1-s2.0-S0264410X14000085-main.pdf

Vaccine 32 (2014) 27402747

Contents lists available at ScienceDirect

Vaccine

j o ur na l ho me page: www.elsev ier .com

Effectiveness of rotavirus vaccine against hospitadiarrhea: A casecontrol study

Maria Y.T. Ichiharaa,, Laura C. Rodriguesb, Carlos A.S. Teles SaMaria da Gloria L.C. Teixeiraa, Sandra R. De Jesusd, Sheila M. AJose P. Gagliardi Leitee, Mauricio L. Barretoa

a Institute of Co trio dBrazilb Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, Keppel Street, WC1H 7HT London, United Kingdomc Department of Exact Sciences, University of the State of Feira de Santana, Avenida Transnordestina, Novo Horizonte, CEP 44036-900, Feira de Santana,Bahia, Brazild Institute Multidisciplinar of Health, Campus Ansio Teixeira, Federal University of Bahia, Vitoria da Conquista, Bahia, Brazile Avenida Brasi

a r t i c l

Article history:Received 20 SeReceived in re13 December Accepted 2 JanAvailable onlin

Keywords:Rotavirus diarVaccine effectAcute diarrheaChild diarrheaVaccineHospitalizatio

Corresponsitrio do CaneTel.: +55 71 32

E-mail addlaura.rodrigue(C.A.S. Teles Sa(S.R. De Jesus)jpgleite@gmai

0264-410X http://dx.doi.ol, 4365-Pav. Hlio & Peggy Pereira, CEP 21040-360, Rio de Janeiro, Manguinhos, Brazil

e i n f o

ptember 2013vised form2013uary 2014e 5 February 2014

rheaiveness

hospitalization

n

a b s t r a c t

Rotavirus is one of the leading cause of hospitalization and outpatients visits among children underve years. This study evaluated overall and genotype-specic vaccine effectiveness of oral monovalentrotavirus vaccine (G1P[8] strain) in preventing hospital admission of Brazilian children with rotavirusacute diarrhea.

A hospital based casecontrol study was conducted in ve Regions of Brazil using the National RotavirusAcute Diarrhea Surveillance System from July 2008 to August 2011. A total of 215 cases (aged 424months) admitted with conrmed rotavirus diarrhea were recruited and 1961 controls hospitalizedwithout diarrhea were frequency matched by sex and age group to cases.

Two-dose adjusted vaccine effectiveness (adjusted by year of birth and the frequency matching vari-ables) was 76% (95%CI: 5886) lasting for two years. Effectiveness controlled by the available potentialconfounders was 72% (95%CI: 4485), suggesting no appreciable confounding by those factors for whichadjustment was made. In a half of the cases the rotavirus genotype was G2P[4] and in 15% G1P[8].Genotype-specic VE (two doses) was 89% (95%CI: 7895), for G1P[8] and 76% (95%CI: 6484) for G2P[4].For all G1, it was 74% (95%CI: 3590), for all G2, 76% (95%CI: 6384), and for all non G1/G2 genotypes,63% (95%CI: 2799). Effectiveness for one dose was 62% (95%CI: 3997).

Effectiveness of two-dose monovalent rotavirus vaccine in preventing hospital admission withrotavirus diarrhea was high, lasted for two years and it was similar against both G1P[8] and G2P[4]. Basedon the ndings of the study we recommend the continued use of rotavirus in the Brazilian National Immu-nization Program and the monitoring of the early emergence of unusual and novel rotavirus genotypes.

2014 The Authors. Published by Elsevier Ltd.

ding author at: Rua Baslio da Gama, s/n, 4 . Andar, Campus Univer-la, Canela CEP 40110-040, Salvador, Bahia, Brazil.837445; fax: +55 71 33360695; mobile: +55 71 88453456.resses: [email protected] (M.Y.T. Ichihara),[email protected] (L.C. Rodrigues), [email protected]), [email protected] (M.d.G.L.C. Teixeira), [email protected], [email protected] (S.M. Alvim De Matos),l.com (J.P. Gagliardi Leite), [email protected] (M.L. Barreto).

1. Introduction

Acute diarrhea (AD) is a frequent cause of child hospitalizationand outpatient visits in children under 5 years [1]. In Brazil, beforeintroduction of the rotavirus vaccine in 2006, about 120.000 hos-pitalizations a year occurred due to AD in children under ve years(DATASUS/Ministry of Health of Brazil, 2006).

Rotavirus is the leading cause of severe acute diarrhea in chil-dren in developed and in developing countries and is the majorcause of death in poor countries [2,3]. Seven groups of rotavirushave been identied (A to G) and group A (RV-A) is responsible formore than 90% of human rotavirus infections [4]. RV-A has greatgenetic diversity due almost 60 serotypes (G and P) and the most

2014 The Authors. Published by Elsevier Ltd. rg/10.1016/j.vaccine.2014.01.007

Open access under CC BY-NC-SA license.

Open access under CC BY-NC-SA license.llective Health, Federal University of Bahia, Rua Baslio da Gama, s/n, Campus Universi/ locate /vacc ine

lized rotavirus

ntosc,lvim De Matosa,

o Canela, CEP 40110-040, Salvador, Bahia,

M.Y.T. Ichihara et al. / Vaccine 32 (2014) 27402747 2741

common strains are: G1P[8], G2P[4], G3P[8], G4P[8] and G9P[8] [5].In Brazil, between 12% and 42% of children under 5 years with diar-rhea had poof the RV-Ahospitalizedwere reporG1P[8], G9P

Vaccinatand its adopzation [10].Rotarix) aG4 and P[8was introdu(BNIP) in 20co-adminisvaccines.

RV-A vamore than 9Africa to 49ried out in middle-incotwo-dose vand a one-against hostrols studiehospitalizattion in vacto antigen enteric pathtion among

The introtion in chilSalvador antion in morand in Mex

This stumonovalenBrazilian choverall andnation (up t

2. Method

2.1. Study d

This wamatched bywhich receia similar geenrolled in to participaRV-A positithe unit anon logistica(SVS/MS), tthe states, tratories, 10capital citie

3. Particip

3.1. Eligible

Childrenadmitted i(and therefrotavirus vbefore adm

were listed and screened to exclude children who had any healthcondition presumed to reduce vaccine effectiveness (immunode-

, gastplasmnd syeive

ion, ot full wer

thaiseaivers

tenti

usiod as ion)sitivcludciatusioals otal, nd prls whea t reces).pote

selels hasex a24 mol ofistri

in wncy mon ofial coersio

fectiv

es: Apositncludtrols

sel stoof theanspcated

of P/SVSEIA)ndati

bora

posit a reis w

Inst Lut

Jande gesitive stool samples for RV-A before the introduction vaccine. This increased from 22% to 38% in children

for AD [6,7]. More than 51 genotype combinationsted and the most common genotypes described were[8] and G2P[4] [8].ion is the better measure to prevent rotavirus [1,2,9]tion has been recommended by World Health Organi-

An attenuated monovalent human RV-A (G1P[8] strain;nd a pentavalent bovine-human reassortant (G1,G2,G3,] strains; RotaTeq) are licensed worldwide. Rotarix

ced in the Brazilian National Immunization Program06 in a two-dose schedule at 2 and 4 months of age andtered with tetravalent, pneumococcal and poliovirus

ccine efcacy against severe RV-A AD varied between0% Europe and Asia, 85% in Latin America, 72% in South% in Malawi [1114]. Three casecontrol studies car-a high income country (Belgium) [15] and in low tome countries (El Salvador and Bolivia) [16,17] found aaccine effectiveness of 90%; 76% and 77% respectivelydose effectiveness of 91%; 51% and 56% respectivelypitalization by RV-A AD. In Brazil, two small case con-s showed a range of 4085% effectiveness in preventingion caused by G2P[4] [18,19]. The reason for varia-cine protection is not clear and has been attributeddiversity, malnutrition and higher incidence of otherogens [20]. There is strong suggestion of cross protec-

genotypes [1114].duction of RV-A vaccination was followed by a reduc-

d hospitalization due to all causes of AD in Brazil, Eld Mexico ranging from 17 to 51% [2123] and a reduc-tality from AD in children under 5 years in Brazil of 22%ico of 41% [24].dy will evaluate the overall effectiveness of the oralt vaccine, used in routine health services, in preventingild hospitalization with RV-A AD. It will also evaluate

genotype-specic VE by time since second dose vacci-o two years), and genotype-specic VE.

s

esign

s a hospital based casecontrol study, frequency- sex and age group. Hospitals were general hospitalsved children with a large range of diseases coming fromographical catchment area. Seventeen of the hospitalsthe RV-A AD National Surveillance System were invitedte in the study, based on having had a large number ofve samples in 2007, adequate level of organization ofd data accessibility. After consultation and agreementl arrangements with the Federal Health Surveillancehe epidemiological surveillance of the hospitals and ofhe Central Public Health and National Reference Labo-

hospitals located in ve macro-regions of Brazil (6 states and 4 municipalities) were selected.

ants

children

were eligible to participate in the study if they weren the study hospitals, were aged 4 to 24 monthsore old enough to have received their second dose ofaccine) and did not have diarrhea up to three weeksission or during hospitalization. All eligible children

ciencyor neosigns ahad rectalizatAll thacontroulationother dthe un

3.2. Po

Incl(deneadmissand powere inno asso

InclhospitloskelemastoiControA diarrdid novaccin

All furthercontroables (and 12the posame duationfrequeselectipotentStata v

3.3. Ef

Casstools were i

Conrandom

Onepart oand trwas lonation(CGLABAssay (omme

3.4. La

All sent totion, thChagas(Adolfo[Rio derylamirointestinal disease (e.g. diverticulitis), malformations conditions related to vaccine effectiveness, generalmptoms, infectious and parasitic diseases), those whod the second dose of vaccine in the 15 days before hospi-r whose vaccination did not follow the BNIP schedule.lled the specic criteria for either effectives case ore included. This aimed to select controls from the pop-t produced the cases, as cases hospitalized by AD or byses were likely to come from the same population givenal health care system in Brazil.

al cases and controls

n criteria for potential cases were: admission with ADthree or more liquid stools in 24 h, up to 14 days before, stool sample was collected until 48 h after admissione for RV-A and stay in hospital for at least 24 h. Childrened in the study in the rst hospitalization only and hade disease.n criteria for controls were: admission from the samef the cases with respiratory, genitourinary, muscu-ervous systems, skin and subcutaneous tissue, ear and

ocesses, eye and adnexa diseases, and external causes.ere not included if they had a previous history of RV-or had a vaccine-preventable disease (as children whoeive one vaccine are more likely to not receive other

ntial controls fullling the criteria above undergone action for frequency matching, so that the all effectived the same distribution of the main confounding vari-nd age group on admission: 46 months; 711 monthsonths) as the cases. This approach aimed to select from

potential controls, an effective control group with thebution of confounders as the effective cases; in the sit-hich more controls than needed were available in theatched groups they were selected at random. Random

frequency matched effective controls from the pool ofntrols was done using the sample command of then 11.0

e cases and controls

ll potential cases fullling the criteria above and hadive for rotavirus conrmed by the reference laboratoryed.: All potential controls fullling the criteria above andected for frequency matching were included.l sample was collected up to 48 h after admission as

RV-A AD Surveillance System. Samples were storedorted to the LACENs of each State where the hospital, according to the guidelines of the General Coordi-

ublic Health Laboratories/Ministry of Health of Brazil/MS). RV-A investigation was done by Enzyme Immune

, using commercial kits, following the manufacture rec-on (Dako or Oxoide).

tory investigation of potential cases

ive samples for RV-A and 25% of negative samples wereference laboratory. According to the LACEN localiza-as either the National Reference Laboratory (Evandroitute [Belm, PA], or a Regional Reference Laboratoryz Institute [So Paulo, SP], and Oswaldo Cruz Instituteeiro, RJ]). Results were conrmed by EIA and polyac-l electrophoresis (PAGE) according to Leite et al. [25].

2742 M.Y.T. Ichihara et al. / Vaccine 32 (2014) 27402747

Fecal suspensions and nucleic acids extraction were carried outaccording to Leite et al. [25] and Boom et al. [26], respectively.The RV-Genotyping was conducted using RT-PCR as described byDas et al. [27] (G genotype) and Gentsch et al. [28] (P geno-types). RV-Ato the Insti(ISC/UFBa).

4. Data col

Informaers who visMedical recstandard qulution, socistatus of thcination stacard, askedhome visitsdence of thepresence anpitalizationand its dethe Nationa[29]. Mothe

QuestionISC/UFBa ancontrols by

To compbirth we coThis systemadmission breastfeedision. Casesaccording t1224 mon

5. Sample

The mincases and 8condence cases and 1vaccine effe

6. Statistic

Vaccine tional logismatching iage both uscoverage ofJackknife, wincluded intion was


8. Vaccine

The twofrequency mFig. 1. Study population.

Fig. 2. Genotypes circulating in Brazil from 20

effectiveness

-dose adjusted VE (adjusted for year of birth and theatching variables) was 76% (95%CI: 5886) (Table 1).

Effectivenewas very sble confoumade.08 to 2011.

ss controlled by the available potential confoundersimilar (72%, 95%CI: 4485), suggesting no apprecia-nding by those factors for which adjustment was


Table 1Effectiveness of oral monovalent rotavirus vaccine in preventing hospital admission in Brazilian children with rotavirus acute diarrhea by number of doses received.

Vaccination by Case Control ORa VEa ORadjb VEb

Number of doses n n (95%CI) (95%CI) (95%CI) (95%CI)

Unvaccinated 67 201 1 1 Fully vaccinated (two doses) 115 1481 0.24 (0.140.42) 76 (5886) 0.28 (0.150.56) 72 (4485)Partially vaccinated (one dose) 33 279 0.38 (0.230.61) 62 (3977) 0.40 (0.250.63) 60 (3775)AICc 1318.253 1194.454

a Odds ratio adjusted by year of birth and variables (sex and age group) used for frequency matching, and robust variance estimation of Jaccknife, where the clusters werethe hospitals. The vaccine effectiveness (VE) was calculated by (1 OR) 100%.

b Odds ratio adjusted by year of birth and variables (sex and age group) of frequency matching and confounders (mothers schooling, mothers absence from home, smokingduring pregnancy, type and regularity of water supply, number of AD hospitalizations before current admission and exclusive breastfeeding) and robust variance estimationof Jaccknife, where the clusters were the hospitals.cAkaike information criteria for measuring the goodness of t of the statistic model.

We excluded a similar proportion of cases (5.7%) and controls(5.3%) because they did not have cards. Sensitivity analysis showedthat if they were included as vaccinated, VE (two doses) would be66% (95%CI: 4280) and if included as unvaccinated VE would be74% (95%CI: 5386).

The VE (adjusted for year of birth and the frequency matchingvariables) for one dose was 62% (95%CI: 3997) and one dose VEadjusted for other potential confounders was 60% (95%CI: 3775).Table 2 shows that VE was similar in those with time since sec-ond dose vaccination until hospitalization stratied by one year(71%; 95%CI: 5482) and two years (78%; 95%CI: 5290). The VEfor G1P[8] and G2P[4] by time since second dose vaccination wasmarginally higher for G1P[8](90%; 95%CI:-0.92-100 for one yearand 89%; 955788 for o

Table 3 (two doses)for G2P [4];all G2 and 6

Estimateed by yeaincreasing v

9. Discussi

Two-dossity of rotavof G2P[4] ggenotypes

lasted for two years after second dose vaccination and it was higherfor G1P[8] than G2P[4].

Variation of RV-A vaccine efcacy and effectiveness have beenreported in the literature: efcacy was higher in Europe (96.4%against RV-A severe AD) [11] than in a low income country (Malawi,49.2% against all diarrhea and 57.5% against hospitalized diarrhea)[13] and in countries with high mortality (63%) [33]. In the middleincome countries of Latin America [12], efcacy was 84.8% againstsevere AD; in South Africa it was 72.2% against all diarrhea [13]. Thisstudy showed similar effectiveness to that found in El Salvador [16]and Bolivia [17] (73% and 76% for severe diarrhea) and in a smallerstudy in Brazil [18] (75.8% against hospitalized diarrhea), but lowerthan in Belgium (90%) [15].

o-dosies wigh mation

theith

veneloreies et sevld beseconr thect the couD anhelp

Table 2Effectiveness n chistratied by ti

Time since tSecond dose

OverallUnvaccina


Table 3Genotype-specic vaccine effectiveness of oral monovalent rotavirus vaccine by number of doses received.

Vaccination by Case Control ORa VEa

Rotavirus genotype n n (95%CI) (95%CI)

G1P[8]UnvaccinaFully vaccPartially v

G2P[4]UnvaccinaFully vaccPartially v

G1UnvaccinaFully vaccPartially v

G2UnvaccinaFully vaccPartially v

Non G1/G2UnvaccinaFully vaccPartially v

a Odds ratio ching,the hospitals.

b Akaike inf

A singleerature (altBolivia [17]

The gootion in the rfollowing th

Genotyplower for tion. Anima(RVA) presRVA infectitective immdevelopmeepitope-blothe humanvalent, Rotapresented different RV[12,19,36,3

Genotypcontrast wifor all G1 tytype (63%), G2P[4] is simto 77% to G2and G2P[4]

There is replacemenG2P[4] suggcyclical patously repor

This studtiveness bywith diarrhPAGE and RRV-A.

datarol fo

Afteed.

wero ye

withsamevity

not e usre.ted 9 201 inated (two doses) 7 1481 accinated (one dose) 8 279

ted 25 201 inated (two doses) 41 1481 accinated (one dose) 14 279




adjusted by year of birth and variables (sex and age group) used for frequency matThe vaccine effectiveness (VE) was calculated by (1 OR) 100%.ormation criteria for measuring the goodness of t of the statistic model.

dose offered some protection, consistent with the lit-hough the VE was higher than in El Salvador [16] and

and lower than in Belgium (91%)) [15].d effectiveness identied is consistent with the reduc-ate of child hospitalization and mortality by AD in Brazile introduction of vaccine in Brazil [21].e-specic VE was high for G1P[8] (89%) and slightlyG2P[4] (76%) indicating a degree of cross protec-l models shown that immunity to group A rotavirusent homotypic and heterotypic components. Repeatons acquired naturally or by vaccination, increase pro-unity to include multiple serotypes, as indicated by

Theto contall VE.identi

Weafter twaction at the sensitibias istion; wexposunt of cross-neutralizing antibodies and cross-reactivecking antibodies specic for VP7 and VP4 antigens. In

vaccine clinical trials (monovalent, Rotarix; penta-Teq) as well as in the follow-up studies, both vaccineshomotypic as well as heterotypic protection againstA genotypes, including G2P[4] and G9P[8] genotypes

7].e specic VE also remained high in the second year, inth the ndings for middle income countries. VE was 74%pes, 76% for all G2 types and lower for the non G1/G2although numbers were small. The result of VE againstilar to the two small studies carried out in Brazil (75.4%

P[4]) but unlike them, effectiveness against both G1P[8] did not fall in the second year [18,19].a discussion as to whether vaccine use leads to serotypet [19]. The high effectiveness against both G1P[8] andests that the predominance of G2P[4] is most likely atern of rotavirus strains occurrence in Brazil as previ-ted [38,39].y avoided the possibility of articially reducing effec-

using controls without diarrhea rather than controlsea and (potential false) no rotavirus in stool. Using EIA,T-PCR we conrmed that all cases were true cases of

Only 73tied. This cwe were abmon circula

In concloral monovBrazilian chProtection lG2P[4] andconferred s

The ndin the Braziing for earlygenotypes.

Since thadministerethe benetdoses mighlead to inco

It mighies to infother effecthe obserVE.1 0.11 (0.050.22) 89 (7895)0.69 (0.301.57) 31 (5770)AIC b(236.815)

1 0.24 (0.160.36) 76 (6484)0.43 (0.220.85) 57 (1578)AIC b (641.170)

1 0.26 (0.100.65) 74 (3590)0.62 (0.221.79) 38 (7978)AIC b (350.510)

1 0.24 (0.160.37) 76 (6384)0.44 (0.260.74) 56 (2674)AIC b (743.862)

1 0.37 (0.111.27) 63 (2799)0.47 (0.723.09) 53 (2.0928)AIC b (137.232)

and robust variance estimation of Jaccknife, where the clusters were

collection strategy allowed us to obtain individual data,r possible confounding and verify interactions in over-r controlling for seven variables, no confounding was

e unable to investigate either if effectiveness declinesars of second dose vaccine or whether there is an inter-

oral poliovirus vaccine as the two vaccines are given time. We assumed non differential missingness in theanalysis. Although this was a case control study recallrelevant because we did not rely on recall of vaccina-ed a record (vaccine card) for establishment of the main% of genotypes of the RV-A positive sample were iden-ould hide the circulation of other genotypes, although,le to estimate genotype-specic VE for the most com-ting strains.usion, we showed consistent effectiveness of two-dosealent vaccine in preventing hospital admissions ofildren with RV-A AD, closer to European than Africa VE.asted for two years and it was similar against G1P[8] and

slightly lower against non G1/G2.The rst dose alreadyome protection.ings of the study supports the continued use of rotaviruslian National Immunization Program and the monitor-

detection of emergence of unusual and novel rotavirus

is vaccine (which requires only two doses and is co-d with other vaccines) provides adequate protection,

s of a change to a multivalent vaccine requiring threet are questionable: this may not increase protection andmplete vaccination schemes.t be useful to conduct cost-effectiveness stud-orm national immunization policy. In addition,tiveness studies should investigate what is behindved variation in monovalent rotavirus vaccine


Finally, it is important to identify early emergence of unusualand novel rotavirus genotypes so that the vaccine effectiveness canbe veried.

Conict of

All authoassociated wnancial su

Contributo

MYTI deinterpretedanalysis ancontributedcontributedsis of the daJPGL contriof the data;sis and inteedit the paall named aised the cauthors listAll authors lectual propimpedimenwith respecrm that thconcerning

Ethical app

This stuCollective 08/CEP/ISC-participate pating child

Funding

This worHealth of Brole in studin the writifor publicat

Acknowled

We recoincludes alllance SysteAlessandra Souza Alvesdo Carmo HBrazil); Ale(Regional RRita Cssia Adolfo LutzLuana da Silgas, Belm,Silva NascimRodrigues FJanete Xavi

Cardoso Arajo, Marco Aurelio de Oliveira Goes, Maria Elisa Paulade Oliveira, Marlia Reichelt Barbosa, Maria Cristina Toledo Coelho,Sandra Cristina Deboni (AD Surveillance System of the States andMunicipalities, Brazil); Ivana R. S. Varella, Elenice Brando Cunha,

on Hma R

MargalhBraz

Leonnda Araof the

dix A

plem in th1.00

nces

ss RI, Bes: tarp://dxashar avirusp://dxyan nce 6p://dxah Ghana

ldren.el MMrotav2:561

CR, ctrophranh5200nfordlecula7;38:

rgel Rng in 8;43:ashar 8;4:5O. M, Oct9;84:ikari inst rnts: r

p://dxhares l live durinble-b

p://dxdhi Se on sp://dxa KB

avirusised,

p://dxeckmation ong yp://dxPalmaldhoop://dxinterest statement

rs conrm that there are no known conicts of interestith this publication and there has been no signicant

pport for this work that could inuence its outcome.

rs

signed the study, managed the eld work, analyzed and the data and wrote the paper. LCR contributed to thed the interpretation of the data and wrote up. CASTS

to analysis and interpretation of the data; MdaGLCT to interpretation of the data; SR did the initial analy-ta; SMAM contributed to prepare the data to analysis;buted with the design of the study and interpretation

MLB contributed with the design of the study, analy-rpretation of the data. All the authors contributed toper. The manuscript has been read and approved byuthors and that there are no other persons who sat-riteria for authorship but are not listed. The order ofed in the manuscript has been approved by all of us.have given due consideration to the protection of intel-erty associated with this work and that there are nots to publication, including the timing of publication,t to intellectual property. In so doing the authors con-ey have followed the regulations of their institutions

intellectual property.

roval

dy was approved by the Committee of Institute ofHealth, Federal University of Bahia (Protocol 017-2008), by four local ethics committees. Consent towas obtained from all the hospitals. Carers of partici-ren signed written an informed consent form.

k was supported by Health Surveillance of Ministry ofrazil who collaborated in recruitment of sites but noy design, in collection, analysis, interpretation of data,ng of the report or in the decision of submit the articleion.

gments

gnize the contribution of the ROTAVAC Group which the professionals enrolled in the rotavirus AD Surveil-m who participated in the conduction of the study:Arajo Siqueira, Greice Madeleine, Rejane Maria de, Viviane Martins, Marli Costa, Ernani Renoir, Eduardoage (Health Surveillance of Ministry of Health, Brasilia,xandre Madi Fialho, Rosane Santos Maria de Assiseference Laboratory, FIOCRUZ, Rio de Janeiro, Brazil);Compagnoli Carmona (Regional Reference Laboratory,, Sao Paulo, Brazil); Joana DArc Pereira Mascarenhas,va Soares (National Reference Laboratory/Evandro Cha-

Brazil); Accia Perolina Resende Setton, Adelaide daento, Ana Gabriela de Andrade Carreira, ngela Maria

erreira, Fabula Maria de Almeida de Holanda Tormenta,er dos Santos, Teonlia Loula Dourado, Mara Espndola

Emersde FtiNiedjade Mapitals, Matos,RaimuMaria tories

Appen

Supfound,2014.0

Refere

[1] Glacinhtt

[2] Parrothtt

[3] ORmahtt

[4] Armin Gchi

[5] Patof 201

[6] LuzeleMa466

[7] Mumo200

[8] Gulati200

[9] Par199

[10] WHtion200

[11] Vesagainfahtt

[12] Linoratis douhtt

[13] Macinhtt

[14] Phurotdomhtt

[15] Bracinamhtt

[16] de chihttenklain Ferruzz, Marcia de Macedo Mory Kuroki, Mariaezende Dria Pinto, Maria Roseilda B. Barreto da Silva,ia de Oliveira Costa Medeiros Netto, Sandra Mendeses da Silva (Technicians of Core Surveillance of hos-il); Ana Accia Pereira, Auxiliadora Novais, Lina Mariaardo Bertollo, Maria do Rosrio Machado Lisboa Luna,

Carvalho, Renilda Betone Moura, Breno Brito, Sandrajo Menezes Cavalcante (Public Health Central Labora-

States).

. Supplementary data

entary data associated with this article can bee online version, at http://dx.doi.org/10.1016/j.vaccine.7.

resee JS, Turcios R, Fischer TK, Parashar UD, Steele AD. Rotavirus vac-geting the developing world. J Infect Dis 2005;192(Suppl 1):S1606,.doi.org/10.1086/431504.UD, Breese JS, Widdownson MA, Gentsch JR. New breaths for

vaccine. Drug Discov Today Ther Strateg 2006;3(2):15963,.doi.org/10.1016/j.ddstr.2006.06.005.M, Lucero Y, Linhares AC. Rotarix(R): vaccine perfor-

years postlicensure. Expert Rev Vaccines 2011;10:164559,.doi.org/10.1586/erv.11.152.E, Steele AD, Binka FN, et al. Changing patterns of rotavirus genotypes: emergence of human rotavirus G9 as a major cause of diarrhea in

J Clin Microbiol 2003;41:231722., Glass R, Desai R, Tate JE, Parashar UD. Fulling the promise

irus vaccines: how far have we come since licensure? Lancet79.Mascarenhas JD, Gabbay YB, et al. Rotavirus serotypes anderotypes identied among hospitalised children in So Luis,o, Brazil. Rev Inst Med Trop Sao Paulo 2005;47:28793, doi:S0036-5000500009.

V, Souza EC, Caruzo TAR, Martinez MB, Rcz ML. Serological andr diversity of human rotavirus in So Paulo, Brazil. Braz J Microbiol45966.Q, Cunliffe NA, Nakagomi O, Cuevas LE. Rotavirus genotypes circu-Brazil before national rotavirus vaccination: a review. J Clin Virol18, http://dx.doi.org/10.1016/j.jcv.2008.04.010.UD, Bresee JS, Gentsch JR, Glass RI. Rotavirus Emerg Infect Dis6170.eeting of the Strategic Advisory Group of Experts on immuniza-ober 2009-conclusions and recommendations. Wkly Epidemiol Rec518.T, Karvonen A, Prymula R, et al. Efcacy of human rotavirus vaccineotavirus gastroenteritis during the rst 2 years of life in Europeanandomised, double-blind controlled study. Lancet 2007;370:175763,.doi.org/10.1016/s0140-6736(07)61744-9.AC, Velazquez FR, Perez-Schael I, et al. Efcacy and safety of anattenuated human rotavirus vaccine against rotavirus gastroenteri-g the rst 2 years of life in Latin American infants: a randomised,lind, placebo-controlled phase III study. Lancet 2008;371:11819,.doi.org/10.1016/s0140-6736(08)60524-3.A, Cunliffe NA, Steele D, et al. Effect of human rotavirus vac-evere diarrhea in African infants. N Engl J Med 2010;362:28998,.doi.org/10.1056/NEJMoa0904797., Lim FS, Lau YL, et al. Safety and efcacy of human

vaccine during the rst 2 years of life in Asian infants: ran- double-blind, controlled study. Vaccine 2009;27:593641,.doi.org/10.1016/j.vaccine.2009.07.098.an T, Van Herck K, Meyer N, et al. Effectiveness of rotavirus vac-in prevention of hospital admissions for rotavirus gastroenteritisoung children in Belgium: case-control study. BMJ 2012;345:e4752,.doi.org/10.1136/bmj.e4752.

O, Cruz L, Ramos H, et al. Effectiveness of rotavirus vaccination againstd diarrhoea in El Salvador: case-control study. BMJ 2010;340:c2825,.doi.org/10.1136/bmj.c2825.


[17] Patel MM, Patzi M, Pastor D, et al. Effectiveness of monovalentrotavirus vaccine in Bolivia: case-control study. BMJ 2013;346:f3726,http://dx.doi.org/10.1136/bmj.f3726.

[18] Justino MC, Linhares AC, Lanzieri TM, et al. Effectiveness of the monova-lent G1P[8] human rotavirus vaccine against hospitalization for severe G2P[4]rotavirus gastroenteritis in Belm, Brazil. Pediatr Infec Dis J 2011;30:396401,http://dx.doi.org/10.1097/INF.0b013e3182055cc2.

[19] Correia JB, Patel MM, Nakagomi O, et al. Effectiveness of monovalentrotavirus vaccine (Rotarix) against severe diarrhea caused by serotyp-ically unrelated G2P[4] strains in Brazil. J Infect Dis 2010;201:3639,http://dx.doi.org/10.1086/649843.

[20] Wood D. WHO informal consultation on quality, safety and ef-cacy specications for live attenuated rotavirus vaccines MexicoCity, Mexico, 8-9 February 2005. Vaccine 2005;23:547887,http://dx.doi.org/10.1016/j.vaccine.2005.07.035.

[21] do Carmo GM, Yen C, Cortes J, et al. Decline in diarrhea mortal-ity and admissions after routine childhood rotavirus immunizationin Brazil: a time-series analysis. PLoS Med 2011;8:e1001024,http://dx.doi.org/10.1371/journal.pmed.1001024.

[22] Yen C, Armero Guardado JA, Alberto P, et al. Decline in rotavirus hospi-talizations and health care visits for childhood diarrhea following rotavirusvaccination in El Salvador. Pediatr Infect Dis J 2011;30(1 Suppl):S610,http://dx.doi.org/10.1097/INF.0b013e3181fefa05.

[23] Richardson V, Hernandez-Pichardo J, Quintanar-Solares M, et al. Effect ofrotavirus vaccination on death from childhood diarrhea in Mexico. New Engl JMed 2010;362:299305, http://dx.doi.org/10.1056/NEJMoa0905211.

[24] Desai R, Oliveira LH, Parashar UD, Lopman B, Tate JE, Patel MM. Reductionin morbidity and mortality from childhood diarrhoeal disease after species Arotavirus vaccine introduction in Latin America - a review. Mem Inst OswaldoCruz 2011;106:90711.

[25] Leite JP, Aleri AA, Woods PA, Glass RI, Gentsch JR, Rotavirus G. P types circu-lating in Brazil: characterization by RT-PCR, probe hybridization, and sequenceanalysis. Arch Virol 1996;141:236574.

[26] Boom R, Sol CJ, Salimans MM, Jansen CL, Wertheim-van Dillen PM, van derNoordaa J. Rapid and simple method for purication of nucleic acids. J ClinMicrobiol 1990;28:495503.

[27] Das BK, Gentsch JR, Cicirello HG, et al. Characterization of rotavirusstrains from newborns in New Delhi, India. J Clin Microbiol 1994;32:18202.

[28] Gentsch JR, Glass RI, Woods P, et al. Identication of group A rotavirus gene 4types by polymerase chain reaction. J Clin Microbiol 1992;30:136573.

[29] De Onis M. WHO child growth standards: length/height-for-age, weight-for-age, weight-for-length, weight -for-height and body mass index-for-age:methods and development. Geneva: World Health Organization; 2006.

[30] Akaike H. A new look at the statistical model identication.IEEE Transactions on Automatic Control 1974;19(6):71623,http://dx.doi.org/10.1109/TAC.1974.1100705.

[31] Linhares AC, Stupka JA, Ciapponi A, et al. Burden and typing of rotavirus groupA in Latin America and the Caribbean: systematic review and meta-analysis.Rev Med Virol 2011, http://dx.doi.org/10.1002/rmv.682.

[32] Banyai K, Laszlo B, Duque J, et al. Systematic review of regionaland temporal trends in global rotavirus strain diversity in the prerotavirus vaccine era: insights for understanding the impact ofrotavirus vaccination programs. Vaccine 2012;30(Suppl 1):A12230,http://dx.doi.org/10.1016/j.vaccine.2011.09.111.

[33] Soares-Weiser K, Maclehose H, Bergman H, et al. Vaccines for preventingrotavirus diarrhoea: vaccines in use. The Cochrane Database of Systematic Rev2012;2:CD008521, http://dx.doi.org/10.1002/14651858.CD008521.pub2.

[34] Patel M, Pedreira C, de Oliveira LH, Umaha J, Tate J, Lopman B, et al. Dura-tion of Protection of pentavalent rotavirus vaccination in Nicargua. Pediatrics2012;130:e36572.

[35] Lopman BA, Pitzer VE, Sarkar R, Gladstone B, Patel M, Glasser J, et al. Under-standing reduced rotavirus vaccine efcacy in low socio-economic settings.PLoS One 2012;7:e41720.

[36] Vesikari T, Matson DO, Dennehy P, Van Demme P, Santosham M, RodriguezZ, et al. Safety and efcacy of a pentavalent human-bovine (WC3) reassortantrotavirus vaccine Rotavirus efcacy and safety trial (REST) study team. N EnglJ Med 2006;354:2333.

[37] Estes MK, Kapikian AZ. Rotaviruses. In: Knipe DM, Howley PM, Grifn DE, et al.,editors. Field Virology. 5th edition PA, USA: Lippincot Williams & Wilkins; 2007.p. 191774.

[38] Carvalho-Costa FA, Arajo IT, Santos de Assis RM, et al. Rotavirus genotypedistribution after vaccine introduction, Rio de Janeiro, Brazil. Emerg Infect Dis2009;15:957, http://dx.doi.org/10.3201/eid1501.36 0711.

[39] Gomez MM, de Mendonc a MC, Voloto Ede M, et al. Rotavirus A genotypeP[4] G2: genetic diversity and reassortment events among strains circulat-ing in Brazil between 2005 and 2009. J Med Virology 2011;83:1093106,http://dx.doi.org/10.1002/jmv.22071.

Effectiveness of rotavirus vaccine against hospitalized rotavirus diarrhea: A casecontrol study1 Introduction2 Methods2.1 Study design

3 Participants3.1 Eligible children3.2 Potential cases and controls3.3 Effective cases and controls3.4 Laboratory investigation of potential cases

4 Data collection5 Sample size6 Statistical analysis7 Results7.1 Study population

8 Vaccine effectiveness9 DiscussionConflict of interest statementContributorsEthical approvalFundingAcknowledgmentsAppendix A Supplementary dataAppendix A Supplementary data

Documents

1-s2.0-S0264410X14000085-main.pdf