAN-017andH26-SHINKOGYOSEI- SHITEI-003).
Conflictofinterest
SO reports personal fees from speaking and/or teaching arrangements,outsidethesubmittedwork;TKreportspersonal feesfromtheBIKENFoundation,outsidethesubmittedwork;WF reportspersonalfeesfromroyalties,personalfeesfromconsulting, personal fees from speaking and/or teaching arrangements, personal fees from scientific advisory committee, and grants outsidethesubmittedwork;YHreportsgrantsfromtheMinistryof Health,Labor,andWelfare,duringtheconductofthestudy;all otherauthorsdeclarenoconflictsofinterest.
Authorcontributions
SO contributed to the study design, data management, statisticalanalysis,datainterpretation,anddraftingofthework orrevisingitcriticallyforimportantintellectualcontent.MD,DT, MK,TT,andallmemberslistedintheAppendixcontributedtodata acquisitionanddatainterpretation.TYandAUcontributedtothe study design and data management. KI, TK, AM, KK, and WF contributed to the study design and data interpretation. YH contributedtotheconceptionofthedesign,overallmanagement, datainterpretation,andmanuscriptediting.Allauthorsprovided commentson thedrafts andhave readandapproved the final manuscript.
Acknowledgements
Wethankallthemedicaldoctorsinvolvedforparticipatingin this study despite their busy schedules in medical practice, education,andresearch.
AppendixA.
OthermembersoftheOsakaPregnantWomenInfluenzaStudy Groupareasfollows(showninalphabeticalorderofaffiliation):
ShiroImai(DepartmentofGynecologyandObstetrics,Aizenbashi Hospital),EikoAkagaki(AkagakiLadiesClinic),MarikoAkai(Akai Maternity Clinic), Yoshitsune Azuma (Azuma Ladies Clinic), ShinichiHamada(DepartmentofObstetricsandGynecology,Bell Land General Hospital), Satoru Motoyama (Department of Obstetrics and Gynecology, Chibune General Hospital), Hiroko Chimori(ChimoriMedicalClinic),ShokoNakagawa(Departmentof Obstetrics and Gynecology, Fuchu Hospital), Takehiko Fukuda (Fukuda Lady’s Clinic), Masahisa Hagiwara (Hagiwara Clinic), Hideto Okuda (Hamada Women’s Hospital), Takuro Hamanaka (HamanakaObstetricsandGynecology),SeiichiYamamasu(Ob- stetrics and Gynecology, Hannan Chuo Hospital), Kenji Hirota (ObstetricsandGynecology,HanwasumiyoshiGeneralHospital), Masataka Oku (Obstetrics and Gynecology, Higashi Osaka City General Hospital), Keizo Hiramatsu (Hiramatsu Obstetrics and Gynecology Clinic), Masanori Hisamatsu (Hisamatsu Maternity
Clinic),YasushiIijima(IijimaWomen’sHospital),MikioTakehara (DepartmentofObstetricsandGynecology,IkedaCityHospital), Somei Ikeda(IkedaOB/GYN Clinic),TakeshiInoue(InoueLady’s Clinic), Eriko Yamashita (Ishida Hospital), Aisaku Fukuda (The CentreforReproductiveMedicineandInfertility,IVFOsakaClinic), ItsukoIwata(IwataClinic),JunkoNishio(DepartmentofObstetrics and Gynecology, Izumiotsu Municipal Hospital), Tateki Tsutsui (Department of Obstetrics and Gynecology, Japan Community HealthcareOrganizationOsakaHospital),KenjiYamaji(Kajimoto Clinic),TakaoKamiya(KamiyaLadiesClinic),AtsushiKasamatsu (Department of Obstetrics and Gynecology, Kansai Medical UniversityHirakata Hospital),TatsuyaNakajima(Departmentof Obstetrics and Gynecology, Kansai Medical University Takii Hospital), KanjiKasahara (KasaharaClinic), KenjitsuKasamatsu (Kasamatsu Obstetrics and Gynecology/Pediatrics), Kawabata Ryoichi (Kawabata Lady’s Clinic), Kazume Kawabata (Kawabata Women’sClinic),Kozo Kadowaki(DepartmentofObstetricsand Gynecology,KawachiGeneralHospital),HiroshiNomura(Kawa- shima Ladies Clinic), Tomoyuki Kikuchi (KikuchiLadies Clinic), AyakoSuzuki (Departmentof Obstetricsand Gynecology, Kinki University), Tadayoshi Nagano (Department of Obstetrics and Gynecology,KitanoHospital),YoshitsuguKomeda(KomedaLadies Clinic), Ryousuke Kondo (Kondo Ladies Clinic), Shinjin Konishi (KonishiLadiesClinic),HideoTakemura(KosakaWomen’sHospi- tal), Masako Kasumi (Masako Ladies Clinic), Kazuo Masuhiro (MasuhiroMaternityClinic),RyojiIto(DepartmentofObstetrics and Gynecology, Matsushita Memorial Hospital), Yoshiki Saka- moto (Department of Obstetrics and Gynecology, Mimihara GeneralHospital),KouzoHirai(Minami-MorimachiLadiesClinic), YoshimitsuYamamoto(DepartmentofObstetricsandGynecology, Minoh City Hospital), Yoshitaka Kariya (Minoh Ladies Clinic), Osamu Misaki (Misaki Clinic), Akira Miyake (Miyake Clinic), YasukoOsako(MomWomen’sClinicOsako), MasaoMori(Mori ObstetricsandGynecologyClinic),KeizoNaka(NakaLadiesClinic), YasumasaTokura(NakaiClinic),JunYoshimatsu(Departmentof PerinatologyandGynecology,NationalCerebralandCardiovascu- larCenter),KeijiTatsumi(DepartmentofObstetricsandGynecol- ogy, National Hospital Organization Osaka National Hospital), Takayoshi Kanda (Department of Obstetrics and Gynecology, National Hospital Organization Osaka Minami Medical Center), MasahiroNishikawa (NishikawaLadies Clinic),Sekio Nishimoto (Nishimoto Ladies Clinic), Yoshihiro Nishioka (Nishioka Clinic), TakaoFunato(DepartmentofObstetricsandGynecology,Nissay Hospital), Kouichi Nozaki (Nozaki Ladies Clinic), Gengo Ohira (OhiraLadiesClinic),YoshiyukiOkamura(OkamuraLadiesClinic), YuzoOga(OgaClinic),OsamuNakamoto(DepartmentofObstetrics and Gynecology, Osaka City General Hospital), Shinichi Nakata (Department of Obstetrics and Gynecology, Osaka City Juso Hospital), Tetsuo Nakamura (Department of Obstetrics and Gynecology,OsakaCitySumiyoshiHospital),MasahikoTakemura (DepartmentofObstetricsandGynecology,OsakaGeneralMedical Center),ToshiyukiSadou(DepartmentofObstetricsandGynecol- ogy,OsakaGyoumeikanHospital),NobuakiMitsuda(Department of Obstetrics, Osaka Medical Centerand Research Institute for Maternal and Child Health), Daisuke Fujita (Department of ObstetricsandGynecology,OsakaMedicalCollege),KojiHisamoto (DepartmentofObstetricsandGynecology,OsakaPoliceHospital), ShinobuAkada(DepartmentofObstetricsandGynecology,Osaka PrefecturalMedicalCenterforRespiratoryandAllergicDiseases), TakafumiNonogaki,ChinamiHoriuchi(DepartmentofObstetrics and Gynecology, Osaka Red Cross Hospital), Yasuhiko Shiki (DepartmentofObstetricsandGynecology,OsakaRousaiHospi- tal),TadashiKimura(DepartmentofObstetricsandGynecology, OsakaUniversityGraduateSchoolofMedicine),KoutaroKitamura (ObstetricsandGynecology,PLHospital),KazuhideOgita(Depart- mentofObstetricsandGynecology,RinkuGeneralMedicalCenter),
