Orientador: Jane Megid / Banca: Marcio Garcia Ribeiro / Banca: Silvio Arruda Vasconcellos / Resumo: A brucelose canina causada pela Brucella canis (B. canis) é uma das principais causas infecciosas de desordens reprodutivas em cães. O presente trabalho teve como objetivo comparar as técnicas de Cultivo Microbiológico e Soroaglutinação Rápida em Cartão (SAR) com e sem o emprego de 2-Mercaptoetanol (2-ME) no diagnóstico da brucelose canina. Adicionalmente foram avaliados sangue, urina, swab vaginal, prepucial e sêmen como materiais a serem processados no diagnóstico microbiológico. Foram avaliados 236 cães quanto à infecção por B. canis, muitos com histórico de problemas reprodutivos, dos quais 24,2% resultaram positivos na SAR, 10,2% na SAR-2ME, 28% na hemocultura, 9,2% na urocultura, 2,1% na cultura de swab vaginal, 13,6% na cultura dos swabs prepuciais e 28,6% no cultivo de sêmen. Dos 71 animais positivos no cultivo microbiológico, 92,9% apresentaram-se positivos na hemocultura. Nos demais materiais foram obtidos percentuais menores. A sensibilidade relativa da SAR resultou em 66,2% e a especificidade relativa 93,9%. A SAR-2ME apresentou sensibilidade relativa de 29,5% com ligeiro aumento na especificidade relativa 98,2%. Quando associados SAR e hemocultura foram observados 97,6% e 93,9% de sensibilidade e especificidade, respectivamente. Os resultados sugerem a utilização associada da SAR à hemocultura sem realização em paralelo com 2-ME. / Abstract: Canine brucellosis caused by Brucella canis (B. canis) is one of the major infectious causes of reproductive disorders in dogs. The present study aimed to compare the performance of bacteriological methods and Rapid Slide Agglutination Test (RSAT) with or without 2-Mercaptoethanol (2-ME) in canine brucellosis diagnosis. Additionally, blood, urine, vaginal and prepucial swabs and semen were evaluated regarding their use in bacteriological diagnostic. Two hundred thirty six dogs, many of them showing reproductive problems, were submitted to investigation for diagnosis of B. canis infection. Among them, 24.2% were positive in RSAT, 10.2% 2ME-RSAT, 28% blood culture, 9.2% urine culture, 2.1% vaginal swab culture, 13.6% prepucial swab culture and 28.6% semen culture. Among the 71 positive dogs in bacteriological culture, 92.9% were positive in blood culture. The percentage of positivity was variable in others samples. The relative sensitivity of RSAT was 66.2% and the relative specificity was 93.9%. The 2ME-RSAT presented lower sensitivity 29.5% and greater specificity 98.2%. The use of blood culture and RSAT associated presented 97.6% e 93.9% of sensibility and specificity, respectively. These results suggest that RSAT should be used with blood cultures independently of the use of 2ME-RSAT. / Mestre

Made available in DSpace on 2014-06-11T19:27:15Z (GMT). No. of bitstreams: 0 Previous issue date: 2006-08-27Bitstream added on 2014-06-13T20:55:59Z : No. of bitstreams: 1 salgado_vr_me_botfmvz.pdf: 342250 bytes, checksum: b60d6784559654ba5442f18a243ca0e7 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / A brucelose canina causada pela Brucella canis (B. canis) é uma das principais causas infecciosas de desordens reprodutivas em cães. O presente trabalho teve como objetivo comparar as técnicas de Cultivo Microbiológico e Soroaglutinação Rápida em Cartão (SAR) com e sem o emprego de 2-Mercaptoetanol (2-ME) no diagnóstico da brucelose canina. Adicionalmente foram avaliados sangue, urina, swab vaginal, prepucial e sêmen como materiais a serem processados no diagnóstico microbiológico. Foram avaliados 236 cães quanto à infecção por B. canis, muitos com histórico de problemas reprodutivos, dos quais 24,2% resultaram positivos na SAR, 10,2% na SAR-2ME, 28% na hemocultura, 9,2% na urocultura, 2,1% na cultura de swab vaginal, 13,6% na cultura dos swabs prepuciais e 28,6% no cultivo de sêmen. Dos 71 animais positivos no cultivo microbiológico, 92,9% apresentaram-se positivos na hemocultura. Nos demais materiais foram obtidos percentuais menores. A sensibilidade relativa da SAR resultou em 66,2% e a especificidade relativa 93,9%. A SAR-2ME apresentou sensibilidade relativa de 29,5% com ligeiro aumento na especificidade relativa 98,2%. Quando associados SAR e hemocultura foram observados 97,6% e 93,9% de sensibilidade e especificidade, respectivamente. Os resultados sugerem a utilização associada da SAR à hemocultura sem realização em paralelo com 2-ME. / Canine brucellosis caused by Brucella canis (B. canis) is one of the major infectious causes of reproductive disorders in dogs. The present study aimed to compare the performance of bacteriological methods and Rapid Slide Agglutination Test (RSAT) with or without 2-Mercaptoethanol (2-ME) in canine brucellosis diagnosis. Additionally, blood, urine, vaginal and prepucial swabs and semen were evaluated regarding their use in bacteriological diagnostic. Two hundred thirty six dogs, many of them showing reproductive problems, were submitted to investigation for diagnosis of B. canis infection. Among them, 24.2% were positive in RSAT, 10.2% 2ME-RSAT, 28% blood culture, 9.2% urine culture, 2.1% vaginal swab culture, 13.6% prepucial swab culture and 28.6% semen culture. Among the 71 positive dogs in bacteriological culture, 92.9% were positive in blood culture. The percentage of positivity was variable in others samples. The relative sensitivity of RSAT was 66.2% and the relative specificity was 93.9%. The 2ME-RSAT presented lower sensitivity 29.5% and greater specificity 98.2%. The use of blood culture and RSAT associated presented 97.6% e 93.9% of sensibility and specificity, respectively. These results suggest that RSAT should be used with blood cultures independently of the use of 2ME-RSAT.

A presente tese estudou o diagnóstico e tratamento da mastite bovina pela avaliação do uso de um teste prático de CCS, pela utilização de cultura bacteriológica a campo e pela definição de protocolos de tratamento. Para rápida determinação da CCS, o teste Somaticell® foi usado em amostras de leite tendo o resultado comparado à contagem eletrônica e avaliado por tipo de amostra e pessoa. O Somaticell determinou corretamente a CCS de amostras frescas de leite de quartos mamários. A correlação registrada entre o Somaticell e a CCS eletrônica foi 0,92 e o coeficiente de concordância 0,82. O teste mostrou adequada validade para determinar infecções intramamárias (sensibilidade 91,3%; especificidade 96,0%) e apresentou contagens mais elevadas em amostras contendo patógenos. Pequena variação foi verificada nos resultados do teste quando realizado em duplicata. Entretanto na análise geral dos dados, a variação observada não foi significativa nem afetou a quantidade de amostras com mastite subclínica. Amostras de leite conservadas a 4 ºC por até 5 horas não influenciaram os resultados do Somaticell, mas amostras congeladas ou adicionadas do conservante bronopol não devem ser usadas. Quanto ao diagnóstico da mastite clínica, as infecções causadas por bactérias Gram positivas foram em maioria isoladas de casos clínicos com alteração visual do leite e edema de úbere, de vacas reincidentes e com CCS mensal média maior de 200.000 céls/mL. Por outro lado, as mastites causadas por bactérias Gram negativas apresentaram em maioria casos clínicos com comprometimento sistêmico em vacas de alta produção. Enquanto isso, as amostras mastíticas com ausência de crescimento na cultura laboratorial apresentaram grau da mastite e perfil do animal variado. A cultura bacteriológica realizada na propriedade leiteira com o conjunto dos meios Sangue Base Azida, MacConkey e Vogel-Johnson, mostrou adequada recuperação dos patógenos causadores de mastite clínica. Do total de 203 cultivos realizados na fazenda e em laboratório 79,3% mostraram concordância. A taxa de concordância das culturas foi afetada pelo número de unidades formadoras de colônia presente na amostra de leite. A sensibilidade e especificidade registradas para a cultura a campo foram 83,0 e 76,5%, respectivamente, as quais foram positivamente influenciadas pelo isolamento de bactérias Gram positivas, e negativamente pelo isolamento de bactérias Gram negativas e resultados com ausência de crescimento. O cultivo feito na fazenda mostrou vantagens em usar amostras de leite frescas, porém a leitura precoce do crescimento bacteriano e a metodologia simples reduzem sua acurácia. Para o tratamento das mastites clínicas com base no resultado da cultura a campo, foi encontrada taxa geral de cura bacteriológica de 69,7%. A CCS da maioria dos tetos que apresentaram mastite se manteve elevada até 21 dias pós-detecção da doença. Mastites causadas por bactérias Gram negativas mostraram taxas de cura mais altas. Já as mastites causadas por Staphylococcus aureus tiveram as menores taxas de cura apesar da utilização de tratamento antibiótico estendido. As mastites de grau 1 e 2, com isolamento de bactérias Gram negativas e ausência de crescimento na cultura bacteriológica, não apresentaram diferença em cura sendo ou não tratadas com antibiótico intramamário. / The present thesis investigated the diagnosis and treatment of bovine mastitis by using a simple test for somatic cell count, on-farm bacteriological culture and guided treatment protocols. To rapidly determine SCC, Somaticell® test was used on milk samples having the results compared to electronic count and evaluated by sample type and reader. The Somaticell test correctly determined the SCC in fresh milk samples of mammary quarters. The correlation between Somaticell and electronic SCC was 0.92, being kappa coefficient equals to 0.82. This test presented good reliability to determine intramammary infections using a threshold of 205.000 cells/mL (sensitivity = 91.3% and specificity = 96.0%) and showed greater SCC in samples containing major mastitis pathogens. Minor intra-individual variation was detected when performing the test. Probably, the homogenization procedure of the test is the most likely explanation for the observed variation. However, the final analysis indicated that this variation was not significant and did not affect the amount of samples classified as having subclinical mastitis. Milk samples preserved at 4 ºC up to 5 hours did not change test results. Nevertheless, frozen or bronopol preserved samples were not suitable for this test. Clinical mastitis data indicated that infections caused by Gram positive bacteria were mainly seen on clinical cases with visual milk abnormalities and udder edema of clinical recurrent animals showing monthly average SCC greater than 200,000 cells/mL. In contrast, clinical mastitis caused by Gram negative bacteria was frequently associated with systemic signs in high producing cows. In addition, mastitic samples without bacterial growth in the laboratory did not present a defined pattern in relation to mastitis grade and animal characteristic. The on-farm bacteriological culture using Azide Blood Agar Base, MacConkey Agar and Vogel-Johnson Agar, showed adequate recovery of mastitis causing pathogens. In 79.3% of the cultures (n = 203) the on-farm results agreed with the standard laboratory culture. Interestingly, the concordance rate was affected by the number of colony forming units in the milk sample. The sensitivity and specificity of on-farm culture were 83.0 and 76.5%, respectively. These results were positively influenced by growth of Gram positive bacteria and negatively influenced by growth of Gram negative bacteria and samples with no growth. The use of fresh milk samples in on-farm culture seemed advantageous for bacterial recovery, although premature plate reading of bacterial growth and simplicity of this methodology might reduce its accuracy. Mastitis treatment guided by on-farm culture showed an overall bacteriological cure rate of 69.7%. In the majority of quarters, SCC remained elevated within 21 days after detection of clinical mastitis. Gram negative bacteria presented greater cure rates. On the contrary, Staphylococcus aureus mastitis displayed the lowest cure rate even by using extended antibiotic treatment. Intramammary antibiotic treatment did not show effect on cure rates of grade 1 and 2 clinical mastitis caused by Gram negative bacteria or with negative growth in the on-farm culture.

A presente tese estudou o diagnóstico e tratamento da mastite bovina pela avaliação do uso de um teste prático de CCS, pela utilização de cultura bacteriológica a campo e pela definição de protocolos de tratamento. Para rápida determinação da CCS, o teste Somaticell® foi usado em amostras de leite tendo o resultado comparado à contagem eletrônica e avaliado por tipo de amostra e pessoa. O Somaticell determinou corretamente a CCS de amostras frescas de leite de quartos mamários. A correlação registrada entre o Somaticell e a CCS eletrônica foi 0,92 e o coeficiente de concordância 0,82. O teste mostrou adequada validade para determinar infecções intramamárias (sensibilidade 91,3%; especificidade 96,0%) e apresentou contagens mais elevadas em amostras contendo patógenos. Pequena variação foi verificada nos resultados do teste quando realizado em duplicata. Entretanto na análise geral dos dados, a variação observada não foi significativa nem afetou a quantidade de amostras com mastite subclínica. Amostras de leite conservadas a 4 ºC por até 5 horas não influenciaram os resultados do Somaticell, mas amostras congeladas ou adicionadas do conservante bronopol não devem ser usadas. Quanto ao diagnóstico da mastite clínica, as infecções causadas por bactérias Gram positivas foram em maioria isoladas de casos clínicos com alteração visual do leite e edema de úbere, de vacas reincidentes e com CCS mensal média maior de 200.000 céls/mL. Por outro lado, as mastites causadas por bactérias Gram negativas apresentaram em maioria casos clínicos com comprometimento sistêmico em vacas de alta produção. Enquanto isso, as amostras mastíticas com ausência de crescimento na cultura laboratorial apresentaram grau da mastite e perfil do animal variado. A cultura bacteriológica realizada na propriedade leiteira com o conjunto dos meios Sangue Base Azida, MacConkey e Vogel-Johnson, mostrou adequada recuperação dos patógenos causadores de mastite clínica. Do total de 203 cultivos realizados na fazenda e em laboratório 79,3% mostraram concordância. A taxa de concordância das culturas foi afetada pelo número de unidades formadoras de colônia presente na amostra de leite. A sensibilidade e especificidade registradas para a cultura a campo foram 83,0 e 76,5%, respectivamente, as quais foram positivamente influenciadas pelo isolamento de bactérias Gram positivas, e negativamente pelo isolamento de bactérias Gram negativas e resultados com ausência de crescimento. O cultivo feito na fazenda mostrou vantagens em usar amostras de leite frescas, porém a leitura precoce do crescimento bacteriano e a metodologia simples reduzem sua acurácia. Para o tratamento das mastites clínicas com base no resultado da cultura a campo, foi encontrada taxa geral de cura bacteriológica de 69,7%. A CCS da maioria dos tetos que apresentaram mastite se manteve elevada até 21 dias pós-detecção da doença. Mastites causadas por bactérias Gram negativas mostraram taxas de cura mais altas. Já as mastites causadas por Staphylococcus aureus tiveram as menores taxas de cura apesar da utilização de tratamento antibiótico estendido. As mastites de grau 1 e 2, com isolamento de bactérias Gram negativas e ausência de crescimento na cultura bacteriológica, não apresentaram diferença em cura sendo ou não tratadas com antibiótico intramamário. / The present thesis investigated the diagnosis and treatment of bovine mastitis by using a simple test for somatic cell count, on-farm bacteriological culture and guided treatment protocols. To rapidly determine SCC, Somaticell® test was used on milk samples having the results compared to electronic count and evaluated by sample type and reader. The Somaticell test correctly determined the SCC in fresh milk samples of mammary quarters. The correlation between Somaticell and electronic SCC was 0.92, being kappa coefficient equals to 0.82. This test presented good reliability to determine intramammary infections using a threshold of 205.000 cells/mL (sensitivity = 91.3% and specificity = 96.0%) and showed greater SCC in samples containing major mastitis pathogens. Minor intra-individual variation was detected when performing the test. Probably, the homogenization procedure of the test is the most likely explanation for the observed variation. However, the final analysis indicated that this variation was not significant and did not affect the amount of samples classified as having subclinical mastitis. Milk samples preserved at 4 ºC up to 5 hours did not change test results. Nevertheless, frozen or bronopol preserved samples were not suitable for this test. Clinical mastitis data indicated that infections caused by Gram positive bacteria were mainly seen on clinical cases with visual milk abnormalities and udder edema of clinical recurrent animals showing monthly average SCC greater than 200,000 cells/mL. In contrast, clinical mastitis caused by Gram negative bacteria was frequently associated with systemic signs in high producing cows. In addition, mastitic samples without bacterial growth in the laboratory did not present a defined pattern in relation to mastitis grade and animal characteristic. The on-farm bacteriological culture using Azide Blood Agar Base, MacConkey Agar and Vogel-Johnson Agar, showed adequate recovery of mastitis causing pathogens. In 79.3% of the cultures (n = 203) the on-farm results agreed with the standard laboratory culture. Interestingly, the concordance rate was affected by the number of colony forming units in the milk sample. The sensitivity and specificity of on-farm culture were 83.0 and 76.5%, respectively. These results were positively influenced by growth of Gram positive bacteria and negatively influenced by growth of Gram negative bacteria and samples with no growth. The use of fresh milk samples in on-farm culture seemed advantageous for bacterial recovery, although premature plate reading of bacterial growth and simplicity of this methodology might reduce its accuracy. Mastitis treatment guided by on-farm culture showed an overall bacteriological cure rate of 69.7%. In the majority of quarters, SCC remained elevated within 21 days after detection of clinical mastitis. Gram negative bacteria presented greater cure rates. On the contrary, Staphylococcus aureus mastitis displayed the lowest cure rate even by using extended antibiotic treatment. Intramammary antibiotic treatment did not show effect on cure rates of grade 1 and 2 clinical mastitis caused by Gram negative bacteria or with negative growth in the on-farm culture.

The objectives of this paper were (i) to perform a systematic review of the literature over the last 21 yr and (ii) to evaluate the efficacy of selective dry cow treatment (SDCT) vs. blanket dry cow treatment (BDCT) in dairy cows regarding the risk of intramammary infection (IMI) after calving, new IMI risk after calving, cure risk during the dry period, and a reduction in antibiotic use at drying-off by meta-analysis. The final number of included studies was n = 3 for IMI risk after calving and n = 5 for new IMI risk after calving, cure risk during the dry period, and antibiotic use. The relative risk (RR) levels for IMI, new IMI, and cure did not differ significantly between SDCT and BDCT. This meta-analysis provides evidence that SDCT seems to be an adequate alternative to BDCT regarding udder health with a simultaneous reduction in antibiotic use. Limitations might arise because of the small number of studies included.:1. Introduction 2. Material and Methods 2.1 Search Strategy 2.2 Study Selection 2.3 Data Extraction 2.4 Assessment of Bias Risk 2.5 Statistical Analysis 3. Results 3.1 Study Selection 3.2 Study Characteristics 3.3 Outcomes 3.4 Risk of Bias 3.5 Pairwise Meta-Analysis 3.6 Subgroup Meta-Analysis 4. Discussion 4.1 Risk of IMI after Calving 4.2 Risk of new IMI after Calving 4.3 Cure Risk during Dry Period 4.4 Antibiotic Use at Drying-Off 4.5 Strengths and Limitations / Ziele dieser Studie waren (i) ein systematisches Review über die Literatur der letzten 21 yr zu erstellen sowie (ii) die Effizienz von selektivem Trockenstellen (SDCT) im Vergleich zu konventionellem Trockenstellen (BDCT) bei Milchkühen anhand von Risiko an intramammären Infektionen (IMI) post partum (pp), Neuinfektionenrisiko pp, Heilungsrisiko während dem Trockenstehen und Reduktion des Antibiotikaeinsatzes zum Trockenstellen mittels einer Meta-Analyse zu untersuchen. Die finale Anzahl eingeschlossener Studien war n = 3 für das IMI-Risiko pp sowie n = 5 für Neuinfektionenrisiko pp, Heilungsrisiko während dem Trockenstehen und Antibiotikaeinsatz zum Trockenstellen. Die relativen Risiken für IMI pp, Neuinfektionen pp und Heilung während der Trockenstehperiode unterschieden sich dabei nicht signifikant zwischen SDCT und BDCT. Diese Meta-Analyse erbringt Evidenz, dass SDCT eine adäquate Alternative zu BDCT im Hinblick auf Eutergesundheit und Antibiotikareduktion ist. Limitationen entstehen durch die geringe Anzahl an eingeschlossenen Studien.:1. Introduction 2. Material and Methods 2.1 Search Strategy 2.2 Study Selection 2.3 Data Extraction 2.4 Assessment of Bias Risk 2.5 Statistical Analysis 3. Results 3.1 Study Selection 3.2 Study Characteristics 3.3 Outcomes 3.4 Risk of Bias 3.5 Pairwise Meta-Analysis 3.6 Subgroup Meta-Analysis 4. Discussion 4.1 Risk of IMI after Calving 4.2 Risk of new IMI after Calving 4.3 Cure Risk during Dry Period 4.4 Antibiotic Use at Drying-Off 4.5 Strengths and Limitations

Huit cent trente et un troupeaux de vaches laitières répartis dans 5 états américains ont été enrôlés dans une étude de cohorte prospective. Un modèle d’équations d'estimation généralisées a été utilisé pour étudier l'association entre les signes cliniques et la détection de salmonelles dans les fèces des animaux soupçonnés de salmonellose clinique. La sensibilité et la spécificité de la culture bactériologique ont été estimées à l’aide d’un modèle de classes latentes. Dix-huit pour cent des 874 échantillons provenant de veaux et 29% des 1479 échantillons de vaches adultes étaient positifs pour Salmonella spp. Il n’a pas été possible d’établir une association claire entre les différents signes cliniques observés et la détection de salmonelles. Les 2 sérotypes les plus fréquemment isolés étaient Typhimurium et Newport. La probabilité de détecter des salmonelles était plus élevée chez les veaux où un autre agent entéropathogène était également détecté. La proportion d’échantillons positifs était plus élevée parmi les vaches ayant reçu des antibiotiques dans les jours précédant l’échantillonnage. La sensibilité de la culture a été estimée à 0,48 (intervalle de crédibilité à 95% [ICr95%]: 0,22-0,95) pour les veaux et 0,78 (ICr95%: 0,55-0,99) pour les vaches. La spécificité de la culture était de 0,94 (ICr95%: 0,87-1,00) pour les veaux et de 0,96 (ICr95%: 0,90-1,00) pour les vaches. Malgré une sensibilité imparfaite, la culture bactériologique demeure utile pour obtenir une meilleure estimation de la probabilité post-test de salmonellose clinique chez un bovin laitier, par rapport à la probabilité estimée suite au seul examen clinique. / Eight hundred and thirty-one dairy herds in 5 states in the northeast USA were enrolled in a prospective cohort study. A generalized estimating equations model including herd as a random effect was built to study the association between clinical signs and detection of Salmonella spp. in the faeces of animals suspected of clinical salmonellosis. The sensitivity and specificity of the bacteriological culture were estimated using a latent class model. Eighteen percent of the 874 samples from calves and 29% of the 1479 samples from adult cows were positive for Salmonella spp. It was not possible to establish a clear association between the various clinical signs and detection of Salmonella spp. in the faeces. The two most frequently isolated serotypes were Typhimurium and Newport. The probability of detecting salmonellas was higher for calves when another enteric pathogen was also detected. The proportion of positive samples was higher among cows that received antibiotics in the days preceding the sampling. The sensitivity of the bacteriological culture was estimated at 0.48 (95% credibility interval [95%CrI]: 0.22 to 0.95) for calves and 0.78 (95%CrI 0.55 to 0.99) for cows. The specificity of the culture was 0.94 (95%CrI: 0.87 to 1.00) for calves and 0.96 (95%CrI: 0.90 to 1.00) for cows. Despite imperfect sensitivity, bacterial culture remains useful to obtain a better estimate of the post-test probability of clinical salmonellosis in dairy cattle, compared to the estimated probability following the clinical examination alone.

Huit cent trente et un troupeaux de vaches laitières répartis dans 5 états américains ont été enrôlés dans une étude de cohorte prospective. Un modèle d’équations d'estimation généralisées a été utilisé pour étudier l'association entre les signes cliniques et la détection de salmonelles dans les fèces des animaux soupçonnés de salmonellose clinique. La sensibilité et la spécificité de la culture bactériologique ont été estimées à l’aide d’un modèle de classes latentes. Dix-huit pour cent des 874 échantillons provenant de veaux et 29% des 1479 échantillons de vaches adultes étaient positifs pour Salmonella spp. Il n’a pas été possible d’établir une association claire entre les différents signes cliniques observés et la détection de salmonelles. Les 2 sérotypes les plus fréquemment isolés étaient Typhimurium et Newport. La probabilité de détecter des salmonelles était plus élevée chez les veaux où un autre agent entéropathogène était également détecté. La proportion d’échantillons positifs était plus élevée parmi les vaches ayant reçu des antibiotiques dans les jours précédant l’échantillonnage. La sensibilité de la culture a été estimée à 0,48 (intervalle de crédibilité à 95% [ICr95%]: 0,22-0,95) pour les veaux et 0,78 (ICr95%: 0,55-0,99) pour les vaches. La spécificité de la culture était de 0,94 (ICr95%: 0,87-1,00) pour les veaux et de 0,96 (ICr95%: 0,90-1,00) pour les vaches. Malgré une sensibilité imparfaite, la culture bactériologique demeure utile pour obtenir une meilleure estimation de la probabilité post-test de salmonellose clinique chez un bovin laitier, par rapport à la probabilité estimée suite au seul examen clinique. / Eight hundred and thirty-one dairy herds in 5 states in the northeast USA were enrolled in a prospective cohort study. A generalized estimating equations model including herd as a random effect was built to study the association between clinical signs and detection of Salmonella spp. in the faeces of animals suspected of clinical salmonellosis. The sensitivity and specificity of the bacteriological culture were estimated using a latent class model. Eighteen percent of the 874 samples from calves and 29% of the 1479 samples from adult cows were positive for Salmonella spp. It was not possible to establish a clear association between the various clinical signs and detection of Salmonella spp. in the faeces. The two most frequently isolated serotypes were Typhimurium and Newport. The probability of detecting salmonellas was higher for calves when another enteric pathogen was also detected. The proportion of positive samples was higher among cows that received antibiotics in the days preceding the sampling. The sensitivity of the bacteriological culture was estimated at 0.48 (95% credibility interval [95%CrI]: 0.22 to 0.95) for calves and 0.78 (95%CrI 0.55 to 0.99) for cows. The specificity of the culture was 0.94 (95%CrI: 0.87 to 1.00) for calves and 0.96 (95%CrI: 0.90 to 1.00) for cows. Despite imperfect sensitivity, bacterial culture remains useful to obtain a better estimate of the post-test probability of clinical salmonellosis in dairy cattle, compared to the estimated probability following the clinical examination alone.