Moreover, otitis media frequently marks the starting point for surgical intervention [74,213]
.Since38–60ofyoungchildrenareexposedtoETSintheirparents’homes, passivesmokingisanimportantfactorinthedevelopmentofthiscondition
[218–221] .
Thefrequencyofotitismedia,surgeryandantimicrobialtherapyduringthepreceding12 monthswasassessedhaircotininemeasurements,physicianmedicalrecords,homevisits
inastudyin227casesand398healthycontrolswhosatisiedtheinclusioncriteriaforthe studyFig.
9.10
[212] .Otitismediadevelopedin23.9ofthechildrenand9.8hadto
undergo myringotomy. Children exposed to ETS had a 2–3 times higher risk of otitis media than non-exposed children Table
9.8 . These results are consistent with those
reportedinotherstudies [222–225]
OR,1.80;CI,1.1–3.0 [222]
. To estimate the relative risk for otitis media OM in children from ETS, maternal
smoking during pregnancy gestational exposure, or both, a national cross-sectional healthsurveywasanalysedbytheuseofquestionnaireinformationandserumcotinine
measurements [226]
.Childrenyoungerthan12yearsN=11,728wereexaminedwho participatedintheThirdNationalHealthandNutritionExaminationSurveyNHANESIII,
conductedfrom1988to1994.Theoutcomeofthestudywastheoccurrenceandrecur- renceofearinfections.Itwasfoundthatthecumulativeincidenceofearinfectionswas69.
Ofallparticipants,38wereexposedtopassivesmoke,23wereexposedtogestational smoke,and19wereexposedtocombinedpassiveandgestationalsmoke.Theoccurrence
ofanyearinfectionwasnotincreasedbypassivesmokeexposureadjustedriskratio[RR], 1.01;95CI,0.95–1.06,butwasslightlyincreasedbygestationaladjustedRR,1.08;95
CI,1.01–1.14andcombinedadjustedRR,1.07;95CI,1.00–1.14smokeexposures. Theriskofrecurrentearinfections³6lifetimeepisodeswassigniicantlyincreasedwith
combinedsmokeexposureadjustedRR,1.44;95CI,1.11–1.81.Otherriskfactorsfor earinfectionidentiiedinmultivariableanalysiswereraceethnicity,poverty–incomeratio
of2.00ormore,attendanceindaycare,historyofasthmaandpresenceofallergicsymp- toms.Fromthisstudy,itcanbeconcludedthatpassivesmokeexposurewasnotassociated
withanincreasedriskofeverdevelopinganearinfection.Theincreasedriskfoundwith gestationalandcombinedsmokeexposureshasmarginalclinicalsigniicance.Forrecur-
rentearinfections,however,combinedsmokeexposurehadaclinicallyandstatistically signiicanteffect
[226] .
Continine Level [ng mg] No. of Current Household Smokers
1 0.60
0.48 0.30
n = 74 n = 13
n = 5
≥2 0.5
1.0 1.5
2.0
Fig. 9.10
Correlationbetween haircotinineconcentration
andnumberofsmokersin household,accordingto
parentalself-report.Medians including25thand75th
percentiles.Dataoutsidethe conidencerange.Signiicant
differencesbetweenthe groupsp=0.03
[212]
9.8 Meningococcal Infections
SepsisandmeningitiscausedbyNeisseriameningitidisfrequentlyhaveafataloutcome, andtobaccosmokeshouldberegardedasavehicleforthenasopharyngealtransportofthe
bacteria [227–229]
.Inparticular,NO
2
fromtobaccosmokeincreasestheincidenceofviral respiratorytractinfections
[230] ;byimpairingdefencemechanisms,thegasencourages
viralspreadonthemucosa.Cigarettesmokeinhibitsmucociliaryclearance,increasesbacte- rialadhesionandrupturesrespiratorytractepithelium
[231–233] .Invitroexperimentsindi-
catethattobaccosmokeharmsneutrophilmigration,andinhibitsthephagocyticactivityof macrophagesandtheproductionofimmunoglobulins
[231,232,234] .
PassivesmokersaremorelikelythanindividualsnotexposedtoETStobeinfectedby pathogenicmeningococci.Alongsidenumerousothercauses
[235] ,ETSisanimportant
factorinpromotingmeningococcalinfectioninyoungchildren [236–238]
.Inacompari- sonof129caseswith274controlsmatchedforage,sex,ethnicgroupetc.,havingamother
whosmokedwasthestrongestindependentriskfactorforinvasivemeningococcaldisease inchildren18yearsofage,comparedwithnon-smokingmothersOR,3.8;CI,1.6–8.9;
p0.01.Thisriskwasadditionallyincreasedwherethechildrenwereinschoolclasses of30pupilsOR,5.7;CI,1.3–24.2;p=0.02.Similarly,theriskofmeningococcalinfec-
tionshowedadose–responserelationshipwithincreasingnumberofmaternalpack-years ofsmoking.Ultimately,accordingtothisstudytheincidenceofmeningococcalinfection
inETS-exposedchildrenwas37higherthaninnon-exposedchildren.Youngerchildren 5yearsoldwereatgreaterriskthanolderchildren5–17yearsold
[235] .
Inviewoftheriskofmeningococcalinfection,frequentlyafatalcondition,itisthere- forerecommendedthatmotherswithyoungchildrenshouldgiveupsmoking.
9.9 Breast Cancer
Althoughacausalassociationbetweenpassivesmokingandanincreasedriskofbreast cancerisnotimmediatelyevident
[239] seeSect.7.8.4in
Chap.7 ,initialindingsindicate
thatanassociationdoesexist.Evidently,ingenetically“sensitive”individuals [240–242]
theinhaledETSproduceshormonalchangesconsistentwithananti-oestrogeneffectand
oestrogen-inducedmitogenesis [243,244]
.Accordingtoastudyconductedin334women withbreastcancer,ever-activesmokershadanORof2.0CI:1.1–3.6whencompared
withnever-active,never-passivesmokers.Risklevelswerehigherinwomenwhosmoked onlybeforetheirirstpregnancyOR,5.6;CI,1.5–21andinwomenwhoquitsmoking
5–15yearsbeforetheirindexyearOR,3.9;CI,1.4–10.Passive-onlysmokershadanOR of2.0CI:1.1–3.7.AmongthosewomenwhowereexposedtoETSbeforetheageof
12years,theORswere4.5CI:1.2–16forpassive-onlysmokers,and7.5CI:1.6–36for ever-activesmokers
[245] .ItispossiblethatETSexposureneedstooccurataveryearly
stage,shortlybeforethedevelopmentofbreasttissue,inordertoinducemitogenicchanges consistentwithcarcinogenesis
[246] .Moreextensivestudiesarecertainlyrequiredinthis
areabeforeadeinitiveverdictcanbegiven. JohnsonandGlantzcomparedthestrengthofevidencefromepidemiologicstudiesof
secondhandsmokeoftheUSSurgeonGeneral’s1986conclusionthatsecondhandsmoke caused lung cancer with the California Environmental Protection Agency’s CalEPA
similar2005conclusiononbreastcancerinyounger,primarilypremenopausalwomen [247]
.Theyreviewedeachreportforcriteriausedtoassesscausality:numberofstudies, statisticallysigniicantincreasesinrisk,andpooledsummaryriskestimates.Theauthors
showedthatboththeSurgeonGeneralandCalEPAusedupdatedBradfordHillcriteriafor assessingcausalityandfoundthattheevidencemetthosecriteria.Sixof13lungcancer
studies46hadstatisticallysigniicantincreasesoneofthreecohortstudies.Pooled risk estimates for lung cancer for spousal exposure were 1.53 for ten combined case-
controlstudiesand1.88forsevenstudieswithdose–responseresults.TheCalEPAreported 10of14studies71hadstatisticallysigniicantincreasesinbreastcancerrisktwoof
fourcohortstudies.Pooledrelativeriskestimatesforyounger,primarilypremenopausal womenwere1.6895CI:1.33,2.12forallexposedwomenand2.1995CI:1.68,
2.84forivestudieswithbetterexposureassessment.Itwasconcludedthattheevidence fromepidemiologicstudiesofpassivesmokein2005forbreastcancerinyounger,primar-
ilypremenopausalwomenwasstrongerthanforlungcancerin1986.
9.10 Psychosocial Changes
ChildrenexposedtoETSdisplaybehaviouralchangesthatcanbelinkedwithETS.Withina family setting, lifestyle habits evolve e.g. smoking, fat consumption, sedentary lifestyle
increasedalcoholconsumptionwhichinturnmayberegardedasriskfactorsforthechildren becausetheyadopttheselifestylehabits
[248,249] .Bodymeasurementswereperformedin
astudyof804childrenaged10–12years;additionally,smokingbehaviouranddietarychar- acteristicswereidentiied.Amongthechildrenfromthesefamilies,19ofboysand10of
girlswerealreadysmokers.Amongthechildrenwhosmoked,57boysand68girls camefromfamiliesinwhichatleastoneparentsmokedOR,2.1;CI,1.2–3.8.Inthelonger
term,however,girlswerelesslikelytoadoptthesmokinghabitsoftheirparentsOR,0.4; CI,0.2–0.6.Parentalsmokingbehaviourwasanadditivepredictorinchildrenoflower
physicalactivityandmoretelevisionwatching,regardlessofwhichparentsmoked.Children’s fatintakewassigniicantlygreaterifeitherparentsmoked,whereaschildren’sbodymass
indexandwaist-to-hipratioweresigniicantlygreaterifmotherssmoked [250]
. Overall,childrenwhogrowupin“smokerhouseholds”takehealthrisksuponthem-
selves regardless of the risks already present as a result of smoking during pregnancy [251]
.Parentalsmokingisformativeinencouraginganunhealthylifestyleinchildren,