EUROPEAN COMMISSION HEALTH & CONSUMER PROTECTION DIRECTORATE-GENERAL

EUROPEAN COMMISSION HEALTH & CONSUMER PROTECTION DIRECTORATE-GENERAL Directorate F - Food and Veterinary Office DG(SANCO)/8198/2006 MR Final FINAL REPORT OF A MISSION CARRIED OUT IN BRAZIL FROM 4 TO 17 JULY 2006 IN ORDER TO EVALUATE ANIMAL HEALTH CONTROLS IN PLACE IN PARTICULAR OVER FOOT AND MOUTH DISEASE AND PUBLIC HEALTH CONTROL SYSTEMS AND CERTIFICATION PROCEDURES 28/11/06-60263

TABLE OF CONTENTS 1. INTRODUCTION...4 2. OBJECTIVES OF THE MISSION...4 3. LEGAL BASIS FOR THE MISSION...5 4. BACKGROUND...5 4.1. Previous mission reports...5 4.2. Background to present mission...5 5. MAIN FINDINGS...6 5.1. Competent authority performance...6 5.2. Holding registration, animal identification and movement controls...9 5.3. Management of FMD...12 5.4. Food business operator s responsibilities...19 5.5. Official controls at establishment level...21 5.6. Certification of meat...22 5.7. Miscellaneous...24 6. CONCLUSIONS...24 6.1. Competent authority performance...24 6.2. Holding registration, animal identification and movement controls...25 6.3. Management of FMD...25 6.4. Food business operator s responsibilities...26 6.5. Official controls at establishment level...26 6.6. Certification of meat...27 6.7. Miscellaneous...27 7. OVERALL CONCLUSION...27 8. CLOSING MEETING...28 9. RECOMMENDATIONS TO THE COMPETENT AUTHORITIES OF BRAZIL...28 10. COMPETENT AUTHORITY REPONSE TO RECOMMENDATION...29 ANNEX- LEGAL REFERENCES...30 2

ABBREVIATIONS & SPECIAL TERMS USED IN THE REPORT ACB CA Cadastro CCA CSN DIA DIPOA DSA EITB ELISA 3ABC EU FMD FVO GO GTA HACCP LVU MAPA MS MT BND OIE OV PCR PR RS SDA SIF SIPAG SISBOV SP SVS VIAA Accredited Certifying Body for animal identification and data registration in SISBOV (Entidades Certificadora) Competent Authority Compulsory holding registers kept at the LVU Central Competent Authority National Sanitary Certificate (Certificado Sanitário Nacional) Animal Identification Document (Documento de Identificação do Animal) Department of Inspection of Products of Animal Origin (Departamento de Inspeção de Produtos de Origem Animal) Department of Animal Health (Departamento de Saúde Animal) Enzyme-linked ImmunoelectroTransfer Blot Indirect Enzyme-Linked Immunosorbent Assay for the detection of antibodies against 3ABC non-structural protein of the FMD virus European Union Foot and Mouth Disease Food and Veterinary Office Goias Animal Movement Permit (Guia de Transito Animal) Hazard Analysis of Critical Control Points Local Veterinary Unit Ministry of Agriculture, Livestock and Food Supply (Ministério da Agricultura, Pecuária e Abastecimento) Mato Grosso do Sul Mato Grosso National Bovine Database (Base Nacional de Dados) World Organisation for Animal Health Official veterinarian Polymerase Chain Reaction Paraná Rio Grande do Sul Secretariat of Livestock Defence (Secretaria de Defesa Agropecuária) Federal Inspection Service (Serviço Inspeção Federal) Animal and Plant Products Inspection Service Brazilian system for identification and certification of origin of bovine and buffaloes (Sistema Brazileiro de Identificação de Origem Bovina e Bubalina) São Paulo State Veterinary Service Virus Infection-Associated Antigen (agar gel immunodiffusion test for detection of antibodies against 3D non structural protein) 3

1. INTRODUCTION The mission took place in Brazil from 4 to 17 July 2006. The mission was undertaken as part of the Food and Veterinary Office's (FVO) planned mission programme. The mission team comprised four inspectors from the FVO. The inspectors operated in two sub-teams during the mission, one sub-team mainly covering aspects related to Foot and Mouth Disease (FMD) eradication and controls and the other sub-team focused on veterinary public health aspects including traceability. The inspection team was accompanied during the mission by representatives from the Central Competent Authority (CCA) in Brasilia, the Secretariat of Livestock Defence (SDA), Departments of Animal Health (DSA) and Inspection of Products of Animal Origin (DIPOA), under the Ministry of Agriculture, Livestock and Food Supply (MAPA). An opening meeting was held on 4 July 2006 with the CCA in Brasilia. At this meeting the inspection team confirmed the objectives of, and itinerary for the mission, and additional information required for the satisfactory completion of the mission was requested. 2. OBJECTIVES OF THE MISSION The objectives of the mission were to evaluate the measures taken by the Brazilian authorities to control FMD and provide adequate guarantees for export of fresh meat to EU, taking into account the conclusions and recommendations of FVO mission reports DG(SANCO)/7589/2005 and DG(SANCO)/8301/2006 (see section 4.1). The mission team reviewed in particular: (1) the FMD situation in all susceptible species in the light of recent events including regionalisation controls and other measures; (2) the implementation of the vaccination programme undertaken by the Brazilian authorities, including its coverage and efficacy; (3) the implementation of the sero-surveillance programme in all FMD susceptible species undertaken by the Brazilian authorities; (4) the systems for certification of animals and meat in relation to the requirements of Council Directive 96/93/EC 1 ; (5) the system for the control and recording of animal movements, including those controls necessary for certification in accordance with the requirements of Council Decision 79/542/EEC; (6) the controls in place over the production of fresh meat; 1 Reference to Community legislation cited in this report are listed in the Annex. 4

In pursuit of these objectives, the following sites were visited: COMPETENT AUTHORITIES Competent authorities Comments Central 1 Opening meeting and closing meeting in Brasilia, 1 additional meeting with CCA during the mission State 3 1 in SP, 1 in MS, 1 in GO Local 4 3 LVUs in the risk area in MS, 1 LVU in MT LIVE ANIMAL & ANIMAL PRODUCT CONTROL SITES Cattle holdings 13 3 in SP, 11 in MS, 1 in MT Feed lots 3 1 in SP, 1 in GO, 1 in MS Bovine semen collection centre 1 In SP Vaccine distribution centre 1 In MS Border inspection point 1 MS to Paraguay Inter-state border post 1 Between MS and PR Internal control posts 2 In the risk area in MS Accredited certifying bodies 1 In GO FOOD PRODUCTION/PROCESSING / DISTRIBUTION - ACTIVITIES EU approved slaughterhouses 6 EU approved coldstores 2 2 in MT, 2 in GO, 2 in MS (suspended approval) all with integrated cutting plants 2 in SP exporting products from slaughterhouses in approved states 3. LEGAL BASIS FOR THE MISSION The mission was carried out under the provisions of Article 46 of Regulation (EC) No 882/2004. 4. BACKGROUND 4.1. Previous mission reports A number of missions have been carried out by the FVO to Brazil in recent years. The reports of these missions are available on the Health and Consumer Protection Directorate General web-site at: http://ec.europa.eu/comm/food/fvo/index_en.htm 4.2. Background to present mission 4.2.1. Animal health information (FMD) Until the outbreaks in 2005, the country was divided, according to the World Organization for Animal Health (OIE), into an FMD free area with 16 States and two municipalities in the state of Amazonas, and an FMD infected area with 11 States, in the north of the country. The majority of the FMD free area was approved for EU export of beef and beef products. Following the FMD outbreak in the state of Mato Grosso do Sul (MS), the status of the FMD free zone with vaccination for the zone of Brazil 5

comprising the states of Mato Grosso do Sul, Tocantins, Minas Gerais, Rio de Janeiro, Espirito Santo, Bahia and Sergipe was suspended with effect from 30 September 2005. Following a suspected FMD outbreak in the state of Paraná (PR), the status of the FMD free zone with vaccination for the zone of Brazil comprising the states of Paraná, São Paulo (SP), Goias (GO), Mato Grosso (MT) and Federal District of Brasilia was suspended with effect from 21 October 2005. At present the FMD free zone with vaccination recognised by OIE only comprises the states of Acre along with two adjacent municipalities of Amazonas, Rio Grande do Sul, Santa Catarina and Rondonia. The latest reported outbreak occurred on 12 April 2006 in the municipality of Japorá, MS. More details concerning the outbreaks can be found in the OIE web-site: 4.2.2. Imposition of EU restrictions http://www.oie.int/eng/en_index.htm As a consequence of the outbreaks of October 2005, the imports of fresh meat from FMD susceptible animals into the EU from the states of PR, SP and MS (the municipalities previously approved for EU export) were suspended from animals slaughtered on or after 30 September 2005. Imports of de-boned meat, produced and certified in conformity with the requirements laid down in Council Decision 79/542/EEC, are permitted from the remaining areas of Brazil approved for EU export. 5. MAIN FINDINGS 5.1. Competent authority performance 5.1.1. General Brazil is divided into 26 states and the Federal District of Brasilia. In each state there is a competent authority (CA) for the animal and public health veterinary services (SVS) subdivided into regional and local services. In each state there is also a delegate of the CCA in charge of supervising the activities of the SVS. The structure and responsibilities of the CAs are described in detail in previous mission reports. Legislation has been produced both at federal and state level to cover the range of activities involved. 5.1.2. Animal health The CA in the animal health area is well organised and duties regarding FMD controls are clearly allocated. 6

Amendments to the FMD legislation, including a new definition of an FMD outbreak following OIE standards, and changes to the contingency plan, are still under preparation and are expected to be finalised by the end of 2006. An inter-ministerial working group at federal level, headed and coordinated by the Minister of Agriculture, has been established in order to achieve an overarching approach to crises such as FMD with economic, social and environmental impact. This includes planning for an emergency task force for FMD. The regional CA visited in the state of MS started in June 2006 to carry out audits of the local CAs and reports from these were available. On several occasions the CA could demonstrate good co-operation between the veterinary services and other services involved (military police, police, border inspection staff and civil defence forces). A protocol signed by the Brazilian and Paraguayan CCAs establishing a common FMD policy in the border areas is in place but there is little practical co-operation between the Brazilian veterinary service and the Paraguayan veterinary service, even at local level in the event of an FMD outbreak close to the border. 5.1.3. Public health In the framework of public health (slaughterhouses, cutting plants, processing plants etc.) the same structure is maintained throughout Brazil with the Federal Inspection Service (SIF) present at establishment level. Audits of the Competent Authorities: Through Circular No. 249/2006/CGPE/DIPOA of 7 April 2006 the CCA highlighted the legal basis for audits by DIPOA and inspections from the state level (Service of Animal and Plant Products Inspection, SIPAG) in exporting establishments and defined the minimum frequencies and procedures to be followed during these control activities. The minimum requirement defined in the legislation is for DIPOA to inspect (audit) at least 40% of the exporting establishments in each state twice annually. However, the CCA stated that the aim is to audit 100% of the exporting establishments twice per year. The main purpose of the DIPOA audits will be to evaluate the performance of the SIPAG management system through assessment of establishment records (own-checks programmes) and results of the SIF inspection in the establishment (monitoring records), and comparing them to the SIPAG inspection and action reports. It is intended to compile the results of the audits in each state into a final report with conclusions and recommendations. 7

Audits have been carried out during the first half of 2006 in all the establishments visited by the mission and reports were available. Audits have revealed that coldstores in general have not introduced ownchecks systems in line with the requirements of circulars 175/2005 and 176/2005. According to the CCA, coldstores that do not comply with the requirements at the latest by 1 January 2007 will be de-listed for exports to the EU. Staff performing official controls: The veterinary curriculum is considered by the CCA to be sufficient for the control tasks in relation to meat. No further theoretical or practical training is required before veterinarians can take up duties in meat establishments. Auxiliaries receive only a very basic training (approximately 4 weeks in a slaughterhouse) before starting work. They receive further training on how to carry out specific tasks by the official veterinarian in the establishment. There is a modular system of continued further training for auxiliaries on an annual basis (e.g. relevant topics such as traceability, HACCP systems etc.) allowing them to gradually extend their competences and ability to take on further tasks. In general the staffing situation was satisfactory. No probation period was required when inexperienced staff took up duties in complex establishments. In one large slaughterhouse visited the two official veterinarians had been appointed to work independently with only very limited practical training. No plans have been made for a training programme equivalent to that laid down in Regulation (EC) No 854/2004. Documented control and verification procedures: The system is gradually being changed from a traditional inspection system with fixed inspection frequencies and routine checks to a partly risk based control system based on audits, supervision and inspection. In order to gradually apply the procedure of basing the official inspection on the establishment s own checks programmes, the CCA issued Circulars No 175/2005 and 176/2005/CGPE/DIPOA of 16 May 2005 regarding the verification procedures for the FBO s obligations in relation to Good Manufacturing Practices, HACCP based procedures, microbiological testing and certification for exports. Circular 348/2005/CGPE/DIPOA of 27 July 2005 lays down the procedures for official audits of the certification of exported meat. 8

More recently the CCA has issued Circulars No 433/2005 and 444/2005/CGPE/DIPOA of 15 September 2005 regarding traceability and certification procedures for fresh meat export to the EU 2. According to Circular No. 249/2006/CGPE/DIPOA, the state inspection body (SIPAG) is responsible for programming and performing inspections in exporting establishments, focusing on the requirements of the importing markets, and in the case of those certified for exporting to EU at a frequency of 4 times per year. Previously the requirement was 2 state inspections per year in exporting establishment. Reports from the state inspections and federal audits were seen in the establishments visited. However, in some establishments the prescribed frequency had not been complied with. In one slaughterhouse visited only one inspection report was available for 2005 and 2006 due to lack of availability of staff at state level in the past. 5.2. Holding registration, animal identification and movement controls The system in place for animal identification and movement controls has been described in detail in previous mission reports DG(SANCO)/7589/2005 and DG(SANCO)/7185/2004. 5.2.1. Bovine identification system (SISBOV) A specific system (SISBOV) is in place for farm registration, animal identification and movement control, including the compulsory individual identification and registration in a national database (BND) of cattle intended to be slaughtered for the export of meat to the EU. According to the CCA, 34 million live cattle on 79 000 holdings are registered in the BND. The holdings are inspected by 70 different accredited certifying bodies (ACB). New legislation concerning the re-organisation of SISBOV was adopted on 13 July 2006. This new, more stringent system will be implemented gradually in co-existence with the present system. Most of the provisions will be implemented by the end of 2007. However, the identification of bovines based on the old system will be forbidden from 1 December 2006. The legal framework for the new system includes requirements for; o Approval of holdings intending to raise animals for export of their meat to the EU market, by a certifying body authorised by the CCA. o Compulsory individual identification of all cattle in the holding, at the latest within 10 months from birth and always before the first movement from the holding. 2 A more detailed description of circulars 443 and 444/2005/CGPE/DIPOA can be found in FVO mission report DG(SANCO)/8301/2006. 9

o At least two on-the spot inspections per year by the ACB to ensure enforcement of identification requirements. o Each farm can only be certified by one ACB. o After 31 December 2008 it will be a closed system which means that animals purchased must come from another SISBOV farm. Registration in SISBOV is compulsory for 90 days in EU approved areas and 40 days in the last holding before slaughter for all bovines intended to be slaughtered for meat export to the EU. When new animals are received in a SISBOV holding, all animals on that holding are blocked for further movements for 40 days or for 90 days when they are coming from non EU approved areas. The number of animals recorded in the BND was frequently much higher than the number of animals present on the holdings visited, amounting in several cases to double the number of animals present on the farm. There is no procedure in place to update the information and delete dead or slaughtered animals (killed earlier or outside EU approved slaughterhouses) or animals moved to another farm. The control system in place within one ACB visited in GO was weak, particularly concerning the controls on the operators in the field by the regional office. In one SISBOV farm visited in SP no ear tags had been applied. Despite this the ACB had certified that the animals were tagged, and had registered 250 animals in the BND (the tags and the specific documents for identification of the animals (DIA) were still stored on the farm). This farm was producing pedigree animals for breeding. The CA took immediate action by issuing a circular whereby the animals will be ineligible for slaughter for export to the EU and initiating a thorough investigation. In one SISBOV farm visited in SP the owner stated that he had removed the SISBOV eartags and replaced them with his own system. This farm had 150 animals registered in the BND and the last documented visit by the ACB had been carried out in May 2005. This farm was involved in embryo transfer and the owner stated that he had no intention of staying in the SISBOV system. Furthermore, one SISBOV eartag found on the ground in this farm in SP belonged to an animal that, according to the BND, was located on another holding in PR. On some occasions farmers within the SISBOV system stated that they did not notify movement of animals to the ACB. The controls done by the ACBs in SISBOV farms were not always carried out at an appropriate frequency. Furthermore, on several occasions these controls had not identified the shortcomings described above. 10

In one farm visited there were three different owners (3 different holding registers, cadastros) and two ACBs involved in all three, which made it almost impossible to evaluate the implementation of the SISBOV system. The CA initiated a head count of all bovines on the farm and the results of this revealed an even more complex ownership and a high number of animals registered in the BND but no longer present on the farm. 5.2.2. Holding registers and animal movement documents The compulsory system for animal movement documents (GTAs) is also under review, mainly in respect of electronic means of communication. The new system will however, not include identification of the individual animals, route plan and time of transport. The validity of the present GTAs can vary and in MS was up to 7 days. As regards the cadastros, it was noted that newborn animals are only entered in this in the course of the notification of the vaccination of the animals. The cadastro, and in most cases the GTAs supporting the movement of animals, could be provided by the CA for each farm visited. However, in one case, the GTAs for incoming animals could not be provided and several GTAs regarding exits had not been introduced into the cadastro in 2005. In several cases the movements of animals were entered in the cadastros after a considerable delay, in one case 3 months after the event. In addition, the date in the cadastro is not the date of the movement itself but the date of the entry of the data into the register which leads to limitations in its use for tracing of animals. In the course of the FMD-outbreak in Mato Grosso do Sul (MS) the CA initiated action to ensure that the cadastros reflect the true situation in the holdings. This included verification by the CA of the number of animals present on the holdings and an amnesty until 15 April 2006 for farmers in order to make it possible to update the cadastros without the appropriate movement documents. Generally, the number of registered animals went down during this exercise, and in one municipality from 40 000 to 34 000. In one farm visited in MS with land on both sides of the international border, the manager clearly stated that 200 animals had been moved to Paraguay. The CA underlined that this was done without authorisation and stressed that the owner looked upon his property as one unit, albeit with land on both sides of the border. These animals were still registered in the Brazilian cadastro and indicated as vaccinated there. In one farm visited in MS a second herd of dairy cows and sheep had been detected in the course of the clinical surveillance of that farm. This farm was still not properly registered and the records were kept in a separate register. 11

5.3. Management of FMD 5.3.1. Action taken to control FMD outbreaks and Epidemiological data As described in FVO report DG(SANCO)/8301/2006 the FMD eradication strategy applied was based on clinical examination, virological testing, culling of the infected herd, preventive stamping out of contact herds and movement restrictions. Reports of investigations into the source of the FMD outbreak in MS in 2005 were shown to the mission team. These reports gave some indication of the origin but could not provide final evidence. The CA stated that by the end of this series of FMD outbreaks a final epidemiological summary will be produced with an official statement in respect of the source of the outbreak. The mission team was informed that, in addition to the 34 outbreaks in MS notified by Brazil to the OIE, another 348 holdings had been culled ( preventive stamping out ). In 36 of these, lesions had been observed on the animals after culling. The CA has sufficient equipment and power to take any action in order to combat FMD: One farmer was heavily penalized for non-notification of a suspicion of an FMD outbreak. The mission team was informed that in SP one vaccine distribution centre was penalised for storing vaccines at a too high a temperature (24 o C). The filing systems in the offices of local CAs visited varied considerably between the individual offices. In one LVU visited it was very difficult to find all the documents requested by the mission team and sometimes the documents provided contained significant errors (FMD outbreak indicated on the wrong holding). Clinical examinations were carried out on suspect FMD cases and on contact farms and farms with epidemiological links. The mission team was provided with documented evidence (Formulario de investigaçao de doenças; FORM IN) of the clinical examinations and epidemiological investigations of these farms. However, significant delays were sometimes noted in the investigations of contact farms, which were situated close to the FMD focus (up to 39 days). All herds in the infected area were subject to clinical examinations. These visits were not carried with a predetermined frequency and reports were not always generated after these visits. Furthermore, these visits did not always include clinical examination of individual animals. There is no written procedure in place on how to carry out a clinical examination and how to report the details of the investigation. 12

The details of the epidemiological investigations recorded in FORM IN were often very limited. On the holding where the last FMD outbreak occurred in April 2006, no epidemiological investigation was carried out. The CA considered that inside the infected area these efforts were superfluous. In Eldorado, in one farm visited the culling of the animals was delayed because the farmer challenged the decision on culling in court. However, the reason for this was related to the Brazilian definition of a FMD outbreak as described in FVO mission report DG(SANCO)/8301/2006. The main focus of the FMD outbreak in MS was the municipality of Japorá, where the cattle population decreased from 45 500 before the outbreak to 20 000 as a consequence of culling. Very few animals died of FMD. In the infected area of MS, in total 33 066 cattle, 629 sheep and goats and 637 pigs in 384 properties were culled by trained operatives from the veterinary services, abattoirs and police. The individual burial places were authorised by the environmental authorities. Officially appointed teams composed of representatives from the Federal level, State level and farmer associations valued the animals before culling. Compensation for the lost animals (market price before the outbreak) was paid to the farmers on average 29 days after culling. The CA provided data on the chronology of events for the 34 individual outbreaks in MS: In 9 out of 34 cases culling of the infected animals was performed before the laboratory results were obtained. The time period between notification of the outbreak and the final disinfection of the affected holding varied remarkably between the three municipalities: Mundo Novo: 17 days; Eldorado: 36 days and Japorá 54 days. In Japorá, where the majority of the outbreaks occurred, the reason for this was mainly related to delays in culling of the herd (up to 53 days) and the final disinfection of the holding (up to 61 days). In Japorá, in one holding with 5 590 cattle the time period between notification and final disinfection was only 20 days. In all holdings, sentinel animals (cattle) were introduced in April 2006 and May 2006. The shortest time period between final disinfection of the outbreak holding and the introduction of sentinel animals was 15 days. In 16 herds, restocking has started but is not finished yet. 5.3.2. Regionalisation measures and controls The CA has issued numerous decrees containing measures for regionalisation and for tackling the FMD outbreak. Related to the 13

establishment of restricted zones in MS the following basic legislation was generated: On 9 October 2005, the CA imposed a movement ban and sanitary emergency measures for the municipalities of Eldorado, Itaquiraí, Iguatemi, Japorá and Mundo Nova in MS (risk area) (PORTARIA/IAGRO/MS N o 932/2005). On 18 November 2005, in MS movement restrictions were imposed within a 25 km radius around the outbreaks in MS (PORTARIA/IAGRO/MS N o 943/2005). On 25 November 2005, main parts of the municipalities of Itaquiraí and Iguatemi were excluded from the risk area (PORTARIA/IAGRO/MS N o 945/2005). On 15 March 2005, intra-state movements of cattle and products originating in the risk area were authorised subject to certain requirements and under supervision of the CA (INSTUÇÃO NORMATIVA N o 9/2006). Following the recommendation of FVO report DG(SANCO)/8301/2006, among other requirements premovement testing was introduced for susceptible animals leaving the risk area. In August and September, before the movement ban was imposed on 9 October 2005, 39 000 cattle originating in the risk area were moved within MS ( inter-municipal transit ). 26 728 (69%) of them were slaughtered. 6 241 (16%) out of the remaining 12 272 cattle were traded outside the risk area, but only 1 324 (3%) came from the municipalities of Eldorado, Mundo Novo and Japorá. The CA stated that these animals were individually traced back, separated and clinically examined but no blood samples were taken to exclude FMD infection. None of the animals showed clinical signs. In August and September, before the movement ban was imposed on 9 October 2005, 2 758 cattle originating in the risk area were moved outside MS ( inter-state transit ). 2 297 (83%) of them were slaughtered. 314 (11%) out of the remaining 461 cattle came from the municipalities of Eldorado, Mundo Novo and Japorá. These animals were traded from the municipalities of Japorá and Mundo Novo to Rio Grande do Sul (RS) (29 cattle) and from the municipality of Eldorado to PR (285 cattle). One of the latter caused the FMD outbreak in PR (epidemiological link from an outbreak farm in Eldorado to an auction in PR in October 2005 before the movement ban was imposed). In PR, actions were taken as described in FVO mission report DG(SANCO)/8301/2006. SP received 19 animals from Itaquiraí, which was later excluded from the risk area. However, cattle from PR were also moved to SP during the time period before the movement ban was imposed. In SP, three holdings were visited that had received animals from areas with different levels of risk: 14

A bovine semen collection centre received in August one bull from the risk area of MS (Itaquiraí). The CA stated that the operator was not informed about the possible risk. No clinical examination was carried out, no separation of the animal was imposed and no blood sampling was conducted. The semen collected from this bull and from all other bulls of the centre was not traced and detained. On another farm that had received 10 heifers from the infected area in PR the farmer was informed about the possible introduction of FMD virus. A movement ban was imposed on the herd and the 10 animals were put under quarantine. The farm was selected for serosurveillance in 2006. In one farm that had received animals from the non-infected area of PR, there was no notification given to the farmer and no visit was carried out by the CA and no movement restrictions were applied. However, the farm was later selected for sero-surveillance in 2006. As described above, 29 cattle from the risk area were traded to RS. In one farm visited within the infected area in MS (Japorá) with an owner possessing several holdings in Brazil and Paraguay, animals had been sent to RS in July 2005. The dossiers provided by the CA of RS indicated that clinical surveillance had been carried out, separation of the animals and movement restrictions were imposed but no serological testing was conducted in order to exclude the FMD infection. None of the animals showed clinical signs. Concerning the control system in place on the movement ban for susceptible animals and the products thereof within the infected zone, the CA stated that movements from infected holdings were totally banned. Movements of animals for slaughter originating from uninfected holdings were allowed within the infected zone. The CA stated that no susceptible animals or products thereof have left the risk area since the movement restrictions were imposed in October 2005. No authorisation was given for intra-state movement within MS following INSTUÇÃO NORMATIVA N o 9/2006. The movement ban was enforced by a number of permanent and mobile control posts in the infected area. The mission team was informed that these controls did not find any evidence for illegal movement. In Japorá, various disinfection points for vehicles were installed, which could not be passed without receiving a shower with disinfectant. The CA stated that since the international border to Paraguay in this area is crossed by many roads, it is impossible to thoroughly control illegal movements, in particular during the night. Furthermore, the CA of the border inspection post (BIP) on the international border with Paraguay stated that for several years no livestock had been traded through this BIP. Consequently, leaflets in several languages providing information on import restrictions did not include livestock. The mission team was informed that at this BIP 15

private cars were not controlled for illegal imports of fresh meat and products. 5.3.3. Sero-surveillance programme The programme has been applied in the FMD-free area for several years and has been implemented as described in FVO reports DG(SANCO)/8301/2006 and DG(SANCO)/7589/2005. In 2005, in the course of the sero-surveillance programme in MS the results of 14 samples (out of 1479) turned out positive as described in FVO report DG(SANCO)/8301/2006. After re-testing of these animals 60 days later, 2 out of these 14 samples had positive results. However, the results of ELISA 3ABC tests showed low reactivity indicating absence of virus circulation. Nevertheless, probang samples were taken but have not yet been examined. In November 2005, the sero-surveillance programme was carried out in MS as usual with the exception of the five municipalities at risk due to the outbreak of FMD. Following the recommendation of FVO report DG(SANCO)/8301/2006, the sero-surveillance programme in MS, PR and SP was intensified in 2006 and the sampling time was brought forward from November 2006 to the time period from April to June 2006 (MS) and July 2006 (PR and SP). In these states procedures were established for selecting cattle for additional testing inside and outside the risk areas. Generally, the animals selected for sero-surveillance were 6 to 12 months old, not vaccinated and individually identified. In 2006, 11.445 samples were collected in MS outside the risk area and 7 348 samples inside the risk area. The results are still pending. In PR 7 853 samples were collected outside the risk area and 9 826 samples inside the in risk area. 95 samples reacted positive (0,97%). Follow-up has been carried out with the new collection of 2 083 samples but the results are still pending. Additionally, 431 sheep were sampled inside the infected area in PR. All turned out negative to VIAA tests. In SP 9 177 samples were taken from cattle and the results are also still pending. All sampled animals in MS, PR and SP have not yet been vaccinated in 2006. In MS 1 380 and in PR 347, weaned calves derived from the noninfected areas were used as sentinel animals after negative testing for the presence of antibodies against the FMD virus (O, A and C). They were not vaccinated and were serologically tested 15 and 30 days after introduction into the holding. In some cases the serological tests were ELISA 3ABC positive but all of these were EITB negative. In addition to the intensified sero-surveillance programmes in MS, PR and SP, an FMD monitoring programme was instituted in 12 further 16

states in order to re-establish the FMD status free without vaccination following the OIE Guidelines for the surveillance of FMD. 5.3.4. Vaccination programme Details of the FMD vaccination programme in place have been described in FVO reports DG(SANCO)/8301/2006 and DG(SANCO)/7589/2005. During the vaccination campaign carried out in November 2005 the percentage of herds vaccinated was reported to be 99.5% in MS, 97.6% in PR and 98.8% in SP. These figures regarding vaccination coverage are based on the number of holdings which were notified as being vaccinated by the farmers themselves meaning that in these States around 1 to 2% of the herds were not vaccinated by the farmers but later subject to vaccination assisted by the CA. Following the recommendation of FVO report DG(SANCO)/8301/2006, 19866 bovine blood samples were taken throughout Brazil in order to assess the efficacy of FMD vaccination (in MS 2710, in PR 958 and in SP 961). The results of this sampling programme were only partly available (5%) and showed that 70 % of the animals in the age from 6-12 months, 85% of the animals between 13-24 months and 96% of the animals over 24 months have sufficient levels of antibodies against FMD virus. The CA stated that in all holdings, where the cattle developed low levels of antibody as response to the FMD vaccination, investigation in respect of verification of the vaccination was carried out immediately. No emergency vaccination was carried out around and in the risk area. What is more, in November 2005, the CA forbade vaccinating cattle in the perifocal area (25 km in diameter) in MS in order to avoid gathering of the animals, which could cause the spread of FMD virus (PORTARIA/IAGRO/MS N o 935 of 27 October 2005). Whilst in Japorá cattle (18000) were only vaccinated in May 2006, cattle in the four remaining municipalities in the risk area were vaccinated in December 2005 and in May 2006. During the FMD vaccination campaign in May 2006 the CA did not assist vaccination carried out by farmers to a greater extent than usual (2% assisted vaccination ). In addition to the 18000 unvaccinated cattle in Japorá described above, another 2000 cattle were not vaccinated in order not to jeopardise the results of the sero-surveillance campaign for the proof of absence of viral circulation from April to June 2006 (false positive results). The latter is related to difficulties in interpreting the serological positivity for 3ABC antibodies due to the problem of 3ABC contamination of the vaccine as described in FVO report DG(SANCO)/8301/2006. These animals are still not vaccinated since positive results would require retesting after 60 days. Vaccination is generally carried out by the farmers themselves. The farmers, however, did not receive written instructions from the CA in 17

respect of the vaccination procedures (cold chain, administration technique, exclusion of sick animals). 5.3.5. Laboratory testing Laboratory testing is carried out as described in previous FVO reports DG(SANCO)/8301/2006 and DG(SANCO)/7589/2005. In response to the outstanding recommendation of FVO report DG(SANCO)/7589/2005, the CA stated that: The procedure for verification of vaccine potency has been changed. Each submitted batch of the vaccines is now assessed with the ELISA LPC (liquid-phase competition) against the three different virus strains (O, A and C). At least 15 leading technical personnel at the official laboratory network have been trained in performing ELISA LPC. Concerning the assessment of FMD genome in biological materials, both PCR (polymerase chain reaction) and real time - polymerase chain reaction (RT - PCR) are being validated. Both methods will be fully operational in December 2006. There is an accelerated procedure in place for processing epithelial and blood samples for virus detection. However, the CA stated that serological testing of infected herds, contact herds and herds with epidemiological links was not systematically carried out since it was considered as too labour intensive and time consuming. Severe delays were, however, noted as regards the processing of samples from the sero-surveillance and vaccine efficacy programmes. The delays related to all stages of the chain including sampling, centrifugation, screening and data entry at the state laboratory as well as forwarding of the samples as batches to one of the federal laboratories. Furthermore, the analysis as such was also delayed and there had been a strike at the laboratories involved from 10 May to 10 July 2006 which reduced the capacity. The monthly capacity of the 3 MAPA laboratories in Belém, Recife and Porto Alegre for testing FMD is limited to 45000 samples for ELISA 3ABC, 10000 for EITB and 50000 for VIAA. In addition to that, 8200 EITB tests and 19000 VIAA tests can be performed in three accredited laboratories. Taking into account that more than 45000 blood samples were tested in course of the sero-surveillance in MS, PR and SP in spring 2006 only, some delays in performing the FMD diagnosis (ELISA 3ABC and EITB) were also due to capacity problems in the laboratories. 18

5.4. Food business operator s responsibilities 5.4.1. General and specific hygiene requirements In general the situation was satisfactory as regards the proper application of the general and specific requirements as laid down in Regulations (EC) No. 852/2004 and No. 853/2004. The FBOs demonstrated adequate production flows and operational procedures with regard to slaughter hygiene, hygiene of de-boning and storage. However, in some cases, the maintenance of facilities was not fully satisfactory. 5.4.2. HACCP systems HACCP-based own-check systems were available in all slaughterhouses visited and Control Points (CPs) and Critical Control Points (CCPs) were established in line with international standards. However, in the coldstores such systems are only in the process of being implemented (deadline 31 December 2006). In general, the HACCP systems were applied as described in the slaughterhouses visited. 5.4.3. Separation of EU/non-EU eligible animals and products In response to the deficiencies identified during the latest FVO mission DG(SANCO)/8301/2006, the CA has provided all SIF veterinarians in the approved slaughterhouses access to the National Bovine Database (BND) of SISBOV (Circular No 271/2006/CGPE/DIPOA of 18 April 2006). The CA has also issued new legislation 3 in order to improve the assurances on the origin of animals and traceability in general. Furthermore, the CA has organised workshops for the veterinary staff and meetings with the Exporting Association to highlight the traceability requirements. The veterinarians in the slaughterhouses visited all had access to the BND of the SISBOV and demonstrated the ability to use this facility to check the data on individual animals. 3 Circulars No 443/2005 and 444/2005/CGPE/DIPOA (see footnote 2). Circular No 087/2006/CGPE/DIPOA amending circular 443/2005 (checking information regarding lots and segregation of non-compliant animals) Circular No 231/2006/CGPE/DIPOA on 90 days declaration in GTA regarding animals entering holding from non-approved areas. 19

The slaughterhouses visited only received EU eligible animals, i.e. all animals received were individually identified with SISBOV eartags and accompanied by an Animal Movement Permit (GTA) and individual Animal Identification Certificates (DIA) guaranteeing at least 40 days presence on the last holding and 90 days presence within an EUapproved region. In general, the systems in place for the segregation of non-compliant animals at reception and separation between EU/non-EU eligible animals and meat throughout the whole production chain was found to be reliable and well documented in all the establishments visited. 5.4.4. FMD controls (maturation) All EU-eligible carcases must be submitted to maturation in separate chilling rooms at a temperature above +2 C for at least 24 hours and ph tested electronically after maturation and before de-boning. Only carcases with a ph value below 6.0 after maturation are eligible for de-boning and export to the EU. However, the previous FVO mission still identified some weaknesses in relation to the separation of EU and non-eu eligible carcases during maturation and pointed out the need for improving the maturation records in some establishments. The maturation controls were found to be in compliance with the requirements in all the slaughterhouses visited. The documentation has been improved after the previous mission and now includes slaughter numbers and ph measurements for each half carcass. A procedure was in place to calibrate the electronic equipment for measuring ph for each 100 carcases and the procedure was permanently supervised by the CA (auxiliary). 5.4.5. Identification marking and labelling Numbered seals with the establishment s SIF-number (identification marks) are still handled as health marks under full veterinary control. Labels with compulsory beef labelling information are enclosed within the vacuum packaging of each piece of de-boned beef. The labelling and identification system was satisfactory in all establishments, providing guarantees equivalent to the requirements laid down in Annex II, Section I to Regulation (EC) No 853/2004. 5.4.6. Traceability Circular No 348/2005/CGPE/DIPOA lays down the procedure to evaluate all elements of the establishment s traceability system during audits or 20

inspections. The system is based on two alternative methods for tracing back from either a box of meat or from an export certificate. Circulars No 443/2005 and 444/2005/CGPE/DIPOA prescribes detailed procedures and documentation requirements regarding traceability to be applied by both the FBO and the CA in charge of supervision. The procedures described cover all steps in the production chain from arrival to dispatch. The traceability systems were tested in all establishments visited and they were considered to be operating adequately. The traceability code (lot identification) used for meat is only linked to the slaughterhouse and the slaughter date and not to a maximum of one day's cutting (of meat from one slaughterhouse only). However, full traceability can be assured when using other compulsory information on boxes (production date, cutting establishment). 5.5. Official controls at establishment level 5.5.1. Ante-mortem and post-mortem inspection Ante-mortem controls are carried out by the official veterinarian after arrival the day before slaughter and again immediately before slaughtering starts in the morning. The official veterinarian in all establishments was assisted by an auxiliary who checked the documentation (GTA and DIA) when animals arrived. The auxiliaries also supervise and certify the cleaning and disinfection of trucks after unloading. Post-mortem inspections are carried out by auxiliaries under veterinary supervision and include inspection of feet and mouth for FMD lesions. The latest FVO mission identified shortcomings in regard to post-mortem records (missing carcass identification on detained or rejected carcases). Satisfactory ante-mortem records were available in all the slaughterhouses visited. The controls over truck wash and disinfection were satisfactory. The post-mortem inspection procedures applied in the slaughterhouses visited were equivalent to the inspection procedures laid down in Annex I, Section IV, Chapter I to Regulation (EC) No 854/2004 in terms of providing the necessary basis for deciding on the fitness of meat for human consumption. Post-mortem records included in all cases the slaughter sequence number of detained or rejected carcases. 21

5.5.2. Health marking The health marking system used in EU approved slaughterhouses is described in details in FVO report DG(SANCO)/8301/2006. Procedures were in place, and verified by the mission team, ensuring that only EU eligible animals without remarks from the post-mortem inspection would be health marked with the oval stamp after the maturation process had been satisfactorily completed. 5.5.3. Verification of food business operator s compliance Detailed guidelines for how to verify the FBOs compliance (responsibilities mentioned in section 5.4.) are laid down in Circulars No 175/2005 and 176/2005/CGPE/DIPOA. Documented verification of the FBO compliance was carried out by the official veterinarians as audits based on the requirements laid down in the guidelines in most of the slaughterhouse visited. In one slaughterhouse visited, no verification had yet been done due to the official veterinarian s lack of training on this subject. As HACCP-based own-checks systems only are in the process of being implemented in coldstores, no verification exercises have yet taken place in these establishments. 5.5.4. Action in case of non-compliance Non-compliance reports (Corrective Action Requests) had been created in three of the slaughterhouses visited, based on the documented verification of the FBO s compliance. The reports all included deadlines for completion and documentation for the CA s follow-up to verify that the corrective action has been taken within the specified (agreed) deadline. 5.6. Certification of meat The procedures for certification of export of meat and meat products are laid down in Circular No 234/2000/DIPOA. The circular describes the controls to be carried out in relation to exports from coldstores. In addition, Circular 238/2005/CGPE/DIPOA lays down guidelines for how to verify the certification procedures in relation to exports by using audit procedures. The CA has recognised that certification constitutes a special problem in relation to export of meat. Certification has therefore received special 22

attention during the latest CA audits where a high number of deficiencies were detected leading to corrective action already having been taken in most cases. In the two coldstores visited, certification was in principle carried out as described in Circular No 234/2000/DIPOA. However, in one of the coldstores the official veterinarian had in a number of cases certified data of which he had no personal knowledge or which could not be ascertained by him. In most cases the internal meat certificates now include the BR-code required for EU certification. However, national sanitary certificates from one establishment had still been issued until very recently without this code. In this case the official veterinarian in the receiving coldstore had requested, and received, a high number of replacement certificates. The code BR-1 was in one case applied to the export certificate for meat products despite the code on the national sanitary (replacement) certificate being BR-2. After receiving photocopies in a coldstore of a cancelled export certificate the CA produced 3 different copies of one of the pages (ORIGINAL stamp placed differently signature different). This indicates that 3 original documents must have been issued. The CCA declared after the mission that the first investigation into the case provided inconsistent explanations and that further investigations have been initiated. Certification for exports to Sweden and Finland constitute a particular problem 4 : The sampling and microbiological testing is not carried out in line with the requirements of Commission Regulation (EC) No 1688/2005. Certification signed by the official veterinarian is based on the carcass testing programme required by the US authorities and not on individual samples from the actual consignment; Despite having informed the veterinary staff about the new EU Regulation through Circular No 186/2006/CGPE/DIPOA of 17 March 2006, this was not known by the official veterinarian in one of the establishments visited certifying exports for Sweden; Some of the official veterinarians considered all beef eligible for export to Sweden and Finland unless it clearly was stated on the national sanitary certificates that it was not eligible (EXCETO SUÉCIA E FINLANDIA); 4 The CCA issued during the mission Circular No 484/2006/CGPE/DIPOA of 17 July 2006 regarding Salmonella testing procedures to be applied to exports of bovine meat to Sweden and Finland. 23