Respiratory syncytial virus infection in infants and correlation with meteorological factors and air pollutants.

0.56 MB·English

by  Vandini, Silvia

Vandinietal.ItalianJournalofPediatrics2013,39:1
http://www.ijponline.net/content/39/1/1 ITALIAN JOURNAL
OF PEDIATRICS
RESEARCH Open Access
Respiratory syncytial virus infection in infants and
correlation with meteorological factors and air
pollutants
Silvia Vandini*, Luigi Corvaglia, Rosina Alessandroni, Giulia Aquilano, Concetta Marsico, Marica Spinelli,
Marcello Lanari and Giacomo Faldella
Abstract
Background: Respiratory Syncytial Virus (RSV) is the most important cause of severe respiratory infections in infants
withseasonalepidemics. Environmental factors (temperature, humidity, air pollution) could influence RSV epidemics
through theireffects on virus activityand diffusion.
Methods: We conducted a retrospective study on a paediatric population who referredto our Paediatric
Emergency Unit inorder to analyze thecorrelationbetween weekly incidence ofRSV positive cases during winter
season inBologna and meteorological factors and air pollutants concentration.
Results: We observed a significant correlationbetween theincidence of RSVinfections and themeanminimum
temperature registered during thesame week and the previous weeks.
The weekly numberof RSVpositive cases was alsocorrelated to themean PM concentration of the week before.
10
Conclusions: RSVepidemic trend inBologna(Italy) is related to the mean minimum temperature, and themean
PM concentration.
10
Keywords: Respitatory syncytial virus, Bronchiolitis, Temperature, Humidity,Air pollution
Background Matter<10μm [PM ],ParticulateMatter<5μm [PM ],
10 2.5
Respiratory Syncytial Virus (RSV) is the most important Carbon Monoxide [CO], Nitric Oxide [NO ]) and respira-
2
cause of severe respiratory infections in infants. In tem- torytractinfectionsandRSVinfections[8,9].
perate climates, RSV infection typically has a seasonal To our knowledge no authors examined the correl-
trendandpeaksduringwinter.Someauthors[1-3]identi- ationbetweenRSVactivityandclimaticfactorsinItaly.
fied a correlation between RSV epidemic and climatic fac- Bologna is a city in the northern temperate area cold
tors(temperature,relativeandabsolutehumidityandUV-B and cloudy during winter; January is usually the coldest
radiation). Meerhoff et al [1]. analyzed RSV epidemics in month of the year and the temperature is higher in au-
differentwinterseasonsandobservedacorrelationbetween tumnthaninspring.
RSVactivity and relative humidity, minimum temperature Relative humidity presents less variable mean values
andcloudcoverintheNetherlands,withtheeffectofrela- throughdifferentmonths,rangingbetween65%registered
tive humidity being more consistent. Yusuf et al [2]. inJulyand84%inNovemberandDecember(meanvalues
analyzedRSVactivityinnine citieswithdifferent meteoro- of the last 30 years). Bologna has almost 400.000 inhabi-
logicalcharacteristicsandobservedapeakduringwinterin tantswithonlytwo hospitalswithaPaediatricEmergency
temperate climates, while in colder regions virus activity Unit.Anairpollutionmonitoringnetworkhasbeeninac-
was nearly continuous. Moreover, some authors [4-7] tive since many years in the city area. Approximately
reported a correlation between air pollutants (Particulare 20.000 children are admitted to the Pediatric Emergency
Unit of S.Orsola-Malpighi Hospital each year with ap-
proximately1600infantsyoungerthan2yearsreferringto
*Correspondence:[email protected]
Neonatology-S.Orsola-MalpighiHospital-UniversityofBologna,Via
Massarenti11,Bologna40138,Italy
©2013Vandinietal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative
CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and
reproductioninanymedium,providedtheoriginalworkisproperlycited.
Vandinietal.ItalianJournalofPediatrics2013,39:1 Page2of6
http://www.ijponline.net/content/39/1/1
Table1RSVactivityandmeteorologicalvariablesbywinterseasonfortheperiod2007-2010
Year RSVcases(n) Wkonset Wkpeak Wkoffset MinimumTemperature(°C) Relativehumidity(%)
Onset Peak Onset Peak
2007-08 125 46 1 16 1,5 −0,1 65 85
2008-09 74 45 7 16 11 −1,4 85 55
2009-10 128 1 8 16 0.5 5,1 93 83
the Paediatric Emergency Unit during the winter season Air pollution data (PM and PM mean weekly con-
10 2,5
foracuterespiratorytractinfections. centration) of the same area were recorded from Bologna
RSV epidemic in Italy presents a seasonal pattern with San Felice monitoring station by Bologna Department of
onset in November, peak in February or March and off- theRegionalEnvironmentalProtectionAgency(ARPA).
setinApril[10]. Ethics approval was not required for this observational
We examined the incidence of RSV infections in chil- andanonymousstudy.Consentwasobtainedforallpatients.
dren admitted to the Paediatric Emergency Unit of S. Statistical analysis was performed with Microsoft Excel
Orsola-Malpighi Hospital and we hypothesized a correl- 2010. Pearson’s correlation was used to correlate the
ationwithclimaticfactorsandairpollutantsconcentration weekly number of RSV positive cases with meteorological
recordedduringdifferentwinterseasonsinBologna. parameters (mean minimum temperature, mean relative
humidity)andairpollutants(PM andPM )meancon-
10 2,5
Methods centrations. A value of r>0,3 was considered statistically
Allpatientsyoungerthan2yearswhoreferredtoPaediatric significant.
EmergencyUnitofS.Orsola-MalpighiHospitalofBologna The number of RSV infections was also correlated to
(Italy) for suspected acute RSV infection during 3 con- the climatic parameters of the same week (no lag), or
secutive winter seasons from September to April (2007– one (1-week lag), two (2-week lags) or three (3-week lag)
2008,2008–2009,2009–2010)wereincluded. weeks before, aiming to identify a delay in the effect of
RSVinfectionwassuspectedinpresenceoffever,cough, temperatureorhumidityonRSVepidemics.
respiratorydistresswithtachypnea,cyanosis,feedingdiffi- Considering air pollutants, we correlate the weekly
culties; RSV rapid test was performed on nasopharyngeal number of RSV infections with the mean weekly concen-
secretions to all infants admitted to the unit presenting tration (μg/m3) of PM and PM measured in the same
10 2,5
thesesymptoms. week (no lag) and in the previous week (1-week lag) aim-
Every week we recorded the number of infants who ing to enlight the short-term effect of air pollutants
weretestedwithBINAXNOW,arapidimmunochroma- exposure.
tographic assay for the qualitative detection of RSV fu-
sionproteinantigen innasopharyngeal secretions. Results
Meteorological data of the geographic area of Bologna Meteorological parameters and RSV activity during the
(minimum temperature,°C; relative humidity, %) were three seasonsaresummarizedinTable1.
recordedfromBolognaBorgoPanigalemeteorologicalsta- Meanrelativehumidityandmeanminimumtemperature
tion;dataweresummarizedasmeanvaluesforeachweek. were similar during the three seasons; temperature was
Figure1RSVpositivedetectionsandmeanminimumtemperatureinthewinterseasons2007–2010inBologna.
Vandinietal.ItalianJournalofPediatrics2013,39:1 Page3of6
http://www.ijponline.net/content/39/1/1
Figure2RSVpositivedetectionsandmeanrelativehumidityinthewinterseasons2007–2010inBologna.
usually lower around RSV peak, while relative humidity Discussion
remained more stable than temperature during the whole RSV activity in Bologna presents the pattern described
season. in temperate areas, with onset and peak during winter
The peak occurred at a mean minimum temperature season.Meteorological conditions inthe 4weeksofpeak
range of −1,4-5,1°C and at a mean relative humidity of activity for each season were similar to data presented
55-85%. byotherauthors [1-3].
We considered as the onset of RSV activity the first These conditions are the mean minimum temperature
week RSV was detected; patients younger than 2 years 2-6°C and mean relative humidity 45-65% and are asso-
referring to Paediatric Emergency Unit with symptoms ciated tothe greatest activity ofthe virus [11-13].
compatible with RSV infection were tested all the year In2009–2010theonsetoccurredlaterthantheprevious
round but tests resulted positive only during the epi- years; a delay in RSV epidemics during the season 2009–
demicperiodssummarizedinTable 1. 2010hasbeenobservedinFrancebyCasalegnoetal[14].
The onset of RSV activity in the three seasons oc- These authors observed that RSV epidemic started more
curred at week 46, 45 and 1 respectively, with peak ac- gradually with a delayed peak and hypothesized that this
tivity atweek 1,7and8,asshown inFigures1and2. delay might be partially caused by pandemic H1N1 Influ-
Pearson’scorrelation showed anegative correlationbe- enza in 2009 through viral interference and increased hy-
tween the number of RSV positive cases and the mean gienemeasures.
minimum temperature (r >−0,3) at different time lags as A study performedinCroatia[15]presented abiennial
shown in Table 2; the correlation was not observed for cycle of RSV outbreaks with one large and one small
themeanrelativehumidity. season; the correlation between RSV epidemic and cli-
The correlation between RSV detections and air pollu- maticfactorswasobservedonlyfor thelarger season.
tants concentration in the different time lags is summar- Inthepresentstudywe cannotdemonstratetheeffects
ized in Table 3. Pearson’s correlation between number of of other respiratory viruses on RSV delayed peak since
RSV positive casesand airpollutants showed a significant weincludedonlyRSVdetectionwhichcanbeperformed
correlation (r=0,3) between RSV infections and PM in a few minutes at Paediatric Emergency Unit to all
10
meanconcentrationoftheweekbefore(Figure3)asignifi- infantswithrespiratorysymptoms.
cantcorrelationbetweenRSVinfectionsandPM2.5mean Moreover, a study period including several additional
concentrationwasnotfound(Figure4). winter seasons would probably be warranted in order to
Table2Pearson’scorrelationbetweennumberofRSVdetectionsandmeanminimumtemperatureandmeanrelative
humidityfordifferenttimelags
Meanminimumtemperature r pvalue Meanrelativehumidity r pvalue
Nolag −0,3 0,012 Nolag −0,05 0,68
1weeklag −0,39 0,001 1weeklag −0,007 0,47
2weeklag −0,43 0,0002 2weeklag 0,13 0,14
3weeklag −0,53 <0,0001 3weeklag 0,17 0,08
Vandinietal.ItalianJournalofPediatrics2013,39:1 Page4of6
http://www.ijponline.net/content/39/1/1
Table3Pearson’scorrelationbetweennumberofRSVdetectionsandmeanconcentrationofairpollutantsPM
10
andPM fordifferenttimelags
2,5
MeanPM (μg/m3) r pvalue MeanPM (μg/m3) r pvalue
10 2,5
Nolag 0,2 0.051 Nolag 0,14 0,25
1weeklag 0,34 0,004 1weeklag 0,26 0,03
Figure3RSVpositivedetectionsandmeanPM concentration(μg/m3)inthewinterseasons2007–2010inBologna.
10
Figure4RSVpositivedetectionsandmeanPM concentration(μg/m3)inthewinterseasons2007–2010inBologna.
2.5
Vandinietal.ItalianJournalofPediatrics2013,39:1 Page5of6
http://www.ijponline.net/content/39/1/1
determine the overall patterns of RSV epidemic onset in Conclusions
Bologna and Emilia-Romagna and to confirm whether a Environmental data collected in Bologna during 3 winter
biennial cycle of RSV outbreak is detectable in our geo- seasons confirm a correlation between RSVseasonality and
graphic area. meanminimumtemperatureandmeanPM concentration.
10
RSVactivity is greater at cold temperature because the
virus is more stable in secretions by which it is transmit- Abbreviations
RSV:RespiratorySyncytialVirus;PM :ParticulateMattermeasuring10μmor
ted. Moreover, cold temperature might drive populations less;PM :FineParticulateMatterw10ithadiametersmallerthan2.5μmor
2,5
indoor whereRSVspreadsmore readily. less.
The correlation between mean minimum temperature
Competinginterests
and RSV epidemic activity persists in the three different
Allauthorsdeclaretheydonothaveanyfinancialandnon-financial
time lags. We hypothesize that a decrease in the mean competinginterest.
minimum temperature may determine a greater RSV ac-
Authors’contributions
tivity in the following weeks associated with more pro-
MLparticipatedinthedesignofthestudyandreviewedthemanuscript;GF
longed permanence indoor, sometimes in overcrowded
coordinatedthestudyandcollaboratedtothedraftofthemanuscript;RA,
rooms; these conditions may determine an increase in GA,MSandCMcollecteddataaboutRSVinfections;LCperformedstatistical
RSV diffusion, which is followed by viral incubation (2–7 analysis;SVdraftedthemanuscriptandpreparedthedatabase.Allauthors
readandapprovedthefinalmanuscript.
days)andleadtoanincreaseinnumberofRSVinfection.
The weak relationship between RSV epidemics and Received:19July2012Accepted:28December2012
Published:11January2013
relativehumiditycanbeexplainedbythesmallvariability
andtheirregularweeklypatternofthisparameterduring
References
the whole winter season in Bologna. This result is differ- 1. MeerhoffTJ,PagetJW,KimpenJL,SchellevisF:Variationofrespiratory
ent from data collected in other countries such as the syncytialvirusandtherelationwithmeteorologicalfactorsindifferent
winterseasons.PediatrInfectDisJ2009,28(10):860–866.
Netherlands [1], where relative humidity has greater sea-
2. YusufS,PiedimonteG,AuaisA,DemmlerG,KrishnanS,VanCaeseeleP,
sonal fluctuations and the effect of relative humidity SingletonR,BroorS,ParveenS,AvendanoL,ParraJ,Chavez-BuenoS,
is more consistent. We also analyzed the correlation be- MurguíaDeSierraT,SimoesEA,ShahaS,WelliverR:Therelationshipof
meteorologicalconditionstotheepidemicactivityofrespiratory
tween RSV detection and air pollutants concentration. syncytialvirus.EpidemiolInfect2007,135(7):1077–1090.
Many authors [5-7,9] showed a positive correlation be- 3. NoyolaDE,MandevillePB:Effectofclimatologicalfactorsonrespiratory
tween fine particulate matter (PM and PM ) and mor- syncytialvirusepidemics.EpidemiolInfect2008,136(10):1328–1332.
10 2,5
4. OrazzoF,NespoliL,ItoK,TassinariD,GiardinaD,FunisM,CecchiA,Trapani
bidity for respiratory infections and other respiratory
C,ForgeschiG,VigniniM,NosettiL,PignaS,ZanobettiA:Airpollution,
conditions such as asthma and chronic obstructive pul- aeroallergens,andemergencyroomvisitsforacuterespiratorydiseases
monarydisease.Thiscorrelationisexplainedbyincreased andgastroentericdisordersamongyoungchildreninsixitaliancities.
EnvironHealthPerspect.2009,117(11):1780–1785.
respiratory symptoms, reduced lung function and bron-
5. BarnettAG,WilliamsGM,SchwartzJ,NellerAH,BestTL,PetroeschevskyAL,
chialreactivityrelatedtoairpollutionexposure. SimpsonRW:Airpollutionandchildrespiratoryhealth.AmJRespCrit
Moreover, these effects are increased in paediatric Care2005,171:1272–1278.
6. BedeschiE,CampariC,CandelaS,ColliniG,CaranciN,FrascaG,GalassiC,
population, especially young infants, because of their
FrancescaG,VigottiMA:Urbanairpollutionandrespiratoryemergency
higher respiratory rate that increases air pollutants per visitsatpediatricunit,ReggioEmilia,Italy.JToxicolEnvironHeal2007,
kilogram of bodyweightexposure. 70:261–265.
7. LinM,StiebDM,ChenY:Coarseparticulatematterandhospitalizationfor
ThecorrelationwithRSVinfection[8,9]wasfoundboth
respiratoryinfectionsinchildrenyoungerthan15yearsinToronto:
for acute exposure (one week) and for subchronic and acasecrossoveranalysis.Pediatrics2005,116:e235–e240.
chronic exposure (30–60 days). We observed a similar 8. KarrC,LumleyY,SheperdK,DavisR,LarsonT,RitzB,KaufmanJ:Acase-
crossoverstudyofwintertimeambientairpollutionandinfant
correlation in our study between the weekly number of bronchiolitis.EnvironHealthPerspect2006,114:277–281.
RSV detections and the mean particulate matter concen- 9. KarrCJ,RudraCB,MillerKA,GouldTR,LarsonT,SathyanarayanaS,Koenig
tration of the week before. We examined only acute air JQ:Infantexposuretofineparticulatematterandtrafficandrisk
hospitalizationforRSVbronchiolitisinaregionwithlowerambientair
pollution exposure (considering PM10 and PM2,5 concen- pollution.EnvironRes2009,109:321–327.
tration in the same week or in the week before) and not 10. RossiGA,MediciMC,MerollaR;OsservatorioVRSStudyGroup:Incidenceof
subchronic and chronic exposure because we considered respiratorysyncytialviruspositivityinyoungItalianchildrenreferredto
theemergencydepartmentsforlowerrespiratorytractinfectionover
the whole population who referred to our Paediatric twoconsecutiveepidemicseasons.Infection2005,33(1):18–24.
Emergency Unit without considering differences in their 11. WelliverRCSr:Temperature,humidity,andultravioletBradiationpredict
residential addressand thedistancefrom theairpollution communityrespiratorysyncytialvirusactivity.PediatrInfectDisJ2007,
26(11Suppl):S29–S35.
monitoringstation.
12. WelliverR:Therelationshipofmeteorologicalconditionstotheepidemic
Furthermore, the reduction of air pollution in urban activityofrespiratorysyncytialvirus.PaediatrRespirRev2009,
areas could lead to an improvement in infants morbidity 10(Suppl1):6–8.
13. duPrelJB,PuppeW,GröndahlB,KnufM,WeiglJA,SchaaffF,SchmittHJ:
asitcoulddetermineareductioninrespiratorysymptoms
Aremeteorologicalparametersassociatedwithacuterespiratorytract
andindividualsusceptibilitytorespiratoryinfection. infections?ClinInfectDis2009,49(6):861–868.
Vandinietal.ItalianJournalofPediatrics2013,39:1 Page6of6
http://www.ijponline.net/content/39/1/1
14. CasalegnoJS,OttmannM,Bouscambert-DuchampM,ValetteM,MorfinF,
LinaB:Impactofthe2009influenzaa(H1N1)pandemicwaveonthe
patternofhibernalrespiratoryvirusepidemics,France,2009.EuroSurveill
2010,15(6).
15. Mlinaric-GalinovicG,WelliverRC,Vilibic-CavlekT,Ljubin-SternakS,
DrazenovicV,GalinovicI,TomicV:Thebiennialcycleofrespiratory
syncytialvirusoutbreaksincroatia.VirolJ2008Jan28,5:18.
doi:10.1186/1824-7288-39-1
Citethisarticleas:Vandinietal.:Respiratorysyncytialvirusinfectionin
infantsandcorrelationwithmeteorologicalfactorsandairpollutants.
ItalianJournalofPediatrics201339:1.
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
www.biomedcentral.com/submit