Influensautbrottet i Sverige 2007

INTERVET SYMPOSIUMMed fokus på hästinfluensan 2007Stockholm Göteborg MalmöSolvalla 16/10 Åby 17/10 Jägersro 18/10

Influensautbrottet i Sverige 2007Gittan Gröndahl och Maria Eriksson, SVA

EquiFluNet och internationella rekommen-dationer

Louise Treiberg Berndtsson, SVA

Influensautbrottet 1993 – Historik och vad lärde vi oss av detta?

Peter Forssberg, STC

Paneldebatt/frågestundFöredragshållarna samt Stig Hägglund, STC

Intervet AB Box 123182 12 DanderydTel 08 - 775 76 50www.intervet.se

INTERVET SYMPOSIUM - Med fokus på hästinfluensan 2007

Solvalla 16 okt - Åby 17 okt - Jägersro 18 okt


Solvalla 16 okt - Åby 17 okt - Jägersro 18 okt

INTERVET SYMPOSIUMMed fokus på hästinfluensan 2007

Solvalla 16 okt – Åby 17 okt – Jägersro 18 okt

INTERVET SYMPOSIUM –Med fokus på hästinfluensan 2007

Program

18.45-19.30 Influensautbrottet i Sverige 2007 (Gittan Gröndahl och Maria Eriksson, SVA)

19.30-19.50 EquiFluNet och internationella rekommendationer (Louise Treiberg-Berndtsson)

19.50-20.00 Bensträckare

20.00-20.20 Influensautbrottet -93 – Historik och vad lärde vi oss av detta? (Peter Forssberg, STC)

20.20-21.00 Paneldebatt/frågestund (föredragshållarna samt Stig Hägglund, STC)

Equilis® Prequenza

Antigen→ Subenhetsvaccin

Adjuvans→ ISCOM-Matrix

Vaccinets skyddande effekt visat→ Challenge mot både Nm/05/03och SA/04/03

Förpackningar→ Sprutor eller ampuller, båda medpeel off-etikett


Solvalla 16/10 - Åby 17/10 - Jägersro 18/10

INTERVET SYMPOSIUMMed fokus på hästinfluensan 2007

Solvalla 16 okt – Åby 17 okt – Jägersro 18 okt

Med fokus på hästinfluensan 2007

• Varför fick detta utbrott så stor spridning?

• Var startade det, och hur/var skedde spridningen?

• Typ av hästar som drabbades, vaccinationsstatus pådrabbade?

• Vad kan vi göra för att förhindra ett liknande utbrott i framtiden?

• Vad lärde vi oss vi förra stora utbrottet -93?

• Orsaken till att vaccinationsobligatoriet inom travet upphävdes?

• Kommer regler/rutiner att ändras, vad händer/gäller utomlands?

• Hur fungerar det globala nätverket för övervakning av hästinfluensa?

• Hur väljs vaccinstammar ut, hur görs uppföljningar?

Hästinfluensa…

Varför ska vi vaccinera..?



Resultat från studien som ges i denna presentation är preliminära.

För slutgiltiga siffror och resultat hänvisas till kommande publikation i Svensk Veterinärtidning.

Oktober 2007

Gittan Gröndahl

GittanGittan GrGröönndahldahl,, tf statsveterinär, VMDMaria Eriksson, vet stud

SVA, 2007

EpideminEpidemin avavhhäästinfluensastinfluensa

i i SverigeSverige 20072007

ävenSydamerikansk variantSydafrikansk variant

Europeisk variant

Amerikansk variant

Sedan 1980-talet:

Hästinfluensaorsakas av

influensavirus av olika subtyper

Tf statsvet Gittan Gröndahl, SVA



Hästinfluensans utbredning

• A1: inte sedan 1979• A2: nya varianter• Hästinfluensa förekommer i Sverige årligen• Mer epidemiskt vissa år• De flesta länder har h ästinfluensa• Nya Zeeland och Island fria• Australien var fritt tills augusti 2007

• Första fallet augusti 2007• Tidigare fritt land,

inga vaccinerade hästar• Kraftig spridning på en månad:

2653 Infected Properties, 323 Dangerous Contact Properties and 356 Suspect Properties. (070928)

• Mål: utrota EI inom 6 månader

Australiens utbrott 2007

05

1015202530354045

19971998

19992000 2001

20022003

20042005

2006

Hästinfluensa

Svenska rapporterade utbrott till Jordbruksverket 1997-2006


1 ”utbrott”= första fallet, men kan vara många hästar drabbade



Symptom på hästinfluensa

Symptom på hästinfluensa A2

• OBS: Vaccinerade hästar får lindrigare symtom – om de över huvudtaget får symtom. Sprider mindre mängd virus, och återhämtar sig snabbare än ovaccinerade


Isoleringsrekommendationervid influensa

• Stallet isoleras 10 dagar efter senaste insjuknade hästs första febertopp

OBS - Spridningen av smitta kan begränsas om man har m öjlighet att separera/isolera den först insjuknade hästen i stallet



Stor svensk epidemi av hästinfluensa 2007

• Spreds från södra Sverige i januari 2007januari 2007• Nådde hela Sverige, janjan--junijuni 20072007• Mest travstallar drabbade,

inkl de flesta banor• 2 unga h ästar akut döda i sviternasviterna• Bristande vaccination, ååsidossidosäättande avttande av

smittskyddsregler, mörkande?

Travronden rapporterade om A2

• [070924]: Kriterieauktionen direktsänds • [070613]: V64-trippel för Kari• [070613]: Den nye Gidde...• [070613]: V64: Formkusk ligger lågt• [070605]: Fler A2-sjuka på Solvalla• [070604]: A2 på Solvalla• [070509]: Smittoläget• [070507]: Misstänkt A2 hos Svensson• [070426]: A2-nytt• [070420]: Russel, Chaplin och USA-importer• [070418]: Ännu mera A2• [070417]: Peter Forssberg om A2-influensan • [070417]: Nurmos-stall isolerat• [070417]: Misstänkt A2 i Boden• [070403]: A2 på Vermo• [070326]: A2 även på Romme• [070325]: "I påsk vill jag vara med igen"• [070321]: Konstaterad A2 i Gävle

• [070320]: Gävle-stall A2-isolerade• [070314]: Hallå där Charlotta B...• [070305]: Inbrott hos Turja igen• [070301]: Nya fall av A2-virus• [070220]: "Det måste gå fort i början" • [070216]: Laursens överl ägsna 101-oddsare• [070210]: Heiskanen tränare i Italien • [070206]: Återbud från Adielsson• [070123]: A2-läget stabiliserat• [070119]: A2-läget: inga nya fall idag• [070118]: A2-läget just nu • [070117]: Dags för vaccinationstvång?• [070117]: Två hästar döda på Axevalla • [070116]: Per ställer in Axevallapremiären?• [070115]: A2 på Axevalla• [070104]: A2 i Skaratrakten• [061209]: Champagne i Århus



Det påstås att…….

• ”Vaccinerade hästar blir också sjuka och det pågår mycket, mycket l ängre än hos ovaccinerade” (Travronden blogginlägg av Pelle, april 2007)

• ”Att vaccinera är att spruta in sjukdom i kroppen (…) och vilken idrottsman gör det frivilligt?” (Pelle igen)

• ”Det är bättre att hästarna blir ordentligt sjuka såman inte råkar missa det och startar en vaccinerad häst som bara är lite sjuk i influensa, för då kan den få allvarliga komplikationer” (en av våra svarare)

Studie av epidemin av hästinfluensa i Sverige 2007

• Oberoende studie vid Statens veterinärmedicinska anstalt (SVA)

• Stöd från SVAs Forskningsfond

Syfte med studien

• Vilken typ av hästar drabbades?• Blev hästarna sjuka trots att de var

vaccinerade? (ny virusstam/dåliga vaccin)• Eller var det ovaccinerade h ästar som

framför allt blev sjuka?• Skyddade vaccin mot sjukdom/gravare

sjukdom?• Kan rutiner förbättras?



Studie av epidemin av hästinfluensa i Sverige 2007

Underlag:

• 68 rapporter vid SVA om positiva influensaproverpositiva influensaproverfrån ca 100 hästar januari-juni 2007

• Känt att hästinfluensan drabbade Sveriges 32 travbanor/regioner enligt följande:3/3 storbanor, 8/11 mellanbanor och 13/18 småbanor, totalt 24 st.

• Mörkertal antal drabbade banor?• Varje banregion innefattar många tränare

Januari:Axvall, Aneby,

Hishult

Tävling Axevalla 16/1 under isoleringen och 3 hästar dör – Åby,

Mantorp och Jä gersro isolerar dagarna efter.

Februari:Hammarö, Visby,

Vårgårda, Borensberg,

Uppsala

Färjestad o Visby isolerar, tävling på

Färjestad 19/2

Mars: Sala, Järvsö,

Borlänge, St Skedvi, Skattkärr, Saxdalen,

Glanshammar, Gävle, Helsingborg

Tävling Gävle 20 och 27/2 under isoleringen. Hagmyren ställer in.

Romme isolerar också.

April: Orsa, Rättvik, Tumba,

Örebro, Boden, Halmstad, Edsvalla, Eskilstuna,

Nossebro, Haparanda, Älandsbro, Lindesberg,

Edsbyn, Åmål, Borlänge, Näsviken, Ljungbyhed, Bollnäs, Eskilstuna, St

Skedvi, Kimstad, Norberg

Bollnäs, Rättvik, Romme, Dannero, Solänget, Åmål,

Skellefteå, Boden, Oviken, Halmstad, Örebro, Solvalla

isolerar. Finland också A2.

Maj:

Tävelsås, Västerljung, Söderköping, Linkö ping,

Skinnskatteberg, Enköping, Säffle, StAnna, Strömsholm, Örebro, Nordingrå

Örebro, Åby, Åmål, Skellefteå, Solvalla

isolerar. Finland också A2.Virus sprider sig över

landet…..

Insamlingsresultat

• 45 av SVA-positiva hästhållare kunde kontaktas varav enkätsvar erhållits från 19 st.

• 13 drabbade travbanor kontaktades.Svar från 8: Axevalla, Färjestad, Visby, Romme, Örebro, Lindesberg, Solvalla samt Halmstad. Inget svar från: Mantorp, Åmål, Rättvik, Gävle, Hagmyren och Boden.

• 63 stallar med 773 hästar analyserade



Enkätfrågor

• Sjuka och friska hästars namn, ålder, kön, ras, träningsstatus

• Datum för senaste 2 vaccinationer mot hästinfluensa

• Tidigare sjuk i influensa• Temperaturkurvor under utbrottet• Antal dagar med hosta +/- näsflöde• Antibiotikabehandlingar

Subjektiva uppfattningar

”Var smittades dina hästar tror du?”

• Många svarar: ”- På travtävlingar för 2-3 dagar sedan”

• Jägersro, Mantorp 2/1, Axevalla, Gävle 12/3, Romme 16/3 + 16/4, Rättvik 26/3, Dannero 15/4 omnämns t ex

Kriterier i studien

• Adekvat vaccinerad = Grundvacc 2 ggreller revacc inom 1 år tillbaka

• Ej adekvat vaccinerad = Vaccineradmen ej som ovan

• Ovaccinerad = Ingen vacc eller endast1 vacc för mindre än 2 veckor sedan



Kriterier i studien

• Feber = Temperatur =38,3°C• Hosta, näsflöde = Enl hästhållaren• Sjukdomsgrad:

lindrig = feber i 1 -2 dagarmåttlig = feber i 3-4 dagarkraftig = feber i 5 dagar eller mer

• Tävlingskondition = Har nyligen startatel ska starta i travlopp

Svar från 63 stallar med totalt 773 friska och sjuka hästar

Tävlingskondition: 233/293 hästar (80%)

Könsfördelning

41%

35%

23%1%

StonValackerHingstarOkänt

Rasfördelning

87%

6%

3%

3%

1%

VarmblodKallblodPonnyHalvblodOkänt

0 – 29 årMedelålder = 4,9 år (±3,2)

0%

5%

10%

15%

20%

0 år

1 år

2 år

3 år

4 år

5 år

6 år

7 år

8 år

9 år

10+ år

okänd

Ålder

An

del

Åldersfördelning i hela materialet



Vaccinationsstatus –känt för 530 hästar (friska och sjuka)

• Ovaccinerade: 213 st (40%)• Adekvat vaccinerade: 169 st (32%)• Ej adekvat vaccinerade: 148 st (28%)• För 1/3 av hästarna visste inte tränarna om

de var vaccinerade eller inte (243 st)• 7% (21/318 svar) hade haft influensa förut

Åldersfördelning (friska+sjuka) per vaccinationskategori

0%

5%

10%

15%

20%

25%

30%

0 år

1 år

2 år

3 år

4 år

5 år

6 år

7 år

8 år

9 år

10+ år okä

nt

Ålder

Ovaccinerade

Adekvat vaccinerade

Ej adekvat vaccinerade

0 %

5 %

10%

15%

20%

0 år1 år

2 år3 år

4 år5 å

r 6 år

7 år8 år

9 år

10+ å

rokä

nd

Ålder

An

de

l

Tränare A inne på Axevalla. Adekvat vaccinerade hästar (25%)

36,5

37

37,5

38

38,5

39

39,5

40

40,5

2007-01-08

2007-01-13

2007-01-18

2007-01-23

2007-01-28

2007-02-02

BB 7 år

CML 6 år

TL 5 år

Tränare A inne på Axevalla. Ej adekvat vaccinerade hästar (42%)

36,5

37

37,5

38

38,5

39

39,5

40

40,5

2007-01-0 8

2007-01-1 3

2007-01-1 8

2007-01-2 3

2007-01-2 8

2007-02-0 2

CH 2 å r

EK 4 år

MR 2 år

RR 2 å r

HC 2 å r

Tränare A inne på Axevalla. Ovaccinerade hästar (33%) (3 är vacc, men väldigt

nära/vid utbrottet)

36,5

37

37,5

38

38,5

39

39,5

40

40,5

2007-01-0 8

2007-01-13

2007-01-18

2007-01-23

2007-01-28

2007-02-0 2

RA 2 år

RH 2 år

HJ 2 år

BFR 3 år

Normal temp är upp till 38°C.

Ovaccinerade hästar f år högre feber och i fler dagar (upp till 2 v).

1 vaccination är bättre än ingen. Men för att hinna ge skydd bör det ha gått några veckor.




Tränare B utanför Axevalla. Hästar som är adekvat vaccinerade (18%)

36,5

3737,5

38

38,539

39,5

4040,5

2007-01-03

2007-01-08

2007-01-13

2007-01-18

2007-01-23

QS 4 årSJ 3 år

Tränare B utanför Axevalla. Ovaccinerade hästar/okänt vaccinationsstatus (72%)

36,5

37,5

38,5

39,5

40,5

2007-01-03 2007-01-08 2007-01-13 2007-01-18 2007-01-23

CL 4 år

LJS 2 år

LJ 2 år

CLA 4 år

HG 2 år

HP 7 år

LM 6 år

LP 10 år

LL 2 år

Ju fler ovaccinerade hästar,

desto högre smittryck (högre virusmängd i stallet)

och desto fler sjuka (sjukare) hästar.


21%

79%

Ovaccinerade(n=168)

21%79%

68%

32%

Adekvat vaccinerade(n=91)

68%32%

45%

55%

Ej adekvat vaccinerade(n=80)

45%55%

Ovaccinerade: majoriteten fick feber

• Av samtliga hästar fick 311/562 st feber (=38,3°C) minst en dag, uppdelat på vaccinationsstatus ser det ut så här:

Ej feberFeber

Ovaccinerade: högre febertoppar och fler dagar sjuka

Vaccinerade (n=91)

Ej adekvat vaccinerade (n=80)

Ovaccinerade (n=168)

Antal feberdagar hos de hästar som hade feber (medel ± SD)

0

1

2

3

4

56

7

*****

36,5

3 7

37,5

3 8

38,5

3 9

39,5

4 0

40,5

4 1

Högsta febertopp hos hästar med feber (medel ± SD)

*** *



0 %

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Vaccinerade(n=91)

Ej adekvatvaccinerade

(n=80)

Ovaccinerade(n=168)

Ingen feber

Kortare feberperiod(1-2 d)

Måttlig feberperiod(3-4 d)

Längre feberperiod(5 d el mer)

Feberperiod

1 häst

Hälften av alla ovaccinerade

hästar fick feber 3 d eller längre

(25% >5d)

Hur länge tar det för ett stall att bli feberfritt?

• 2 – 22 dagar• Medelvärde 7 dgr (±3,4)• När började mätningarna?

– Sannolikt längre perioder i verkligheten

Hosta & Näsflöde

Hosta• 202 st (45 %) hostade• 1-23 dagar• Medel 6,1 dgr (±3,7)• n=453 svar

Näsflöde• 208 st (45 %) hade

näsflöde• 1-15 dagar• Medel 6,1 dgr (±3,0)• n=460 svar



0%

10%

20%

30%

40%

50%

60%

70%

80%

Ovaccinerade Adekvatvaccinerade


Hosta

Näsflöde

Ovaccinerade: fler hostar och snorar och under längre tid

0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

Ovaccinerade Adekvatvaccinerade


An

tal d

agar

Medel hostdagar

Medel näsflödesdagar

Hur många hästar har hosta eller näsflöde?

Hur många dagar har dessa hosta eller näsflöde i

snitt?

Antibiotikabehandlingar

• 9 % behandlades sekundärt (30/319)– vanligast penicillin (18) och trimsulfa (7)– behandling 5,9 dgr (±1,03)– genomsnittlig ålder 3,7 år (±1,91)***

jmfr 4,9 år för hela materialet.

• Inga av dessa 30 hästar var adekvatvaccinerade…

Vaccinerade h ästar jämfört med ovaccinerade:• Färre hästar med feber

• Färre feberdagar/häst• Lägre feber• Färre hästar får hosta och näsflöde

• Färre dagar/häst med hosta resp. näsflöde• Färre (inga!) hästar behövde behandlas

Take home message

Adekvat vaccinerade (n=91)

68%

32%

Ovaccinerade (n=168)

21%

79%



Konklusion

Hästar som vaccineras mot influensa på ett adekvat s ätt blir inte sjuka eller

får en mildare sjukdom under kortare tidjämfört med ovaccinerade individer.

Förebygg influensa!• Vaccinera fölen

efter 6 månaders ålder (2 ggr)• Vaccinera alla hästar i stallet• Vaccination helst var 6 mån.

(unga hästar) t.o.m. 4 års ålder• Årlig vaccination fr.o.m. 5 år• FEI, internationell ridsport:

Vaccination var 6 mån.• Vaccinera när utbrott närmar sig!• Inför obligatorisk vaccination av

svenska travhästar igen?

Att diskutera…

• Hur länge är prestationen nedsatt?• 1 veckas vila för varje feberdag

rekommenderas• Varje dag i ett travträningsstall kan kosta

ca 300 kr. 10 dagars isolering/sjukdom = 3000 kr/häst i träningsavgift till spillo för travhästägaren

• Vaccination kostar mindre än 1 kr/dag• Bättre isoleringsåtgärder vid utbrott?



RRäätt sktt sköötsel tsel och och

hhäästhsthåållningsllnings--rutinerrutiner

med med smittskydd i smittskydd i

fokusfokusEkonomiskt Ekonomiskt mycket att vinnamycket att vinna

Kortare avbrott i Kortare avbrott i verksamhetenverksamheten

BBäättre djurskyddttre djurskydd



Equi Flu Net och internationella rekommendationer

Louise T BerndtssonAvd för virologi

Influensa hos häst

• Först isolerad i Tjeckoslovakien 1956 A1, H7N7(A/eq1/Prague/56)

• Isolerad i Miami 1963 A2, H3N8(A/eq2/Miami/63)

Influensavirus



Hemagglutinin ochNeuraminidas

HA – för införselav virus in i cellen

Na – för att ta sigut ur cellen efter virusreplikationen

Faktorer som vidmakthåller epizootier av hästinfluensa

• Antigen driftNär existerande antikroppar inte längrekänner igen HA-gp och därmed ej virus.

• Kort duration av immunitet• (Sprids över speciesgränserna?)

Amerikansk och Europeisk variant



Hästinfluensa fylogenetiskt träd

Kentucky/98

Kentucky/94Newmarket/1/93Newmarket/1/93

Newmarket/5/03

South Africa/4/03

Amerikansk

A1Prag/56† ?

A2Miami/63

Europeisk

A2

Internationell övervakning av hästinfluensa

• Expert Surveillance Panel– SVA ett av 6 laboratorier globalt– Möts varje år för genomgång av

utbrott och karakterisering av stammar

– På grundval av utvecklingen rekommenderas bibehållande eller utbyte av vaccinstammar

– Rekommendationer för harmonisering och standardi-sering av metoder för antigen-bestämmning i vacciner

ESP 2007 - antigen karakterisering

• South Africa/4/03 täcker väl stammar av amerikanska varianten

• Newmarket/2/93 täcker väl de flesta stammar av europeiska varianten, men några isolat från 2002 ”outgroups”

• Antigenic Cartography en metod för framtida karakterisering?



H3N8 stammar som används i nuvarande amerikanska och

europeiska vaccin

• Europeisk variant:– Suffolk/89– Borlänge/91– Newmarket/2/93

• Amerikansk variant:– Newmarket/1/93– Kentucky/92– Kentucky/94– Kentucky/95– Kentucky/97– Kentucky/98

2006 OIE recommendations

Update vaccines to contain :

• an A/equine/South Africa/4/03 (H3N8)-like virus (American lineage) and

• an A/equine/Newmarket/2/93 (H3N8)-like virus (European lineage) remains

Olika möjligheter för influensavacciner

• uppgradera nuvarande vacciner -kan ta flera år

• Testa existerande vaccin -challenge studie mot Syd Afrikanska varianten



Conclusions and recommendations from the Expert Surveillance Panel on Equine

Influenza Vaccines - January 2006

These recommendations relating to the composition of vaccines for 2006 were made

following review of the data arising from equine influenza surveillance by the panel of

international collaborators for the period January 2005 – January 2006. The

recommendations for vaccine strains remain as for 2005.

Influenza activity 2005

Outbreaks of equine influenza in Denmark, France, Sweden, Tunisia, United Kingdom, and

the USA were reported during 2005. Some outbreaks occurred in vaccinated animals but

disease was generally mild.

All influenza activity was associated with H3N8 viruses. There were no reports of serological

or virological evidence of H7N7 (equine-1) subtype viruses circulating in the equine

population. Nevertheless, diagnostic laboratories should continue serological and virological

monitoring and when using polymerase chain reaction (PCR) for rapid diagnosis, should

ensure that primers specific for H7N7 virus as well as H3N8 virus are used.

Characteristics of recent isolates

All viruses characterised antigenically and/or genetically from Europe and North America

during 2005 belonged to the ‘American' lineage with the exception of one isolate in the UK.

In haemagglutination inhibition (HI) tests using post infection ferret antisera American

Lineage viruses isolated in Europe and North America were closely related to the prototype

vaccine strain A/South Africa/4/2003 and the A/eq/Newmarket/5/2003 reference strain. The

HA1 sequences of American lineage viruses isolated since 2003 in America, Europe and

South Africa all fall within a single phylogenetic sub-group, previously referred to as the

‘Florida' lineage (Lai et al., 2001; 2004). The sequences of viruses isolated in America since

2003 and represented by A/eq/South Africa/4/2003 (and A/eq/Ohio/2003) are characterised

by two further amino acid changes in antigenic sites compared with the HA1 sequences of

viruses isolated in Europe; these additional changes appear to contribute to greater antigenic

drift from A/eq/Newmarket/1/93-like viruses currently included in vaccines. The European

lineage virus isolated in 2005 reacted well in HI tests with ferret antisera against the

European lineage reference strain A/eq/Newmarket/2/93.

Recommendations for the composition of equine influenza vaccines

During the period January 2005 to January 2006, H3N8 viruses of the ‘American’ lineage

continued to circulate in Europe and North America with some vaccinated horses affected.

These viruses, together with those responsible for the 2003/4 outbreaks in South Africa and

circulating in North America were antigenically closely related to the currently recommended

vaccine strains, A/eq/South Africa/4/2003-like. Only one virus belonging to the ‘European’

lineage was characterised during 2005 and no serious clinical episodes have been attributed

to these viruses. Nonetheless, the recommendation remains that a European lineage virus be

included in vaccines and surveillance for European lineage viruses be continued.



It is recommended, therefore, that vaccines contain the following:

• an A/eq/South Africa/4/2003 (H3N8)-like virus (American lineage)1

1A/eq/Ohio/2003 is as equally acceptable as A/eq/South Africa/4/2003.

• an A/eq/Newmarket/2/93 (H3N8)-like virus (European lineage)2

2A/eq/Suffolk/89 and A/eq/Borlänge/91, currently used vaccine strains, continue to be

acceptable.

Reference reagents

Reference reagents specific for the recommended European lineage vaccine strains are

available for standardisation of vaccine content by single radial diffusion (SRD) assay and

can be obtained from the National Institute for Biological Standards and Control (NIBSC).

Preparation of reagents for the 2005 recommendation is under review.

Three equine influenza horse antisera (anti-A/eq/Newmarket/77 [H7N7], anti-

A/eq/Newmarket/1/93 [H3N8] and anti-A/eq/Newmarket/2/93 [H3N8]) are available as

European Pharmacopoeia Biological Reference Preparations (EP BRPs) for serological

testing of equine influenza vaccines by the single radial haemolysis assay. These antisera are

also available from the Office International des Epizooties International Reference

Laboratory in Newmarket (UK) for use as primary standards in diagnostic serological testing.

Pooled equine serum obtained post infection with A/eq/South Africa/4/2003 (H3N8) virus is

currently the subject of an international collaborative study to establish this serum as an EP

BRP / OIE primary standard to supersede the anti-A/eq/Newmarket/1/93 (H3N8) serum.

SRD reference reagents EP BRPs for serological

testing of equine influenza

vaccines

OIE primary standards

for diagnostic serological

testing

NIBSC, Blanche Lane,

South Mimms, Potters Bar,

Herts, EN6 3QG, UK

Fax: +44 (0)1707 646730

e-mail:

[email protected]

European Directorate for

the Quality of Medicines,

BP 907, F-67029

Strasbourg Cedex, France

Website:

http://www.pheur.org

Animal Health Trust,

Lanwades Park, Kentford,

Newmarket, Suffolk,

CB8 7UU, UK

Fax: +44 (0)8700 50 24 61

e-mail: [email protected]

References:

Lai A.C.K., Chambers T., Holland R.E., Morley P.S., Haines D.M., Townsend H.G.G. &

Barrandeguy M. (2001). Diverged evolution of recent equine-2 influenza (H3N8) viruses in

the Western Hemisphere. Arch. Virol., 146 , 1063–1074;

Lai A.C.K., Rogers K.M., Glaser A., Tudor L. & Chambers T. (2004). Alternate circulation

of recent equine-2 influenza viruses (H3N8) from two distinct lineages in the United States.

Virus Res., 100 , 159–164.



1

Influensautbrottet 1993- Historik och vad lärde vi oss av detta?

Peter Forssberg, STC

A2-influensa

amerikansk

stam

europeisk

stam

Historik

Större utbrott

1979, 1985, 1992, 2006



2

Vaccinationsobligatorium

Ej i Finland, ej i Sverige

Framställan om obligatorium

Nordisk djurskyddskommitté

Banveterinärkonferens

Utbrott A2

november 1992 – maj 1993

< 1 år efter föregående utbrott



3

Pia Törnqvist et.al 1991

effekt 81%

intervaller?

• Förväntningar• Biverkningar• Ekonomi• Avvecklat obligatorium

Borttaget obligatorium

• Lokala smittskyddsgrupper

• Ändrade intervaller

• ”öppnare” isolering



4

1993

färre sjuka 19% 8%vaccinationsgrad dålig:Solvalla 90-100%Mantorp 27%Gävle 34%Östersund 15%

Fördelning kallblod

25%

25%

25%

25%

Påstådda biverkningar

12%

15%

18%

55%



411Vet. Res. 35 (2004) 411–423© INRA, EDP Sciences, 2004DOI: 10.1051/vetres:2004023

Review article

Current perspectives on control of equine influenza

Janet M. DALY*, J. Richard NEWTON, Jennifer A. MUMFORD

Centre for Preventive Medicine, Animal Health Trust, Lanwades Park, Kentford, Newmarket, Suffolk, CB8 7UU, United Kingdom

(Received 7 July 2003; accepted 31 October 2003)

Abstract – Influenza A viruses of the H3N8 subtype are a major cause of respiratory disease in horses.Subclinical infection with virus shedding can occur in vaccinated horses, particularly where there isa mismatch between the vaccine strains and the virus strains circulating in the field. Such infectionscontribute to the spread of the disease. Rapid diagnostic techniques are available for detection ofvirus antigen and can be used as an aid in control programmes. Improvements have been made tomethods of standardising inactivated virus vaccines, and a direct relationship between vaccinepotency measured by single radial diffusion and vaccine-induced antibody measured by single radialhaemolysis has been demonstrated. Improved adjuvants and antigenic presentation systems extendthe duration of immunity induced by inactivated virus vaccines, but high levels of antibody arerequired for protection against field infection. In addition to circulating antibody, infection withinfluenza virus stimulates mucosal and cellular immunity; unlike immunity to inactivated virusvaccines, infection-induced immunity is not dependent on the presence of circulating antibody toHA. Live attenuated or vectored equine influenza vaccines, which may better mimic the immunitygenerated by influenza infection than inactivated virus vaccines, are now available. Mathematicalmodelling based upon experimental and field data has been applied to examine issues relating tovaccine efficacy at the population level. A vaccine strain selection system has been implementedand a more global approach to the surveillance of equine influenza is being developed.

equine influenza / epidemiology / vaccine strain selection / surveillance

Table of contents

1. Introduction...................................................................................................................................... 4122. Epidemiology................................................................................................................................... 4123. Vaccine potency............................................................................................................................... 4134. Natural immunity and live vaccines ................................................................................................ 4145. Optimising vaccination schedules .................................................................................................. 4166. Vaccine strain selection ................................................................................................................... 4177. Diagnosis ......................................................................................................................................... 4198. International control......................................................................................................................... 4209. Conclusion ....................................................................................................................................... 420

* Corresponding author: [email protected]



412 J.M. Daly et al.

1. INTRODUCTION

Management procedures aimed at limit-ing the severity of disease and the spread ofinfection, whether on a local or interna-tional basis, require sensitive diagnostictechniques for rapid detection of clinicaland subclinical infection. Equine influenzavaccines were first developed in the 1960s[4], and are used widely for control ofequine influenza however, in spite of inten-sive vaccination programmes in some groups,equine influenza infections remain a seri-ous problem. The H3N8 component ofinactivated vaccines has been the subject ofintense investigation with a view to identi-fying the reasons for vaccine breakdownagainst this subtype. Research has focussedon vaccine potency, adjuvants, vaccinationschedules and antigenic drift. During thelast decade, progress has been made in allthese areas of investigation, providing newapproaches to the control of equine influ-enza.

2. EPIDEMIOLOGY

Equine influenza was first recognised in1956, when influenza was recovered duringa widespread epidemic of respiratory dis-ease among horses in Eastern Europe [58].The virus (A/eq/Prague/56), which has anH7 haemagglutinin (HA) and an N7 neu-raminidase (NA), was designated as theprototype equine influenza virus, histori-cally referred to as equine subtype 1. Thelast confirmed outbreak caused by an H7N7subtype virus was in 1979; however H7-specific antibody has been reported inhorses believed to be unvaccinated, sug-gesting that the virus may still circulate ina subclinical form.

In 1963, an equine influenza virus of adifferent antigenic subtype (H3N8), origi-nally designated as equine subtype 2,caused a major epidemic in the USA [64].The prototype virus, A/eq/Miami/63, wasintroduced into the equine population ofFlorida with the importation of horses from

Argentina [57]. Field evidence suggestedthat regular vaccination provided protec-tion against H7N7 infections, but that theH3N8 component of the vaccine was lesseffective [53]. For example, in January1976 a localised outbreak of H3N8 occurredin Thoroughbred horses in Newmarket(UK) at a time when many animals hadrecently been vaccinated [59]. Clinicalinfluenza affected unvaccinated and somevaccinated horses, with the severity of dis-ease corresponding with the period sincevaccination. Stables in which over 75% ofhorses were vaccinated were not affectedseriously [59]. Between 1978 and 1981,widespread epidemics of H3N8 viruseswere reported in Europe and North Americawith infections occurring in vaccinated aswell as unvaccinated horses [7, 28, 30, 52,62]. In Britain in 1979, influenza was con-fined to unvaccinated horses during the firstsix months of the year, but spread to vacci-nated Thoroughbreds in June 1979, provid-ing clear evidence that the vaccines did notprovide immunity against field infectionfor the full year between “booster doses”[6]. Racing was affected, and this led to thesubsequent introduction of mandatory vac-cination in the UK and Ireland in 1981.

In 1989, there was again a major epi-demic of influenza H3N8 in Europe affect-ing not only unvaccinated but also largenumbers of vaccinated horses [33]. Thisrepresented the first major outbreak in Brit-ain since 1979. Outbreaks of equine influ-enza have occurred sporadically in Europeand on the American continent since the1989 epidemic.

In the last 15 years, there have also beena number of serious outbreaks of H3N8influenza in populations with no previoushistory of the disease. In 1986 and 1987, theinfection was introduced into South Africaand India, respectively. The source of theseoutbreaks could be traced to the transporta-tion of infected horses by air from areaswhere influenza was endemic. Inadequatequarantine at the port of entry allowed theintroduction of infected horses into the local



Control of equine influenza 413

susceptible populations with subsequentexplosive spread of disease and some mor-tality. Analysis of the HA genes of the SouthAfrican and Indian viruses have confirmedtheir close relationship to viruses circulat-ing in the USA and Europe at the time. In1989, an influenza epidemic was reportedin horses in China with morbidity rates ashigh as 80% and mortality rates reaching20% in some herds. Fatal cases were alwaysassociated with bacterial infection [21].The origin of this outbreak was not tracedto the importation of equidae and indeed theantigenic characteristics of this virus appearmarkedly different from other equine H3N8isolates [22]. On the basis of sequenceinformation, it was proposed that this viruswas derived from an avian source and assuch represented a new interspecies trans-mission event [65]. Although this avian-derived virus successfully transmitted tohorses and lost its ability to infect ducks, itdid not spread beyond China and did notpersist in the local horse population beyond1990 [23]. Further outbreaks in Hong Kongin 1992 [54], Dubai in 1995 [66], and thePhilippines in 1997 highlighted the easewith which equine influenza outbreaks canbe introduced into susceptible populationsas a result of international movement ofhorses.

3. VACCINE POTENCY

Currently, the principal markers forresistance to and recovery from influenzavirus infection are circulating antibodiesspecific for the HA and NA glycoproteins[1]. These glycoproteins are the principledeterminants for cell entry in infection(HA) and for exit from the cell after virusreplication (NA). Progress in assessing theprotective efficacy of early vaccines washampered by a lack of reliable methods tomeasure the HA content of vaccines and thehost’s antibody response to the HA. Addi-tionally, there was no reproducible chal-lenge method in horses for assessing theprotection provided by vaccination. The

HA content of vaccines was measured inchick cell agglutination (CCA) units andantibody responses to the HA were meas-ured by the haemagglutination inhibition(HI) test. In some instances these methodsare both still used. Early attempts to analysethe relationship between vaccine-inducedantibody and protection against infectionwere confused by technical problems, andHI titres ranging from 8 to 128 were quotedas being protective [5, 31, 55, 59]. Improvedmethods of measuring vaccine potency,antibody responses and protection againstinfection have since been developed, facil-itating progress in vaccine standardisationand design. A reliable in vitro potency test,the single radial immunodiffusion (SRD)test, has been introduced for measurementof immunologically active HA in equineinfluenza vaccines and has been evaluatedin an international collaborative study [68].A further international collaborative studydemonstrated that the single radial haemo-lysis (SRH) assay is more reproducible thanthe HI test for measuring antibody to HA[38]. Furthermore, there is a direct relation-ship between vaccine potency, in terms ofmicrogrammes of HA, and antibody to HAstimulated by inactivated vaccines as meas-ured by SRH [41, 67].

Vaccine evaluation by experimental chal-lenge infection of horses was slow to progressbecause of difficulties encountered in repro-ducing clinical disease [8, 31, 35, 56]. Thesedifficulties have been overcome by usingnebulised aerosols. This delivery systemmimics a natural infection by producinginfectious droplets (diameter < 5 mm) capa-ble of reaching the upper and lower airways(D. Hannant, unpublished data) and avoidsa concentration of challenge inoculum atthe site of sampling. Using this challengemethod, a series of experiments to measurethe protection afforded by inactivated virusvaccines with a variety of adjuvants andantigen presentation systems have been per-formed. A number of experiments haveused the SRD test to standardise inactivatedvaccines, the SRH test to measure antibodyresponses in the horse and challenge infections




to assess protection from infection and dis-ease. These studies have determined therelationships between vaccine potency, cir-culating antibody to HA and protectionagainst infection and disease. Levels of anti-body required for virological protection againstchallenge with an antigenically similarvirus were between 120 to 154 mm2, withevidence that a higher threshold was requiredfor protection with increasing doses of neb-ulised virus [39]. The influenza epidemic inSouth Africa in 1986 provided a rare oppor-tunity to examine vaccine efficacy in thefield in a population where no naturalimmunity exists. From pre-infection anti-body levels it was possible to estimate thatan SRH value of around 160 mm2 was con-sistent with a 90% protection rate based onthe proportion of horses that seroconvertedwhen exposed to infection [39].

The majority of current equine influenzavaccines contain inactivated whole virus(with adjuvants, which include oil, alhydrogelor carbomer) or subunit vaccines (ISCOMsor micelles combined with Quil A). It wasfound that antibody responses stimulatedby vaccines containing aluminium phos-phate or hydroxide were more durable thanthose induced by aqueous vaccines ofequivalent antigenic content. Antibody never-theless declined to low levels by 16 to20 weeks after the second and third dose ofvaccine. In contrast, the incorporation of apolymer adjuvant was found to stimulateantibody that remained at a high level for atleast six months after the third dose of vac-cine [43]. Similarly, vaccination with threedoses of ISCOMs containing 15 mg HAresulted in the level of SRH antibody per-sisting at around 70 mm2 for 15 months fol-lowing the third dose [42].

The historical lack of standardisation ofvaccines from different sources, and theundemanding standards of some licensingauthorities, has resulted in the use of prod-ucts with inadequate potency in terms ofability to stimulate antibody to the HA.Morley et al. [37] described a large double-blind field trial using a commercial killed

vaccine that failed to demonstrate a significantdifference in the rate of disease betweenvaccinated and unvaccinated animals in theface of a naturally occurring outbreak ofdisease in a population of horses stabled ata racetrack. The situation is improving withthe establishment of European Pharmaco-poeia international reference preparationsto standardise serological tests for potencyevaluation of vaccines, and the introductionof federal regulations on equine influenzavaccines in Europe [17] and, more recently,in the USA (9CFR parts 112 and 113).

4. NATURAL IMMUNITY AND LIVE VACCINES

Immunity provided by inactivated influ-enza virus vaccines, is dependent on highlevels of circulating antibody to HA and, inthe absence of such antibody, vaccinatedhorses are susceptible to infection. In con-trast, infection with influenza induces long-term immunity independent of circulatingantibody against HA. For example, ponieswith low or undetectable anti-HA antibod-ies were clinically and virologically pro-tected from challenge infection more thanone year after natural infection [26]. Thissuggests an important difference in theimmune response following infection com-pared with vaccination using inactivatedvirus. Additional components of the immuneresponse that may be involved are the cel-lular immune and mucosal antibody responseslocal to the site of infection.

Cellular immune responses to influenzaare well defined in man. The key cell-medi-ated immune response is the developmentof MHC class I restricted CD8+ cytotoxicT lymphocytes (CTL), which are usuallydetectable within 3 to 4 days after infection.CD8+ CTL lyse virus-infected host cells[70]. The epitopes recognised by CTL onthe HA, nucleoprotein (NP), matrix (M1)and polymerase PB2 proteins are morehighly conserved than those involved inhumoral immunity. MHC class II-restrictedCD4+ T helper cells facilitate both humoral




and cellular immune responses and canexert cytolytic effects, though to a lesserextent than CD8+ CTL. Whereas antibodiesreduce virus load and restrict re-infection,cellular immune mechanisms probably play amore important role in clearance of virusduring the convalescent period [12, 36].Less is known about cellular immuneresponses in horses. Experimental infectionof ponies with influenza induces a geneti-cally restricted, antigen-specific CTL responsethat persists for at least six months [25].Generation of CTL in this case probablyoccurs through endogenous antigen process-ing followed by peptide presentation viaMHC class I molecules. In contrast, inacti-vated virus vaccines fail to stimulate a sig-nificant CTL response because the antigensundergo exogenous processing and presen-tation via MHC class II.

Equine influenza virus infection has beendemonstrated to generate virus-specific mucosalIgA and serum IgGa and IgGb responses,whereas an inactivated virus vaccine inducedonly a serum IgG(T) response [44].

The qualitative differences between theimmune responses that follow infection orvaccination with inactivated virus suggestthat improvements can be made in vaccinedesign. Ideally, vaccines should inducebroadly reactive, local and systemic, anti-body and cellular immune responses, estab-lish memory and consequently generate arapid anamnestic response upon field expo-sure to equine influenza virus. The inci-dence of free and cell-associated virus isthereby reduced and recovery enhanced.Live attenuated and live, vectored equineinfluenza vaccines that should more closelymimic natural infection are available. TheMerial vaccine PROTEQ Flu is a liverecombinant vaccine that uses canarypox asthe vector to express the HA genes ofequine influenza viruses. The recombinantvirus undergoes an abortive infection inmammalian cells so that no progeny virusesare made but the expressed viral antigensare processed endogenously and presentedas peptides via MHC class I by the host cell

in the same manner as occurs in naturalinfection but without associated infectionrisks. There is a wealth of evidence forcanarypox vaccines inducing cellular immuneresponses to human immunodeficiencyvirus in man [18, 20], but this has yet to bedemonstrated for the PROTEQ Flu vaccine.

A cold-adapted, temperature-sensitive,modified-live virus equine influenza vac-cine (FluAvert IN Vaccine), which is deliv-ered intranasally, is now licensed for sale inthe USA. The safety and efficacy of the vac-cine has been demonstrated in experimentalstudies, however the vaccine does not pro-vide sterile immunity [10, 34, 60, 71]. Nocorrelation was found between the concen-tration of serum antibody induced by vac-cination and protection against infection,though an anamnestic response was dem-onstrated at seven days post infection [61].Although there is evidence to show thatprimed animals will develop a serologicalresponse [71], it appears that the use ofserum antibody response as a measure oflive virus mucosal vaccines in naïve ani-mals is inappropriate. Our ability to meas-ure alternative correlates of immunity haslagged behind the development of thesealternative vaccination strategies.

Induction of a cellular immune responseto a conserved protein such as NP maypotentially provide protection when theviral strains incorporated in the vaccine donot match circulating strains. Such cross-reactive immunity may even extend to par-tial protection against infection with a virusof a different subtype (heterosubtypic immu-nity). Infection of mice with a human influ-enza A virus of one subtype can induce partialprotection against infection with virus of adifferent subtype [47], and a similar studyin pigs suggested that CD8+ T lymphocyteshave a role in this heterosubtypic immunity[27]. Generation of such cross-reactiveimmunity in the horse could be advanta-geous in the event of a new subtype of influ-enza A virus emerging (or re-emerging) inthe horse population.




5. OPTIMISING VACCINATION SCHEDULES

The early vaccination schedules forinactivated virus vaccines required two pri-mary doses 4 to 6 weeks apart followed byannual booster doses. The current mini-mum requirements imposed for competi-tion animals by the Federation EquinesInternational are a primary course of twodoses 4 to 6 weeks apart and a booster sixmonths later followed by annual boosters.Mathematical models validated against exper-imental and field data have demonstratedthat vaccination dramatically reduces boththe incidence and size of epidemics, withlarger outbreaks of equine influenza beingexceptional amongst groups of vaccinatedanimals [19]. Thus the vaccination policyensures a sufficient level of herd immunityto prevent large-scale outbreaks that arelikely to lead to cancellation of race meet-ings and other equestrian events. Howeverit is questionable whether the preliminaryprogramme of three doses followed byannual vaccination provides sufficient immu-nity to protect young horses from the dis-ease or individual training yards from smalloutbreaks of influenza. The short-livedimmunity provided by inactivated vaccineshas been acknowledged for some years, andit is apparent from various studies [13, 61,63] that vaccination in accordance with theminimum requirements of Jockey Clubrules and the vaccine manufacturer’s rec-ommendations leaves horses with low anti-body titres for several months between theirsecond and third vaccination. Newton et al.[46] found that SRH antibody levels inyearling Thoroughbreds on studs in New-market declined below a protective levelwithin four months of a booster vaccina-tion. Importantly, this also coincided withthe autumn sales, a recognised risk periodfor transmission of influenza in youngThoroughbreds [45]. Later observations inyearlings entering training yards in New-market confirmed that antibody levels atthis time were influenced by both timeelapsed since the last vaccination and the

total number of vaccines that had been pre-viously administered [46]. Cullinane et al.[13] demonstrated that an additional 6-monthly booster would benefit horses thatmay be at high risk during this interval.Intensive vaccination regimes, involvingbooster doses every 30 to 60 days, havebeen practised in the USA. However, littleis known about the potential adverse effectsof administering a potent vaccine too fre-quently, which may attenuate the immuneresponse. Using a stochastic model to assessthe risk of an outbreak occurring in a Thor-oughbred population in a typical flat racingtraining yard, Park et al. [51] suggested thatincreasing the frequency of vaccination inhorses aged 2-years and upwards to includesix monthly boosters would offer a signifi-cant increase in protection over annual vac-cination.

Timing of the first vaccination may becritical to the subsequent development ofantibody. Although it is recognised that mater-nal antibody generally inhibits the develop-ment of neonatal antibody synthesis, it hasoften been assumed that these antibodieshave decayed to an insignificant level by 3to 4 months. The temptation is to vaccinateelite stock prior to the loss of maternal anti-bodies to avoid any window of susceptibil-ity. Foals born to mares vaccinated duringthe gestation period have high levels ofmaternal antibody within two days of birth[13, 61, 63]. In contrast to Liu et al. [32],who reported that maternal antibody per-sisted for only a short period, severalauthors [13, 61, 63] found that the majorityof foals they tested had detectable (HI) anti-body titres at three months of age but thesehad virtually disappeared at six months.Cullinane et al. [13] suggested that not onlydoes vaccination in the face of maternalantibody interfere with the development ofactive immunity but that repeat vaccinationin the face of maternal antibodies may inducetolerance. On the basis of their findings,they recommended that mares should bevaccinated against equine influenza in thelast 6 to 4 weeks of pregnancy to ensure thetransfer of protective levels of antibody in




the colostrum, and that foals should not bevaccinated until their maternal antibodieshave waned (i.e. not until six months of ageor they are seronegative).

6. VACCINE STRAIN SELECTION

Surveillance of antigenic drift is a cor-nerstone of influenza control programmesbased on vaccination. As with other RNAviruses, influenza virus replication is highlyerror-prone, therefore newly synthesisedviral genes have a high frequency of muta-tion. Many of these mutations are eitherinconsequential or detrimental to the virus,but mutations affecting the antigenic sitesof the HA (and NA) can lead to the virus notbeing recognisable by pre-existing antibod-ies generated by infection or vaccinationwith an earlier strain, a process known as“antigenic drift”. The formulation of humaninfluenza vaccines is reviewed on an annualbasis and in most years is changed to reflectthe virus strains most representative ofthose in worldwide circulation.

Historically, antigenic drift in equineH3N8 viruses has been examined in HI testsemploying post infection or post vaccina-tion sera prepared in a number of differentspecies. Conclusions about the antigenicrelatedness of equine H3N8 viruses and thesignificance of observed differences withrespect to the immunity induced have var-ied. For example, Hinshaw et al. [28] con-cluded than the majority of viruses isolatedbetween 1979 and 1981 were substantiallydifferent from the prototype virus, Miami/63 included in the vaccine when comparedusing post infection ferret sera in HI assays,and that representatives of the new variantshould be included in the vaccines. On theother hand, Burrows et al. [6, 7] concludedthat the minor antigenic drift that theydetected in viruses isolated between 1963and 1979 did not justify a change in vaccinestrains because post vaccination sera fromhorses immunised with Miami/63 viruswere highly cross-reactive in HI tests withviruses from 1979. This conclusion did not

take into account the findings of Haaheimand Schild [24] that strain-specific anti-body is more effective than cross-reactiveantibody in conferring protection.

Horse sera are relatively cross-reactive,particularly when taken from repeatedlyvaccinated animals whereas ferrets developa more strain-specific antibody response [39].

During the 1989 outbreak of influenza inthe UK, only horses with very high levelsof vaccine-induced antibody were pro-tected against infection, raising the possi-bility that there had been significant anti-genic changes in the 1989 isolate thatprevented its neutralisation by antibodystimulated by vaccines containing Miami/63,Fontainebleau/79 or Kentucky/81. Sequenc-ing of the HA1 gene and antigenic analysisusing monoclonal antibodies suggested thatthere were significant differences betweena representative 1989 strain and the vaccinestrains in current use at the time [2]. Thehypothesis was tested by vaccinating groupsof ponies with monovalent vaccines con-taining either of the vaccine strains or a1989 strain and experimentally challengingthem with a 1989 virus [15]. Although allvaccines provided clinical protection, vac-cine efficacy in terms of ability to eliminatevirus excretion correlated directly with thedegree of antigenic relatedness betweenvaccine and challenge strain. Following ameeting of OIE and WHO experts on newlyemerging strains of equine influenza, it wasrecommended that equine influenza vac-cines be updated to include a 1989 isolate,and that efforts be made to increase surveil-lance and virus characterisation [40].

Phylogenetic analysis of HA sequencesrevealed that equine H3N8 viruses, whichhad been evolving as a single lineage [29],apparently diverged into two distinct line-ages during the mid-1980s [14] and, todate, both lineages continue to co-circulateindependently (Fig. 1). Viruses in one lin-eage were predominantly isolated fromhorses in Europe, with the exception of onevirus isolated in Canada in 1990, whereas




viruses in the other lineage were predomi-nantly from horses on the American conti-nent. It was apparent, however, that Americanlineage viruses had been introduced intoEurope on at least one occasion. Thegenetic divergence of American and Euro-pean lineage viruses was reflected in theirantigenic reactivity, raising the question ofthe potential importance of geographicalvariations in antigenic character for vaccineefficacy. Further vaccination and experi-mental challenge studies in ponies sug-gested that vaccines containing virus from

the American lineage may not be as effec-tive in protecting against infection as thehomologous vaccine against challenge withvirus from the European lineage [69]. Fieldobservations have supported the hypothesisthat antigenic differences between virusesof the American and European lineages aresufficient to adversely affect vaccine efficacy.During an outbreak caused by a Europeanlineage virus in vaccinated Thoroughbredhorses in the UK in 1995, horses with anti-body levels of more than around 140 mm2 wereprotected against infection [46]. However,

Figure 1. Phylogenetic tree constructed from equine influenza H3 HA1 amino acid sequences usingparsimony method.




during an outbreak caused by an Americanlineage virus in 1998, when the vaccinesused contained only European lineage viruses,a quarter of horses with antibody levelshigher than 140 mm2 became infected [45].

The co-circulation of antigenic variantsmeans that it is important to base the selec-tion of new vaccine strains on knowledge ofthe dominant virus circulating in the field.Following a further consultation of OIE andWHO experts in 1995, a more formal sur-veillance system was established for equineinfluenza [48]. An international panel ofexperts including representatives from OIEand WHO influenza reference laboratoriesreviews data collected on outbreaks ofinfluenza, vaccine performance in the field,and antigenic and genetic characteristics ofnew virus isolates annually. The expert sur-veillance panel make recommendations onthe need to update vaccine strains, whichare published in the OIE Bulletin. The cri-teria used for deciding on the need to updateequine influenza vaccine strains are basedlargely on those used for human influenzavaccine strain selection, i.e. detection ofchanges in the HA as characterised by HItests using ferret and horse antisera, geneticsequencing of the HA1 gene and vaccinebreakdown in the field. Improved surveil-lance in the field, standardisation of thepotency of vaccines and the introduction ofa vaccine strain selection system has ena-bled the development of a fast-track licens-ing system for vaccines containing updatedstrains [16].

7. DIAGNOSIS

We have demonstrated that vaccinatedhorses are often only partially immune toinfluenza (particularly if vaccine strains area poor match for circulating viruses) andmay shed virus in the absence of clinicalsigns. Such animals present a significantrisk for the spread of infection. Thus ourability to diagnose both clinical and subclinical infections in partially immune ani-

mals is critical in attempts to control equineinfluenza.

For many years the diagnosis of equineinfluenza has relied on culture of virus inembryonated hens’ eggs (and more recentlyMadin-Darby canine kidney cells) andmeasurement of antibody responses to theHA. Although a useful epidemiologicaltool, serological diagnosis of equine influ-enza tends to be retrospective because aconvalescent sample taken around twoweeks after an acute sample is required fora definitive diagnosis. This is becauseinfection-induced antibody detected in anacute sample cannot be distinguished fromvaccine-induced antibody.

An ELISA to detect antibody to the non-structural protein NS1 has been developed[3, 50]. As this protein is produced duringan infection but is not incorporated intoinactivated whole virus vaccines, it theoret-ically enables differentiation of antibodyresponses to infection from responses tovaccination with a traditional vaccine. Withthe introduction of live attenuated equineinfluenza vaccines, the potential usefulnessof this test for confirmation of infection invaccinated animals will probably be con-siderably reduced. However, the currenttrend towards genetically engineered vac-cines may facilitate the development ofDIVA (differentiation of infected from vac-cinated animals) vaccines in which a spe-cific gene encoding a highly immunogenicprotein is modified or removed.

Detection of the presence of infectiousvirus by culture of virus in nasal secretionscan take a minimum of 2 or 3 days, and ifmultiple passages are required confirma-tion of diagnosis is delayed further. Anumber of alternative assays based upon theuse of a monoclonal antibody to detectnucleoprotein in nasal swab abstract pro-vide a diagnosis within 24 h. An equineinfluenza-specific ELISA has been described[11]. When used in parallel with virus iso-lation during the 1989 equine influenza epi-demic in Britain, the ELISA enhanced thevirus detection rate by 44% [33]. Kits for




detection of human influenza are commer-cially available, and, because of the highdegree of conservation of the nucleoproteinamong influenza A viruses, one of these, theDirectigen Flu-A assay, has been shown tobe applicable to the diagnosis of equineinfluenza [9]. These direct detection meth-ods are useful in the application of controlmeasures, as they can be used as a basis forisolating horses excreting virus in order toreduce infection pressure and for a decisionon curtailing exercise, which may exacer-bate disease. They are also a useful adjunctto virus isolation, which remains essentialfor characterising new viruses and to pro-vide future vaccine strains, as they permitvirus isolation efforts to be focussed onsamples known to be positive for equineinfluenza.

8. INTERNATIONAL CONTROL

The ever-increasing international move-ment of horses for competition and breed-ing purposes presents a challenge withregard to the control of equine influenza.Several explosive outbreaks of equine influ-enza attributable to the introduction ofinfected animals into susceptible indige-nous populations have been described dur-ing the last 20 years [54, 66]. Due to eco-nomic and competitive issues, it is desirablefor the disruption to training programmescaused by quarantine to be kept to a mini-mum when horses are moved. There is,therefore, a reliance on surveillance ofinfluenza in the population that animals areleaving and on the effectiveness of vaccinesto prevent viral shedding. When thesemeasures fail, and subclinically infectedhorses shedding virus are transported, theshort quarantine periods that are often usedfail to prevent introduction of infection.

Regulations relating to the movement ofanimals based on the use of improved diag-nostic techniques and vaccination policiesthat recognise the limitations of current prod-ucts are now in place. The Code Commis-sion of the OIE recommends that importing

countries that are free of equine influenzashould require that all horses travellingfrom endemic areas are fully vaccinatedand have received their last booster dosewithin 2 to 8 weeks of travel [49]. A simpleadditional measure that can be imple-mented is the screening of antibody usingthe SRH assay, which can identify poten-tially susceptible animals that require re-vaccination to boost their antibody levelsbefore travelling. The advent of more rapiddiagnostic tests for equine influenza meansthat animals can be screened for viral shed-ding while still in quarantine at their desti-nation before being released into potentiallysusceptible local populations.

9. CONCLUSION

There are still important goals to be metin the control of equine influenza. Theseinclude increased surveillance, virus recov-ery and characterisation from large equinepopulations in the Americas and Far East,and international harmonisation of vaccinestandards and licensing procedures. How-ever, many of the activities are now in placeto provide vaccine manufacturers with thenecessary information for production ofeffective vaccines containing epidemiolog-ically relevant strains, and the developmentof rapid diagnostic assays has increased ourability to monitor equine influenza activityworldwide and avoid transmission of infec-tion via movement of horses from areaswhere the infection is active.

ACKNOWLEDGEMENTS

Much of the data presented in this paper havebeen generated through the collaborative effortsof research teams at the Animal Health Trust,which currently includes A. Park, L. Spencer andD. Hannant, and at the Gluck Equine ResearchCentre, University of Kentucky, USA, headedby T. Chambers. The authors greatly appreciatethe continued financial support of the HorseraceBetting Levy Board, Animal Health Trust, andthe equine influenza vaccine manufacturers.




REFERENCES

[1] Askonas B.A., McMichael A.J., Webster R.G.,The antibody response to influenza virus, in:Beare A.S. (Ed.), Basic and Applied InfluenzaResearch, CRC Press, Boca Raton, Florida,1982, pp. 164–181.

[2] Binns M.M., Daly J.M., Chirnside E.D.,Mumford J.A., Wood J.M., Richards C.M.,Daniels R.S., Genetic and antigenic analysisof an equine influenza H3 isolate from the1989 epidemic, Arch. Virol. 130 (1993) 33–43.

[3] Birch-Machin I., Rowan A., Pick J., MumfordJ., Binns M., Expression of the nonstructuralprotein NS1 of equine influenza A virus:detection of Anti-NS1 antibody in post infec-tion equine sera, J. Virol. Methods 65 (1997)255–263.

[4] Bryans J., Control of equine influenza, in:Bryans J.T. (Ed.), Proc. 1st Int. Conference onEquine Infectious Diseases, Stresa, Italy,1966, pp. 157–165.

[5] Bryans J.T., The antibody response of horsesto two inactivated oil adjuvanted equine influ-enza virus vaccines, in: International sympo-sium on influenza vaccines for men andhorses, Paris, France, 1973, pp. 311–317.

[6] Burrows R., Denyer M., Antigenic propertiesof some equine influenza viruses, Arch. Virol.73 (1982) 15–24.

[7] Burrows R., Denyer M., Goodridge D., HamiltonF., Field and laboratory studies of equineinfluenza viruses isolated in 1979, Vet. Rec.109 (1981) 353–356.

[8] Cameron T.P., Alford R.H., Kasel J.A., HarveyE.W., Byrne R.J., Knight V., Experimentalequine influenza in Chincoteague ponies,Proc. Soc. Exp. Biol. Med. 124 (1967) 510–515.

[9] Chambers T.M., Shortridge K.F., Li P.H.,Powell D.G., Watkins K.L., Rapid diagnosisof equine influenza by the Directigen FLU-Aenzyme immunoassay, Vet. Rec. 135 (1994)275–279.

[10] Chambers T.M., Holland R.E., Tudor L.R.,Townsend H.G.G., Cook A., Bogdan J., LunnD.P., Hussey S., Whitaker-Dowling P.,Youngner J.S., Sebring R.W., Penner S.J.,Stiegler G.l., A new modified live equineinfluenza virus vaccine: phenotypic stability,restricted spread and efficacy against heterolo-gous virus challenge, Equine Vet. J. 33 (2001)630–636.

[11] Cook R.F., Sinclair R., Mumford J.A., Detec-tion of influenza nucleoprotein antigen innasal secretions from horses infected with A/equine influenza (H3N8) viruses, J. Virol.Methods 20 (1988) 1–12.

[12] Couch R.B., Kasel J.A., Immunity to influ-enza in man, Annu. Rev. Microbiol. 37 (1983)529–549.

[13] Cullinane A., Weld J., Osborne M., Nelly M.,McBride C., Walsh C., Field studies on equineinfluenza vaccination regimes in Thorough-bred foals and yearlings, Vet. J. 161 (2001)174–185.

[14] Daly J.M., Lai A.C.K., Binns M.M., ChambersT.M., Barrandeguy M., Mumford J.A., Anti-genic and genetic evolution of equine H3N8influenza A viruses, J. Gen. Virol. 77 (1996)661–671.

[15] Daly J.M., Yates P.J., Browse G., Swann Z.,Newton J.R., Jessett D., Davis-Poynter N.,Mumford J.A., Comparison of hamster andpony challenge models for evaluation of effectof antigenic drift on cross-protection affordedby equine influenza vaccines, Equine Vet. J.35 (2003) 458–462.

[16] EMEA/CVMP/112/98, Notes for guidance:harmonisation of requirements for equineinfluenza vaccines-specific requirements forsubstitution or addition of a strain or strains,1998.

[17] European Pharmacopoeia, Monograph onequine influenza vaccines, 3rd ed., 1998.

[18] Evans T.G., Kallas E.G., Campbell M., AndrewsJ.D., Schwartz D., Keefer M., Caudrelier P.,Evaluation of canarypox-induced CD8(+)responses following immunization be measur-ing the effector population IFNgamma pro-duction, Immunol. Lett. 77 (2001) 7–15.

[19] Glass K., Wood J.L.N., Mumford J.A., JessettD., Grenfell B.T., Modelling equine influenza 1:a stochastic model of within-yard epidemics,Epidemiol. Infect. 128 (2002) 491–502.

[20] Gorse G.J., Patel G.B., Belshe R.B., HIV type1 vaccine-induced T cell memory and cyto-toxic T lymphocyte responses in HIV type 1-uninfected volunteers, AIDS Res. Hum. Ret-roviruses 17 (2001) 1175–1189.

[21] Guo Y., Wang M., Zheng S., Wang P., Ji W.,Chen Q., Aetiologic study on an influenza-like epidemic in horses in China, Acta Virol.35 (1991) 190–195.

[22] Guo Y., Wang M., Kawaoka Y., Gorman O.,Ito T., Saito T., Webster R.G., Characteriza-tion of a new avian-like influenza A virus fromhorses in China, Virology 188 (1992) 245–255.

[23] Guo Y., Wang M., Zhang G-S., Li W.-K.,Kawaoka Y., Webster R.G., Seroepidemio-logical and molecular evidence for the presenceof two H3N8 equine influenza viruses inChina in 1993–1994, J. Gen. Virol. 76 (1995)2009–2014.




[24] Haaheim L.R., Schild G.C., Antibodies to thestrain-specific and cross-reactive determinantsof the haemagglutinin of influenza H3N2viruses. 2. Antiviral activities of the antibod-ies in biological systems, Acta Pathol. Micro-biol. Scand. B 88 (1980) 335–340.

[25] Hannant D., Mumford J.A., Cell mediatedimmune responses in ponies following infec-tion with equine influenza virus (H3N8): theinfluence of induction culture conditions onthe properties of cytotoxic effector cells, Vet.Immunol. Immunopathol. 21 (1989) 327–337.

[26] Hannant D., Mumford J.A., Jessett D.M.,Duration of circulating antibody and immu-nity following infection with equine influenzavirus, Vet. Rec. 122 (1988) 125–128.

[27] Heinen P.P., de Boer-Luijtze E.A., BianchiA.T.J., Respiratory and systemic humoral andcellular immune responses of pigs to a heter-osubtypic influenza A virus infection, J. Gen.Virol. 82 (2001) 2697–2707.

[28] Hinshaw V.S., Naeve C.W., Webster R.G.,Douglas A., Skehel J.J., Bryans J., Analysis ofantigenic variation in equine 2 influenza Aviruses, Bull. World Health Organ. 61 (1983)153–158.

[29] Kawaoka Y., Bean W.J., Webster R.G., Evo-lution of the hemagglutinin of equine H3influenza viruses, Virology 169 (1989) 283–292.

[30] Klingeborn B., Rockborn G., Dinter Z., Sig-nificant antigenic drift within the influenzaequi 2 subtype in Sweden, Vet. Rec. 106(1980) 363–364.

[31] Kumanomido T., Akiyama Y., Immuno-effect of serum and nasal antibody againstexperimental inoculation with influenza A-Equi-2 virus, Exp. Rep. Equine Health Lab. 12(1975) 44–52.

[32] Liu I.K.M., Pascoe D.R., Chang L.W.S., ZeeY.C., Duration of maternally derived antibod-ies against equine influenza in newborn foals,Am. J. Vet. Res. 46 (1985) 2078–2080.

[33] Livesay G.J., O’Neill T., Hannant D., YadavM.P., Mumford J.A., The outbreak of equineinfluenza (H3N8) in the United Kingdom in1989: diagnostic use of an antigen captureELISA, Vet. Rec. 133 (1993) 515–519.

[34] Lunn D., Hussey S., Sebring R., Rushlow K.,Radecki S., Whitaker-Dowling P., Youngner J.,Chambers T., Holland R., Horohov D., Safety,efficacy, and immunogenicity of a modified-live equine influenza virus vaccine in poniesafter induction of exercise-induced immuno-suppression, J. Am. Vet. Med. Assoc. 218(2001) 900–906.

[35] Martens J.G., Development and evaluation ofaerosol delivery of antivirals for the treatment

of equine virus induced respiratory infections,Diss. Abstr. Int. 47 (1986) 69.

[36] McMichael A.J., Gotch F.M., Cullen P.,Askonas B.A., Webster R.G., Cytotoxic T cellimmunity to influenza, N. Engl. J. Med. 309(1983) 13–17.

[37] Morley P.S., Townsend H.G.G., Bogdan J.R.,Haines D.M., Efficacy of a commercial vac-cine for preventing disease caused by influ-enza virus infection in horses, J. Am. Vet.Med. Assoc. 215 (1999) 61–64.

[38] Mumford J., Collaborative study for the estab-lishment of three European PharmacopoeiaBiological Reference Preparations for equineinfluenza horse antiserum, Pharmeuropa(2000) 7–21.

[39] Mumford J., Progress in the control of equineinfluenza, in: Plowright W., Rossdale P.D.,Wade J.F. (Eds.), Proc. 6th Int. Conference onEquine infectious diseases, Cambridge, 1992,pp. 207–218.

[40] Mumford J.A., Wood J., Conference report onWHO/OIE meeting: Consultation on newlyemerging strains of equine influenza, Vaccine11 (1993) 1172–1175.

[41] Mumford J.A., Wood J.M., Folkers C., SchildG.C., Protection against experimental infec-tion with influenza virus A/equine/Miami/63(H3N8) provided by inactivated wholevirus vaccines containing homologous virus,Epidemiol. Infect. 100 (1988) 501–510.

[42] Mumford J.A., Jessett D.M., Rollinson E.A.,Hannant D., Draper M.E., Duration of protec-tive efficacy of equine influenza immunos-timulating complex/tetanus vaccines, Vet.Rec. 134 (1994) 158–162.

[43] Mumford J.A., Wilson H., Hannant D., JessettD.M., Antigenicity and immunogenicity ofequine influenza vaccines containing a Car-bomer adjuvant, Epidemiol. Infect. 112 (1994)421–437.

[44] Nelson K.M., Schram B.R., McGrergorM.W., Sheoran A.S., Olsen C.W., Lunn D.P.,Local and systemic isotype-specific antibodyresponses to equine influenza virus infectionversus conventional vaccination, Vaccine 16(1998) 1306–1313.

[45] Newton J.R., Verheyen K., Wood J.L.N.,Yates P.J., Mumford J.A., Equine influenza inthe United Kingdom in 1998, Vet. Rec. 145(1999) 449–452.

[46] Newton J.R., Lakhani K.H., Wood J.L.N.,Baker D.J., Risk factors for equine influenzaserum antibody titres in young Thoroughbredracehorses given an inactivated vaccine, Prev.Vet. Med. 46 (2000) 129–141.




[47] Nguyen H.H., Moldeveanu Z., Novak M.J.,van Ginkel F.W., Ban E., Kiyono H., McGheeJ.R., Mestecky J., Heterosubtypic immunityto lethal influenza A virus infection is associ-ated with virus-specific CD8+ cytotoxic T lym-phocyte responses induced in mucosa-associ-ated tissues, Virology 254 (1999) 50–60.

[48] OIE, Conclusions and recommendations fromthe consultation meeting of OIE and WHOexperts on equine influenza, Newmarket,United Kingdom, September 18–19, 1995,OIE Bull. 108 (1996) 482–484.

[49] OIE Code, International Animal Health Code,7th ed., Office International des Épizooties,Paris, France, 1998.

[50] Ozaki H., Sugiura T., Sugita S., Imagawa H.,Kida H., Detection of antibodies to the non-structural protein (NS1) of influenza A virusallows distinction between vaccinated andinfected horses, Vet. Microbiol. 82 (2001)111–119.

[51] Park A.W., Wood J.L.N., Newton J.R., DalyJ., Mumford J.A., Grenfell B.T., Optimisingvaccination strategies in equine influenza,Vaccine 21 (2003) 2862–2870.

[52] Plateau E., Crucière C., Virat J., Benazet P.,Grippe équine : isolement, caractérisation etétude sérologique dans divers foyers au coursde l’épizootie 1978-1979, Bull. Acad. Vet. Fr.52 (1979) 189–194.

[53] Powell D.G., Burrows R., Spooner P., MumfordJ., Thomson G., Field observations on influ-enza vaccination among horses in Britain,1971–1976, Dev. Biol. Stand. 39 (1977) 347–352.

[54] Powell D.G., Watkins K.L., Li P.H., ShortridgeK.F., Outbreak of equine influenza amonghorses in Hong Kong during 1992, Vet. Rec.136 (1995) 531–536.

[55] Rouse B.T., Ditchfield J.B., The response ofponies to Myxovirus influenzae A-Equi-2 III.The protective effect of serum and nasal anti-body against experimental challenge, Res.Vet. Sci. 11 (1970) 503–507.

[56] Rouse B.T., Ditchfield W.J.B., The responseof ponies to Myxovirus influenzae A-equi 2 I.Serum and nasal antibody titres followingexposure, Can. J. Comp. Med. 34 (1970) 1–6.

[57] Scholtens R.G., Steele J.H., U.S. epizootic ofequine influenza, 1963: epizootiology, PublicHealth Rep. Washington 79 (1964) 393–398.

[58] Sovinová O., Tumová B., Pouska F., NemecJ., Isolation of a virus causing respiratory dis-ease in horses, Acta Virol. 2 (1958) 51–61.

[59] Thomson G.R., Spooner P.R., Powell D.G.,The outbreak of equine influenza in England:January 1976, Vet. Rec. 100 (1977) 465–468.

[60] Townsend H.G.G., Penner S.J., Watts T.C.,Cook A., Bogdan J., Haines D.M., Griffin S.,Chambers T., Holland R.E., Whitaker-DowlingP., Youngner J.S., Sebring R.W., Efficacy ofa cold-adapted, intranasal equine influenzavaccine: challenge trials, Equine Vet. J. 33(2001) 637–643.

[61] Van Maanen C., Bruin G., de Boer-Luijtze E.,Smolders G., de Boer G.F., Interference ofmaternal antibodies with the immune responseof foals after vaccination against equine influ-enza, Vet. Q. 14 (1992) 13–17.

[62] Van Oirschot J.T., Masurel N., HuffelsA.D.N.H., Anker W.J.J., Equine influenza inthe Netherlands during the winter of 1978–1979; antigenic drift of the A-equi 2 virus,Vet. Q. 3 (1981) 81–84.

[63] Van Oirschot J.T., Bruin G., de Boer-LuytzeE., Smolders G., Maternal antibodies againstequine influenza virus in foals and their inter-ference with vaccination, Zentralbl. Veteri-narmed. B 38 (1991) 391–396.

[64] Waddell G.H., Teigland M.B., Sigel M.M., Anew influenza virus associated with equinerespiratory disease, J. Am. Vet. Med. Assoc.143 (1963) 587–590.

[65] Webster R.G., Yuanji G., New influenza virusin horses, Nature 351 (1991) 527.

[66] Wernery R., Yates P.J., Wernery U., MumfordJ.A., An equine influenza outbreak in a poloclub in Dubai, United Arab Emirates in 1995/96, in: Wernery U., Wade J.F., Mumford J.A.,Kaaden O.-R. (Eds.), Proc. 8th Int. Confer-ence on Equine Infections Diseases, Dubai,1998, pp. 342–346.

[67] Wood J.M., Mumford J., Folkers C., ScottA.M., Schild G.C., Studies with inactivatedequine influenza vaccine: I. Serological responsesof ponies to graded doses of vaccine, J. Hyg.(Lond.) 90 (1983) 371–384.

[68] Wood J.M., Schild G.C., Folkers C., MumfordJ., Newman R.W., The standardization ofinactivated equine influenza vaccines by sin-gle-radial immunodiffusion, J. Biol. Stand. 11(1983) 133–136.

[69] Yates P., Mumford J.A., Equine influenzavaccine efficacy: the significance of antigenicvariation, Vet. Microbiol. 74 (2000) 173–177.

[70] Yewdell J.W., Hackett C.J., The specificityand function of T lymphocytes induced byinfluenza A viruses, in: Krug R. (Ed.), Theinfluenza viruses, Plenum Press, New York/London, 1989, pp. 361–429.

[71] Youngner J.S., Whitaker-Dowling P., ChambersT.M., Rushlow K.E., Sebring R., Derivationand characterization of a live attenuatedequine influenza vaccine virus, Am. J. Vet.Res. 62 (2001) 1290–1294.



Influensautbrottet i Sverige 2007

Documents

Transcript of Influensautbrottet i Sverige 2007