Search

Alternative pig farms, which do not raise animals in closed buildings with slatted and/or concrete floors, have critical points that need particular attention. Internal parasitism is one, as the farming conditions in such structures are more favorable to the development and survival of parasites. The objectives of this study, carried out on 70 alternative farms in continental France, were to (i) estimate the frequency and level of infestation by the main internal parasites on these farms, and (ii) define their typology according to the level of parasitism. For this purpose, fecal samples were taken for coprological analysis from 10 sows, 10 pigs aged 10-12 wk, and 10 pigs at the end of the fattening period. Blood samples were also taken for serological analysis (targeting Ascaris suum and Toxoplasma gondii) from 10 sows and 10 pigs at the end of the fattening period. Of the 70 farms, only 5 had no helminth egg or coccidian oocyst. Coccidia oocysts were observed in 79% of the farms, while eggs of Oesophagostomum spp./Hyostrongylus rubidus, Ascaris suum, and Trichuris suis were found in 47%, 16%, and 36% of the farms, respectively. On each infested farm, an average of 56.8% of sows, 23.8% of grower pigs, and 38.9% of finisher pigs were parasitized. At least 1 Ascaris suum-seropositive finisher pig was found on 91% of the farms, and at least 1 Toxoplasma gondii-seropositive finisher pig or sow on 60% of the farms. Data on housing, animal management, and health management (particularly parasite control) were collected to characterize the typology of farms according to their level of parasitism. The variables defining these farm typologies differed according to the parasites. Access to the outdoors for breeding stock was a characteristic of the farms most heavily infested with helminths or T. gondii. Conversely, the farms with the lowest frequency of coccidia oocyst infestation were characterized by free-range farrowing facilities and also by the presence of slatted floors, mostly plastic in our study, rather than straw bedding in the farrowing rooms. The level of biosecurity concerning the storage of straw for pig bedding was another discriminating factor for parasitism level of helminths and T. gondii. Farms with the highest levels of helminth parasitism were more likely not to practice an all-in-all-out postweaning system and to deworm their grower/finisher pigs less frequently than farms with the lowest levels of helminth parasitism.

This report describes the detection of a triple reassortant swine influenza A virus of H1avN2 subtype. It evolved from an avian-like swine H1avN1 that first acquired the N2 segment from a seasonal H3N2, then the M segment from a 2009 pandemic H1N1, in two reassortments estimated to have occurred 10 years apart. This study illustrates how recurrent influenza infections increase the co-infection risk and facilitate evolutionary jumps by successive gene exchanges. It recalls the importance of appropriate biosecurity measures inside holdings to limit virus persistence and interspecies transmissions, which both contribute to the emergence of new potentially zoonotic viruses.

Highlights • Maternally-derived neutralizing antibodies impair PRRS vaccination immune responses in piglets. • Maternally-derived neutralizing antibodies reduce PRRS vaccine efficacy in piglets. • IFNa may also interfere with PRRS vaccination., Modified live virus (MLV) vaccines are commonly used to reduce the impact of porcine reproductive and respiratory syndrome (PRRS) but limited efficacy is achieved in field conditions. Here, we evaluated the impact of maternally-derived neutralizing antibodies (MDNAs) on vaccine efficacy after PRRS virus (PRRSV) challenge. Piglets with low (A−) or high (A+) MDNA levels derived from a commercial pig herd were moved to experimental facilities to be vaccinated (V+) or not (V−) with a PRRSV-1 MLV vaccine at 3 weeks of age (woa). Because of unexpectedly low vaccine detection in A−V+ piglets post-vaccination (pv), all V+ piglets received a second vaccination at 4 woa. Five weeks (W5) pv, piglets were inoculated with a PRRSV-1 field strain to evaluate vaccine protection, and were mingled 24 h later with non-inoculated piglets of similar immune status to assess viral transmission. Vaccine strain was detected at W2 pv in 69% and 6% of A−V+ and A+V+ piglets, and at W5 pv in 50% and 25% of A−V+ and A+V+ piglets, respectively. At W5 pv, 94% of A−V+ and 44% of A+V+ piglets seroconverted, with a significant IFNg response induction in the A−V+ group only. After challenge, compared to the V− inoculated group, viremia was 100-fold lower at 10 days post-infection in A−V+ whereas viremia was not significantly reduced in A+V+ piglets. A lower transmission rate was estimated for the A−V+ group: 0.15 [0.07–0.29] versus 0.44 [0.18–1.76] and 0.32 [0.14–0.68] for the A+V+ and V− groups, respectively. Investigations about the low vaccine strain detection after the first vaccination suggested a relationship between IFNa levels and vaccine strain detection in A−V+ piglets. We showed that MDNAs impair vaccine efficacy against PRRSV both in inoculated and contact piglets, probably by reducing vaccine replication. IFNa may also interfere with PRRSV vaccination. These new data could help improving vaccination protocols.

Understudied, coinfections are more frequent in pig farms than single infections. In pigs, the term “Porcine Respiratory Disease Complex” (PRDC) is often used to describe coinfections involving viruses such as swine Influenza A Virus (swIAV), Porcine Reproductive and Respiratory Syndrome Virus (PRRSV), and Porcine CircoVirus type 2 (PCV2) as well as bacteria likeActinobacillus pleuropneumoniae,Mycoplasma hyopneumoniaeandBordetella bronchiseptica. The clinical outcome of the various coinfection or superinfection situations is usually assessed in the studies while in most of cases there is no clear elucidation of the fine mechanisms shaping the complex interactions occurring between microorganisms. In this comprehensive review, we aimed at identifying the studies dealing with coinfections or superinfections in the pig respiratory tract and at presenting the interactions between pathogens and, when possible, the mechanisms controlling them. Coinfections and superinfections involving viruses and bacteria were considered while research articles including protozoan and fungi were excluded. We discuss the main limitations complicating the interpretation of coinfection/superinfection studies, and the high potential perspectives in this fascinating research field, which is expecting to gain more and more interest in the next years for the obvious benefit of animal health.

In order to assess influenza D virus (IDV) infections in swine in France, reference reagents were produced in specific pathogen free pigs to ensure serological and virological analyses. Hemagglutination inhibition (HI) assays were carried out on 2090 domestic pig sera collected in 2012&ndash, 2018 in 102 farms. Only 31 sera from breeding sows sampled in 2014&ndash, 2015 in six farrow-to-finish herds with respiratory disorders contained IDV-specific antibodies. In two of them, within-herd percentage of positive samples (73.3% and 13.3%, respectively) and HI titers (20&ndash, 160) suggested IDV infections, but virus persistence was not confirmed following new sampling in 2017. All growing pigs tested seronegative, whatever their age and the sampling year. Moreover, PB1-gene RT-qPCR performed on 452 nasal swabs taken in 2015&ndash, 2018 on pigs with acute respiratory syndrome (137 farms) gave negative results. In Corse, a Mediterranean island where pigs are mainly bred free-range, 2.3% of sera (n = 177) sampled on adult pigs in 2013&ndash, 2014 obtained low HI titers. Finally, 0.5% of sera from wild boars hunted in 2009&ndash, 2016 (n = 644) tested positive with low HI titers. These results provide the first serological evidence that sows were exposed to IDV in France but with a limited spread within the swine population.

Aims A new multiplex qPCR, targeting Mycoplasma (M.) hyopneumoniae, M. hyorhinis and M. flocculare, was developed and the relationship between detection of those mycoplasma species and the extent of gross pneumonia-like lesions in slaughtered pigs lungs were investigated. Methods and results The multiplex qPCR method targets the p102, p37 and fruA genes and has detection limits of 14, 146 and 16 genome equivalents μl-1 for M. hyopneumoniae, M. hyorhinis and M. flocculare, respectively. In all, 671 lungs were collected and analysed, among them 666 were scored for macroscopic pneumonia and categorized according to the extent of the lesions (no or minor lesions, moderate lesions and extensive lesions). According to results of multiplex qPCR, 59·5% were positive for M. hyopneumoniae, 3·4% for M. hyorhinis and 34·7% for M. flocculare, with on average, 3·1 × 107 , 9·7 × 106 and 5·7 × 106 genome equivalents of mycoplasma ml-1 , respectively. More results showed that no or minor lesions were associated with multiplex qPCR-negative results or qPCR-positive results for M. flocculare. Moderate to extensive lesions were positively correlated with qPCR-positive results for M. hyopneumoniae. Extensive lesions were associated with qPCR-positive results for at least two mycoplasma species (M. hyopneumoniae and M. hyorhinis). Conclusion The findings also indicated that M. hyopneumoniae and M. hyorhinis significantly increased the odds for a lung to have macroscopic pneumonia. No relationship was found between the extent of lesions and the mycoplasma genome load. Significance and impact of the study This new multiplex qPCR appears to be specific, sufficiently sensitive and repeatable. The validation of this method with field samples guarantees its use for field epidemiological investigations, particularly to gain more insight into the aetiology of the porcine respiratory disease complex.

Respiratory diseases are a major health issue for pigs reared under confined conditions in intensive systems. Pneumonia and pleuritis are the most frequent lung alterations. The aetiology of these diseases is complex and multifactorial. A better knowledge of their epidemiology is needed to implement adequate control strategies. The paper describes the results of an epidemiological study designed to identify infectious and non-infectious factors associated with pneumonia and pleuritis in farrow-to-finish herds. According to these findings, preventive measures related to the building design, animal husbandry and medicine may be proposed.

This study was designed to assess the diagnostic characteristics of two PCV2 ELISAs without a gold standard. Four hundred and sixty-five serum samples from finishing pigs (25 herds) not vaccinated against PCV2 were used. Samples were tested by two ELISAs: an in-house ELISA (I-ELISA) and the commercial SERELISA®PCV2 Ab Mono Blocking kit (S-ELISA). A ROC curve was used to assess the S-ELISA's optimal threshold by taking the I-ELISA as a reference and using the cut-off previously determined by comparison to an cccmonolayer assay (IPMA). This led to an S-ELISA result ≥170 being considered as positive. The sensitivity (Se) and specificity (Sp) of each ELISA were then estimated without a gold standard using a Bayesian approach. The mean Se and Sp values of the I-ELISA were slightly higher than those of the S-ELISA (mean Se I-ELISA=0.90 vs. mean Se S-ELISA=0.86; mean Sp I-ELISA=0.92 vs. mean Sp S-ELISA=0.85). However, the 95% credibility intervals (CI95%) overlapped (Se I-ELISA CI95%=0.85-0.95 vs. Se S-ELISA CI95%=0.82-0.90; Sp I-ELISA CI95%=0.82-0.98 vs. Sp S-ELISA CI95%=0.75-0.94). Both ELISAs appeared to be valuable tools for detecting PCV2 antibodies.

Background Pigs are frequently colonised with Salmonella enterica, and this constitutes a major risk for human salmonellosis. The infection can be assessed by the serological response of pigs to S enterica. A longitudinal study was undertaken on-farm to correctly describe this serological response and investigate factors associated with age at Salmonella seroconversion. Methods Three pig farms and in each farm three successive batches were considered. Per batch, 40 piglets were selected at random from 10 sows (four piglets per sow). Blood was sampled from sows one week after farrowing and from piglets at weeks 1, 6, 10, 14, 18 and 22 and at the slaughterhouse. Salmonella antibodies were detected in serum using a commercial ELISA test. Factors related to farm characteristics, batch management system, porcine reproductive and respiratory syndrome infection, and sows’ Salmonella serological status were recorded to assess their effect on age at seroconversion. Results At week 1 after farrowing, 96.5 per cent of the sows had antibodies against Salmonella. The serological results of piglets at weeks 1 and 6 only were positively correlated with those of the sows. The average age at Salmonella seroconversion was 137±2.2 days (confidence interval at 95 per cent). The first seroconversions occurred from weeks 10 to 14, but most of the pigs (54.6 per cent) were seropositive at the end of the fattening period, with variations between farms and batches (28.9–75.7 per cent). Herd/farm was significantly associated with age at seroconversion. Conclusion This longitudinal study allowed the authors to follow precisely the evolution of Salmonella seroconversion from maternity to slaughterhouse and confirm the relationship between the seroconversion of sows and serology of their piglets. Moreover, factors related to farm practices and management as a whole are more influential than individual factors (at the pig level) on age at Salmonella seroconversion.

Mycoplasma hyorhinis and M. flocculare are commonly co-isolated with M. hyopneumoniae (primary agent of swine enzootic pneumonia) in gross pneumonia-like lesions, but their involvement in the disease process remains unknown. T cells play an immuno-pathological role during mycoplasmal infections. Dendritic cells (DCs) are major antigen-presenting cells involved in T cell activation and differentiation. In this study, we investigated cytokine (IL-6, IL-8, IL-10, IL-12, and TNF-&alpha, ) production by porcine bone-marrow-derived DCs (BM-DCs) stimulated by M. hyopneumoniae, M. hyorhinis, and/or M. flocculare. Results showed that cytokine production levels were relatively homogenous for all evaluated M. hyopneumoniae strains in contrast to M. hyorhinis and M. flocculare strains. The most noteworthy inter-species differences were the overall (i) lower IL-12 production capacity of M. hyopneumoniae, and (ii) higher TNF-&alpha, production capacity of M. flocculare. Co-stimulation of BM-DCs showed that M. hyorhinis dominated the IL-12 production independently of its association with M. hyopneumoniae or M. flocculare. In addition, a decreased BM-DC production of TNF-&alpha, was generally observed in the presence of mycoplasma associations. Lastly, M. flocculare association with M. hyopneumoniae increased BM-DC ability to secrete IL-10. A higher cytotoxicity level in BM-DCs stimulated by M. hyorhinis was also observed. Overall, this study demonstrated that the combination of M. hyorhinis or M. flocculare with M. hyopneumoniae may participate to the modulation of the immune response that might affect the final disease outcome.

The porcine reproductive and respiratory syndrome virus (PRRSV) is plaguing porcine production. Previously piglets were immunized with a PRRSV-1 commercial modified live virus vaccine (MLV1), a PRRSV-2 MLV (MLV2) or a Western European strain (Finistere: Fini) to assess and compare the protection brought by these strains upon challenge with virulent Lena strain. Lena viremia was reduced in all the immunized groups with a slightly higher level of protection following immunization with Fini. Using lung samples collected from the same experiment, tissue response to Lena challenge was assessed at the acute and chronic stages of infection. A pre-immunization with any one of the three PRRSV strains globally exacerbated microscopic lung lesions. Ten days post-challenge (DPC), MLV1 group score was higher than unimmunized group score and 42 DPC, MLV2 group score was higher than in unimmunized group. Lowest lung Lena viral loads were measured in Fini group. Using principal component analysis, we showed 10 DPC that the lesion score was positively correlated with chemokine receptors and negatively correlated with viral load. Forty-two DPC, variables for lesion score, IL6, IL8, and CCL20 transcripts were positively correlated together and negatively correlated with CCL28, CXCL6, and CXCR4 transcripts suggesting a role for the latter ones in the tissue recovery process. In conclusions, our study shows a significant impact of the three immunizations on pulmonary tissue with the best protection against Lena challenge conferred by Fini strain. Furthermore, it gives insight into the interactions between vaccine and Fini strains and the lung upon Lena challenge.

In pig production, the widespread conventional indoor system with a slatted floor currently dominates. However, this production system is becoming less socially acceptable. In addition to general environmental protection issues, animal welfare, the absence of suffering and distress, and the management of pain also constitute societal concerns. In this context, alternative production systems are gaining ground. Although they are popular with consumers and other citizens, these alternative systems have their critical points. Here, we reviewed the international scientific literature to establish the state of the art of current knowledge regarding welfare, biosecurity, animal health and pork safety in this type of farming system. In general, alternative farms give pigs the opportunity to express a broader range of behaviours than conventional farms. However, the management of feeding, watering, temperature and predators is often more complicated in these outdoor systems. In addition, biosecurity measures seem to be applied less strictly in alternative farms than in conventional farms, especially in free-range systems, where they are more difficult to implement. On the other hand, pigs kept in these farming systems seem to be less affected by respiratory diseases, but parasitism and piglet crushing (in farrowing units) both remain a real challenge. Furthermore, the higher prevalence of many zoonotic pathogens in these farms may represent a risk for human health.

International audience; An acute epidemic of porcine epidemic diarrhea (PED) has affected the USA since 2013 and spread all around the world. In France, the immune status of the pig population against PED virus (PEDV) was expected to be low due to the absence of circulation of the virus since the 80's and a compulsory notification of PED was set up in 2014. Here, we reported the first case of a PED outbreak in December 2014 in the North of France after a long absence of the disease, the monitoring of the excretion and the control measure implementation. The isolated strain in France in December 2014 was a PEDV "S-InDel" strain which was close to the "S-InDel" German PEDV strain isolated in May 2014. The individual shedding duration of PEDV in feces was estimated around 20 days for pigs of different ages. Biosecurity measures implemented allowed the limitation of PEDV spread to fattening and farrowing rooms without dissemination to the nursery block. Using strict biosecurity measures, direct shipment of infected fatteners to the slaughterhouse, strict decontamination protocols with a quarantine of 6 weeks for replacement gilts without voluntary contamination helped PEDV fade out within the herd and avoided the spread to other herds. PEDV presence in manure was investigated as well as the inactivation treatment of the virus present in the liquid manure. An increase to a pH 12 of liquid manure by liming led to the absence of PEDV detection by RT-PCR after seven days.

The monophasic variant of Salmonella Typhimurium is highly prevalent in human and in pork. However, little is known about colonization dynamics and serology in pigs. We orally inoculated 24 seven-week-old piglets with 109 CFU/pig of a porcine strain of monophasic Salmonella Typhimurium in an experimental trial. Three groups of eight piglets were orally inoculated and monitored for 21, 49, or 84 days post-inoculation until necropsied. From 3 days post-inoculation to necropsy, individual feces were sampled twice weekly and blood once weekly. At necropsy, the tonsils, mesenteric lymph nodes, and the contents of the duodenum, jejunum, ileum, and cecum were collected from each pig. We determined the number of CFU/g in all the samples and measured also Salmonella antibodies in OD% in all blood samples. At different times during the trial, we tested by MLVA (Multilocus Variable Number Tandem Repeat Analysis) the genomic stability of the strain after passing through the intestinal tract. Salmonella was continuously excreted by pigs, ranging from 1.4 to 5.8 log10 CFU/g. At necropsy, Salmonella was present in all samples, but the tonsils were particularly infected. Salmonella antibodies were detected in five pigs 7 days post-inoculation. At 49 days post-inoculation, all the pigs were seropositive. We observed new MLVA types for 3.3% of the isolates tested over the trial. Our study allowed us to show the serovar's ability to persist in pigs after infection up to 84 days post-inoculation. We demonstrated that Salmonella seroconversion appeared earlier than in naturally infected pigs and that the strain's genome can evolve after passing through the digestive tract of pigs.

Background Growing and finishing performances of pigs strongly influence farm efficiency and profitability. The performances of the pigs rely on the herd health status and also on several non-infectious factors. Many recommendations for the improvement of the technical performances of a herd are based on the results of studies assessing the effect of one or a limited number of infections or environmental factors. Few studies investigated jointly the influence of both type of factors on swine herd performances. This work aimed at identifying infectious and non-infectious factors associated with the growing and finishing performances of 41 French swine herds. Results Two groups of herds were identified using a clustering analysis: a cluster of 24 herds with the highest technical performance values (mean average daily gain = 781.1 g/day +/− 26.3; mean feed conversion ratio = 2.5 kg/kg +/− 0.1; mean mortality rate = 4.1% +/− 0.9; and mean carcass slaughter weight = 121.2 kg +/− 5.2) and a cluster of 17 herds with the lowest performance values (mean average daily gain =715.8 g/day +/− 26.5; mean feed conversion ratio = 2.6 kg/kg +/− 0.1; mean mortality rate = 6.8% +/− 2.0; and mean carcass slaughter weight = 117.7 kg +/− 3.6). Multiple correspondence analysis was used to identify factors associated with the level of technical performance. Infection with the porcine reproductive and respiratory syndrome virus and the porcine circovirus type 2 were infectious factors associated with the cluster having the lowest performance values. This cluster also featured farrow-to-finish type herds, a short interval between successive batches of pigs (≤3 weeks) and mixing of pigs from different batches in the growing or/and finishing steps. Inconsistency between nursery and fattening building management was another factor associated with the low-performance cluster. The odds of a herd showing low growing-finishing performance was significantly increased when infected by PRRS virus in the growing-finishing steps (OR = 8.8, 95% confidence interval [95% CI]: 1.8–41.7) and belonging to a farrow-to-finish type herd (OR = 5.1, 95% CI = 1.1–23.8). Conclusions Herd management and viral infections significantly influenced the performance levels of the swine herds included in this study. Electronic supplementary material The online version of this article (10.1186/s40813-018-0082-9) contains supplementary material, which is available to authorized users.

The outputs of epidemiological models are strongly related to the structure of the model and input parameters. The latter are defined by fitting theoretical concepts to actual data derived from field or experimental studies. However, some parameters may remain difficult to estimate and are subject to uncertainty or sensitivity analyses to determine their variation range and their global impact on model outcomes. As such, the evaluation of immunity duration is often a puzzling issue requiring long-term follow-up data that are, most of time, not available. The present analysis aims at characterizing the kinetics of antibodies against Porcine Reproductive and Respiratory Syndrome virus (PRRSv) from longitudinal datasets. The first dataset consisted in the serological follow-up of 22 vaccinated gilts during 21 weeks post vaccination. The second one gathered the maternally derived antibody (MDA) kinetics in piglets from 3 different farms up to 14 weeks of age. The peak of the post-vaccination (PV) serological response against PRRSv was reached 6.9 weeks post-vaccination on average withan average duration of antibodies persistence of 26.5 weeks. In the monitored cohort of piglets, the duration of passive immunity was found relatively short, with an average duration of 4.8 weeks. The level of PRRSv-MDAs was found correlated with the dams’ antibody titer at birth and the antibody persistence was strongly related to the initial MDAs titers in piglets. These results evidenced the importance of PRRSv vaccination schedule in sows, to optimize the delivery of antibodies to suckling piglets. These estimates of the duration of active and passive immunity could be further used as input parameters of epidemiological models to analyze their impact on the persistence of PRRSV within farms.

The feasibility of using individual and pen-based oral fluid samples to detect PRRSV antibodies in growing-finishing pigs and group-housed sows was investigated. The diagnostic performances of a commercial oral fluid ELISA (OF-ELISA) and a serum ELISA (SER-ELISA) performed on individual or pooled samples from 5 or 10 pigs and sows was evaluated. The performance of the OF-ELISA was also assessed for pen-based oral fluids. Eight hundred and thirty-four pigs and 1598 sows from 42 PRRSV-infected and 3 PRRSV-negative herds were oral fluid sampled and bled. PRRSV antibodies were detected by an OF-ELISA performed at individual, pool (5 or 10 samples) and pen levels. Serum samples were tested by a SER-ELISA at individual and pool levels. The sensitivity and specificity of ELISAs for individual samples were assessed by Bayesian analysis. The relative diagnostic performance for the pools was calculated by taking individual samples as the gold standard. SER-ELISA and individual OF-ELISA results were used as references for estimating OF-ELISA performance for pen-based samples. Individual oral fluid collection was feasible in all kinds of pigs, whereas pen-based samples were unsuccessful in 40% of the group-housed sow pens. High levels of sensitivity comparable to those of the SER-ELISA were found for the OF-ELISA when performed on individual, 5-sample pool or pen-based samples from pigs or sows. The OF-ELISA lacked specificity for individual samples from sows. Pooling 5 individual oral fluid samples or using pen-based samples increased test specificity.

The porcine reproductive and respiratory syndrome virus (PRRSV) causes huge economic losses for the swine industry worldwide. In the past several years, highly pathogenic strains that lead to even greater losses have emerged. For the Western European swine industry, one threat is the possible introduction of Eastern European PRRSV strains (example Lena genotype 1.3) which were shown to be more virulent than common Western resident strains under experimental conditions. To prepare for the possible emergence of this strain in Western Europe, we immunized piglets with a Western European PRRSV field strain (Finistere: Fini, genotype 1.1), a new genotype 1 commercial modified live virus (MLV) vaccine (MLV1) or a genotype 2 commercial MLV vaccine (MLV2) to evaluate and compare the level of protection that these strains conferred upon challenge with the Lena strain 4 weeks later. Results show that immunization with Fini, MLV1 or MLV2 strains shortened the Lena-induced hyperthermia. In the Fini group, a positive effect was also demonstrated in growth performance. The level of Lena viremia was reduced for all immunized groups (significantly so for Fini and MLV2). This reduction in Lena viremia was correlated with the level of Lena-specific IFNγ-secreting cells. In conclusion, we showed that a commercial MLV vaccine of genotype 1 or 2, as well as a field strain of genotype 1.1 may provide partial clinical and virological protection upon challenge with the Lena strain. The cross-protection induced by these immunizing strains was not related with the level of genetic similarity to the Lena strain. The slightly higher level of protection established with the field strain is attributed to a better cell-mediated immune response.

Herd-level factors associated with European H1N1 or H1N2 swine influenza virus (SIV) infections were assessed by mean of a cross-sectional study carried out in 125 herds in France. Serum samples from 15 fattening pigs in each herd were tested by haemagglutination inhibition. Data related to herd characteristics, biosecurity, management and housing conditions were collected by questionnaire during the farm visit. Climatic conditions in the post-weaning and fattening rooms, where the sampled pigs were housed, were measured over 20 h. Factors associated with H1N1 or H1N2 sero-positive status of the herd were identified by logistic regressions for binary outcome. For both subtypes, the odds for a herd to be SIV sero-positive increased if there were more than two pig herds in the vicinity (OR=3.2, 95% confidence interval (95% CI): 1.4-7.6, p

The aims of this study were to assess the feasibility of individual and pen-based oral fluid sampling (OFS) in 35 pig herds with group-housed sows, compare these methods to blood sampling, and assess the factors influencing the success of sampling. Individual samples were collected from at least 30 sows per herd. Pen-based OFS was performed using devices placed in at least three pens for 45min. Information related to the farm, the sows, and their living conditions were collected. Factors significantly associated with the duration of sampling and the chewing behaviour of sows were identified by logistic regression. Individual OFS took 2min 42s on average; the type of floor, swab size, and operator were associated with a sampling time >2min. Pen-based OFS was obtained from 112 devices (62.2%). The type of floor, parity, pen-level activity, and type of feeding were associated with chewing behaviour. Pen activity was associated with the latency to interact with the device. The type of floor, gestation stage, parity, group size, and latency to interact with the device were associated with a chewing time >10min. After 15, 30 and 45min of pen-based OFS, 48%, 60% and 65% of the sows were lying down, respectively. The time spent after the beginning of sampling, genetic type, and time elapsed since the last meal were associated with 50% of the sows lying down at one time point. The mean time to blood sample the sows was 1min 16s and 2min 52s if the number of operators required was considered in the sampling time estimation. The genetic type, parity, and type of floor were significantly associated with a sampling time higher than 1min 30s. This study shows that individual OFS is easy to perform in group-housed sows by a single operator, even though straw-bedded animals take longer to sample than animals housed on slatted floors, and suggests some guidelines to optimise pen-based OFS success.

Factors associated with porcine reproductive and respiratory syndrome virus (PRRSV) infection were investigated in 109 herds. Serums from four batches of pigs (4, 10, 16 and 22 weeks, 15 pigs/batch) were tested by ELISA for PRRSV antibodies. Infection by Mycoplasma hyopneumoniae (Mhp), Actinobacillus pleuropneumoniae, H1N1 and H1N2 swine influenza A viruses (swIAV) and PCV2 were detected by specific serological or PCR tests. Data related to herd characteristics, biosecurity, management housing and climatic conditions were collected during a herd visit. Factors associated with the herd's PRRSV seropositive status were identified by logistic regression. Large herd size, the lack of disinsectisation in the gestation facilities, on-farm semen collection, a short time-period for gilt quarantine and a low temperature setpoint for the ventilation controller in the fattening room significantly increased the odds of a herd being seropositive for PRRSV. Infection by Mhp and H1N2 swIAV were associated with a PRRSV seropositive status. A Cox proportional hazards model was used to identify the factors associated with the age-time to seroconversion in infected herds. Joint housing for the gilts and sows when lactating, a large nursery pen, a small number of pens per fattening room and lack of all-in all-out management in the fattening section significantly reduced the age-time to seroconversion. A small range of temperatures controlling ventilation rate in the nursery room was also associated with time to PRRSV seroconversion. Infection by Mhp and a high PCV2 infection pressure were associated with a shorter time to seroconversion. Biosecurity measures minimising the risk of introducing PRRSV into the herd, management practices reducing contacts between animals from different batches and within batches and favourable climatic conditions should be implemented to better control PRRSV infection.

The influence of maternally-derived antibodies (MDAs) on the post-vaccination humoral and cellular immune responses in piglets vaccinated against PRRS was studied. The piglets came from a vaccinated breeding herd. Thirty piglets with a low (A-) or high level (A+) of PRRSV-neutralizing MDAs were vaccinated (V+) with a modified live vaccine at 3 weeks of age. Blood samples were collected before vaccination and then at 2, 4, 8 and 14 weeks post-vaccination (WPV). The samples were analysed to detect the vaccine viraemia (RT-PCR) and quantify the post-vaccination humoral (ELISA and virus neutralisation test) and cellular (ELISPOT IFNγ) immune responses. PRRSV vaccine strain was detected in 60%, 64%, 36% and 0% of A-V+ piglets 2, 4, 8 and 14 WPV respectively. No virus was detected in A+V+ piglets during the first four WPV but 32% and 6% of A+V+ piglets were PCR-positive at 8 and 14 WPV. Eighty-five percent of A-V+ piglets and 0% of A+V+ piglets seroconverted (ELISA) between 2 and 4 WPV. Neutralising antibodies appeared 4 WPV in the A-V+ piglets and 14 WPV in the A+V+ piglets. The number of PRRSV-specific IFNγ-secreting cells was significantly higher in A-V+ piglets at 2 and 4 WPV than in A+V+ piglets. These results show that MDAs can affect both post-vaccination humoral and cellular immune responses in piglets. Further studies are required to assess the impact of MDAs on vaccine efficacy following a PRRSV challenge and its ability to reduce viral transmission.

A longitudinal study was carried out in five French farrow-to-finish herds differently affected by respiratory diseases to describe the carrying and infection patterns of batches of sows to various respiratory pathogens during gestation and lactation. An entire batch of sows was followed during two successive reproduction cycles. Nasal, tonsillar and oro-pharyngeal swabs and blood samples were taken from each sow 9 and 4 weeks before farrowing and 1 and 4 weeks after farrowing. Mycoplasma hyopneumoniae, Actinobacillus pleuropneumoniae, Pasteurella multocida, Haemophilus parasuis and Streptococcus suis were detected from swab samples using PCR assays. Blood samples were tested for antibodies against M. hyopneumoniae, A. pleuropneumoniae serotypes 1-9-11 and 2, Porcine Circovirus type-2 (PCV-2) and Porcine Reproductive and Respiratory Syndrome virus (PRRSV) by ELISA tests. Antibodies against H(1)N(1), H(1)N(2) and H(3)N(2) Swine Influenza Viruses (SIV) of European lineages were tested by hemagglutination inhibition assay. The results indicated that S. suis is widespread among sows (67.1% of PCR-positive sows). A. pleuropneumoniae, P. multocida, and H. parasuis were detected by PCR in 30.9%, 24.6% and 23.4% of the sows, respectively. Antibodies against M. hyopneumoniae were recovered from more than 55% of the sows in all herds whereas the micro-organism was detected in 2.4% of the sows. Although PCV-2 and SIV infections were highly prevalent, the PRRSV infection patterns ranged from no infection in farms mildly affected by respiratory diseases to active circulation in more severely affected herds. The sow population thus constitutes a reservoir for a continuous circulation of respiratory pathogens and needs to be properly considered in control strategies.

AIMS A triplex real-time PCR assay to quantify Mycoplasma hyopneumoniae in specimens from live and dead pigs was developed and validated. The minimal dose of Myc. hyopneumoniae required to induce pneumonia in specific pathogen-free pigs was determined. METHODS AND RESULTS This TaqMan test simultaneously detected three genes encoding the proteins P46, P97 and P102. All Myc. hyopneumoniae strains analysed were detected, including strains isolated in three countries (France, England and Switzerland) and from several pig farms (n = 33), and the test was specific. The estimated detection thresholds were 1.3 genome equivalents (microl(-1)) for the targets defined in p97 and p102 genes and 13 genome equivalents (microl(-1)) for the segment defined in the p46 gene. This test was used to quantify Myc. hyopneumoniae in specimens sampled from experimentally infected pigs. In live pigs, c. 10(7), 10(8) and 10(10) genome equivalents (ml(-1)) of Myc. hyopneumoniae were detected in the nasal cavities, tonsils and trachea samples, respectively. In dead pigs, 10(8)-10(10) genome equivalents (ml(-1)) of Myc. hyopneumoniae were detected in the lung tissue with pneumonia. The estimated minimal dose of Myc. hyopneumoniae required to induce pneumonia was 10(5) colour-changing units (CCU) per pig (corresponding to 10(8) mycoplasmas). CONCLUSION The triplex RT-PCR test was validated and can be used for testing samples taken on the pig farms. SIGNIFICANCE AND IMPACT OF THE STUDY This test should be a very useful tool in pig herds to control enzootic pneumonia or healthy carrier pigs and to study the dynamics of Myc. hyopneumoniae infections.

International audience; We did a prospective observational 9-month long study to quantify risk factors of managerial and hygiene practices, and pig-health status for Salmonella seroconversion of fattening pigs reared in subclinically infected French farrow-to-finish farms. During the fattening phase, 2 649 pigs belonging to the same batch of contemporary pigs, from 89 conventional farrow-to-finish farms were individually followed and regularly blood sampled on a monthly basis. Farm recruitment was based on the farmer's willingness to cooperate. Pig status was assessed using an indirect ELISA test. Evolution of the serological status was studied by means of survival analysis. A Cox proportional-hazards model, taking into account the clustering of animals at the farm level, was used to examine the effects of explanatory variables on the time to Salmonella seroconversion of pigs. Applying group level antibiotic treatment to the pigs during the fattening period (Hazard Ratio (HR) = 2.4; 95% CI: 1.7, 3.4) was identified as a risk factor for Salmonella seroconversion, as the presence of residual Salmonella contamination in the fattening pen before placing the pigs into the pens (HR = 1.9; 95% CI: 1.2, 2.9). Porcine reproductive and respiratory syndrome virus (PRRSV) seropositivity during the fattening period also indicated an increased hazard for seroconversion (HR = 1.6; 95% CI: 1.1, 2.5). The batch size was identified as a risk factor for Salmonella seroconversion: the higher the number of pigs was in the fattening room followed, the higher was the risk (HR(+10 pigs) = 1.05 for a 10-pig increment; 95% CI: 1.03, 1.06). The biosecurity measures of wearing specific clothes before entering the facilities (HR = 0.5; 95% CI: 0.3, 0.9) and enclosing the pig farm facilities were protective (HR = 0.4; 95% CI: 0.2, 0.8).

We report the first and only case of a porcine epidemic diarrhea (PED) outbreak occurring in December 2014 in northern France, and we show using the full-length genome sequence of the French PED virus (PEDV) isolate that it was a PEDV indel strain close to German PEDV strains recently isolated.

Porcine respiratory disease complex (PRDC) is one of the main causes of economic losses for swine producers. This complex is due to a combination of different pathogens and their interactions. Two major pathogens involved in PRDC are Mycoplasma hyopneumoniae (Mhp) and porcine reproductive and respiratory syndrome virus (PRRSV). The objectives of this study were (i) to develop an experimental model of dual Mhp/PRRSV infection in SPF pigs with European strains of Mhp and PRRSV and (ii) to assess and compare the effects of single Mhp, single PRRSV or combined Mhp/PRRSV vaccination against this dual infection. Pigs dually infected with Mhp and PRRSV showed a combination of symptoms characteristic of each pathogen but no significant exacerbation of pathogenicity. Thus, the co-infected pigs displayed coughing and pneumonia typical of Mhp infection in addition to PRRSV-related hyperthermia and decrease in average daily gain (ADG). Hyperthermia was reduced in PRRSV vaccinated animals (single or combined vaccination), whereas ADG was restored in Mhp/PRRSV vaccinated pigs only. Regarding respiratory symptoms and lung lesions, no vaccine decreased coughing. However, all vaccines reduced the pneumonia score but more so in animals receiving the Mhp vaccine, whether single or combined. This vaccine also decreased the Mhp load in the respiratory tract. In conclusion, combined vaccination against both Mhp and PRRSV efficiently pooled the efficacy of each single PRRSV and Mhp vaccination and could be an interesting tool to control PRDC in European swine production.

Fattening-pigs carriers of Salmonella enterica are believed to be a main source of carcass and pork contamination at the later steps of the meat process. We did a prospective study in 2000-2001 in 105 French farrow-to-finish pig farms. In each farm, a batch of contemporary fattening pigs housed in the same room was followed throughout the fattening period. Salmonella shedding was assessed on environmental samples of faecal material (taken by means of pairs of gauze socks) analysed by classical bacteriological methods. 36.2% of the batches studied had at least one contaminated environmental sample and therefore were classified as Salmonella-shedding batches. Logistic regression was used to assess the association between managerial and hygiene practices and health status and the shedding risk at the end of the finishing period. Emptying the pit below the slatted floor after the previous batch of sows was removed and frequent removal of sow dung during the lactation period were protective. Presence of residual Salmonella contamination of the floor and pen partitions in the fattening rooms before loading the growing pigs also was a risk factor. The risk for Salmonella shedding at the end of the fattening period was increased when dry feed (versus wet feed) was provided during the fattening period. Lastly, Lawsonia intracellularis seroconversion and PRRSV seropositivity during the fattening period also was a risk factor for Salmonella shedding.

A study was carried out in 125 farrow-to-finish pig herds to assess the relationships between pathogens involved in respiratory disorders and to relate these findings to clinical signs of respiratory diseases and pneumonia and pleuritis at slaughter. Clinical examination and sampling were carried out on four different batches in each herd (pigs aged 4, 10, 16 and 22 weeks). Mycoplasma hyopneumoniae, Actinobacillus pleuropneumoniae, swine influenza viruses (SIV), porcine reproductive and respiratory syndrome virus (PRRSV) and porcine circovirus type 2 (PCV2) were detected by serological or PCR tests. Pneumonia-like gross lesions and pleuritis were scored at the slaughterhouse. The results indicate that the percentage of pigs PCR-positive for PCV2 at 4, 10 and 16 weeks old was associated with the percentage of pigs PCR-positive for M. hyopneumoniae at these ages. On the other hand, the percentage of pigs with antibodies against PRRSV at 10, 16 and 22 weeks was positively correlated with the percentage of pigs seropositive for M. hyopneumoniae at 22 weeks, with the percentage of pigs with antibodies against SIV H1N1 and SIV H1N2 and the percentage of pigs sero-positive for A. pleuropneumoniae serotype 2. The findings also indicate that, within the five studied pathogens, M. hyopneumoniae, PRRSV and SIV H1N1 are the major pathogens involved in pneumonia-like gross lesions even though PCV2 may play a role. A. pleuropneumoniae serotype 2, in association with PRRSV, is significantly associated with extensive pleuritis. Respiratory diseases could be significantly reduced by implementing measures including appropriate management practices to control these pathogens.

Relationships between macroscopic lesions and Polymerase Chain Reaction (PCR) detection of Mycoplasma hyopneumoniae (Mhp), Pasteurella multocida (Pm), Actinobacillus pleuropneumoniae (App), Haemophilus parasuis (Hps) and Streptococcus suis (Ssuis) of the lungs of 3731 slaughter pigs from 125 herds were assessed in France. Pneumonia and pleuritis were the most frequent lesions (69.3% and 15% of the lungs, respectively). Mhp, Pm, App, Ssuis and Hps were detected in 69.3%, 36.9%, 20.7%, 6.4% and 0.99% of the lungs, respectively. Mhp and Pm were associated with pneumonia at both the pig and herd levels. Pleuritis was not associated with any pathogen at the pig level, but was associated with a high percentage of pigs PCR-positive for App at the herd level. Measures focused on control of Mhp, Pm and App should significantly reduce the occurrence of both pneumonia and pleuritis.

A cross-sectional study involving 143 farrow-to-finish herds was carried out to identify herd-level noninfectious factors associated with pneumonia and pleuritis in slaughter pigs. Data related to herd characteristics, biosecurity, management and housing conditions were collected by questionnaire during a farm visit. Climatic conditions were measured over 20 h in the post-weaning and finishing rooms where the slaughter pigs were kept. After these on-farm investigations, the finishing pigs were examined at slaughter for lung lesions. A sample of 30 randomly selected pigs per herd was scored for pneumonia and pleuritis. Herds were grouped into three categories according to their pneumonia median score (class 1: ≤ 0.5; class 2: 0.5score ≤3.75; class 3:3.75). For pleuritis, a herd was deemed affected if at least one pig had a high pleuritis score (≥ 3). A multinomial logistic regression model was used to identify factors associated with pneumonia classes 2 and 3. A logistic regression for binary outcome was used to identify risk factors for severe pleuritis. An interval of less than four weeks between successive batches (OR=4.5, 95% confidence interval (95% CI): 1.5-13.6, p0.01), large finishing room size (OR=4.3, 95% CI: 1.6-11.6, p0.01) and high mean CO(2) concentration in the finishing room (OR=4.2, 95%CI: 1.6-11.3, p0.01), significantly increased the odds for a herd to be in class 2 for pneumonia. The same risk factors were found for class 3 and, in addition, a direct fresh air inlet from outside or from the corridor in the post-weaning room vs an appropriate ceiling above the pigs (OR=5.1, 95% CI: 1.4-18.8, p=0.01). The risk for a herd to have at least one pig with a high pleuritis score was increased when the farrowing facilities were not disinsected (OR=2.7, 95% CI: 1.2-5.8, p=0.01), when tail docking was performed later than 1.5 days after birth (OR=2.6, 95% CI: 1.2-5.7, p=0.01) and if the piglets were castrated when more than 14 days old (OR=2.7, 95%CI: 1.1-6.8, p=0.03). A temperature range of less than 5°C for the ventilation control rate in the farrowing room (OR=2.7, 95% CI: 1.2-5.9, p=0.01), a mean temperature in the finishing room below 23°C (OR=3.0, 95% CI: 1.3-6.8, p0.01) and large herd size (OR=3.1, 95% CI: 1.4-6.9, p0.01) were also associated with increased risk of pleuritis. The factors affecting pneumonia and pleuritis seemed to be different. All rearing steps from farrowing to finishing must be taken into account in any health programme aimed at controlling pneumonia and pleuritis and lung health may be improved through several pathways, i.e. correcting managerial and hygienic factors, implementing an appropriate and well-functioning ventilation in order to offer favorable climatic conditions.

International audience; In the absence of commercial kits for serological studies specifically for hepatitis E virus (HEV) infection of swine, the capsid protein of a swine genotype 3 HEV was expressed using a recombinant baculovirus. This antigen was used in an enzyme-linked immunosorbent assay (ELISA) for detection of anti-HEV antibodies. Estimation of sensitivity and specificity was carried out in comparison with a commercial serological assay using sera from pigs infected experimentally with HEV genotype 3, negative sera from SPF pigs and swine sera collected at slaughterhouses. Since a "gold standard" is not available, a latent-class Bayesian approach for correlated tests was used to estimate the sensitivity and specificity of both tests. External field and experimental data were used to determine the parameters of Beta prior distributions. The estimated mean sensitivity of the commercial test was 0.47 with a 95% credibility interval of 0.39-0.55, whereas the estimated mean sensitivity of the genotype 3-based assay was 0.92 [0.81-0.99]. The estimated specificities were 0.98 [0.93-0.99] and 0.98 [0.95-0.99] for the genotype 3-based test and commercial test, respectively. in conclusion, genotype 3-antigen derived from swine HEV is a better candidate for assessing hepatitis E serology in swine.

Four sampling techniques for Mycoplasma hyopneumoniae detection, namely nasal swabbing, oral-pharyngeal brushing, tracheo-bronchial swabbing and tracheo-bronchial washing, were compared in naturally infected live pigs. In addition, a quantitative real-time PCR assay for M. hyopneumoniae quantification was validated with the same samples. 60 finishing pigs were randomly selected from a batch of contemporary pigs on a farm chronically affected by respiratory disorders. Each pig was submitted to nasal swabbing, oral-pharyngeal brushing, tracheo-bronchial swabbing and tracheo-bronchial washing. Nested-PCR and real-time PCR assays were performed on all samples. A Bayesian approach was used to analyze the nested-PCR results of the four sampling methods (i.e. positive or negative) to estimate the sensitivity and specificity of each method. M. hyopneumoniae was detected by nested-PCR in at least one sample from 70% of the pigs. The most sensitive sampling methods for detecting M. hyopneumoniae in live naturally infected pigs were tracheo-bronchial swabbing and tracheo-bronchial washing, as compared to oral-pharyngeal brushing and nasal swabbing. Swabbing the nasal cavities appeared to be the least sensitive method. Significantly higher amounts of M. hyopneumoniae DNA were found at the sites of tracheo-bronchial sampling than in the nasal cavities or at the oral-pharyngeal site (p0.001). There was no difference between the tracheo-bronchial washing and the tracheo-bronchial swabbing results (p0.05). Our study indicated that tracheo-bronchial swabbing associated with real-time PCR could be an accurate diagnostic tool for assessing infection dynamics in pig herds.

Aims: To examine the variability among Pasteurella multocida strains isolated from pigs (nasal, tonsil and lung specimens) and humans in France. Methods and Results: The genetic diversity of 117 French isolates of P. multocida, obtained from pigs (n = 101) and humans (n = 16) and three reference strains, was evaluated by pulsed-field gel electrophoresis (PFGE) after macrorestriction with ApaI. Sixty-four patterns were detected. The genetic relationships revealed five clusters (Aa1, Aa2, Aa3, Ab and B). The pig isolates obtained from pneumonic lungs and nasal cavities were clustered in groups Ab and Aa1, respectively (P 0·95 in both populations of human and pig isolates. Therefore, ApaI-PFGE seems to be a useful tool for epidemiological tracing of P. multocida infections.

23 pages; The purpose of this study was to compare in SPF pigs, the pathogenicity of an Actinobacillus pleuropneumoniae serotype 9 strain 21 (isolated from the palatine tonsils of a healthy gilt on a French nucleus pig farm, with no clinical signs or lung lesions but a highly positive reaction to A. pleuropneumoniae serotype 9 antibodies) with a pathogenic A. pleuropneumoniae strain 4915 serotype 9 (isolated in France from an outbreak of porcine pleuropneumonia). The pathogenicity of one Mycoplasma hyopneumoniae strain alone or associated with A. pleuropneumoniae strain 21 was also compared. Eight groups of 7 pigs were infected (at 6 or 10 weeks of age) and a control group was kept non-infected. Results showed that sensitivity to A. pleuropneumoniae was related to the age of the pig (6 weeks vs 10 weeks) whatever the strain. Surviving pigs infected at 6 weeks of age developed severe clinical signs, lung lesions typical of A. pleuropneumoniae and they seroconverted. In contrast, symptoms and lung lesions were almost non-existent in pigs infected with strain 21 at 10 weeks of age, but a seroconversion was observed with very high ELISA titres. These results were in accordance with those observed in the nucleus pig farm. Infection with M. hyopneumoniae alone induced typical mycoplasmal symptoms, pneumonia and seroconversion. Symptoms and lung lesions were the most noticeable in pigs infected with M. hyopneumoniae at 6 weeks of age and with A. pleuropneumoniae 4 weeks later. Our results show that the presence of A. pleuropneumoniae serotype 9 in a pig herd may be clinically unnoticed and that M. hyopneumoniae may potentiate A. pleuropneumoniae infection.

An analytic epidemiological survey was carried out in 105 French farms to identify factors associated with Salmonella shedding by finishing pigs. This study gave out a list of 7 risk factors using a logistic model. The aim of the present survey was to validate this model on a second sample of batches of pigs in order to estimate their Salmonella status. The validation study was carried out from April 2003 to August 2005 on 64 finishing pig batches distinct from those used originally to generate the logistic model. In each farm, Salmonella shedding of a batch of pigs at the end of the finishing phase was assessed using swabs as described in the analytical study. Questionnaires were filled in with the farmer to collect data related to management routines. Blood samples from10 growing and 10 finishing pigs were taken to assess sanitary risk factors: status vs Lawsonia intracellularis and Porcine Respiratory Coronavirus. Salmonella contamination status of a finishing room before loading, a further identified risk factor, was tested by environmental swabbing procedure. The estimated risk with the standard error, of Salmonella shedding was calculated using the logistic model and compared to the bacteriological Salmonella status of each batch. Several thresholds are proposed and sensitivity, specificity, positive and negative predictive values related to each cut-off value were calculated. A cut-off value of 0.34 maximised both sensitivity (76.9%) and specificity (68.6%) of the model. Whatever the threshold, the accuracy of the Salmonella non-shedding predicted status is better than the Salmonella shedding predicted status. In a bacteriological sampling programme, this model could be a useful tool to identify batches with low risk of Salmonella shedding and to focus attention on those getting a high probability for being positive.

An analytic epidemiological survey was carried out in France to assess the circum- stances for Salmonella shedding of finishing pigs. This study gave out a combination of risk fac- tors. The aim of our study was to evaluate that the implementation of a programme based on these risk factors reduced Salmonella carriage. A farm was selected according to the positive Salmonella shedding status of fattening pigs. A first profile of the farm according to the risk fac- tors was obtained. On the basis of these results, decisions were taken with the farmer in order to reduce the risks on 5 batches of pigs. At the end of the fattening phase, the Salmonella status is measured. The profile according to the risk factors is obtained. No one of the 5 batches was found to be Salmonella contaminated. The results indicate that this type of multifactorial approach is an appropriate way to prevent Salmonella contamination.