EU FMD-NSP workshop 1 TAIEX, EUFMD (FAO), EPIZONE & EU COORDINATION ACTION FMD-CSF Workshop on the design and interpretation of post Foot-and-Mouth Disease (FMD)-vaccination serosurveillance by NSP tests Part III (Scandinavian and Baltic region) October 23-25, 2007 Report VAR-CODA-CERVA Leuvensesteenweg 17 B-3080 Tervuren Belgium VAR Belgium EUFMD Belgian Food Agency DG RESEARCH DG SANCO
145
Embed
FMD-NSP 3 report final · EU FMD-NSP workshop 2 Contents Page Abstract 3 EU FMD-NSP workshop organising committee 6 EU FMD-NSP workshop supervisory team 6 Participants 6 Observers
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
EU FMD-NSP workshop 1
TAIEX, EUFMD (FAO), EPIZONE &
EU COORDINATION ACTION FMD-CSF
Workshop on the design and interpretation of post Foot-and-Mouth Disease (FMD)-vaccination serosurveillance by
NSP tests Part III (Scandinavian and Baltic region)
October 23-25, 2007
Report
VAR-CODA-CERVA Leuvensesteenweg 17
B-3080 Tervuren Belgium
VAR Belgium
EUFMD Belgian Food Agency
DG RESEARCH DG SANCO
EU FMD-NSP workshop 2
Contents Page Abstract 3 EU FMD-NSP workshop organising committee 6 EU FMD-NSP workshop supervisory team 6 Participants 6 Observers 6 Sponsors 6 Objectives 7 Report 8
Opening session 8 Session I: Review on FMD vaccination and NSP testing 10 Session II: Review on statistical computer tools, OIE guidelines, EU Directives and carrier problem
19
Session III: Serosurveillance post-vaccination: theory and practice 36 Session IV and V: Results from working groups 71 Session VI: Discussion of results and models for serosurveillance – Summary, Conclusions, Recommendations and Observations
84
Acknowledgments 91 Annex A: FMD-NSP workshop program 92 Annex B: presentation of scenario 1 by group 1 93 Annex C: presentation of scenario 1 by group 2 102 Annex D: presentation of scenario 2 by group 3 108 Annex E: presentation of scenario 2 by group 4 116 Annex F: presentation of scenario 3 by group 5 122 Annex G: presentation of scenario 3 by group 6 129 Annex H: Post NSP-workshop thoughts – Dr. David Paton 144
EU FMD-NSP workshop 3
Abstract Given the current FMD-free status without vaccination in Europe and the possibility of a future outbreak with vaccinate-to-live used as an emergency measure, followed-up by a post-vaccination serosurveillance, to return to the status ‘free from infection without vaccination’, this workshop, for countries from the Scandinavian, Baltic and neighbouring area , had two main objectives, viz.: I. on day 1, to make participants familiar with:
1. NSP tests available, other relevant tests and their use 2. computer tools for the calculation of the required number of samples to be taken 3. the legislation relevant for NSP testing 4. the detection of carrier animals: relevant potential control measures 5. the requirements to personnel (field workers) and lab resources (tests and personnel).
II. and on day 2 and 3, concentrate on: 1. The design and implementation of a survey to substantiate free from infection with a certain
degree of confidence after vaccination has been performed. 2. The guidance to the interpretation on the follow-up on seropositive animals/herds/flocks; 3. The guidance to the use of laboratory test results in decision-making; 4. The identification of the resources (laboratory, veterinarians) required.
Three different exercise scenarios were written, including:
(a) a scenario with a limited number of outbreaks where only bovine and pigs were vaccinated, (b) a second scenario with a limited number of outbreaks, one in a very peculiar and big pig holding,
where only bovine, and pigs were vaccinated, and (c) a third scenario with a large number of outbreaks where only bovine were vaccinated.
Seventeen invited countries, including Belarus, Czech Republic, Estonia, Finland, Iceland, Israel, Latvia, Lithuania, Luxemburg, Moldova, Norway, Poland, Russian Federation, Slovakia, Sweden, Switzerland and Ukraine (of which Belarus, Iceland, Lithuania and Russian Federation did not participate) and observers of OIE, EFSA and EPIZONE were divided in 6 groups, each group having to make the best possible survey design (which was not necessarily the design proposed by OIE Guidelines or EU Directives), to follow-up the seropositive herds/animals/flocks and to do an identification of the necessary resources for their own scenario. Summary of workshop conclusions: The approaches taken by the different working groups showed a clear degree of similarity. Summary of conclusions:
1. The vaccination-to-live policy with subsequently substantiating freedom from infection by a survey system including NSP testing is a realistic and achievable option in FMD control. However, stamping out will always remain part of the control policy;
2. Because NSP assays are not sensitive enough (especially for carriers), conclusions on the infection status of the vaccinated herds can only be based on a combination of clinical and serological surveys and epidemiological investigations such as cluster analysis;
3. Proving freedom from infection for vaccinated animals is impossible, in contrast to substantiating freedom from virus circulation or freedom from infection in non-vaccinated animals;
EU FMD-NSP workshop 4
4. The current EU Directive (2003/85/EC) for the control of FMD mentions two surveys, i.e.: (a) A survey for detecting the presence of FMD virus (Article 56) in the vaccination zone, which
should be a combination of clinical, epidemiological and serological investigations with high overall system sensitivity, which includes: • a survey of the non-vaccinated animals; • a serosurveillance of all herds with vaccinated animals (NSP tests). Within herds
sampling of all vaccinated ruminants and their non-vaccinated off-spring; For large numbers of other species, sampling should be based on a 5% prevalence with 95% confidence.
(b) A survey, to regain freedom from infection after emergency vaccination(Article 61), which must have a high specificity. This survey might include a second serosurveillance, but the first serosurvey could serve the purpose of this one (cf. OIE Guidelines in App. 3.8.7. of the TAHC);
5. The follow-up of herds with seroreactors by serological investigation has to be based on NSP assays (Paton et al., 2006) with well-defined performance characteristics.
6. If specificity of the serological test system were known, only seroreactor rates above the Herd Cut Point could be considered, but this is not compatible with EU directive 2003/85/EC.
7. A clinical surveillance combined with paired serology can detect holdings where virus circulation is ongoing. Since there is no possibility of detecting each and every carrier within sub-clinically infected herds, all ruminants should be tested. Evidence of virus circulation would lead to herd slaughter, but evidence of carriers would lead to slaughter of these reactor animals only;
8. Testing all animals in the vaccinated population as prescribed in the EU Directive 2003/85/EC, is considered as not achievable in areas with a dense pig population or within big pig herds, if such pigs have been vaccinated.
9. Vaccination of small herds remains a controversial item. Two possible options were discussed: (a) a non-vaccination policy for small herds with integration in the survey system as sentinels or (b) a vaccination policy for small herds because of their contribution in achieving the necessary
level for population protection and because of political reasons. 10. This kind of workshop should be done for other veterinary diseases, like CSF and AI also (for all
EU members, EUFMD countries and EU neighbours).
Summary of recommendations: 1. Conclusions on the infection status of the herds after FMD outbreaks in a vaccinated population
should only be based on a survey system, including at least clinical, serological and epidemiological investigations;
2. The performance characteristics of the survey system should be determined; 3. The term ‘demonstrate absence’ should be replaced by ‘substantiate absence’; 4. Contingency plans should include a clear flow chart for the follow-up of seropositive herds
(minimum requirements of App. 3.8.7. of TAHC) 5. All large ruminants should be tested to substantiate freedom from infection in a vaccinated
population after FMD outbreaks. While evidence of virus circulation must lead to the declaration of an outbreak, consensus should be sought on the slaughter of reactor animals only, in case it is evident that these animals are carriers;
6. A change in the definition of an outbreak in OIE guidelines and EU Directives is needed where carriers are concerned;
EU FMD-NSP workshop 5
7. The relative confidence attainable with “herd-based” and “individual” certification needs to be explored for different herd sizes and prevalence;
8. Consideration should be given to an amendment of the Directive in order to allow a within-herd sampling scheme based on a 5% prevalence and 95% confidence for vaccinated pigs;
9. The vaccination of small herds should be further discussed; 10. To refine the application of NSP tests, more work could be done in predicting the expected
prevalence of infection within and amongst vaccinated herds; 11. Functional FMD expert groups should be created in every country;
Summary of workshop observations:
Following points were specifically highlighted during the second and third workshop:
1. The different species analysed in the survey system should be taken into account to determine the performance characteristics of the survey system.
2. Containment of an infection region is possible since the change of the OIE code. 3. Training for the set-up of an information system for field data is necessary for some countries. 4. The whole scenario assumes a perfect movement control, which is not a real life situation. 5. Probang testing is not the method of choice to substantiate freedom from infection. 6. The introduction of negative animals as sentinels in a vaccinated herd is of limited value due to the
low transmission rate in these herds, as well for cattle as for pigs. 7. Vaccination is not a major option in countries with a lot of small herds 8. The organisation of a workshop on ‘vaccination: how and when, after or in face of an outbreak’
should be considered. 9. SP-tests can also be used to check the efficiency of and coverage obtained through vaccination. 10. If possible, slaughter for consumption should be considered. 11. Sub-clinical transmission in cattle and pigs is unlikely. 12. Surveillance in the surveillance zone could be of help to substantiate absence of infection. 13. It is essential to have a decision-scheme on the follow-up before the start of the outbreak. 14. If the 95% confidence cannot be achieved for a survey then it is still worth doing it.
EU FMD-NSP workshop 6
EU FMD-NSP Workshop Organizing Committee See Annex A: FMD-NSP workshop program
EU FMD-NSP Workshop Supervisory Team See Annex A: FMD-NSP workshop program
Lithuania, Luxemburg, Moldova, Norway, Poland, Russian Federation, Slovakia, Sweden, Switzerland and Ukraine were invited. Except for Belarus, Iceland, Lithuania and Russian Federation, every invited country has sent a team to participate to the workshop.
• Participant list: see Annex A: FMD-NSP workshop program
Observers See Annex A: FMD-NSP workshop program
Sponsors
The workshop was kindly sponsored financially by TAIEX, FAO EUFMD, the EU Coordination action FMD-CSF project, the EU EPIZONE project Work Package 4.3 (DIVA) and logistically by the Veterinary Administration of the Republic of Slovenia, the Federal Agency for the Safety of the Food Chain in Belgium and the Veterinary and Agrochemical Research Centre in Belgium.
EU FMD-NSP workshop 7
Objectives
The objectives of the third FMD-NSP workshop were two-fold: I. To make participants familiar with:
6. NSP tests available, other relevant tests and their use 7. computer tools for the calculation of the required number of samples to be taken 8. the legislation relevant for NSP testing 9. the detection of carrier animals: relevant potential control measures 10. the requirements to personnel (field workers) and lab resources (tests and personnel).
II. Given the current FMD-free status without vaccination in Europe and the possibility of a future outbreak with vaccinate-to-live used as an emergency measure, being followed-up by post-vaccination serosurveillance, to return to the favoured status of ‘free from infection without vaccination’: 1. Design and implementation of a survey to substantiate free from infection (i.e. free from
infection with a certain degree of confidence, according to the EU directives and the OIE general and specific serosurveillance guidelines). So a country has an outbreak (or different outbreaks), vaccination of the surrounding herds is performed, the outbreaks are stopped, what will this country do to regain the ‘free from FMD without vaccination’ status?
2. Guidance to the interpretation on follow-up on seropositive animals/herds/flocks, for example by using 3 different exercise scenario’s, where we will have 3 different conclusions on the last day.
3. Guidance to the use of laboratory test results in decision-making 4. Identification of the resources required
EU FMD-NSP workshop 8
Report Opening session: Wellcome word by Dr. Lea Knopf, Officer in charge of the recognition of countries’ animal disease status, OIE Dear participants, dear organizers, observers and guests It is a great honour for me that I was invited to address a few words to all of you on the occasion of the opening of this third EUFMD workshop on NSP testing. First of all I would like to thank the organizers and sponsors who made it possible to launch this third workshop on NSP testing. Today, I would like to share with you some general thoughts related to the importance of the present workshop: This workshop series is quite a unique event that brings together different people involved in FMD control around a common goal or challenge - to address emergency vaccination also called “vaccinate-to-live strategy” and follow up activities to regain the FMD free status without vaccination in a European context. This workshop on FMD-NSP testing emphasizes hopefully again that creative interdisciplinary problem-solving, hands-on training and the active involvement of all the participants are a fruitful strategy. As officer in charge of countries official disease status recognition at OIE I am in continuous contact with countries around the globe which either suffer from an acute FMD outbreak or which seek to achieve or re-instate an FMD free status after a FMD epidemic. In all cases the appropriate design of surveys and the interpretation of serological survey results play a crucial role, as does the choice of a thoroughly planned and adapted vaccination strategy, if applicable. During the last years major achievements have been made on a scientific level driven by highly dedicated end experienced colleges. Also the regulatory framework has been continuously updated to the newest scientific evidence, both on EU level as on an international level. It should not be omitted to mention that changes in legislation may also be driven by the public perception or economic constraints in the animal health and animal welfare sector. Despite the hard and excellent work of numerous scientists, CVOs and other officers in the veterinary science field, legislation and guidelines still limp a little bit behind the reality in the field. Amongst other factors, this might be due to the fact that some of the recommended implementations are based on experiences in the past or that the legislative body, as well as scientists, could simply not consider all the amazing eventualities that happen to occur in the real world. For these reasons I was several times facing the challenge of having to stick to guidelines or legislations and to try to adjust for the complexity of the real world that goes far beyond any recommendation or legislation laid out on paper! Because when it comes to practical implementation, the real world occasionally and relentlessly strikes back, thereby pin pointing both, a partial lack of available options to implement prescribed procedures and the difficult task of veterinary services to plan and conduct most efficiently vaccination campaigns and serological surveys under emergency conditions. In the best case this reveals open questions to be addressed in future revisions of guidelines or research, in the worst case this leads e.g. to heavy economic losses in an affected country due to extended waiting periods, additional outbreaks, culling of animals that would have been avoidable, lack of veterinary or laboratory staff and simply distress. It made me think a lot about the problems that the attendees of the last NSP
EU FMD-NSP workshop 9
workshop had to struggle with during their simulation exercise and emphasises again the importance of such practical exercises well before the real world problems occur! Based on the great experience of the last workshop on NSP-FMD-testing and being a non-native English speaking person I came further to the conclusion that “workshop” might also means “work to shop”. Let me explain this daily life association: Having the opportunity to intensively work together for a few days across countries, across disciplines and different levels of expertise, has the wonderful potential to create a huge shopping centre. During the workshop this shopping centre will be continuously filled with well tried (the theoretical background) and new products (your creative synthesis) that will draw the attention of the participants. “To shop” could therefore be interpreted as “shopping around” for the products that correspond to your specific needs or to increase your stock of problem solving tools in the context of your country. Its is most likely that in view of the number of products at your sight you may get lost in the labyrinth of this shopping centre, you may not have the chance to check all the shelves or you may be simply spoilt for choice… I encourage all of you to catch this opportunity to actively engage in the simulation exercises, YOUR questions, YOUR discussions can lead to alternative approaches to be considered or in other terms they may lead to a brand new products on the shelve of this FMD- shopping centre! I wish you a very successful shopping tour at VAR!
EU FMD-NSP workshop 10
Session I: Review on FMD vaccination and NSP testing The Science behind Non-Structural Protein (NSP) testing, Dr. Kris De Clercq, Belgium
Differentiation of
infected – vaccinated animals
Kris De Clercq
A virus infects a cell, the RNA is read by the cell system and produces structural proteins (SPs) to make the structure of the virus; other parts of the RNA are decoded in proteins to help to make the structure of the virus or help to duplicate the RNA: these are the non-structural proteins (NSPs)
The virus cycle goes on and more and more virus is made but also more and more NSPs. When the cell is destroyed, virus is liberated as well as the NSPs.
Part comes in the blood, so the animal will make antibodies against the SPs and the NSPs
EU FMD-NSP workshop 11
A vaccine producer mimics in fact what happens in nature by adding virus to a cell culture.
The virus cycle goes exactly in the same way…
but then he filters the vaccine to have a pure virus culture.
The virus is inactivated so that a virus cycle (and the formation of NSPs) becomes impossible.
So, when the animal is vaccinated it will make only antibodies against the SPs.
Summary of the reaction of the animal after infection or vaccination or no
infection/vaccination.
EU FMD-NSP workshop 12
ELISA
StructuralProteins
Non-StructuralProteins
+ +
+ -
- -
Translation to the laboratory using two different ELISAs: one classical ELISA detecting antibodies against SPs and one NSP-ELISA, showing how you can differentiate between infected and/or vaccinated animals.
ELISA
StructuralProteins
Non-StructuralProteins
+ +
+ -
If an animal gets infected after the vaccination, normally the infected animals will react but however some animals will not react, causing a false negative.
If a vaccine is not well purified you will have…
NSPs in the vaccine.
ELISA
StructuralProteins
Non-StructuralProteins
+ +
+ -
So, after one injection you will normally get no reaction against NSPs but after several vaccinations with this kind of vaccine you will get false positives.
EU FMD-NSP workshop 13
Validation and comparison of NSP tests, Dr. Donal Sammin, Ireland
Comparative evaluation and validation of NSPEs
Dónal Sammin, CVRL-DAF, Ireland
Introduction
• 2003/85/EC: vaccination to live and post-vaccination surveillance
• DIVA testing => use of NSP-based tests• OIE-approved method ex. Panaftosa• 5 other NSPEs in Europe• Evaluate comparative performance• Validate for purpose
Studies
• FMD-ImproCon multinational workshop; IZS-Brescia, May 2004; cattle ( �sheep/pigs); Brocchi et al., 2006 [App 01] + Dekker et al. [App 08]
• EUFMD/WRL field study; Zimbabwe, April 2004; cattle, SAT-type infection; Sammin et al., VR, 2007
• EUFMD/WRL field study; HK-SAR, March 2005; (+ experimental study, IRL, 2006); pigs; Paton et al.
• Proposed field and exptl studies on sheep, 2007/8
FMD_ImproCon WS; IZS, Brescia; May 2004Brocchi et al., 2006 [App 01]
MATERIALS & METHODS• 3551 sera [BE, DE, DK, ISR, IT, NL, TK, UK]• 2579 (67%) bovine; 703 ovine; 269 pig• Different vaccination and infection status• All sera tested x6 NSPE in parallel• Panaftosa, IZS-B, Ceditest, Svanovir, Chekit
and UBI ELISAs
WORKSHOP; IZS, Brescia; May 2004MATERIALS & METHODS: BOVINE SERA
• Experimental (n = 1037 sera)425 V+ I-
62 [54] V- I+ 21 [17] C+
550 [285] V+ I+ 225 [67] C+
• Field (n = 1542 sera)672 V- I-
867 V+ I���� (ISR/ZIM)
WORKSHOP; IZS, Brescia; May 2004MATERIALS & METHODS: NSP ELISAs
NOINDIRECT; coated
3BUBI FMDV NS ELISA
YESINDIRECT; coated
3ABCCHEKIT-FMD-3ABC
NOINDIRECT; coated
3ABCSVANOVIR FMDV 3ABC-Ab ELISA
NOBLOCKING; trapping
3ABCCeditest FMDV-NS
YESINDIRECT; trapping
3ABCIZS-Brescia
YESINDIRECT; coated
3ABCPanaftosa
“Grey zone”FormatAntigenELISA
EU FMD-NSP workshop 14
WORKSHOP; IZS, Brescia; May 2004RESULTS: diagnostic specificities
• 1100 bovine sera [675 V+ I- & 425 V+ I-]• 97.2% - 98.5% on first screening test• 98.3% - 99.7% on retesting positives
WORKSHOP; IZS, Brescia; May 2004RESULTS: detection rates
• CATTLE V- I+ (n = 54) & V+ I+ (n = 285)
• Sub-categories of V+ I+ cattle: • (i) infection demonstrated (n = 164); • (ii) carrier status demonstrated (n = 67); • (iii) not carriers (n = 26); • (iv) no evidence of infection (n = 17)
• 7-14, 15-27, 28-100 and >100 days p.i.• Subcategories (i) and (ii), 14-27dpi: Panaftosa,
• Objective: evaluate NSPEs for SAT-type FMD• 403 cattle; 6 herds; April-May 2004• SAT1/SAT2, 5 herds, 1-5 months pi• 12 - 35% of herds were “carriers” (probangs; VI & PCR)
• Overall seroprevalence of 56% - 75% with NSPEs; 81% with SPCE and 91% with VNT
• Carrier detection rate of 70% - 90%
• Sammin et al., Veterinary Record (2007) 160: 647-654
EUFMD/WRL field study; HK-SAR& exptl study on vaccinated pigs, IRL
Objective: evaluate 3 x NSPEs for use in pigsCeditest, UBI and Chekit ELISAs
Field study• 405 pigs; 4 vaccinated herds; Feb-March 2005• Type O infection, 2 herds, 1-2 months pi• All 3 NSPEs detected infection in pigs; different dSE and dSP• Cedi was both sensitive and specific; UBI was less specific
and Chekit was less sensitiveExperimental study• 99 pigs; vaccination x2; sampling 28 dpv• dSP: 100% for Cedi and UBI; 98% for Chekit
Still to do…
• Publish paper on HK & IRL pig studies• Publish paper on LRs and ROC analysis of results
from Brescia WS• Perform field and experimental studies on sheep
during late 2007/early 2008
EU FMD-NSP workshop 15
Sensitivity and specificity of SP and NSP tests – relevance and use, Dr. Aldo Dekker, The Netherlands
Sensitivity and specificity of SP and NSP tests; relevance and use
Aldo Dekker
Outline
� Objective
� Validation of SP tests
� Results validation study NS ELISA's in Brescia� Validation using dichotomised results� Validation using continuous results
Validation SP tests
� Neutralisation tests considered as gold standard� No differentiation between vaccinated and infected
animals (DIVA)� Large variation in results when using different cell lines
� No true validation study available
� LPB and SPC ELISA� Very well validated� Different cut-offs for different purposes
� Definition of FAO cut-off sera unclear since UK outbreak 2001
Validation SP tests
� Neutralisation tests considered as gold standard� Sensitivity close to 100% (by definition)� Very high specificity (> 98% when using one serotype)
� LPB and SPC ELISA� Sensitivity and specificity similar to neutralisation test
Objective Brescia workshop
� Compare different DIVA tests for FMDV
� Provide estimates that can be used for surveillance
� Post test odds = pre test odds x LR� Independent of prevalence in target population
� Pre test odds = (a+c)/(b+d)� Post test odds = a/b
� LR+ = sens/(1-spec)
=# true –/# false positives a b
c d
+
-
+ -true
test
EU FMD-NSP workshop 18
Likelihood ratio on continuous data
-20 -10 0 10 20 30 40 50 60 70 80 90 100
Percentage inhibition
0
20
40
60
80
100
0
20
40
60
80
100
NON-EXPOSED CATTLE
EXPOSED CATTLE
Num
ber
of c
attle
Num
ber
of c
attle
Percentage inhibition Ceditest ELISA
Likelihood ratio analysis Ceditest ELISA
0
5
10
15
20
25
30
0 20 40 60 80 100
Percentage inhibition
Like
lihoo
d ra
tio
Cut-off
� At cut-off already high likelihood for infection� Many sera from exposed
cattle with low response
� Result can be used in decision scheme
� With the same sera different tests produce a different LR
Conclusion� ROC analysis similar to results obtained
previously � The estimation of the Likelihood ratios will help
diagnosticians and decision makers
Recommendation:� The LR should be implemented in decision
scheme and used when ELISA's for FMD virus non-structural proteins are used after an FMD emergency vaccination.
Risk maps with 3 colours:Red: '1 km culling' not
sufficient to avoid large outbreaks (= R0>1)
Orange : '1 km culling' sufficient
Green: 'no ring culling' already sufficient (= EU-measures)
On the basis of slowest decreasing kernel and immediate culling
EU FMD-NSP workshop 19
Session II: Review on statistical computer tools, EU Directives and carrier problem Statistical computer tools: theoretical and practical approach. A. Theory, Dr. Koen Mintiens, Belgium
Statistical computer tools: theoretical and practical approach.
A. Theory
K. MintiensCo-ordination Centre for Veterinary
Diagnostics
Contents• Different approaches for ‘substantiating’
freedom from disease• Two-stage sampling• Discussion
Claiming freedom from Claiming freedom from Claiming freedom from Claiming freedom from diseasediseasediseasedisease
• SPS agreement of World Trade Organization (WTO):
– Current animal health-related rules for international trade;
– Countries need to provide science-based evidence to support their claims to freedom from livestock diseases;
– One way is using survey samples.
Starting point• Population in which a disease may be present or
absent• If disease is present, then: minimal expected
prevalence = design prevalence (p*)• Diagnostic test which can identify the ‘presence
of the disease’• Survey sample which is sufficient to
‘substantiate’ presence/absence of disease with a certain confidence level
Principle• Pick enough balls without
replacement to have at least one red from N balls
• hypergeometricdistribution:
−−
==+
n
N
xn
dN
x
d
xTP )(
Principle• When aiming to substantiate freedom from
disease, x=0:
• Factorial formulae with large numbers are difficult to compute
• Approximations of formula have been developed together with IT capacity
( ) ( )!)!(
!!)0(
NndN
nNdNTp
−−−−==+
EU FMD-NSP workshop 20
• Methods for detecting the presence of a disease
• Assumptions:
– Design prevalence
– Perfect diagnostic tests
• Convenient approximation:
n
nN
dTP
−−−==+
21
1)0(
First approach: Cannon & Roe, 1982
First approach: Cannon & Roe, 1982
• Sample size:
• Confidence level α (probability of finding at least one positive in the sample):
( )( ) 12
111
+
−×−−≈ dNn dα
n
nN
d
−−−≈
−2
)1(11α
• Tables with standard numbers available in epidemiological text books
• Software tools, e.g. WinEpiscope:
http://www.clive.ed.ac.uk/winepiscope /
First approach: Cannon & Roe, 1982
Illustration of sample size calculationIllustration of sample size calculationIllustration of sample size calculationIllustration of sample size calculation:
– FMD outbreak in previously free area
– Outbreak controlled and disease no longer present
– Design prevalence: 30%
– Herd with 265 animals
– Confidence level (1-α)=0.95
=>Output from WinEpiscope: n=9
First approach: Cannon & Roe, 1982
Improvements: Cameron & Baldock, 1998
• Diagnostic tests are not perfect:
• Thus the hypergeometric distribution extends:
SppSepTP
SpppSeTP
)1()1()(
)1)(1()(
−+−=
−−+=−
+
∑∑=
+−−−−
=
+ −×
−−
−
−−
==),min(
00
)1()1()(yx
j
jyxnjxjyjd
y
SpSpjx
ynSeSe
j
y
n
N
yn
dN
y
d
xTP
• Equation is very complex (only small population sizes can be computed)
Illustration of sample size calculation:Illustration of sample size calculation:Illustration of sample size calculation:Illustration of sample size calculation:
– Design prevalence: 30%
– Herd with 265 animals
– Type I error (α )= type II error (β)=0.05
•Confidence level (1-α)=0.95
– ELISA Se=0.95 & Sp=0.98
=> FreeCalc: n= 14 with 1 reactor
Improvements: Cameron & Baldock, 1998
• Addresses the uncertainty in sensitivity, specificity and design prevalence
• Uses hypothesis testing• Bayesian approach
– Prior distributions for Se, Sp and p*
– Focus on the posterior probability that p is below a threshold
Further extension: Johnson et al., 2004
• Procedure also available as freeware: BayesFreeCalc
http://www.epi.ucdavis.edu/diagnostictests/
Further extension: Johnson et al., 2004
Illustration of sample size calculation:– Design prevalence: > 30%
– Herd with 265 animals
– Type I error = type II error = 0.05
– ELISA Se>0.90 & Sp>0.95
=> BayesFreeCalc: n= 18, 2 reactor
Further extension: Johnson et al., 2004
Two-stage sampling• Survey can be used at any level: herd, region,
nation, etc…• Available statistical tools can provide confidence
levels for rejecting the hypothesis of disease presence, given:– Population size N– Sample size n– Design prevalence p*– Se and Sp of diagnostic test– Allowable Type I and II error
EU FMD-NSP workshop 22
Two-stage sampling• For practical reasons often two-stage sampling
strategy• Diseases tend to cluster in the population (e.g.
within farms)• Two-levels for substantiating disease freedom:
– Each herd is classified as diseases or non-diseased: test results of individual animals
– Population of herds is classified as diseased or non-diseased based on results of individual herds
Two-stage sampling• Sensitivity (HSe) and specificity (HSp) of herd
classification is influenced by:– Se + Sp of individual-animal test– # animals tested per herd– Interpretation of individual test results
Two-stage sampling• In this context:
– HSe = • probability that a diseased herd will be classified as diseased• one minus the probability that a diseased herd will be
classified as non-diseased = 1-α
– Similarly HSp =• probability that a non-diseased herd will be classified as non-
diseased • one minus the probability that a non-diseased herd will be
classified as diseased = 1- β
Two-stage samplingSample size• Number of herds:
– Preset Type I and type II error @ herd level– Desired HSe and HSp– Minimal expected herd prevalence
• Number of animals per herds– Type I error = 1-HSe– Type II error = 1-HSp– Se and Sp of assay for animal testing– Minimal expected within-herd prevalence
Two-stage samplingAnalysis of results• Classification of herds
– Each herd is classified as diseased or non-diseased, based on• number of reactors• Se + Sp of animal test• Na, na, type I and type II error
– Result: number of reactor herds
• Classification of population– Based on number of reactor herds, and
• Se + Sp of animal test• Nh, nh, type I and type II error
Two-stage samplingHypothetical example• Survey in 8532 herds to substantiate freedom from
disease• Screening test: Se=0.94, Sp=0.90• 95% confidence and power• Minimal expected herd-p*= 5%• Minimal expected within-herd-p*=20%• Logistical constrain: max 200 herds to be tested
EU FMD-NSP workshop 23
Two-stage samplingHypothetical example: FreeCalc• Trail and error: 193 herds to be tested if:
– Type I = Type II =0.05– Herd-p= 5%– HSe= 90%– HSp= 98.4%– 6 reactors allowed
• Within herd testing sample size:– Type I error= 10%– Type II error= 1.6%– Within-herd-p= 20%– Sample size can be calculated for each herd
Design prevalence• Either minimal expected prevalence• Or, small enough to be considered negligible
• Can often not be defined:– Wildlife populations or emerging diseases– Vaccinated populations
• Small design prevalence implies large sample size
• The application of design prevalence does not allow ‘zero prevalence’
Discussion
DiscussionNew approach (Mintiens et al, 2005)• Bayes Theorem:
– The posterior distribution of the probability of freedom from disease (F) given the observed test results (T)
)(
)()()(
TP
FPFTPTFP =
)(
),,,(),,,(),,,(
TP
SpSeprevFPSpSeprevFTPTSpSeprevFP =
Discussion
New approach (Mintiens et al, 2005)• Probability for positive test result:
• The likelihood function takes into account that the observed prevalence only occurs when the population is not free from disease.
Prev.Vet.Med. 52, 227-249.Greiner, M. & Dekker, A., 2005, On the surveillance for animal diseases in small herds.
Prev.Vet.Med. 70, 223-234.Huzurbazar, S., et al., 2004, Sample size calculations for bayesian prediction of bovine viral-
diarrhoea-virus infection in beef herds. Prev.Vet.Med. 62, 217-232.Johnson, W.O., et al., 2004, Sample size calculations for surveys to substantiate freedom of
populations from infectious agents. Biometr. 60, 165-171.Martin, P.A. et al., 2007, Demonstrating freedom from disease using multiple complex data
sources 1: A new methodology based on scenario trees. Prev.Vet.Med. 79, 71-97.Martin, P.A. et al., 2007, Demonstrating freedom from disease using multiple complex data
sources 2: Case study--classical swine fever in denmark. Prev.Vet.Med. 79, 98-115.Mintiens, K. et al., 2005, Estimating the probability of freedom of classical swine fever virus of
the east-belgium wild-boar population. Prev.Vet.Med. 70, 211-222.Ziller, M. et al., 2002, Analysis of sampling strategies to substantiate freedom from disease in
large areas. Prev.Vet.Med. 52, 333-343.
Thank you
EU FMD-NSP workshop 25
Statistical computer tools: theoretical and practical approach. B. Practice, Dr. Koen Mintiens, Belgium
Statistical computer tools: theoretical and practical approach.
B. Practice
K. MintiensCo-ordination Centre for Veterinary
Diagnostics
Aim• To illustrate a few of the statistical
approaches to substantiate disease freedom
• Use of a simulated FMD outbreak in Belgium
• Provide guidance to the workshops
Case• FMD outbreak in Belgium
– 10 km surveillance zone around 6 infected premises:• All infected premises are stamped out• All susceptible animals within 500m radius are culled• All cattle within 10,000m radius are vaccinated
Classification of population• Type I = Type II= 0.05• HSe=99%, HSp=99%• Design prevalence= 1%• N=2188• n=1325• Positive reactors= 21
Analysis of resultsTwo stage approach
Classification of population
Results are not adequate to conclude that population is free from disease (design prevalence 1%). The confidence level is only 75.815%. We may conclude that population is diseased at confidence level 98.564%.
Discussion• Two-stage sampling:
– More practical– Increases sample size– More confident results (accounts for clustering)
• Positive results are marginal, but this can be expected in a vaccinated population– Sometimes these marginal positive result borderline or not
acceptable at herd level
• EC directive requires testing of each vaccinated animal
Discussion• Can’t we assume Sp=1 at animal level?
– Reduces sample size– All positives (in the sample) in a herd are culled…– Does that mean that the infection is eliminated from the herd?– What about stamping out of positive herds?– Does that mean that all infected herds are eliminated?
• Sp=1 can be assumed if all efforts are made to exclude false positive results
EU FMD-NSP workshop 30
Discussion• How to increase confidence level and power:
– Increase accuracy: serial testing?– Look at geographical clusters?– Look at non-vaccinated populations, sentinels, buffer zone, …– …
• What about decreasing the power (increasing type II error)?– Decreases sample size– At herd level: less reactors are allowed
• More (false) positive herds are stamped out– Cost of taking more samples per herd vs. stamping out of herds
Thank you
EU FMD-NSP workshop 31
EU Directives for FMD control, Dr. Alf Füssel, EU DG SANCO, Head of Trade and Zootechnics Sector
Dr. Alf-Eckbert FüsselEuropean Commission - DG Health and Consumer Protection
Directorate D - Animal health and welfareUnit D1 - Animal health and Standing Committees
This presentation does not necessarily represent the views of the Commission 2
EmergencyEmergency vaccinationvaccination
►► CouncilCouncil DirectiveDirective 2003/85/EC 2003/85/EC providesprovides forforemergencyemergency vaccinationvaccination in in casecase of of moremore thanthan oneoneoutbreakoutbreak of FMDof FMD
►► TypesTypes of of vaccinationvaccination�� ProtectiveProtective ((vaccinatevaccinate--toto--livelive))
►► VaccinationVaccination isis carriedcarried out in out in accordanceaccordance withwith�� ArticleArticle 52(1) 52(1) -- vaccinationvaccination zonezone
► all the measures provided for in Articles 36, 44, 54, 55, 56 and 57 have been completed,
► at least one of the following conditions applies:� OIE rules as in force, or� 3 months after slaughter of vaccinated animals + serological
surveillance in accordance with OIE guidelines, or� 6 months after the last outbreak and last vaccination + serological
survey for NSP demonstrated absence of infection in vaccinated animals
► a Decision has been adopted to re-establish the free status
8
ModifiedModified recoveryrecovery of of freefree statusstatus
Following emergency vaccination, and by way of derogation, it may be decided byComitology, to withdraw the restrictionsapplied in the vaccination zone after theclinical and serological survey and theclassification of herds have been completedand confirmed the absence of foot-and-mouthdisease virus infection
Risk assessment taking into account different test strategies
9
Objectives Objectives
of postof post--vaccination measuresvaccination measures►►Risks from products from vaccinated Risks from products from vaccinated
animals can be mitigatedanimals can be mitigated
butbut
►►PostPost--vaccination surveillance is carried out vaccination surveillance is carried out to ensure that after the recovery of the to ensure that after the recovery of the status vaccinated ruminants can be moved status vaccinated ruminants can be moved safely out of the previous vaccination zone safely out of the previous vaccination zone and continue their productive life within the and continue their productive life within the vaccinating Member Statevaccinating Member State
10
ConclusionConclusion
►►When the free status is recovered there When the free status is recovered there
should only be herds which are should only be herds which are ►►either either serosero--negative negative
OrOr
►►SeroSero--positive exclusively from the administration of positive exclusively from the administration of
an inactivated vaccine an inactivated vaccine
►►No danger of carriers or undetected disease No danger of carriers or undetected disease
EU FMD-NSP workshop 33
Combining multiple sources of evidence to demonstrate disease freedom, Dr. Tom Murray, APO FAO EUFMD
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 11
Combining Multiple Sources of Combining Multiple Sources of Evidence to Demonstrate FMD Evidence to Demonstrate FMD
FreedomFreedom
Tom Murray, Associate Tom Murray, Associate Professional Officer, FAOProfessional Officer, FAO--
EUFMD CommissionEUFMD Commission
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 22
Key participants in the development of Key participants in the development of work are Ausvet, The International Epilab work are Ausvet, The International Epilab in the Danish Institute for Food and in the Danish Institute for Food and Veterinary Research and The Australian Veterinary Research and The Australian Biosecurity CRC Biosecurity CRC
The objective is to develop a system to The objective is to develop a system to quantitatively evaluate confidence in quantitatively evaluate confidence in Disease FreedomDisease Freedom
Details of methodology available at Details of methodology available at www.ausvet.com.au/freedomwww.ausvet.com.au/freedom
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 33
Summary of PresentationSummary of Presentation
1.1. Overview of The use of the stochastic Overview of The use of the stochastic Scenario Tree ApproachScenario Tree Approach
2.2. Review of various quantitative inputs and Review of various quantitative inputs and outputs in the context of an overall surveillance outputs in the context of an overall surveillance systemsystem
3.3. Example: Evaluation of Random SeroExample: Evaluation of Random Sero--survey survey using the stochastic Scenario Tree approachusing the stochastic Scenario Tree approach
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 44
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 55
Overview of Scenario Tree Overview of Scenario Tree ApproachApproach
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 66
Surveillance System Surveillance System Component(SSCComponent(SSC))
EU FMD-NSP workshop 34
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 77
ExplanationsExplanations
The minimum proportion The minimum proportion of animals within a herd of animals within a herd that you would expect to that you would expect to have FMD if it was have FMD if it was presentpresent
Design Unit PrevalenceDesign Unit Prevalence
The minimum proportion The minimum proportion of herds you would of herds you would expect to have FMD if expect to have FMD if FMD is present in a FMD is present in a regionregion
Design Herd PrevalenceDesign Herd Prevalence
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 88
ExplanationsExplanations
Given that surveillance Given that surveillance systems have not detected systems have not detected FMD, how confident are we FMD, how confident are we that this is a true reflection of that this is a true reflection of the situationthe situation
Negative Negative predictive value predictive value (NPV)(NPV)
The probability that the The probability that the surveillance systems will surveillance systems will detect FMD if it is present in detect FMD if it is present in the regionthe region
Surveillance Surveillance System SensitivitySystem Sensitivity
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 99
ExplanationsExplanations
The estimate of the probability The estimate of the probability of FMD based on prior of FMD based on prior probability and the survey probability and the survey resultsresults
Posterior/PostPosterior/Post--test test probability of probability of diseasedisease
The estimate of the probability The estimate of the probability of FMD presence before the of FMD presence before the survey is carried outsurvey is carried out
Prior/Pretest Prior/Pretest probability of probability of diseasedisease
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1010
SpecificitySpecificity
Objective: To demonstrate disease Objective: To demonstrate disease freedomfreedom
Implies all tests have lead to a negative Implies all tests have lead to a negative outcomeoutcome
Specificity = 1Specificity = 1
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1111
Bayes TheoremBayes Theorem
Inputs and Outputs are based on Inputs and Outputs are based on distributions rather than point values. This distributions rather than point values. This reflects the uncertainty and variability reflects the uncertainty and variability associated with biological dataassociated with biological data
Prior probability of diseases and the actual Prior probability of diseases and the actual survey results are combined to give a survey results are combined to give a posterior probability of diseaseposterior probability of disease
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1212
Example: Evaluation of a Simple Random Example: Evaluation of a Simple Random SeroSero--SurveySurvey(Using Point Values)(Using Point Values)
Followed by combination of various SSCFollowed by combination of various SSC’’s s
EU FMD-NSP workshop 35
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1313
Step 1: Calculate the probability Step 1: Calculate the probability that any one sample tests positivethat any one sample tests positive
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1414
Step 2:Calculate Sensitivity of Step 2:Calculate Sensitivity of Surveillance System Component.Surveillance System Component.If If ““aa”” is the probability that any sample is the probability that any sample tests positivetests positive
Then the probability that the surveillance Then the probability that the surveillance system component detects disease if it is system component detects disease if it is present is present is 11--(1(1--a)a)n n
(where (where ““nn”” is the number of samples taken)is the number of samples taken)
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1515
Step 3: Step 3: Calculate Negative Predictive Value of Calculate Negative Predictive Value of Surveillance System and the Posterior probability of FMDSurveillance System and the Posterior probability of FMD
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1616
Step 4: Carry out similar procedure Step 4: Carry out similar procedure for other surveillance system for other surveillance system
componentscomponentsOther SSCOther SSC’’s may require more s may require more assumptions and the uncertainty assumptions and the uncertainty associated with inputs would be greaterassociated with inputs would be greater
The degree of uncertainty associated with The degree of uncertainty associated with inputs can be reduced with time through inputs can be reduced with time through increased knowledge and the use of increased knowledge and the use of expertiseexpertise
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1717
Step 5:Combine the various SSCStep 5:Combine the various SSC’’s s together to get an overall value of together to get an overall value of
the surveillance systemthe surveillance system
Lack of independence between various Lack of independence between various surveillance system components needs to surveillance system components needs to be accounted forbe accounted for
03 July 200703 July 2007 FAOFAO--EUFMD CommissionEUFMD Commission 1818
Key PointsKey Points
This is a simplified overview of the processThis is a simplified overview of the processThe quantitative evaluation of surveillance systems can The quantitative evaluation of surveillance systems can provide for comparisons between systemsprovide for comparisons between systemsThis work requires the close involvement of expertise This work requires the close involvement of expertise and experience from various fieldsand experience from various fieldsThis work emphasizes the value of current over historical This work emphasizes the value of current over historical data.data.The use of this approach can identify gaps in knowledge The use of this approach can identify gaps in knowledge for further researchfor further researchAn effective system for demonstration of FMD freedom An effective system for demonstration of FMD freedom also means a system which is more likely to detect new also means a system which is more likely to detect new incursions earlierincursions earlier
EU FMD-NSP workshop 36
Session III: OIE guidelines and serosurveillance post-vaccination: practice (1) OIE rules and guidelines on use of serosurveillance to demonstrate freedom from infection with FMDV and (2) use of a model to predict expected prevalence of carriers and to design serosurveillance for their detection. Dr. David Paton, United Kingdom
1) OIE rules and guidelines on use of serosurveillance to demonstrate
freedom from infection with FMDV
2) Use of a model to predict expected prevalence of carriers and to design
serosurveillance for their detection
7/3/2007 2
OIE Terrestrial Animal Health Code
• Chapter 1.3.5.1. Zoning & Compartmentalisation• Chapter 2.2.10. FMD• Appendix 3.8.1. General guidelines for animal
health surveillance• Appendix 3.8.7. Guidelines for the surveillance
of FMD
OIE Diagnostic Manual
7/3/2007 3
OIE Code FMD Chapter 2.2.10.7: Recovery of FMD-free status following outbreaks
(1)
Outbreaks in a country or zone that is “FMD-free with
vaccination”.
Can regain this status by:
– Vaccination, slaughter of infected, serosurveillance to detect virus circulation rather than infection, 6 month
minimum waiting period
– OIE provides guidelines for serosurveillance
– Do not need to detect carriers at all
– Risk of carriers still being present is mitigated by restrictionon trade in live animals if “free with vaccination”, rather than
“free without vaccination”
7/3/2007 4
Outbreaks in a “FMD-free country without vaccination”
Can regain this status by:
– Slaughter of infected, no vaccination, serosurveillance to
demonstrate absence of infection, 3 months minimum wait
– Slaughter of infected, vaccinate-to-kill, serosurveillance to
demonstrate absence of infection, 3 months minimum
– Slaughter of infected animals, vaccinate-to-live, serosurveillance to demonstrate absence of infection, 6 months minimum
OIE Code FMD Chapter 2.2.10.7: Recovery of FMD-free status following outbreaks
– FMD virus isolated or viral RNA or virus antigen
demonstrated in samples from an animal
– NSP antibodies not due to vaccination in one or more animals showing signs of FMD or linked to a
confirmed or suspected outbreak, or giving cause for
suspicion of previous association or contact with FMDV
7/3/2007 6
OIE Code Appendix 3.8.1.General Guidelines for Surveillance
• Article 3.8.1.6. Surveillance to demonstrate freedom from disease/infection
– Demonstration of freedom from infection
• Implies absence of the pathogenic agent
• Complete absence cannot be proven with 100% confidence – rather provide adequate evidence to an
acceptable level of confidence that infection is at less than a certain acceptable prevalence (however, none of these acceptance criteria are defined for FMD within Code)
• However, finding evidence of infection at ANY level automatically invalidates a claim for freedom.
EU FMD-NSP workshop 37
7/3/2007 7
OIE Code Appendix 3.8.7.
Guidelines for FMD Surveillance
• Article 3.8.7.7. The use and interpretation of serological tests– Reference to Manual for recommended tests
– Option to use NSP tests in vacc/unvacc animals and NSP tests in vacc animals
– Need diagnostic follow-up on presumptive positive serological test results
– All herds with seropositive reactors should be investigated –including clinical and epidemiological evaluation and supplementary lab tests• High Sp and Se equivalent to screening assays
– Information needed on performance characteristics and validation of tests used
7/3/2007 8
• Article 3.8.7.7. The use and interpretation of serological tests (cont)– Follow-up if no vaccination
• Where possible virological investigations should be made
• Reactor animal assumed positive unless suspicion ruled out
– Follow-up if vaccination• Procedures described are aimed at detection of virus circulation only and not detection of carriers
• Resample and test initially surveyed animals in reactor unit
• Sample and test additional contacts
• Sample and test other epidemiologically linked units
• Sentinels can be used
• Relate lab results to epidemiological situation
7/3/2007 9
Fig. 1. Schematic representation of laboratory testsfor determining evidence of FMDV infection
prevalence with 95% probability• Have also looked at testing all animals
EU FMD-NSP workshop 40
7/3/2007 25
Epidemic size distribution – 200 runs
7/3/2007 26
Mean no of IPs/day
�Vaccinating small farms
– 66 vaccinated
farms/IP, 71% reduction in cases
�Vaccinating farms >50
cattle – 35 vaccinated farms/IP, 68% reduction
in cases
7/3/2007 27
Distribution of carrier farms by herd size
� Reducing number
of farms as herd
size increases
� But higher
incidence of those farms
7/3/2007 28
Within herd prevalence
� Farms with many initial infs->IPs
� Farms infected late after vaccination have few initial infs
� Therefore carriers occur at low prevalence
7/3/2007 29
Sensitivity of testing (at farm level)
Strategy True positives
Detected Sensitivity
Vacc >50 only
335 118 35%
Vacc all 325 140 43%
• 75% sensitivity if all animals tested (for vacc>50)
7/3/2007 30
Specificity of testing (at farm level)
Strategy True negatives
False positives
Specificity
Vacc >50 only
233,000 1,562 99.32%
Vacc all 397,000 1,890 99.52%
• 98.4% specificity if all animals tested (for vacc>50)
EU FMD-NSP workshop 41
7/3/2007 31
Main results (1)
• Little benefit for vaccinating small herds in terms of total cases
• Carrier farms distributed across all herd sizes – fewer large herds but higher incidence of large herds
• Small number of carriers in positive herds means low herd-level sensitivity (with 5% design prevalence) – therefore need to test all animals in vaccinated herds
7/3/2007 32
Main results (2)
• Vaccinating small herds
– Smaller average herd size of positives
– Higher herd level sensitivity – larger
proportion of animals tested
– Higher herd level specificity
– But far more vaccinated farms to test
•• Same conclusions from results of NSP testing in Cumbria
EU FMD-NSP workshop 42
Presentation of adopted FMD vaccination plans - scenarios prepared for the workshop, Dr. Phillippe Houdart, Belgium
Federal Agency for the Safety of the Food Chain
INTRODUCTION TO THE EXERCISE
Federal Agency for the Safetyof the Food Chain
Federal Agency for the Safety of the Food Chain
Aim of the exercise
• design and interpret a post-vaccination FMD surveillance scheme including the use of NSP tests, based on EU and OIE guidelines
• objective: substantiate a claim to return to the status free of infection
• starting point:– FMD epizootic with 3 clusters of outbreaks– each cluster = one scenario– vaccination to live policy � vaccinated animals
are not killed
Federal Agency for the Safety of the Food Chain
Approach
• different control strategy in each of the clusters, but always with vaccination of one or more species:– cluster 1 = index case + 3 neighbouring outbreaks– cluster 2 = 2 outbreaks
– cluster 3 = 23 outbreaks in different subclusters
• outbreaks between end of February and end of March
• start of the surveillance programme = end of April beginning of May, > 1 month after last outbreak
Federal Agency for the Safety of the Food Chain
Scenario 1 = cluster 1
• PZ and SZ surrounding index case + 3 neighbouring outbreaks
• size = 433 km2
• density: – cattle = medium density (37 / km2)– pigs = high density (300 / km2)
– sheep/goat = low density (2 / km2)
Federal Agency for the Safety of the Food Chain
Scenario 1 = cluster 1
• culling: susceptible animals in 4 outbreaks
• vaccination of cattle and pigs in protection zone
• particularities– high number of pig farms:
• 3 very large pig farms• 85% are small farms
– very low number of sheep– in the north of the SZ the farms are grouped
(villages)
Federal Agency for the Safety of the Food Chain
EU FMD-NSP workshop 43
Federal Agency for the Safety of the Food Chain Federal Agency for the Safety of the Food Chain
Federal Agency for the Safety of the Food Chain Federal Agency for the Safety of the Food Chain
Scenario 2 = cluster 2
• PZ and SZ surrounding 2 neighbouring outbreaks
• size = 380 km2
• FMDV introduced through movement of infected sheep (outbreak 5)
• density:
– cattle = medium density (40 / km2)– pigs = medium density (124 / km2)– sheep/goat = low density (4 / km2)
Federal Agency for the Safety of the Food Chain
Scenario 2 = cluster 2
• culling: susceptible animals in 2 outbreaks
• vaccination of cattle and pigs in protection zone
• particularities:
– one of outbreaks = large pig farm (>40.000 pigs)– only infected pigs in a quarantine unit separated
from the other buildings– rest of holding is vaccinated
Federal Agency for the Safety of the Food Chain
EU FMD-NSP workshop 44
Federal Agency for the Safety of the Food Chain Federal Agency for the Safety of the Food Chain
Federal Agency for the Safety of the Food Chain
Scenario 3 = cluster 3
• vaccination area surrounding 23 outbreaks
• size = 1.440 km2
• FMDV introduced through movement of infected sheep (outbreak 5)
• density:
– cattle = high density (160 / km2)– pigs = high density (>1.000 / km2)– sheep/goat = low density (13 / km2)
Federal Agency for the Safety of the Food Chain
Scenario 3 = cluster 3
• culling: susceptible animals in all outbreaks
• vaccination of all cattle in vaccination area
• particularities:– high density area for pigs and cattle– outbreaks in grouped into subclusters
– several outbreaks with no or unclear epidemiological links
Federal Agency for the Safety of the Food Chain
0,0 2,5 5,0 7,5 10,0
Kilometers
Federal Agency for the Safety of the Food Chain
0,0 2,5 5,0 7,5 10,0
Kilometers
EU FMD-NSP workshop 45
Federal Agency for the Safety of the Food Chain
What is expected of you?
• each of the scenarios is approached by 2 groups of2 countries
• tasks: – design surveillance scheme for one of the given
clusters (both vaccinated and non vaccinated animals)
– interpret set of results of your surveillance scheme given by the supervisory team
– design follow up for positive results– present results to the audience tomorrow
(powerpoint presentation)
Federal Agency for the Safety of the Food Chain
Assignment of the groups
• scenario 11. Moldova + Ukraine Per Have + David Paton2. Norway + Latvia Tom Murray
• scenario 23. Slovakia + Estonia Åse Uttenthal + Philippe Vanier4. Finland + Poland Aldo Dekker
• scenario 3: 5. Czech Rep. + Sweden Dónal Sammin6. Switzerland + Israel Lea Knopf
room
308333
309333
265138
Federal Agency for the Safety of the Food Chain
documentation
• publications about FMD testing and in particular NSP testing
• regarding scenarios:
– background information
– tables with composition of the different clusters
– timeline of the outbreaks
– short information about each of the outbreaks
– maps
Background information about the epizootic At the end of February an outbreak of FMD strain A Iran 96 is observed in a pig holding. The FMD virus has most likely been introduced onto the holding through contaminated leftovers brought back from Turkey by the son of the owner around mid February. The initial clinical signs are ignored by the owner, resulting in a rather late detection of the disease by the farm veterinarian. This outbreak is the first case detected and is the index case of the epizootic. The infected farm is situated at the centre of cluster 1. Three major routes are responsible for the subsequent spread of the disease:
- A diseased pig is slaughtered in the initial stages of the outbreak. The village butcher who kills the diseased pig subsequently spreads the virus to a sheep trader on the opposite site of the village. The brother of owner of the first outbreak assists at the slaughtering and spreads the virus to his own farm in the same village.
- Subsequently, a sheep trader whose farm is infected via the butcher spreads the virus to clusters 2 and 3 through the selling of infected sheep. The trader is very reluctant to disclose information about his comings and goings, amongst others about the movement of infected animals to his colleague trader who is at the origin of cluster 3.
EU FMD-NSP workshop 46
- Where in cluster 2 the infection is detected in an early stage, in cluster 3 the disease is detected only after the virus has already spread to several farms. The common link in the initial spread in this cluster is a second sheep trader that has bought infected animals from the sheep trader in cluster 1. The infected animals have stayed in this trader’s stable in a larger group of sheep. Sheep of this group are sold to several known and unknown clients in the region.
In the weeks following the detection of the first outbreak, a total of 29 outbreaks are recorded grouped into 3 distinct clusters, namely 4 outbreaks in cluster 1 (including the index case), 2 in cluster 2 and 23 in cluster 3. Cluster 3 is the largest of the 3 clusters; the 23 outbreaks in this cluster are for the greater part grouped into smaller clusters of outbreaks. The last outbreak is detected at the end of March, almost 5 weeks after the detection of the index case. All outbreaks are finally managed by a combination of culling and preventive vaccination. The vaccination is always performed in either cattle or pigs or both, but never in small ruminants. Characteristics of cluster 1 Cluster 1 is comprised of the 10 km surveillance zones surrounding the index case and 3 other outbreaks in the same village:
- general composition: high density area for pigs, medium density area for cattle, low density area for small ruminants;
- size: 433 km2;
- composition: see tables. The index case is not immediately detected; the 3 other outbreaks in the cluster pop up within a week of the detection of the index case. Characteristics of cluster 2 Cluster 2 is comprised of the 10 km surveillance zone surrounding 2 neighbouring outbreaks:
- general composition: medium density area for cattle, medium density area for pigs, low density area for small ruminants;
- size: 380 km2;
- composition: see tables. Both outbreaks are detected in an early stage and subsequently no more outbreaks are recorded in this area. The medium density area for pigs is misleading: there is only one big pig holding (one of the outbreaks) in the area and a few very small backyard holdings Characteristics of cluster 3 Cluster 3 is comprised of the fused 10 km surveillance zones surrounding a total of 23 outbreaks in cattle, pigs or small ruminants. Vaccination is performed throughout the inner part of this zone, but is not performed in a 5 km band on the edge; the latter acts as a buffer zone with the non-infected areas. The main characteristics of cluster 3 are:
- general composition: high density area for pigs, high density area for cattle and low density area for small ruminants;
- size: 1.440 km2; - Composition: see tables.
EU FMD-NSP workshop 47
The outbreaks in cluster 3 are grouped into 2 larger sub clusters (7 and 6 outbreaks), 2 smaller sub clusters (3 and 3 outbreaks) and 4 separate outbreaks. Several of the outbreaks have no clear link with any other outbreak or are recorded after vaccination in this cluster has started. Control strategy and vaccination The authorities have opted for a combination of culling, zoning, restriction measures and vaccination since:
- the FMDV has spread rapidly from the orginal cluster to other parts of the country;
- the epidemiological circumstances of the spread are not clear;
- the number of outbreaks and the fact that large professional farms are involved put an enormous strain on the capacity for depopulation.
In each of the scenarios all susceptible animals in the outbreaks are culled. Only in the large pig farm (outbreak 6) in cluster 2, solely the pigs in the affected quarantaine stable are culled. In all outbreaks, the classic 3 km protection zone and 10 km surveillance zone are originally delimited. Since the exercise is situated at least one month after the last outbreak, the protection zones no longer exists; only vaccination zones and the non-vaccinated remainders of the surveillance zones remain. The details of the vaccination strategy in each of the clusters are summarized under “scenarios”. Scenarios of the exercise The time setting for the exercise is the end of April, at least 1 month after the last outbreak has occurred. Since the last outbreak, the surveillance has been based on clinical surveillance. Apart from the holdings that are culled, no serological surveillance has been performed yet. The aim of the exercise is to device and to interpret a post-vaccination FMD surveillance scheme that includes the use of NSP tests. The surveillance scheme must substantiate a claim to return to the status free of infection (i.e. freedom of infection with a certain degree of confidence, according to the EU guidelines and the OIE general and specific serosurveillance guidelines) taking into account the characteristics of the laboratory test or tests chosen. The participants are divided into working groups that will device the surveillance scheme for one of the scenarios.
- Since the exercise focuses on the use of NSP test, the scheme for dealing with both the vaccinated and non-vaccinated animals in the vaccination areas is the principal assignment.
- Secondly, the scheme for the rest of the surveillance zone must also be elaborated.
- Finally, based on the schemes presented, the groups will be presented with results of the serosurveillance scheme chosen. They will have to interpret these results and present a suitable follow up of the seropositive animals/herds/flocks. A sample is considered positive when it will be twice positive in a double, successive testing in the chosen analysis scheme.
EU FMD-NSP workshop 48
Scenario 1 Scenario 1 focuses on cluster 1 that surrounds the 4 outbreaks around the index case. It takes into account the vaccination of all cattle and pigs in a 3.000 m radius around the outbreaks (the former protection zone). The size of the vaccination area is 57 km2. Scenario 2 Scenario 2 focuses on cluster 2 surrounding the large pig farm and a second outbreak. All cattle and small ruminants in the second outbreak and the pigs kept in the affected stable on the large pig farm are culled. The vast majority of the pigs in the affected pig farm are not culled (separate production units). This scenario takes into account the vaccination of all cattle and pigs in a 3.000 m radius around the outbreaks (the former protection zone). The size of the vaccination area is 32 km2. Scenario 3 Scenario 3 focuses on cluster 3 that surrounds 23 outbreaks. It takes into account the vaccination of all all cattle in the whole of the fused surveillance zones around the outbreaks, except an outermost 5 km band that acts as a buffer area around the vaccinated zone. The size of the vaccination area is 706 km2. Annexes The tables in annexe reproduce:
- the surface of the different zones,
- the composition of the various clusters with a stratification by radius,
- the composition of the various clusters with a stratification by size. Several maps visualise the three clusters. General time setting
- The outbreaks occur between the end of February and the end of March. - The vaccination campaign is performed:
- in the first week of March for clusters 1 and 2, starting on 4 respectively 5 March; - in between 14 and 21 March for cluster 3.
- The final monitoring begins at the end of April. Standard policy applied t control the outbreaks
- Delimitation of the classical protection zone (3 km) and surveillance zone (10 km). - Culling of infected herds. - At the detection of the first outbreak, on 27 February, a standstill is imposed in the part
of the country where clusters 1 and 2 are situated. A standstill is also imposed on 10 March in the part of the country comprising cluster 3 after the detection of the first outbreak in this cluster (outbreak 7).
- Vaccination is performed according to the above mentionned dates and the schemes mentionned in the background document.
Time schedule of the outbreaks See timeline
EU FMD-NSP workshop 49
OUTBREAK 1 = pig holder
ID-number 100332373
cluster 1
link to = index case
source of infection contaminated products from Turkey, brought home by owner’s son
date of infection around 13-02
date of confirmation 27-02
localisation municipality of Ljutomer
type of holding fattening pig holding
structure/capacity different small sheds
type of rearing Indoors
number of animals present 84 pigs: - 9 sows - 12 piglets - 60 fattening pigs of 80 kg - 2 gelts - 1 boar
Particularities
- On 20-02 one of the heavier pigs is showing symptoms of inappetite. Since this pig is destined to be slaughtered soon, the owner decides to slaughter it immediately at home. The slaughter is done by the local butcher. The owners’s brother participates in the slaughter and takes some of the meat home.
- On 25-02 the owner notices inappetite in one of the pigpens. The same day, after a quick investigation, the veterinarian suspects traumatic injuries due to the feed the animals have received, allthough some animals have fever.
- On 26-02 in the morning, the veterinarian is recalled since more pigs show the same symptoms of inappetite, salivation, marked fever and lameness as well. On detailed clinical examination of the animals, he establishes changes in the tongue and mucous tissue of the oral cavity, and additionally in the skin of the feet between the claws in the interdigital space, characteristic of foot-and-mouth disease. Consequently, he suspects the presence of foot-and-mouth disease virus. The veterinary service is notified.
EU FMD-NSP workshop 50
The veterinary service visits the farm, takes the necessary samples and sends the samples immediately to the laboratory. In view of the suspicion, the holding is blocked and all farms in the same village are put under surveillance and cannot move cattle, small ruminants and pigs any more.
- On 27-02 FMD is diagnosed in the lab. - On 28-02 the animals in the outbreak are killed and destroyed.
OUTBREAK 2 = brother of the index
ID-number 100344771
cluster 1
link to outbreak 1
source of infection indirect contact due to owner’s contact with infected pig from index case
date of infection 20-02
date of confirmation 27-02
localisation municipality of Ljutomer, 140 m from index
type of holding cattle and sheep
structure/capacity 1 stable
type of rearing indoors
number of animals present 13 cattle, 3 pigs
Particularities
- The FMD virus is brought on the farm by the owner who has helped slaughter the first infected pig at the index case. - The infection in this farm is detected by the veterinary service that visits the farm on 26-02 based on the inquiry at the index case. Two
cattle in this farm show few mild clinical signs of FMD: changes in the tongue and mucous tissue of the oral cavity and fever. Samples are taken and sent to the laboratory.
- On 27-02 FMD is diagnosed in the lab. - On 28-02 the animals in the outbreak are killed and destroyed
EU FMD-NSP workshop 51
OUTBREAK 3 = neighbour of the index
ID-number 100322348
cluster 1
link to outbreak 1
source of infection indirect contact due to probably airborne spread
date of infection around 20-02
date of confirmation 28-02
localisation municipality of Ljutomer, 50 m from index
type of holding cattle
structure/capacity 1 stable
type of rearing indoors
number of animals present 16 cattle
Particularities
- The introduction of the FMD virus is probably due to airborne spread from the index case. - The infection in this farm is detected by the veterinary service that visits the farm on 27-02 after being allerted by the owner that some of
his cattle in the shed next to the index case show fever. Samples are taken and sent to the laboratory. - On 28-02 FMD is diagnosed in the lab. - On 29-02 the animals in the outbreak are killed and destroyed.
EU FMD-NSP workshop 52
OUTBREAK 4 = first case in cluster 2
ID-number 100337021
cluster 2
link to outbreak 5
source of infection direct contact due to infected sheep brought by the sheep trader of outbreak 5
date of infection 24-02
date of confirmation 02-03
localisation municipality of Domzale
type of holding mixed holding
structure/capacity several sheds for maximal 200 sheep
type of rearing outdoors, next to the stables and the quarantaine stable of outbreak 6
number of animals present 8 cattle, 4 sheep, 3 pigs
Particularities
- The introduction of the FMD virus is due to direct contact on 24-02 when the trader of outbreak 5 takes infected sheep to the farm. - The sheep are placed in a shed on a pasture next to large pig farm (outbreak 6) where the cattle and pigs from the same owner are kept. - The infection in this farm is detected by the veterinary service who visits the farm on 01-03 following the notification by the veterinarian of
the farm of a suspicion of FMD in the pigs of the farm. - On 02-03 in the morning FMD is diagnosed in the lab. - On 02-03 the animals in the outbreak are killed and destroyed. - The infected animals are not supposed to be on this pasture so close to the big commercial pig farm, but the owner has decided to ignore
this rule and has temporarily housed the animals on the pasture because of the bad state of his stable.
EU FMD-NSP workshop 53
OUTBREAK 5 = sheep trader – source of spread towards clusters 2 and 3
ID-number 100474259
cluster 1
link to outbreak 1
source of infection indirect contact due to the butcher who slaughtered the first infected pig
date of infection around 20-02
date of confirmation 02-03
localisation municipality of Ljutomer, other part of the village 450 m from index
type of holding sheep trader
structure/capacity several sheds for maximal 200 sheep
type of rearing indoors
number of animals present 30 sheep, 2 pigs
Particularities
- The introduction of the FMD virus is due to indirect contact through the butcher that slaughtered the first infected pig in the index case. On the same day (20-02), the butcher goes slaughter a sheep at the trader’s.
- The infection in this farm is detected by the veterinary service that visits the farm on 02-03 following the inquiry at the first case in the second cluster. Some of the sheep at the trader’s show some changes in the tongue and mucous tissue of the oral cavity and some fever. Samples are taken and sent to the laboratory.
- On 02-03 in the evening FMD is diagnosed in the lab. - On 03-03 the animals in the outbreak are killed and destroyed. - The trader is very reluctant to disclose information about his activities in the last 2 weeks. He does not disclose the selling of sheep to a
sheep trader that later on is declared outbreak 9, the first outbreak in cluster 3.
EU FMD-NSP workshop 54
OUTBREAK 6 = big pig farm
ID-number 100295700
cluster 2
link to outbreak 4
source of infection indirect contact due to airborne spread of FMD from the pigs of outbreak 4 kept in a pasture; the ventilation fan of the infected quarantaine stable sucks air coming from the pasture
date of infection around 28-02/01-03
date of confirmation 04-03
localisation municipality of Domzale, 30 m from outbreak 4
type of holding big commercial pig farm
structure/capacity closed holding with 9 stables for sows/piglets and 36 stables for fattening pigs; separate quarantaine unit
type of rearing indoors
number of animals present 39.844 in total: 3916 sows, 72 boars, 16444 piglets, 18.885 fattening pigs of different age
Particularities
- The introduction of the FMD virus is due to airborne spread from the pigs kept on the pasture next to the quarantaine stable of the farm. The ventilator of the stable sucks air passing over the pasture.
- The quarantaine stable is separated from the other stables on the farm. The animals are only cared for as the last duty in the daily routine. Before entering or leaving the quarantaine stable, staff have to go through strict biosecurite procedures.
- The farm veterinarian raises the suspicion on mild clinical symptomes in the pigs in the quarantaine stable. As a result of outbreak 4, the pigs in this stable (that is the closest to the infected pasture) are inspected twice a day. At the morning inspection on 03-03, 4 out of 69 pigs (all gelts brought in the week before) in the stable show a temperature and reddening in the mouth. The other animals in the stable show no clinical signs of disease.
- The authorities are called in, the suspected pigs in the stable are killed and samples are sent to the laboratory, where FMD is diagnosed on 04-03.
- The same day (03-03), the owner decides to kill the 67 remaining gelts in the quarantaine unit. - It is believed that because of the very early detection of the outbreak, the fact that it is a quarantaine stable with strict biosecurity measures
and the fact that the stables next to the quarantaine stable and the pasture with the infected pigs of outbreak 4 are empty, the disease will
EU FMD-NSP workshop 55
not spread to the rest of the farm. As a precautionary measure, on 05-03, vaccination is started immediately in the rest of the farm, progressing outwards beginning at the stables closest to the quarantaine stable.
EU FMD-NSP workshop 56
OUTBREAK 7 = sheep farm
ID-number 40000672
cluster 3
link to outbreak 9
source of infection direct contact due to infected sheep bought from outbreak 9
date of infection 05-03
date of confirmation 11-03
localisation municipality of Lotenhulle
type of holding backyard sheep holding
structure/capacity -
type of rearing outdoors
number of animals present 3
Particularities
- The introduction of the FMD virus is due to the purchase of 3 sheep from outbreak 9. - The infection is detected by the farm veterinarian who visits the farm on 09-03 after the owner has observed some inappetite in the
purchased sheep. The veterinarian does not suspect FMD at first but reviews his opinion after visiting the next day outbreak 8 where he observes some clearer symptoms of FMD, also in sheep bought from outbreak 9.
- The veterinary service is called in on 10-03 and samples are taken and sent to the laboratory. - On 11-03 FMD is diagnosed in the lab. - On 10-03 the animals in the outbreak are killed and destroyed.
EU FMD-NSP workshop 57
OUTBREAK 8 = sheep farm
ID-number 30030867
cluster 3
link to outbreak 9
source of infection direct contact due to infected sheep bought from outbreak 9
date of infection 05-03
date of confirmation 11-03
localisation municipality of Beernem
type of holding mixed pig and cattle farm with a few hobby sheep
structure/capacity fattening pig in closed stables and dairy cattle
type of rearing indoors
number of animals present 125 dairy cattle, 1.031 pigs of 12 and 18 weeks, 9 sheep
Particularities
- The introduction of the FMD virus is due to the purchase of 2 sheep from outbreak 9. - The farm veterinarian is called in on 10-03 after the owner has observed some inappetite and salivation in the 2 purchased sheep. Since
the day before, the veterinarian has seen similar symptoms in sheep of outbreak 7 and the sheep are coming from the same trader, he investigates the affected animals more closely this time. Based on the changes he finds in the tongue and mucous tissue of the oral cavity and a slight fever in the affected animals, he suspects FMD and calls in the veterinary service.
- Samples are taken and sent to the laboratory. - On 11-03 FMD is diagnosed in the lab. - On 10-03 and 11-03 the animals in the outbreak are killed and destroyed. - At close inspection at culling, a limited number of pigs begin to show the first symptoms of FMD. The samples taken at culling reveal
infected animals. Neither clinical signs nor unfavourable analysis results are observed in the cattle.
EU FMD-NSP workshop 58
OUTBREAK 9 = sheep trader
ID-number 100357660
cluster 3
link to outbreak 5
source of infection direct contact due to the purchase of infected sheep at outbreak 5 (sheep trader in cluster 1)
date of infection 24-02
date of confirmation 11-03
localisation municipality of Aalter
type of holding sheep and cattle trader
structure/capacity different pens for maximal 230 sheep and 25 cattle under one roof
type of rearing indoors and outdoors
number of animals present 122 sheep, 9 cattle
Particularities
- The FMD virus is brought into the premises by purchasing on 24-02 a group of 30 sheep, some of which are infected, at the trader of outbreak 5. These sheep are put in a stable at outbreak 9 with some 150 other sheep. Most of the animals are sold to sheep farmers in the surrounding villages in the 2 weeks following the arrival of the infected group.
- The infection in this farm is detected by the veterinary service that visits the farm on 10-03 following the inquiry at outbreaks 7 and 8, the first detected outbreaks in cluster 3. Only very obscure changes in the tongue and mucous tissue of the oral cavity and some slight fever are observed in some of the sheep present. Samples are taken and sent to the laboratory.
- On 11-03 FMD is diagnosed in the lab. - On 10-03 the animals in the outbreak are killed and destroyed. - The inquiry at the trader’s reveales the link to outbreak 5 (sheep trader in cluster 1) and the selling of animals (cattle and sheep) to some
20 clients in the region since the arrival of the infected animals from cluster 1. Most of the clients are one-time buyers who have bought and immediatley taken away their sheep, without the trader recording their coordinates. The subsequent epidemiological inquiry into these contact herds does not succeed in retracing all the animals sold.
EU FMD-NSP workshop 59
OUTBREAK 10 OUTBREAK 11 OUTBREAK 12
ID-number 30031138 31089581 40023141
cluster 3 3 3
link to outbreak 8 outbreak 8, 9 and 10 unknown
source of infection neigbouring contact neigbouring contact of outbreaks 8 and 10; contact of outbreak 9
unknown
date of infection around 10-03 around 10-03 around 05-03
date of confirmation 14-03 15-03 15-03
localisation municipality of Beernem municipality of Beernem municipality of Nevele
type of holding dairy cattle holding backyard sheep holding cattle holding
type of rearing indoors outdoors indoors
number of animals present 125 cattle 12 sheep 30 fattening cattle
particularities Neighbouring holding of outbreak 8. Family ties and important professional contacts exist between the two premises. The outbreak is detected in its early stages.
Neighouring holding of outbreaks 8 and 10 and direct contact of outbreak 9 following the purchase of 4 sheep. The 12 sheep were kept on a pasture. Suspicion is raised on 14-03 and confirmed on 15-03 by the veterinary service in the framework of inquiry into oubreak 9. The outbreak is detected in its early stages
The suspicion is raised on 15-03 by the farm veterinarian. The cattle show 10-to-12-day-old lesions. The owners, a 75-year-old couple has only observed limping the previous day. The premise is situated within the surveillance zone surrounding outbreak 7. The outbreak is an isolated case: no epidemiological link with any of the existing outbreaks is revealed.
EU FMD-NSP workshop 60
OUTBREAK 13 OUTBREAK 14 OUBTREAK 15
ID-number 30021906 40134423 40038270
cluster 3 3 3
link to unknown outbreak 9 unknown
source of infection unknown neighouring holding of outbreak 9 unknown
date of infection around 05-03 around 10-03 around 05-03
date of confirmation 15-03 15-03 16-03
localisation municipality of Oostkamp municipality of Aalter municipality of Waarschoot
type of holding fattening cattle and pigs closed pig farm cattle + pig
type of rearing indoors + outdoors indoors Indoors + outdoors
number of animals present 57 cattle, 11 pigs 1019 pigs 156 dairy cattle, 973 fattening pigs
particularities The suspicion is anonymously raised to the veterinary service on 14-03. A visit on the next day reveals indeed very mild symptoms in some of the pigs. These symptomes are confirmed by laboratory analysis on the same day. The subsequent indeep inquiry reveals that the owner has been trying to hide the infection: meat (in the freezer) and offal (in the dungheap) of 2 pigs are discovered and turn out to be positive in the laboratory analysis. There are no links to previous outbreaks.
The suspicion is raised by the veterinarian on 15-03 in the framework of the monitoring in the protection zone around outbreak 9. The farm is situated at 600 m of outbreak 9. The infection is detected in its early stages. It might be the result of the culling activities at outbreak 9.
The owner raises the suspicion on 16-03 following the appearance of clinical signs in both pigs (housed indoors) and the calves (housed in a semi-open stable). The premises are situated outside of the existing surveillance zone at about 10 to 12,5 km from the nearest outbreaks (7, 9 and 12). The first infected animal on the farm (a calve that died on 09-03 with, in retrospect, very mild symptoms of FMD) has probably been missed. No links with existing outbreaks exist.
EU FMD-NSP workshop 61
OUTBREAK 16 OUTBREAK 17 OUBTREAK 18
ID-number 40001504 30882006 40038531
cluster 3 3 3
link to outbreak 9 unknown outbreak 15 (at 2,5 km)
source of infection neighouring holding of outbreak 9 unknown unknown
date of infection around 10-03 around 11-03 around 09-03
date of confirmation 17-03 17-03 17-03
localisation municipality of Aalter municipality of Schuiferskapelle municipality of Waarschoot
type of holding fattening cattle + closed pig holding dairy cattle cattle farm
type of rearing indoors indoors outdoors (shed on pasture)
number of animals present 45 cattle, 1.734 pigs 178 cattle 10 fattening cattle
particularities The suspicion is raised by the veterinarian on 16-03 in the framework of the monitoring in the protection zone around outbreak 9. The farm is situated in between the outbreaks 9 and 14. The infection is detected in its early stages. It might be the result of the culling activities at outbreak 9.
The owner raises the suspicion on 17-03 following the appearance of clinical signs in the cattle. The holding is situated on the outskirts of the existing surveillance zones around outbreaks 9 and 7. There is no link with existing outbreaks. The outbreak is detected in its early stages. The premises is culled on 17-03 and 18-03.
Suspicion is raised by the veterinarian following a visit of the pasture in the framework of the clinical surveillance in the protection zone around outbreak 15. There are no known direct or indirect contacts with any of the other outbreaks. The outbreak is detected in its early stages.
EU FMD-NSP workshop 62
OUTBREAK 19 OUTBREAK 20 OUBTREAK 21
ID-number 40038343 30081003 30082050
cluster 3 3 3
link to outbreak 15 (at 4 km) outbreak 17 outbreak 17
source of infection unknown neighbouring holding of outbreak 17 (at 1,6 km)
neighbouring holding of outbreak 17 (at 3 km)
date of infection around 12-03 around 18-03 around 18-03
date of confirmation 18-03 22-03 23-03
localisation municipality of Waarschoot municipality of Ruiselede municipality of Ruiselede
type of holding cattle cattle + pigs diary cattle and closed pig farm
type of rearing indoors indoors indoors
number of animals present 145 cattle 83 fattening cattle, 763 fattening pigs 84 dairy cattle, 1.116 pigs
particularities Suspicion is raised by the veterinary service following a visit of the premises in the framework of the intensified clinical monitoring just outside of the protection zone around outbreak 15. There are no known direct or indirect contacts with any of the other outbreaks. The outbreak is detected in its early stages.
Suspicion is raised by the farm veterinarian following his visit of the premises in the framework of the clinical surveillance of the protection/ vaccination zone around outbreak 17. There are no known contacts with any other outbreak. The outbreak is detected in its early stages with a few pigs showing recent lesions of FMD. The outbreak might be the result of the culling activities at outbreak 17.
The farm is situated on the outskirts of the protection zone around outbreak 17. Suspicion is raised by the farm veterinarian in the framework of his clinical surveillance of the protection/ vaccination zone around outbreak 17. There are no known contacts with any other outbreak. The outbreak is detected in its early stages with 2 dairy cows showing the initial symptoms of FMD.
EU FMD-NSP workshop 63
OUTBREAK 22 OUTBREAK 23 OUBTREAK 24
ID-number 30880641 40193195 40038841
cluster 3 3 3
link to outbreak 17 unknown outbreaks 15, 18 and 19
source of infection neighbouring holding of outbreak 17 (at 1,1 km)
unknown neighbouring holding of outbreaks 15, 18 and 19 (at 1,3 to 2,5 km)
date of infection around 18-03 around 10-03 around 18-03
date of confirmation 24-03 24-03 24-03
localisation municipality of Schuiferskapelle municipality of Lovendegem municipality of Waarschoot
type of holding cattle hobby sheep breeding cattle
type of rearing outdoors outdoors indoors
number of animals present 6 fattening cattle 5 sheep 59 cattle
particularities Suspicion is raised by the farm veterinarian following his visit of the premises in the framework of the clinical surveillance of the protection/ vaccination zone around outbreak 17. There are no known contacts with any other outbreak. The outbreak is detected in its early stages. The outbreak might be the result of the culling activities at outbreak 17.
Suspicion is raised by the owner following the dead of 2 of his sheep on their pasture. All 5 sheep turn out to have lesions consistent with a 2-weeks-old FMD infection. The owner has neglected the compulsory control visits and thus has missed the first symptoms. There are no links with any other outbreak.
Suspicion is raised by the farm veterinarian following his visit of the premises in the framework of the clinical surveillance of the protection/ vaccination zone around outbreak 15. There are no known contacts with any other outbreak. The outbreak is detected in its early stages.
EU FMD-NSP workshop 64
OUTBREAK 25 OUTBREAK 26 OUBTREAK 27
ID-number 31098406 30053190 40038580
cluster 3 3 3
link to outbreak 17 outbreak 25 outbreak 18
source of infection neighbouring holding of outbreak 17 (at 1,6 km)
neighbouring holding of outbreak 25 (at 1,2 km)
neighbouring holding of outbreak 18 (at 400 m)
date of infection around 18-03 around 23-03 around 18-03
date of confirmation 26-03 27-03 27-03
localisation municipality of Wingene municipality of Wingene municipality of Waarschoot
type of holding pigs dairy cattle breeding cattle and breeding pigs
type of rearing indoors indoors indoors
number of animals present 550 fattening pigs 84 cattle 38 cattle and 443 pigs
particularities Suspicion is raised by the farm veterinarian following his visit of the premises in the framework of the clinical surveillance of the protection/ vaccination zone around outbreak 17. There are no known contacts with any other outbreak. About a dozen pigs show clinical signs consistent with a week-old infection of FMD.
Suspicion is raised by the veterinary service following a visit of the premises in the framework of the intensified clinical surveillance of the farms around outbreak 25. There are no known contacts with any other outbreak. The infection is detected in its early stages.
Suspicion is raised by the farm veterinarian following his visit of the premises in the framework of the clinical surveillance of the protection/ vaccination zone around outbreak 15. There are no known contacts with any other outbreak. The outbreak is detected in vaccinated sows showing little symptoms, some around a week old. It might be the result of the culling activities at outbreak 18
EU FMD-NSP workshop 65
OUTBREAK 28 OUTBREAK 29
ID-number 400038262 31027381
cluster 3 3
link to outbreaks 15 and 18 outbreak 25
source of infection neighbouring holding of outbreaks 15 and 18 (at 1 to 1,3 km)
neighbouring holding of outbreak 25 (at 1 km)
date of infection around 18-03 around 23-03
date of confirmation 27-03 28-03
localisation municipality of Waarschoot municipality of Wingene
type of holding fattening calves pigs
type of rearing indoors indoors
number of animals present 243 fattening calves 425 fattening pigs
particularities Suspicion is raised by the farm veterinarian following his visit of the premises in the framework of the clinical surveillance of the protection/ vaccination zone around outbreak 15. There are no known contacts with any other outbreak. The outbreak is detected in 4 calves showing mild symptoms of about a week old.
Suspicion is raised by the veterinary service following a visit of the premises in the framework of the intensified clinical surveillance of the farms around outbreak 25. There are no known contacts with any other outbreak. The infection is detected in its early stages with only a few pigs in one pen showing symptoms.
EU FMD-NSP workshop 66
Size of the clusters:
area surface (km2)
cattle farm
density (per km2)
cattle density
(per km2)
sheep/goat farm
density (per km2)
sheep/goat density
(per km2)
pig farm density
(per km2)
pig density
(per km2)
cluster 1 - vaccination zone 57 2,09 35 0,18 2 5,40 1.423 cluster 1 - surveillance zone (including vaccination zone)
433 2,31 37 0,23 2 6,09 300
cluster 2 - vaccination zone 32 3,28 48 0,06 1 0,16 1.441 cluster 2 - surveillance zone (including vaccination zone)
Resources: G1 G2 Vet teams / VZ/PZ: 8 teams (16 persons) doing 2 visits per day
during 3.8 weeks (331 visits taking 14,454 samples) 8 persons supporting sampling Prioritization of sampling in large herds, larger sheep flocks and if closer to positive cases
Lab Bovine: Cedi NSP, retest all positives with a second Cedi-NS test Pigs: Cedi NSP (Se 0.70 / Sp 0.99), confirmation with UBI test Small ruminants: Cedi NSP, retest all positives with a second Cedi-NS test
12 persons doing lab testing, for a total number of samples of 14,454 + 2,185.
Surveillance zone (SZ) • No vaccination of animals in 3 km – 10 km zone:
SZ: 22 teams (22 persons) doing 5 visits per day during 3.5 weeks (2,477 visits taking ~ 1185 samples + CS Prioritization of sampling in large herds, in sheep flocks and take wind direction and other possible routes for spread into consideration
Lab Cedi-NS + confirmatory VNT
EU FMD-NSP workshop 73
Initial laboratory results (given by the supervisory team, based on the design made by the group) Cluster 1:
G1 G2 • VZ/PZ: 2 big pig herds (ID 327, 325) have 31
(25) and 20 (17) positive results • VZ/PZ: 5 medium pig herds (ID 171, 263, 33,
178, 257) have 16 (10) positive results • VZ/PZ: 2 small pig herds (ID 312, 295) have 5
(1) positive results • VZ/PZ: 1 cattle herd (ID 103) has 1 positive
result • VZ/PZ: no reactors in small ruminants • SZ: 9 reactors in small ruminants (ID 2364,
379, 117), distributed over 3 farms
• VZ/PZ: 70 pig reactors in herd 48 • VZ/PZ: 9 pig reactors in herd 325 • VZ/PZ: 7 pig reactors in herds 33, 171, 178,
296, 327 • VZ/PZ: 8 bovine reactors in herd 103 • VZ/PZ: 1 bovine reactor in herds 99, 270 • VZ/PZ: no reactors in small ruminants • SZ: 3 sheep reactors in herd 1317
Interpretation and follow-up
G1 G2 • Analysis of test results: o No. of reactors higher than expected from
design? Yes in farm 327, 325, 312, 263, 2364, 379, 1317, 103. o Reactors: most pig herds have low level
reactors; one pig herd with strong suspicion of infection; one cattle reactor; one sheep reactor
• Resample and retest groups of animals with reactors to look if positives remain positive and some negatives became positive
• Cull single cattle and sheep reactors (since no probing samplers or virological confirmation tests available)
• Analysis of test results: o VZ/PZ bovine: sample all animals for
serology and virology, cull cattle herd with 8 positive reactors
o VZ/PZ pigs: collect epidemiological information from the herds, sample animals showing clinical signs for virology, if positive PCR: stamping out of the herd
o VZ/PZ: special case – herd no. 48 (60,000 pigs) – 20 strong positive results (all sows and in one section of 600 sows) + 50 weak positives distributed all over
o SZ: positive sheep herd: collect epidemiological information from the herd, sample all animals for virology, cull the flock
pigs, strong suspicion): o Pigs not carriers, so if re-sampling and testing
shows no circulation, the herd provides low risk
o Herd will need to be killed and farm disinfected – can be slaughtered for human consumption if not seropositive
o Consider this an outbreak, but occurred in February, so does not invalidate freedom claim
• VZ/PZ pigs: if negative PCR and if no epidemiological links and if no clinical signs in the herds: apply for FMD freedom
• VZ/PZ special case herd 48: o In suspicious section: clinical
examination of all sows, sample animals showing clinical signs for virology and SP-testing, cull all sows
o In other units: collect epidemiological information, clinical examination of all pigs, sample animals showing clinical signs for virology
o Keep herd under restrictions with additional bio-security measures
EU FMD-NSP workshop 75
Scenario 2: 2 outbreaks – 1 cluster All bovine/small ruminants culled in outbreak herds Only culling of pigs kept in affected stable Vaccination of bovine and pigs
Group 3 = Estonia + Slovakia + Åse Uttenthal + Philippe Vanier (Annex D) Group 4 = Poland + Finland + Aldo Dekker (Annex E)
Vaccination zone/Protection zone (VZ/PZ) • Vaccination of all bovine/pigs:
• No vaccination of small ruminants : 2 herds – 19 animals • Vaccination zone = protection zone Design: G3 G4 Bovine Clinical examination; at the herd level in 1 km
area around the outbreak, collect all animals in all holdings; outside the 1 km zone, test 10% within herd prevalence, dependant on the size of the herds. If herds <26, clinical examination only.
Sample all herds/all animals
Small ruminants Cull all animals or individual sampling + clinical examination
Sample all herds/all animals
Pigs Clinical examination of small holdings, For big pig farm: • Sample sows/ 2% within herd prevalence • Sample fattening pigs/5% within herd
prevalence • Piglets/5% within herd prevalence
Sample all herds/all animals, but for big pig farm, divide in: • 9 separate stables of sows, boars and piglets
- sample all units/ (95:5) animals • 36 separate stables of fattening pigs:
Clinical examination (once per day by farmer, once per week by official vet). Justification: (1) closed holding, (2) FMD is clinical disease and even if not clear signs in vaccinated population, transmission of virus effectively prevented, so no carrier status documented in pigs, (3) fattening pigs going to be slaughtered, so possible to check/sample at slaughter, (4) required a lot of resources
• VZ: backyard pig farm with 2 reactors • VZ: 3 mixed farms with one positive pig each • VZ: 5 small bovine farms have 1-3 reactors • VZ: 27 reactors in 5 different stables in large
pig farm • SZ: 1 sheep farm with 3 reactors
Interpretation and follow-up
G3 G4 • VZ: Positive cattle samples: If positive result, retest the same sample, if positive again, collect a sample from the same animal, if positive cull the animal and retest the whole herd + clinical inspection of all animals; if negative results, previous results are considered as false positive, else, cull the herd. • VZ: Positive sow samples: (Lower than maximum number of reactors allowed) Retest sample, if positive, cull the 4 sows. Clinical inspect and test the sows around these 4 ones depending on the epidemiological picture in the herd. • VZ: Positive fattening pigs: (Lower than maximum number of reactors allowed) Retest sample, if positive, re-sample from the same pig. If positive, sample 130 pigs again. • VZ: Positive piglets: (Lower than maximum number of reactors allowed) Retest sample, if positive, re-sample from the same pig. If positive, sample 130 pigs again. • SZ: positive cattle: Retest the samples with another test system (f.e. VNT); if possible re-sample the same animals+ the ones surrounding the positive ones (epid. investigation) and collect blood from 108 animals (5% prevalence) + clinical examination of the whole herd • SZ: positive pigs: Retest the samples with another test system (f.e. VNT); if positive, clinical examination of the whole herd, including cattle in the mixed herd
• VZ: backyard pig farm has 2 pigs which are both positive → cull (located in same municipality as outbreak no. 4)
• VZ: one mixed farm with 37 negative cattle and 1 positive pig out of 2 → false positive pig
• VZ: one mixed farm with 5 negative cattle and 1 seropositive pig out of 2 → false positive pig
• VZ: one mixed farm with 1 seropositive pig and 16 negative sheep → false positive pig (high specificity of the test in non-vacc. sheep).
• VZ: large pig farm: retesting of positive stables = (1) positive animals + 50% of penmates and pigs in adjacent pens
• VZ: small bovine farms (16-104 bovine): probing sampling and PCR-testing to rule out carrier status
• SZ: 1 sheep farm with 3 reactors → all reactors tested by VNT → 1 positive in VNT (1:45) = singleton reactor
EU FMD-NSP workshop 78
Field follow-up
G3 G4 / • For big pig farm: 23 reactors in the retesting of
the positive sows + 12 reactors out of 2107 samples in the group of penmates and adjacent pens → option for decision-making:
o Culling and disinfection of infected premises (46,000 pigs) + repopulation with vaccinated animals = unrealistic
o Culling of 5 positive stables (3988 sows+piglets) and disinfection + repopulation with vaccinated animals – usefulness?
o Vaccination of piglets from positive stables before transport to empty stables for fattening
o Revaccination of all sows (+ some fattening pigs – youngest stables)
o Do nothing
EU FMD-NSP workshop 79
Scenario 3: 25 outbreaks – 3 clusters Only bovine vaccinated All bovine/sheep/goats/pigs culled 0 - 500 m
Group 5 = Czech Republic + Sweden + Dònal Sammin (see annex F) Group 6 = Israel + Switzerland + Lea Knopf (see annex G)
Vaccination zone (VZ) • Vaccination of all bovine in 500m – 5 km:
� Bovine: 2,188 herds – 144,872 animals • No vaccination of pigs/small ruminants:
Design: G5 G6-EU approach G6-Israeli-Swiss approach Bovine Large survey: sample all
herds/all animals CS all herds /all animals Sample all herds /all animals
CS (95:5) herds and sample 10% of inspected holdings + Two-stage sampling Hse 80%, Hsp 99.5%, (95:5) herds, if herd >=60 and if herd ≤75 animals / all animals, if herd >75 animals/75 animals per herd. (within herd prevalence of 5%)
Small ruminants
Preliminary survey: Sample (95:5) herds/(95:20) animals Large survey: sample (95:2) herds/(95:5) animals, but all animals in flocks <25 animals and 1 large farm (1,200 animals)
CS all herds /all animals Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤30 animals / all animals, if herd >30 animals/30 animals per herd. (within herd prevalence of 10%)
No clinical inspection since no clinical signs Two-stage sampling Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤30 animals / all animals, if herd >30 animals/30 animals per herd. (within herd prevalence of 10%)
Pigs Preliminary survey: Sample (95:5) herds/(95:20) animals Large survey: CS
CS all herds /all animals Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤20 animals / all animals, if herd >20 animals/30 animals per herd. (within herd prevalence of 20%)
CS all herds /all animals and sample 5% of inspected holdings + Two-stage sampling Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤20 animals / all animals, if herd >20 animals/20 animals per herd. (within herd prevalence of 20%)
animals CS all herds /all animals Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤30 animals / all animals, if herd >30 animals/30 animals per herd. (within herd prevalence of 10%)
CS (95:5) herds with Se 95% and Sp 99%/ Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤20 animals / all animals, if herd >20 animals/20 animals per herd. (within herd prevalence of 20%)
Small ruminants
Preliminary survey: Sample (95:5) herds/(95:20) animals Large survey: sample (95:2) herds/(95:5) animals, but all animals in flocks <25 animals
CS all herds /all animals Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤30 animals / all animals, if herd >30 animals/30 animals per herd. (within herd prevalence of 10%)
CS (95:5) herds with Se 95% and Sp 99%/ Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤30 animals / all animals, if herd >30 animals/30 animals per herd. (within herd prevalence of 10%)
Pigs Preliminary survey: Sample (95:5) herds/(95:20) animals Large survey: CS
CS all herds /all animals Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤20 animals / all animals, if herd >20 animals/20 animals per herd. (within herd prevalence of 20%)
CS (95:5) herds with Se 95% and Sp 99%/ Sample in two-stage approach: Hse 90%, Hsp 99%, (95:5) herds, if herd >=10 and if herd ≤20 animals / all animals, if herd >20 animals/20 animals per herd. (within herd prevalence of 20%)
Analysis of test results: o 107 pig farms with reactors
(2 expected) o 6 sheep farms with reactors
(0 expected) o Some farms with
unexpectedly high numbers of reactors
• Actions: o Intensified CS in pigs in
entire VZ, but focus on Waarschoot and neighboring municipalities, prioritize large farms (>1,000)
o CS of cattle in buffer zone o Actions on all infected
(confirmed) farms (culling, tracing, ..)
o Perform large survey
• Analysis of test results: o VZ:
- cattle: 1,159 reactors – 1,159 allowable
- pigs: 404 reactors – 404 allowable
- small ruminants 9 reactors – 9 allowable o SZ:
- cattle: 140 reactors – 140 allowable
- pigs: 275 reactors – 404 allowable
- small ruminants 21 reactors – 21 allowable
• Analysis of test results: o VZ:
- cattle: 111 reactors – 111 allowable
- pigs: 404 reactors – 404 allowable
- small ruminants 9 reactors – 9 allowable o SZ:
- cattle: 140 reactors – 140 allowable
- pigs: 275 reactors – 404 allowable
- small ruminants 21 reactors – 21 allowable
EU FMD-NSP workshop 83
Field follow-up
G5 G6 • Large survey – analysis of test results:
– Sheep: retest positive samples, send vets to flocks where still positives, sample all animals and prioritize SZ.
– Cattle: cull all NSP+ cattle, slaughter rest of herds in 3 municipalities, but only cull NSP+ animals
– Sample all sheep herds in 3 municipalities
• Vaccinated animals (cattle): 1. resample all positive animals 2. epidemiological investigation of herds with
positive results 3. testing the re-samples with (a) Cedi-test (if test
negative = negative) and (b) if Cedi-test positive retesting with Svanova-test
4. If Svanova positive: elimination of positive animals and testing of whole herd
• Non-vaccinated animals: 1. Resample all positive animals 2. Epidemiological investigation of herds with
positive results 3. Testing the resamples: cattle with Svanova-
test and pigs, small ruminants with Cedi-test 4. If animals still positive, retest with SPCE 5. If SPCE positive: elimination of positive
animals and testing of whole herd Group conclusions (Remarks)
G5 G6 • Vaccination:
o The group proposes not to vaccinate in herds <50 animals
o The group proposes to vaccinate pigs on vaccinated cattle holdings
• Testing: o CS only in non-vaccinated cattle herds o Random sampling of vaccinated cattle
herds (126 herds), test all animals in each herd
o Non-vaccinated pigs and sheep should be treated as was proposed in the exercise
• Freedom cannot be substantiated yet
• First approach (following EU guidelines) proposes 156,980 samples for NSP testing and 3,780 samples for virus detection
• Second approach (IL-CH suggestion) proposes 22,365 samples for NSP testing and 590 samples for virus detection
• The confidence of the combined surveys (clinical + serological) needs to be calculated
• Would the second approach with considerably less testing be accepted by neighboring countries?
• After follow-up of positive animals (epidemiological investigation + serological survey): (i) all animals could be described as false positive = freedom of disease demonstrated, (ii) some animal in the follow-up were found positive = freedom of disease is not demonstrated
EU FMD-NSP workshop 84
Session Vi: Discussion of results and models for serosurveillance: Summary, Conclusions, Recommendations and Observations Summary Vaccination zone: Total no. of samples/herds: G1 G2 Bovine (vacc.) Small ruminants (non-vacc.) Pigs (vacc.)
119/1,989 10/88
308/8,990
119/1,989 10/88
308/12,377 G3 G4 Bovine (vacc.) Small ruminants (non-vacc.) Pigs (vacc.)
1,044 2/19 725
105/1,549 2/19
5/2,707 + check 4,270 animals at slaughter
G5 G6 (IL-CH) Bovine (vacc.) Small ruminants (non-vacc.) Pigs (non-vacc.)
2,188/144,872 391/2,920
CS
50+181/250+13,575 55+164/275+4,100
• Vaccinated bovine:
o Groups 1, 2, 4 and 5 propose testing all vaccinated bovine (based on the risk of becoming a carrier), although for group 5 this was only proposed for in the second (large) survey.
o Groups 3 and 6 propose to do a survey and test part of the population (cf. according to EU Directive 2003/85/EC, all herds/all animals have to be sampled). Group 6 makes a distinction between the serological and the clinical survey. Samples taken from the clinical survey will be investigated in serology and virus detection.
• Non-vaccinated pigs: o Only clinical surveillance (CS) of non-vaccinated pigs proposed by group 5. o Group 6 proposes sampling and CS. They make a distinction between the serological and the
clinical survey. Samples taken from the clinical survey will be investigated in serology and virus detection.
• Vaccinated pigs: o Smaller sample size recommended instead of sampling all animals (as according to EU
directive 2003/85/EC) proposed by groups 1 and 2, since it is nearly impossible to sample all animals in big pig holdings.
o Group 4 proposes to sample all small holdings and sample part of the population in the big pig holding and to check 4,270 fattening pigs at slaughter
• Non-vaccinated sheep and goats: o Groups 1, 2, 3 and 4 agree on sampling following EU Directive 2003/85/EC. o Groups 5 and 6 propose sampling part of the population
• Most groups calculated the Herd Cut Point to see if SP positive results should be considered as significant.
EU FMD-NSP workshop 85
• G2 and G4 calculated the number of vet teams necessary to perform the surveys. Surveillance zone: Total no. of samples/herds: G1 G2 Bovine (non-vacc.) Sheep/goats (non-vacc.) Pigs (non-vacc.)
o Groups 1, 4, agree on clinical surveillance only, G2 proposes to take samples from animals showing clinical signs as well, G3 proposes to take samples
• Non-vaccinated sheep and goats: o Group 6 proposes clinical surveillance only o Groups 1, 2 and 3 follow EU Directive 2003/85/EC o Group 4 proposes to sample all animals in sheep holdings without other small ruminants o Group 5 proposes to sample part of the population
• The in advance calculation of necessary lab resources and vet teams to execute the survey was very useful for some groups.
• Calculation of expected positive herds and expected positives per herd was done by most groups
EU FMD-NSP workshop 86
Conclusions The approaches taken by the different working groups/countries showed a clear degree of similarity. The results were extensively discussed and the conclusions taken all three workshops were accepted: 1. The vaccination-to-live policy with subsequently substantiating freedom from infection by a survey
system including NSP testing is a realistic and achievable option in FMD control. However, it must be clear that in order to regain a free status without vaccination within a reasonable time, i.e. in accordance with Article 2.2.10.7 paragraph (1)(c) of the Terrestrial Animal Health Code, a stamping out policy, as defined by OIE, remains part of the control policy such as in the beginning of an epidemic.
2. NSP assays are not sensitive enough to detect all infected animals (especially carriers) in a
vaccinated population and therefore conclusions on the status of the herds can only be on a herd basis and in combination of results from clinical and serological surveys and epidemiological investigations such as cluster analysis.
3. Demonstrating absence of infection, in particular in a vaccinated population, is impossible and
therefore the term ‘demonstrate absence’ should be replaced by ‘substantiate absence’. 4. The current EU Directive (2003/85/EC) for the control of FMD mentions two surveys:
(a) A first survey, for detecting the presence of FMD virus in the vaccination zone (Article 56), should be a combination of clinical, epidemiological and serological investigations with high overall system sensitivity (Martin et al., 20071). Article 56 takes into account that the vaccination status within a vaccination zone may not be homogeneous, i.e. that within the declared vaccination zone there might be vaccinated and non-vaccinated animals. Consequently this survey includes: • a survey of non-vaccinated animals for detecting FMD virus infection in the same manner as
in a protection zone without vaccination; • a serosurveillance of all herds with vaccinated animals for detecting FMD virus infection by
use of NSP tests. Within the herds all vaccinated ruminants and their non-vaccinated off-spring must be sampled and other species kept in large numbers making individual testing impractical shall be sampled based on a 5% prevalence with 95% confidence.
(b) A second survey, to regain the freedom from infection status after emergency vaccination (Article 61), must have a high specificity. This survey might include a second serosurveillance. However, the serosurvey described in Article 56 could serve the purpose of the serosurvey required in Article 61 of the Directive, referring to surveillance guidelines to be laid down and in fact being the OIE Guidelines in Appendix 3.8.7 of the Terrestrial Animal Health Code2.
5. The follow-up of herds with seroreactors by serological investigation has to be based on a
combination of NSP assays (Paton et al., 2006) with well-defined performance characteristics and
1 Martin PA, Cameron AR, Greiner M. Demonstrating freedom from disease using multiple complex data sources 1: a new methodology based on scenario trees. PrevVet Med 2007 May 16;79(2-4):71-97. Epub 2007 Jan 16. Review. 2 In 2003 there were no Guidelines of OIE and it was discussed at OIE not at all to lay down Guidelines. A decision is in preparation to lay down that the OIE Guidelines should be the Guidelines mentioned in Article 61. The current OIE Guidelines also do not specify the period that must elapse between the last vaccination and the beginning of the survey.
EU FMD-NSP workshop 87
those assays should preferably be conditional independent. The final conclusion must take into account the specificity of the overall ‘test system’.
6. If the specificity of the ‘serological test system’ used were known, then one approach would be to
anticipate the rate of false positive test reactions and only consider seroreactor rates above this expected number or proportion as significant (Herd Cut Point - Paton et al., 20063). However, this is not considered as compatible with the OIE Code and the EU Directive 2003/85/EC, which require all herds with seroreactors to be followed up and classified as either containing or free from infection. Therefore re-sampling and testing herds with seroreactors should be applied. The latter approach would enable active virus circulation to be confirmed or ruled out by use of paired serology to look for changes in antibody status or titer rises and/or by evaluating results using Likelihood Ratios (Dekker et al., submitted4). Contingency plans should include a clear flow chart for the follow-up of seropositive herds.
7. A clinical surveillance combined with paired serology can detect holdings where virus circulation is
ongoing. However, there is no possibility of detecting each and every carrier within sub-clinically infected herds if, as seems likely, they are present in few herds and at a low level. Therefore all ruminants should be tested. Evidence of virus circulation would lead to the declaration of an outbreak and the stamping out of the herd, but evidence of carriers (in particular if at a low number within a herd) should, different to the current requirements, lead to slaughter of these reactor animals only, but not of the whole herd.
The advantages of this approach would be:
(a) The test specificity can be lowered since the consequence of false positive results is now individual animal rather than whole herd slaughter.
(b) This in turn leads to an increased test system sensitivity (c) The “small herd” problem is decreased
(cf. Annex H). 8. Testing all animals in the vaccinated population as prescribed in the EU Directive 2003/85/EC, is
certainly the way forward in case of vaccinated bovine, it is however considered as not achievable in areas with a dense pig population or within big pig herds, if such pigs have been vaccinated. In addition is the risk of developing a carrier state in pigs considered to be minimal, if not negligible. It is proposed to replace Article 56(3) (b) of Directive 2003/85/EC by the following: ‘… testing for antibodies against NSP of the FMD virus shall be carried out on samples taken from all vaccinated large ruminants and their non-vaccinated offspring and from all vaccinated pig herds with a within herd sampling based on a 5% prevalence with 95% confidence …’.
3 DJ Paton et al., Application of non-structural protein antibody tests in substantiating freedom from foot-and-mouth disease virus infection after emergency vaccination of cattle, Vaccine. 2006 Oct 30;24(42-43):6503-12. Epub 2006 Jul 5. Review. 4 Dekker A et al., Comparison of ELISAs for antibodies against foot-and-mouth disease virus non-structural proteins in cattle sera based on the continuous results, submitted, see Annex 9 of FMD-NSP program.
EU FMD-NSP workshop 88
9. Vaccination of small herds remains a controversial item. Two possible options were discussed: (a) A non-vaccination policy for small herds and integration in the survey system as sentinels.
This option is based on the fact that the 95% confidence cannot be achieved in small herds (considering a realistic within herd prevalence) and that small herds slow down the emergency vaccination campaign. Spatial cluster analysis should be applied to evaluate the serological survey results of the sentinels.
(b) Vaccinate small herds and test all animals in all of these small herds. This option is considered because vaccination of small herds contributes to an increase in vaccination coverage and thus to achieving the necessary level of protection for the population, considering the fact that vaccinating all animals with a homologous vaccine will only protect about 90% of the animals. It might also be politically difficult to deny vaccination to owners of small herds and thus exposing the animals to infection and possible stamping out.
10. This kind of workshop should also be done for other veterinary diseases such as Classical Swine
Fever or Avian Influenza (for all EU members, EUFMD countries and EU neighbours).
EU FMD-NSP workshop 89
Recommendations 1. Conclusions on the infection status of the herds after FMD outbreaks in a vaccinated population
should only be based on a survey system including at least clinical, serological and epidemiological investigations;
2. The performance characteristics of the survey system should be determined. When the desired confidence level cannot be reached, spatial clusters analysis should be considered;
3. The term ‘demonstrate absence of infection’ should be replaced by ‘substantiate absence of infection’;
4. Contingency plans should include a clear flow chart for the follow-up of seropositive herds, which must take into account at least the requirements of Appendix 3.8.7. of the Terrestrial Animal Health Code. Serological investigation should be based on a combination of NSP assays with well-defined performance characteristics;
5. All large ruminants should be tested to substantiate freedom from infection in a vaccinated population after FMD outbreaks. While evidence of virus circulation must lead to the declaration of an outbreak, consensus should be sought on the slaughter of reactor animals only, in case there should be evidence that these animals are carriers, instead of the whole herd removal as currently required in Article 57(3). Consideration should be given to finding out if there would be a consensus on the latter;
6. A change in the definition of an outbreak in OIE Guidelines and EU Directives is needed where carriers are concerned;
7. The relative confidence attainable with “herd-based” and “individual” certification needs to be explored for different herd sizes and prevalence;
8. Consideration should be given to an amendment of the Directive in order to allow a within-herd sampling scheme based on a 5% prevalence and a 95% confidence for vaccinated pigs;
9. The vaccination of small herds should be further discussed;
10. To refine the application of NSP tests, more work could be done in predicting the expected prevalence of infection within and amongst vaccinated herds;
11. Functional FMD expert groups should be created in every country (cf. article 78 of Council Directive 2003/85/EC for the EU Member States5), including in non-EU countries.
5 In article 78 it is stated that the group shall compose of epidemiologists, veterinary scientists and virologists in a balanced way
EU FMD-NSP workshop 90
Observations of WS2 and WS 3: During the discussions, following points were highlighted:
1. Recommendation 2 specifies that the performance characteristics of the survey system should be determined, but this should take into account the different species analysed in the survey system6.
2. As the OIE code has changed in May 2007, making containment of an infection region possible, the status of the country remains unaffected.
3. Some countries need training for the set-up of an information system for field data. Also, countries with a large population of backyard animals face a lot of problems in the collection, the control and the management of disease information.
4. The scenarios assume a perfect movement control, but it has to be kept in mind that this is not a real life situation.
5. Probang testing will not prove the absence of virus circulation. Therefore it was stated that probang testing brings more problems than solutions to substantiate freedom from infection.
6. It was concluded that the introduction of negative animals as sentinels in a vaccinated herd is of limited value due to the low transmission rate in these herds, as well for cattle as for pigs.
7. Several countries indicated that vaccination will not be their main option, especially if there are a lot of small herds in the country.
8. The organisation of a workshop on ‘vaccination: how and when, after or in face of an outbreak’ should be considered. It was considered important to have a good vaccination coverage. Also the purpose of vaccination was questioned: is it a disease control measure in a rising epidemic curve in combination with the DIVA principle to try to stop disease spread and to reduce the total outbreak number? But the expected residual infection remains to be detected. The vaccination of herds < 50 animals was questioned as well: the detection of 5% prevalence is easier if these herds are not vaccinated, but vaccination coverage is reduced (coverage of herds is probably more important than coverage of animals).
9. SP test can be used for the survey of non-vaccinated animals. SP-tests can also be used to check the efficiency of and coverage obtained through vaccination programmes (Conclusion 2).
10. Always try to slaughter for consumption, if slaughter should be necessary (Recommendation 5). 11. Sub-clinical transmission in cattle and pigs is unlikely. Virus circulation in pigs can be determined
by paired serology, similar to what is done in South America for cattle. Clinical signs in sheep are not absent but very difficult to discern from other diseases and easily missed.
12. It will be easier to prove absence of infection in the surveillance zone and it could be of help to substantiate absence of infection.
13. The WS shows that countries should be prepared for the possible outcomes of the surveillance. Also false positive results can be clustered, so it is essential to have a decision-scheme on the follow-up before the start of the outbreak.
14. If the 95% confidence cannot be achieved for a survey then it is still worth doing it but the confidence level should be determined and specified. The actual confidence level obtained should be taken into account for the final conclusion on the disease status.
15. It should be noted that during the 3 WSs different figures for diagnostic Se and Sp were used as well as different design prevalence although similar situations.
6 Martin PA, Cameron AR, Greiner M. Demonstrating freedom from disease using multiple complex data sources 1: a new methodology based on scenario trees. PrevVet Med 2007 May 16;79(2-4):71-97. Epub 2007 Jan 16. Review.
EU FMD-NSP workshop 91
Acknowledgments
DG RESEARCH
DG SANCO
VAR Belgium Food Agency Belgium
EUFMD
EU FMD-NSP workshop 92
Annex A see FMD-NSP workshop program
EU FMD-NSP workshop 93
Annex B
SCENARIO 1
First outbreak at end of February due to serotype A
The initial clinical signs are ignored by the owner, resulting in a rather late detection of the disease
EU FMD-NSP workshop 94
Characteristics of cluster 1
• Cluster 1 is comprised of the 10 km surveillance zones surrounding the index case and 3 other outbreaks in the same village:
• general composition: high density area for pigs, medium density area for cattle, low density area for small ruminants;
• size: 433 km2;• composition: see tables.• The index case is not immediately detected; the
3 other outbreaks in the cluster pop up within a week of the detection of the index case.
EU FMD-NSP workshop 95
EU FMD-NSP workshop 96
Exercise• The time setting for the exercise is at the end of
April, at least 1 month after the last outbreak has occurred (2 months for cluster 1). Since the last outbreak, the surveillance has been based on clinical surveillance. Apart from the holdings that are culled, no serological surveillance has been performed yet.
• The aim of the exercise is to device and to interpret a post-vaccination FMD surveillance scheme that includes the use of NSP tests.
Scenario 1
• Scenario 1 focuses on cluster 1 that surrounds the 4 outbreaks around the index case.
• It takes into account the vaccination of all cattle and pigs in a 3.000 m radius around the outbreaks (the former protection zone). The size of the vaccination area is 57 km2.
EU FMD-NSP workshop 97
PZ Census
81.11130888101.989119
24.09572.2843-00-001001 to
-00-00-00501 to 1000
3141.2574-00-00251 to 500
1442.59418-00-00101 to 250
722.02028-0091182251 to 100
351.1043229291329142926 to 50
181.1286416322176083511 to 25
74556372137238335 to 10
326996264247201 to 4
average number of pigs
total number of pigs
number of pig holdings
average number of sheep/goat
total number of sheep/goat
number of sheep/goat holdings
average number of cattle
total number of cattle
number of cattle holdings
size
cluster 1 - 0 to 3.000 m = vaccination zone
48.9202.3289258813.909883
1.5513.1022-00-001001 to
6083.0385-007117111501 to 1000
3718.15822-002762761251 to 500
15310.73570-001239828101 to 250
746.07182601813662.0573151 to 100
355.619161311244333.36410226 to 50
165.6343501644427174.27325811 to 25
73.6365288981371.6762305 to 10
32.9271.1082784125702521 to 4
average number of pigs
total number of pigs
number of pig holdings
average number of sheep/goat
total number of sheep/goat
number of sheep/goat holdings
average number of cattle
total number of cattle
number of cattle holdings
size
cluster 1 - 3.000 to 10.000 m = surveillance zone = non-vaccinated zone
SZ Census
EU FMD-NSP workshop 98
PZ Surveillance strategy
• Vaccinated cattle and pigs, unvaccinated sheep/goats– Clinical examination of all species on all holdings– Blood sampling of all vaccinated cattle on all holdings as
numbers not too great (~2,000)– Pigs to be sampled/tested so as to detect a 5% within herd
prevalence of infection with 95% confidence (stratified into three herds-size groups)
– Bleed and test all unvaccinated sheep & goats (very small numbers, n=88)
• Selected Cedi-NS for all screening (retesting all positives with a second Cedi-NS test). For pigs assumed Sp=99%, Se=70% to calculate sample numbers with Freecalc.
• UBI test used for confirmation in pigs• All animals to be marked for follow-up sampling
SZ Surveillance strategy
• No vaccination– Clinical examination of all species on all
holdings
– Blood sampling of sheep and goats only since clinical signs less obvious• Decided to test all animals as numbers relatively
small (n=925)• Used Cedi-NS and then asked CRL to carry out
confirmatory VNT (no SP test available)• All animals to be marked for follow-up sampling
EU FMD-NSP workshop 99
Protection zoneNumber of cattle to be tested (vaccinated) 1989 in 119 holdings Number of sheep and goats (non vaccinated) 88 in 10 holdings Number of pigs (vaccinated) 234 per holding in 3 large holdings
182 per holding in 4 medium holdings 7560 pigs in 301 little holdings (all pigs)
Surveillance zoneNumber of sheep and goats (non vaccinated) 925 in 88 holdings (all animals)
Sampling & testing numbers
Results of testing in PZ
0<21081GRLAVA4108103
01 (0)<21030KRAPJE1030257
02 (0)<1250VERŽEJ2537295
03 (1)060VERŽEJ60312
03 (1) 41820CVEN3143178
03 (2)41820IŽAKOVCI2521033
04 (3)<21190KRAPJE1190263
05 (4)41820CVEN2800171
020 (17)52340CVEN91150325
031 (25)52340LJUTOMER30270327
Pos_sheep
Pos_pigs
Expected
falsepositive
Numberof samples
Pos_cattle
MUNICIPALITYPIGSCATTLEID
EU FMD-NSP workshop 100
Breakdown of Positive results for 2 large pig herds
PZ• Pig herds with low level reactors• One pig herd with strong suspicion of
infection• One cattle reactorSZ• One sheep reactor
Follow-up
• Resample and retest groups of animals with reactors to look if positives remain positive and if some negatives became positive
• No probang samplers or virological tests available for testing cattle and sheep, so if single reactors remain they should be culled
• Problem of pig herd with strong suspicion of infection– Pigs not carriers, so if resampling and testing shows no
circulation, the herd provides low risk– Nevertheless, pigs will need to be killed and farm disinfected–
can be slaughtered for human consumption if not seropositive– Consider this an outbreak, but that this occurred in February and
therefore does not invalidate freedom claim
EU FMD-NSP workshop 102
Annex C
Testing in the vaccination zone
� Clinical examination and
serological testing by NSP-CEDI test
● Cattle
� All herds; all animals
● Sheep and goats
� All herds; all animals
● Pigs
� All animals in small herds (< 1000)
� Sample in large herds (3027; 9115; 60,142)
Within herd design prevalence 1%;
No. tests: 1150 (1135), 1200 (1172), 1200 (1192)
Selection of pigs in the large pig herds
� All animals showing clinical signs of disease
� A systematic sample taking a fixed proportion of the
animals
prop = (sample size) / (herd size)
� Ensure that all sections are represented in the
sample
EU FMD-NSP workshop 103
Testing in the non-vaccinated zone
� Cattle
● Clinical examination of all herds
● Serology/virology of animals showing clinical signs
� Sheep and goats
● Serology all herds; all sheep and goats
� Pigs
● Clinical examination of all herds
● Serology/virology of animals showing clinical signs
Number of visits and samples
2,477
88
2,328
883
Visits
2,18563,75414,454331Total
9259258810Sheep
~98048,92012,377308Pigs
~28013,9091,989119Cattle
SamplesExam.SamplesVisits
Non-vaccinated zoneVaccination zone
EU FMD-NSP workshop 104
Human resources
3.55222477Non-vacc. zone
3.828331Vacc. zone
Weeks *Visits /
Team*day
TeamsVisits
* Assuming 6 days working week
Vacc. Zone 22 teams á 1 person 22Non vacc. Zone 8 teams á 2 persons 16Supporting sampling 8Laboratory personell 12Total 58
Prioritisation of sampling
� In vaccinated zone
● Large herds
● Larger sheep flocks (sheep not vaccinated)
● Closeness to the positive cases
� In non-vaccinated zone
● Large herds
● Sheep flocks
● Take wind direction and other possible routes for
spread into consideration
EU FMD-NSP workshop 105
Results
0300650nv1317
0010074v270
0013044v99
700302700v327
70041100v296
70031403v178
70028000v171
700252010v33
00840108v103
900911500v325
20+50006014200v48
PigsSheepCattlePigsSheepCattleID
No positivesNo animals
Follow up of NSP-positive samples
In non-vaccinated zone (1 sheep flock with 65 animals)
� Collect epidemiological information from the herd
� Sample all animals for virology
� Cull the flock
EU FMD-NSP workshop 106
Follow up of NSP-positive samples
In vaccinated zone
� Cattle
● Sample all animals for serology and virology
● Cull cattle herd with 8 positive reactors
� Pigs
● Collect epidemiological information from the herds
● Sample animals showing clinical signs for virology
● If positive PCR� FMD detected: stamping out
● If negative PCR …
Follow up of herds with NSP positive samples
If all samples examined for PCR is negative
● If no epidemiological links and
● If no clinical signs in the herds
� Apply for freedom of FMD
EU FMD-NSP workshop 107
Follow up of herd with 60,000 pigs
● NSP-results
� 20 strong positives, all sows in one section of 600 sows
� 50 weak positives distributed all over
● Action in suspicious section
� Clinical examination of all sows
� Sample animals showing clinical signs for virology and SP-testing
� Cull all sows
● Action in other units
� Collect epidemiological information
� Clinical examination of all pigs
� Sample animals showing clinical signs for virology
● Keep herd under restrictions with additional bio security measures
EU FMD-NSP workshop 108
Annex D
Group 3Cluster 2
Estonia and Slovakia
EU FMD-NSP workshop 109
46.09851921.549105
46.09146.0911-00-001001 to
-00-00-00501 to 1000
-00-00-00251 to 500
-00-001041041101 to 250
-00-0069411651 to 100
-00-00375531526 to 50
-0016161172011211 to 25
-00-007166255 to 10
-743312114461 to 4
average number of
pigs
total number of
pigs
number of pig
holdings
average number of sheep/goat
total number of sheep/goat
number of sheep/goat
holdings
average number of
cattle
total number of
cattle
number of cattle
holdingssize
cluster 2 - 0 to 3.000 m = vaccination zone
EU FMD-NSP workshop 110
894631.3676613.6971.092
-00-00-001001 to
-00-005955951501 to 1000
-00-00-00251 to 500
159637415546631376855101 to 250
-00621853682.0993151 to 100
-00353189342.9078526 to 50
166341626917164.08625811 to 25
796147741172.3023305 to 10
298412552331.0233821 to 4
average number of
pigs
total number of
pigs
number of pig
holdings
average number of sheep/goat
total number of sheep/goat
number of sheep/goat holdings
average number of
cattle
total number of
cattle
number of cattle
holdingssize
cluster 2 - 3.000 to 10.000 m = surveillance zone = non-vaccinated zone
Description of outbreak
• One vaccinated zone– Divide into a 1km zone close to outbreak: 0- 1
km
– 1-3 km vaccination zone– CEDI test sens 86%; spec 99.5 %
• One non vaccinated zone– CEDI test sens 98%; spec 99.5
– Other confirmatory tests possible (vnt)
EU FMD-NSP workshop 111
Within 1 km zone
• Clinical examination of all herds• Sheep (19): the best is to cull them (welfare
problems?) or individual sampling plus clinical examination.
• Cattle (1.549): 2% of design prevalence in total animal population collection of 391 sera (FreeCalc minimum no of sera to be tested)
• At the herd level, in 1 km area around the outbreak, collect all the animals in all the holdings: 524 tests.
Precautions on pos samples cattle
• For positive results, Retest the same sample, if positive again, collect a second time the blood from the same animal; if positive, cull the animal (to test the animals for virus and abs) and retest the whole herd and inspect clinically all the animals. If negative results, the previous results are considered as false positive. If positive, the herd is culled.
EU FMD-NSP workshop 112
Vacc zone outside the 1 Km
• CATTLE: In all the herds, test depending on the size of the herds; the basis is 10% of within herd prevalence:
• 6 herds with 51 to 100: test 44=(6x44)264• 8 herds with 26 to 50: test 37 (8x37)=256• For the rest of herds lower than 26: clinical
examination only.
Pigs vacc zone 0- 3 kmClinical examination of the 7 pigs in 4 small holdings
• Blood testing in the big herd:• With sub clusters as the transmission of viruses
is not the same in different categories of animals:
• Sows (3916): 2 % within herd prevalence (5 reactors possible): 465 samples: 4 positive results: retest the sample. If positive: cull the 4 sows (to test virus and abs). Clinical inspect and test the sows around these 4 ones depending on the epidemiological picture in the herd.
EU FMD-NSP workshop 113
Surveillance zone• Cattle: Clinical examination in all herds; herds less than 26 only
clinical examination• For the rest 10% within herd prevalence• With tests of 98% of sensitivity and Spe: 99.5%
• Herds of:• 26 to 50: 33 samples in 85 holdings: 2805• 51 to 100: 38 31 1178• 101 to 150: 42 5 210• 595 44 1 44. • 4 positive; retest the samples with another test system such as
neutralization test; if positive, resample the same animals plus the ones surrounding the positive ones (epidemiological investigations) and collect blood from 108 animals (5 % prevalence) plus clinical examination of the whole herd.
Pigs vacc zone 3 km continued
• Fattening pigs: 5% within herd prevalence (2 reactors possible): 130 samples out of 18885; 1 positive, retest the sample; if positive, (normally individually identified at the first sampling): to collect blood from the same pig; if positive, a new sampling of 130 pigs.
• Piglets: 5% within herd prevalence (2 reactors possible): 130 samples out of 16444; 1 positive: the same as for fattening pigs.
• Particular attention to the piglets born after the vaccination and not vaccinated and use as sentinels.
EU FMD-NSP workshop 114
Surveillance zone continued
• For the other herds, one single reactor in the biggest herds;
• result expected; so, same samples should be retested with another test system such as neutralization test; if still positive, clinical examination of herds.
Sheep Surveillance zone
• 5% within herd prevalence• Herds less than 35 sheep: test all• 3 herds of 62 sheep: 54 tested per herd• 3 herds of 155 sheep: 70 tested per herd• All negative. No further action
EU FMD-NSP workshop 115
Pigs surveillance zone
59 small herds: only clinical examination• 4 holdings with average of 159 pigs: 70
pigs are tested per herd (5% within herd prevalence).
• 2 reactors in 2 large herds. So, same samples should be retested with another test system such as neutralization test; if still positive, clinical examination of the herds including cattle in the mixed herd.
We will survive!!
EU FMD-NSP workshop 116
Annex E
1.1.2007alatunnistetiedot
46 992681 3866815 2461 197
894631 3676613 6971 092surveillance zone
46 09851921 549105vaccination zone
totalnumber of pigs
numberof pigholdin
gs
total number of
sheep/goat
number of
sheep/goat
holdings
totalnumber of cattle
numberof
cattleholdin
gs
zone
Cluster 2 by Poland and Finland
• Two outbreaks; – mixed farm culled– large pig farm (n= 46 000 pigs) ->
animals at quarantine unit culled and all the restvaccinated
1.1.2007alatunnistetiedot
Vaccination zone
• All cattle and pigs vaccinated• Small ruminants non-vaccinated
• Survey design:
– All cattle sampled (all animals in all farms)-> 1 549 samples from 105 holdings
– All small ruminants sampled->19 animals 2 farms
– All backyard pigs sampled-> 7 pigs from 4 holdings
EU FMD-NSP workshop 117
1.1.2007alatunnistetiedot
Vaccination zone
• Large pig farm:
– 9 stables with sows, boars and piglets->sampling to detect 5 % within-stable prevalencewith 95% confidence (appr. 300 samples/stable,altogether 2700 samples)– 36 stables with fattening pigs-> clinical
monitoring only (once per day by the farmer, once per week by the official vet)
1.1.2007alatunnistetiedot
Vaccination zone
• Justification for not sampling all the animals in the largepig herd:
1. A closed holding2. FMD is a clinical disease and even if not clear signs
in vaccinated population, transmission of virus effectively prevented-> no carrier status documentedin pigs
3. Fattening pigs going to be slaughtered-> possible to check/sample at slaughter
4. Requires a lot of resources
Altogether 4270 samples from the vaccination zone
EU FMD-NSP workshop 118
1.1.2007alatunnistetiedot
Surveillance zone
• A clinical examination carried out in all farms in the surveillance zone
• In addition, those sheep farms with no ruminants(22 holdings),a survey to detect 5 % prevalencewith 95% confidence would be carried out -> in the mixed herds to rely on the clinical signs in cattle orpigs
• A total of 951 samples
1.1.2007alatunnistetiedot
Surveillance
• Sampling carried out in 134 farms ( 5220 samples)
• Altogether 1200 farms to be monitored
• Farms were sampling would be carried out wouldbe examined first
EU FMD-NSP workshop 119
1.1.2007alatunnistetiedot
Results in the vaccination zone
• Five farms with seropositive animals– One backyard farm with both two pigs
seropositive-> culling (located in the samemunicipality as the outbreak nb 4)
– One mixed farm with 37 cattle (negative) and 1+ve pig out of two -> false positive pig
– One mixed farm with 5 cattle (negative) and 1 seropositive pig out of two-> false positive pig
– One mixed farm with one seropositive pig and 16 negative sheep-> false positive pig (highspecifity of the test in non-vacc. Sheep)
1.1.2007alatunnistetiedot
Results in the vaccination zone
• 27 reactors in 5 different stables in the large pigfarm (3-10/stable)
-> retesting of positive stables: positive animals+ 50 % of penmates and pigs in adjacent pens
-> 23 reactors in the retesting of the positivesows + 12 reactors out of 2107 samples in the group of penmates and adjacent pens
EU FMD-NSP workshop 120
1.1.2007alatunnistetiedot
Options for the decision-making in the large pig farm
• Culling and desinfection of the infected premises(46 000 pigs) + repopulation with vaccinatedanimals –unrealistic
• Culling of the 5 positive stables (3988 sows + piglets) and desinfection + repopulation withvaccinated animals –usefullness?
• Vaccination of the piglets from positive stablesbefore transport to empty stables for fattening
• Revaccination of all the sows (+ some fatteningpigs-youngest stables)
• Do nothing
1.1.2007alatunnistetiedot
Vaccination zone
• In cattle, 1-3 reactors were detected in five smallfarms having 16-104 bovines -> probang-sampledand PCR-tested to rule out the carrier status
EU FMD-NSP workshop 121
1.1.2007alatunnistetiedot
Results in the surveillance zone
• Monitoring for clinical signs negative in cattle and pig farms and mixed farms
• Only sheep/goat farms sampled
• One farm of 232 sheep with 3 reactors -> allreactors tested by virus neutralisation-> 1 positivein virus neutralisation (1:45) = singleton reactor
1.1.2007alatunnistetiedot
Resources for the design and implementation of the survey
• Central Vet. Adm. : 2 persons for the planning etc.• Province: 2 prov. vet officers for coordination and compilation of the field data
• Field staff: 160 veterinary working days– 20 vet.days for sampling the cattle herds + 13
vet.days for sampling the sows (2700 in 9 stables) in the large pig farm + 18 vet.days for cl. monitoring of the 36 stables with 19 000 fattening pigs
– 100 vet.days for cl. monitoring of cattle and pigherds in the surv. zone + 11 vet. Days for sampling the sheep farms
• 6 sheep farms with reactors (0 expected)• Geographical cluster in Waarschoot
• Some farms with unexpectedly high numbersof reactors
EU FMD-NSP workshop 125
Actions
• Intensified clinical surveillance in pigs in entire vacc zone, but focus on Waarschoot and neighbouring municipalities, prioritise large farms (>1000)
• Clinical surveillance of cattle in buffer zone, large farms first (>250 cows)
• Paired serum samples• Surveillance in the buffer zone• Actions on all infected (confirmed) farms (culling,
tracing etc)
Large survey
• Pigs –clinical surveillance only • Sheep SP test to detect 2% infected herds and within-
herd prevalence 5% • Vaccination zone :Random selection of herds (plus 1
large with 1200 animals), all animals in smaller flocks (<25 animals), random samples in larger = 391 flocks, 2920 animals
• Buffer zone: Random selection of herds, all animals in smaller flocks (<25 animals), random samples in larger = 328 flocks, 2509 animals
• Cattle – NSP test in vaccinated, all animals in all herds. Non-vaccinated clinical surveillance only
EU FMD-NSP workshop 126
Results (serology)
• 49 positive sheep holdings in vacc zone (1-2/herd)
• 23 positive sheep holdings in surv zone (1-3/herd)
• 259 positive cattle holdings (1-15/herd)• Herds with >3 pos in 3 municipalities
(Sleidinge, Waarschoot, Lovendegem)
Actions
• Re-test positive sheep samples, send vets to flocks where still positives, sample all animals Prioritise surveillance zone
• Positive cattle herds: Cull all NSP positive cattle, slaughter rest of herd in 3 municipalities, other municipalities only cull NSP pos animals
Vaccinated animals (cattle):1. Resample all positive animals2. Epidemiological investigation of herds with positive
results3. Testing the resamples:
a) Cedi-test (if test negative = negative)
b) If Cedi-test positive retesting with Svanova-test4. If Svanova positive:
- Elimination of positive animals- Testing of whole herd
Follow up of positive animals II
Non-vaccinated animals:1. Resample of all positive animals2. Epidemiological investigation of herds with positive results3. Testing the resamples:
– Cattle with Svanova-test– Pig, sheep, goat with Cedi-test
4. If animals still positive:– Retest with solid phase competitive ELISA (SPCE)
5. If SPCE positive:- Elimination of positive animals- Testing of whole herd
EU FMD-NSP workshop 143
Conclusions
After follow up of positive animals (epidemiologicalinvestigation + serological survey):
• All animals could be described as false positive
= freedom of disease is demonstrated
• Some animals in the follow up were foundpositive= freedom of disease is not demontrated
EU FMD-NSP workshop 144
Annex H Post NSP Workshop thoughts, after the first FMD-NSP workshop in January 2007. The approach to post-vaccination serosurveillance set out in the EU Directive assumes that serological tests for virus infected animals will have an insufficient sensitivity to certify individual animals as infection-free, but that they could be used for herd-based certification. Accordingly, it is required that all vaccinated animals in all vaccinated herds or flocks should be sampled and tested serologically and that if infection cannot be ruled out, then the entire herds must be slaughtered and animals either disposed of into the food chain (in cases where it is considered that the probability of genuine infection is extremely low) or by destruction (in other cases). Available tests can be used to support this strategy (Paton et al., 2006) subject to certain limitations, namely that:
1. The need to slaughter entire herds where seroreactors are found makes it essential that the test system used has a very high specificity which in turn reduces the sensitivity.
2. The strategy will still lead to slaughter of some seroreactor herds of uncertain infection status. 3. The low test system sensitivity means that a low prevalence of infection cannot be reliably
detected and that small herds cannot be certified as infection free, even if they have a moderate (5%) prevalence of infection.
Another problem is that the strategy espoused by the EU Directive does not distinguish between the two different aims of detecting (1) virus circulation and (2) virus carriers, a concept that has recently been introduced into the OIE Code. Demonstrating absence of virus circulation can be substantiated by paired serology and arguably, is barely necessary if there is a properly conducted campaign of vaccination and clinical surveillance. In contrast, the feasibility of demonstrating the presence of carriers is entirely dependent on their prevalence within the population and poses the main challenge for post-vaccination serosurveillance in countries wishing to substantiate freedom from infection without vaccination, after use of emergency vaccination. There is a lack of certainty about the prevalence of infection (circulating virus or carriers) that is likely to be found after use of emergency vaccination within and amongst vaccinated herds. However, limited experience of outbreaks where emergency vaccination has been used (Balkans and The Netherlands) as well as preliminary findings from modelling studies suggest that the number of infected herds or flocks will be very few and within them the prevalence of infection will also be low – perhaps one or two animals only. This is because either immunity blocks transmission or else, where immunity is insufficient such that extensive infection can occur, then clinical signs are likely to be seen allowing the herd to be diagnosed as an infected holding without serology. In summary, clinical surveillance combined with paired serology can detect holdings where virus circulation is ongoing. However, there is no possibility of detecting each and every carrier within sub-clinically infected herds if as seems likely, they are present in few herds and at a low level. For example, even assuming a prevalence as high as 5% for carriers within herds, the approach outlined by Paton et al (2006) is only designed to detect carrier containing herds with 95% confidence, which means that 5% of herds containing carriers could expect to be missed.
EU FMD-NSP workshop 145
If the prevalence of carriers is so low amongst and between herds, that herd based testing with a relatively insensitive test cannot succeed, than an alternative testing strategy may be one based on individual animal certification. This could be combined with herd-based testing to detect circulating virus, under the assumption that evidence of virus circulation would lead to herd slaughter, but that evidence of carriers (particularly if at a low number within a herd) would lead to slaughter of these reactor animals, but not of the whole herd. This has a number of advantages:
(d) Test specificity can be lowered since the consequence of false positive results is now individual animal rather than whole herd slaughter.
(e) This in turn leads to an increased test system sensitivity (f) There is no longer a “small herd” problem nor a problem of inadequate specificity in very
large herds. For example, in cattle, a test system based on Ceditesting with repeat Ceditesting of seroreactors gives a test system sensitivity of 82% and specificity of 99.2%. If all vaccinated animals in all vaccinated herds were tested, then there is now an 82% probability of detecting all carriers, regardless of prevalence. At a specificity of 99.2%, 80 false positive seroreactors would be expected and have to be wrongly slaughtered for every 10,000 cattle tested. However, the advantage of the system is that the “unnecessary” slaughtering is shared amongst the herds so that each only suffers a little! To investigate this option further:
1. More work is needed to refine models predicting the expected prevalence of infection within and amongst vaccinated herds.
2. The relative confidence attainable with “herd-based” and “individual” certification needs to be explored for different herd sizes and prevalences.
3. The serosurveillance needed to substantiate freedom from virus circulation should be considered. 4. Consideration should be given to finding out if there would be a consensus on not slaughtering
individual carrier animals within vaccinated herds where there was no evidence of virus circulation.