EVALUATION OF TUBERCULIN SKIN REACTION IN BUFFALOES (BUBALUS BUBALIS)ABSTRACT
The present work evaluated the tuberculin skin test for the diagnosis of tuberculosis in buffaloes. In Experiment I a comparative study was carried out of tuberculin skin reaction between buffaloes and bovines experimentally sensitized with M.bovis or M.avium that showed reactions 1.8 times greater in caudal fold than in the cervical region of buffaloes and cervical reactions 1.69 times greater in buffaloes than in bovines. The caudal fold test (CFT) reading by paquimeter was suggested, classifying as positive the reactions greater than 10.O mm, suspect between 6.4 and 9.9 mm and negative when smaller than 6.3 mm. In the cervical comparative test (CCT), buffaloes were classified as reagent when skin reaction was greater than 5.0 mm, suspect between 3.0 and 4.9 mm and negative when smaller than 2.9 mm. In Experiment II, the tuberculin tests in buffaloes naturally exposed to M.bovis or other environmental mycobacterias were compared to the anatomic, histopathological, bacteriological and molecular biological analysis which revealed that the CFT showed 43.24% of false positive results and 33.33% of false negative results. If the modified values were used, percentage of false positive results decrease to 10.81% and zero respectively in both tests, although the percentage of false negatives do not change. There was no significant interference of environmental mycobacterias infection in the buffalo tuberculin tests.
KEY WORDS:
Buffaloes; tuberculosis; diagnostic; tuberculin test; M.bovis.
RESUMO
O presente trabalho avalia a resposta imunoalérgica cutânea à tuberculina para o diagnóstico da tuberculose em bubalinos. No Experimento I foi avaliada a reatividade à tuberculina em bubalinos e bovinos experimentalmente sensibilizados com M.bovis ou M.avium, frente aos controles, evidenciando que em bubalinos a intensidade das reações na prega ano-caudal foi 1,8 vezes maior que na região cervical e esta foi 1,69 vezes maior que em bovinos. Foi sugerida a leitura do teste simples caudal (T.S.C.) em bubalinos por paquimetria, considerando-se como positiva a reação maior ou igual a 10,0 mm, suspeita entre 6,4 e 9,9 mm e negativa a reação menor que 6,3 mm. No teste cervical comparativo (T.C.C.), os bubalinos seriam considerados positivos quando apresentassem reação maior ou igual a 5,0 mm, suspeitos entre 3,0 e 4,9 mm e negativos quando menor que 2,9 mm. No Experimento II, os resultados dos testes tuberculínicos em bubalinos naturalmente expostos à infecção por M.bovis ou outras rnicobactérias ambientais foram comparados aos exames anátomo e histopatológicos, bacteriológicos e de biologia molecular, revelando que o T.S.C. apresentou 43,24% de resultados falso-positivos e 33,33% de falso-negativos, enquanto que o T.C.C. apresentou 2,7% de resultados falso-positivos e 66,66% falso-negativos. Aplicando-se os valores modificados para a interpretação dos testes, diminuem os resultados falso-positivos para 10,81% e zero em ambos, sem contudo alterar a porcentagem de falso-negativos. Os resultados não foram influenciados pela infecção por micobactérias ambientais.
PALAVRAS-CHAVE:
Bubalinos; tuberculose; diagnóstico; teste tuberculínico; M.bovis.
INTRODUCTION
Nowadays, tuberculosis reappears in human beings as an infectious disease of great magnitude in Third World countries (KLEEBERG, 1984) and the immune deficiency syndrome-AIDS contributed to increase the number of cases and the risk of infection in man by M. tuberculosis, and probably by M. bovis (SAUBOLLE, 1989;0'REILLY & DABORN, 1995). The zoonotic aspect of the disease is a matter of concem in developing countries, where the knowledge of the problem is scarce. In these countries there must be taken into account the public health aspects conceming M.bovis, especially in risk populations (WORLD HEALTH ORGANIZATION, 1993). Occupational groups, such as butchers and farm workers, are the most frequently affected (O'REILLY & DABORN, 1995).
The water buffalo, the bovine and several other domestic and wild animals might be hosts of M.bovis and the infected animal is the main source of infection to other animals and man (HAAGSMA, 1995; O'REILLY & DABORN, 1995). The agent can be eliminated by the respiration, nasal rheum, milk, faeces, urine, uterine and vagin secretion and semen. The infected milk is the main way of M.bovis transmission to young animals and children (NEILL et al., 1994).
As in bovines, tuberculosis in water buffalo is related in several countries, causing herd fecundity decrease and economic losses by condemnation of carcass. The tuberculin test with slaughter of reagent animals is the most efficient method to control tuberculosis. (DESHPANDE et al., 1966; PORTU-. GAL et al., 1971; MANDAL & IYER, 1971; MANDAL & SINGH, 1975; VACCARO et al., 1982; RATLEDGE & STANFORD, 1983; VANAMAYYA, 1987; HAAGSMA, 1995).
However, the tuberculosis diagnosis by tuberculin test is not effective in any species, NAIN & KAUSHIK (1985) stated that tuberculin test is not a good screening method and it can be_ positive in some animals in which the infection cannot be confirmed at slaughter.
KANTOR et al. (1984) stated that the main advantage of single tuberculin test in the caudal fold was the high sensitivity, similar to the cervical single test, it is easily and fast to apply and read, has a high specificity and it should be used for herd screening. The cervical comparative test use should be limited to "problem-herds". On the other hand, the sensitivity and specificity of tuberculin tests can be modified by the PPD dosage and interpretation of the results.
DOBBELAER et al. (1983) carrying out experimental pattems to evaluate PPD potency using heatkilled BCG in oil medium, verified that the PPD response can be different between bovines and guinea pigs. They suggested that the tuberculin should be tested in the appropriate species.
The water buffalo seems to develop tuberculin skin reactions more intense than bovine and many cases of nonspecific infection have being described (LÁU, 1994; NICOLETTI, 1994). Such nonspecific reactions should be promoted by M.avium, M. paratuberculosis or environmental mycobacterias (O'REILLY & DABORN, 1995).
VALE (1994) pointed out the lack of sanitary conditions on Latin America buffalo farms and stressed that the differences in the immunological aspects between bovines and buffaloes must be taken into account. LÁU (1994) and NICOLETTI (1994) also pointed out that extrapolation of bovine protocols to water buffalo may cause mistakes in the interpretation of tuberculin test results.
In. this work, a comparative study of tuberculin responses in bovine and buffalo was carried out to evaluate the tests for the diagnosis of tuberculosis in water buffaloes.
MATERIALS AND METHODS
Experiment I - tuberculin test in experimentally sensitized buffaloes and bovines
Animals: 27 buffaloes and 21 bovines, young (9 to 23 months) and adults (2 to 12 years old) of different breedings from the Faculdade de Medicina Veterinária e Zootecnia da Universidade de São Paulo, Pirassununga, São Paulo, were screened by the caudal fold test (CFf) and divided into Group A, which received intramuscular dead M.avium in oil injection; Group B, which received intramuscular dead M.bovis in oil injection and Group Control, which received only an intramuscular oil injection.
Inoculum: The inoculum was prepared by the Laboratório de Referência Animal, of the Ministério da Agricultura e Reforma Agrária - LARA-MAARA, Pedro Leopoldo, Minas Gerais, as a M. avium strain D4 or M. bovis strain AN5 according to the CENTRO PANAMERlCANO DE ZOONOSIS (1980). Adult anirnals received 1.0 mL intramuscular injection and young anirnals received 0.5 mL as recommended by LARA-MAARA.
Tuberculin Tests: The cervical comparative test (CCT) was performed 70 days after animal sensitization, injecting 0.1 mL of bovine PPD and 0.1 mL of avian PPD intradermally in the middle thirds of the neck at sites 10 cm apart. The skinfold thickness was recorded at the injection sites at 24, 48 and 72 hours later using a Mitutoyoâ digital paquimeter. After 60 days, the caudal fold test (CFT) was carried out injecting 0.1 mL of bovine PPD intradermally at the left side of the caudal fold approximately 10 cm distal of the base of the tail and the reading was performed by a Mitutoyoâ digital paquimeter at 72 hours later. The bovine PPD concentration were 1.0 mg/mL of protein and avian PPD with 2,500 I.U./0.1 mL produced by the Instituto de Tecnologia do Paraná - TECPAR.
Skin samples: The histological evaluation of the tuberculin reaction was carried out by 2.0 mm skin biopsies performed at 7? hours after PPD inoculation.
Histological Techniques: The skin biopsies were fixed in 10% buffered formalin for at least 48 hours, and paraffin blocks were sectioned and stained with haematoxylin and eosin - H.E. (FERNANDES, 1943).
Statistic Tests: The nonparametric Mann-Whitney test to unpaired samples was perforÍned to evaluate the CFT results and the Wilcoxon matched pairs test to compare CFT and CCT results (VIEIRA, 1942; SIEGEL, 1975).
Experiment II - Tuberculin test in buffaloes naturally sensitized to M.bovis and other environmental mycobacterias
Animais: 43 water buffaloes (Bubalus bubalis) young or adults of different breedings from CEBAEMBRAPA1, Santarém, in the state of Pará were screened by caudal fold test to compose the Group R, with 19 reagent animals and Group N, with 24 nonreagent animais. The CEBA-EMBRAPA buffalo herd had a history of M. bovis infection. The cervical comparative test was carried out 4 days after the CFT to prevent desensitization (ROSWURM & KONYHA, 1973). 40 buffaloes were slaughtered and lymph nodes were collected for histological and bacteriological examination.
Tuberculin Test: The CFf was carried out injecting 0.1 mL of bovine PPD intradermally at the left side of the caudal fold approximately 10 cm distal of the base of the tail. The increase of skin thickness was recorded by a digital paquimeter at 72 hours later and the reaction was classified as positive when there was any detectable increase. At 72 hours after CFf application, the CCT was carried out injecting 0.1 mL of bovine PPD and 0.1 mL of avian PPD intradermally in the middle third of the neck skin at sites 10 cm apart. The skin-fold thickness was recorded at the sites of injection 72 hours later using a Mitutoyoâ digital paquimeter and the animal was considered negative if the increase of skin thickness was less than 1.9 mm, suspect if between 2.0 and 2.9 mm or positive if was greater than 3.0 mm, according to the interpretation protocol of the Brazilian Animal Health authorities for the tuberculosis diagnosis ofbovines.
Skin biopsy and lymph node samples: There were collected cervical and caudal fold skin biopsies from 6 animais of both groups at 72 hours from the inoculation. There were collected at least 3 lymph nodes from each slaughtered animal (retropharyngeal, mediastinal, mesenteric, prescapular and superficial inguinal) for histological and bacteriological examination. Three animals that were sold to another farm could not be examined at the slaughterhouse.
Histological Techniques: The skin biopsies were fixed in 10% buffered formalin for at least 48 hours, and paraffin blocks were sectioned and stained with haematoxylin and eosin - H.E. (FERNANDES, 1943).
Bacteriological Techniques: The lymph nades were decontaminated by Petroff's technique using 5% oxalic acid and inoculated onto Petragnani and Stonebrink media with CO2• The slopes were exarnined weekly for macroscopic growth until 60 days of incubation at 37 ºC and 22 ºC. Suspect M. bovis isolates were inoculated in guinea pigs by intramuscular route (KANTOR, 1979).
Molecular Biological Analysis: The polymerase chain reaction (PCR) technique performed for the differentiation of Mycobacterium tuberculosis complex was carried out by the Departamento de Ciências Biológicas da Faculdade de Ciência Farmacêuticas/UNESP, Campus de Araraquara, using an IS986-based DNA probe (HERMANS et al., 1990).
RESULTS
Experiment 1 - tuberculin test in experimentally sensitized buffalo and bovines
The Group B responses were greater than Group A and Control at CFT in buffalo. The Mann-Whitney test considered significant (p=0.0156 and p=0.0039, respectively) although not significant between Group A and Control (p=0.8750). The buffalo Group B response at the CFT was greater than at the CCT signed by Wilcoxon test (p=0.042), but in the Group A and Control responses were not significant (Table 1). The mean reaction at CFT was 1.80 times greater than at CCT.
The responses to the avian and bovine PPD were both specific in bovines as well as in buffaloes at CCT in M. avium and M. bovis sensitized animals respectively (p<0.05), as showed in Table 2 and 3.
In the Group B, buffaloes responses at the CCT were greater than in bovines (p=0.0165).
There was no significant difference in the CCT response between buffaloes and bovines from Group A and Control (p=0.0694 and p=0.2188, respectively). The tuberculin reactions reached their maximum size at 72 hours of the experiment observation time (Graphic 1).
The Graphic 2 shows the results distribution for both animal species, divided into three classes: nonspecific (Group Control), para-specific (PPD reaction against heterogeneous mycobacterial infection), and specific (PPD reaction against homologous mycobacteria infection). In buffaloes, despite the occurrence of nonspecific and paraspecific superposition of values, the specific reactions were distinct (mean 12.24 and 13.90 mm). In the sarne case, the distribution of bovine specific reactions were less dispersed (mean 7.42 and 8.20 mm).
Bovine PPD reaction in water buffaloes sensitized with M. avium (Group A), M. bovis (Group B) and Control (not sensitized) at caudal-fold and cervical skin.
There were no significant histological differences in the tuberculin skin reaction between buffaloes and bovines, as shown in the Figure 1. In both animal species, tuberculin reaction was a conglomeration of cells of the mononuclear phagocyte series, sometimes with oedema and necrosis.
Experiment II - Tuberculin test in buffaloes naturally sensitized to M. bovis and other environmental mycobacterias
Two buffaloes from Group N (5158 and 5233) presented reactions of 3.36 mm and 4.08 mm, respectively, and two from Group R (3382 and 4675) showed 3.16 mm and 3.23 mm, respectively, for the avian PPD (Table 4 and 5). Three of these animals were slaughtered and only one of them (4675) presented granulomatous lesion in the lymph nodes and udder and isolation of mycobacteria (Fig. 2).
Two buffaloes from Group R were considered as reagent at CFT recording 4.75 mm and 10.12 mm of swelling (4557 and 4592 respectively), but only one of them, a 4-year-old cow (4592), was positive in both tests and showed caseous lesions in the retropharyngeal, mediastinal, mesenteric, prescapular and superficial inguinal lymph nodes and also in the udder (Fig. 2). The histological examinations showed granulomas with necrosis and calcification without Langhans giant cells (Fig. 3).
Cervical comparative test in water buffaloes sensitized with M. avium (Group A), M. bovis (Group B) and Control (not sensitized), recording at 24, 48 and 72 hours after injection.
Mycobacterias were isolated from 3 animals in Group R (4452, 4675 and 5031) and 5 in Group N (5138, 5177, 5185, 5192 and 5478). From the 10 isolates, 4 were identified as Mycobacterium tuberculosis complex by PCR technique (2 isolates from 5478, one from 4592 and one from 5031). In the histological examination these three animals also presented granulomas in the lymph nodes. Only animal 4592 was reagent in the CCT.
The histopathological examination of the lymph nodes revealed 3 positive animals from Group R (16.66%) and 2 from Group N (9.09%).
Only 2 animals from Group R and 1 from Group N revealed mycobacteria and caseous in the sarne lymph node (4592, 4675 and 5031).
Cervical comparative test in cattle sensitized with M. avium (Group A), M. bovis (Group B) and Control (not sensitized), recording at 24, 48 and 72 hours after injection.
There was accordance between the results of CFT, CCT, histological and bacteriological examination in 2 animals from Group R (4592 and 4675).
There was no difference in the environmental mycobacteria isolates number from both groups.
Mycobacterias were isolated mainly from the retropharyngeal lymph nodes (15.78%). Caseous lesions were observed mainly in the prescapular lymph node (10.52%).
DISCUSSION
The avian and bovine PPD reactions were specific in M. avium and M. bovis experimentally sensitized animals respectively, in buffaloes as well as in bovines this technique was able to identify specific, nonspecific and para-specific reactions.
VANAMAYYA et al. (1987), analysing tuberculin and johnin responses, observed that buffaloes were different from bovines and do not develop cross reactions with both reagents, it seems that buffaloes react specifically to the mycobacterias.
At histopathological examinations, the tuberculin reactions were similar in both animal species, although the Brazilian protocol of bovine tuberculin test interpretation seemed to fail when applied to buffalo.
Although the tuberculin tests present 72-96% sensitivity and 70-99% specificity in bovines, it is not the sarne in buffaloes (VANAMAYYA et al., 1987; NAIN & KAUSHIK, 1985; ORGANIZACION Panamericana de la Salud, 1992). The sensitivity and specificity of tuberculin test can be changed by the tuberculin injected dose or concentration and the test cut off (FRANCIS et al., 1978; KANTOR et al., 1984; ORGANIZACION Panamericana de la Salud, 1992).
The Ministério da Agricultura e Reforma Agrária of Brazil states that bovines should be considered positive at CFT when any swelling is observed at 72 hours of PPD injection. In this study, however, the paquimeter record revealed positive reactions in buffaloes greater than 10.0 mm, suspect between 6.40 and 9.9 mm and negative when smaller than 6.30 mm. In fact, the results presented in Table 1 recorded O to 6.47 mm for nonspecific reactions (mean 2.53 mm), O to 9.78 mm for para-specific reactions (mean 4.36 mm) and 6.30 to 43.88 mm for specific reactions (mean 25.09 mm). The mean reaction at CFT was 1.80 times greater than at CCT.
The mean reaction at CCT was 13.07 mm in buffaloes and 7.81 mm in bovines. So, considering the mean reaction, the relation is 1.69 times greater for buffaloes. Conceming this relation, the buffaloes parameter can be considered positive for tuberculin reactions greater than 5.0 mm, suspect between 3.0 and 4.9 mm and negative smaller than 2.9 mm. By the way, only one buffalo (4592) in Experiment II can be considered positive in both CFT and CCT, with agreement in the necroscopic, histological and bacteriological examinations.
Tuberculous bacilli were isolated from one buffalo which was negative at CFT and CCT (5478). This animal was young (1.2 years old) and might be have just been infected and had no time to develop tuberculous lesions and tuberculin response.
The buffalo M. tuberculosis complex isolates grew very slowly, dysgonic as M. bovis, tiny and presented only one or t"'.o colonies in each Stonebrink tube. These isolates wqe-less virulent for guinea pigs 60 days after inoculation and did not grow in subculture procedures as had been previously reported by MOHAN (1968).
CONCLUSIONS
1 - Water buffaloes presented tuberculin reactions greater than bovines.
2 - Buffaloes experimentally sensitized to M. bovis more strongly reacted at CFT and CCT than M. avium inoculated or not mycobacteria inoculated ones.
3 - For buffaloes experimentally sensitized to M. bovis, the mean caudal-fold reaction was 1.80 times greater than the cervical tuberculin reaction.
4 - The bovine PPD reaction in buffaloes was 1.69 times greater than in bovines at CCT.
The natural infection with environmental mycobacterias in buffaloes does not affect the results of tuberculin reaction at CFT and CCT.
Caudal-fold test and cervical comparative test in water buffaloes naturally sensitized to M. bovis and other environmental mycobacterias.
Results of histopathological and ba teriological examination_of lymph nodes from slaughtered water buffaloes.
Mean skin-fold thickness of the neck at cervical comparative test in water buffalo and bovine.
Distribution of skin-fold thickness of the neck at cervical comparative test in experimentally sensitized water buffaloes and bovines.
Vertical section of the skin of the neck showing tuberculin reaction in A) water buffalo and B) bovine. Stained with haematoxylin and eosin. Magnification X 125.
ACKNOWLEDGMENTS
Thanks are due to the technical staff of the Instituto Biológico, Faculdade de Medicina Veterinária da USP, EMBRAPA/Pará and Matadouro Municipal de Santarém
REFERENCES
- CENTRO PANAMERICANO DE ZOONOSIS. Preparación y estandarización de tuberculinas PPD. OPAS/OMS. Nota técnica, n.17, v.l, 1980. 44p.
- COCKRILL, W.R. Present and future of buffalo production in the world. Past and present. In: WORLD BUFFALO CONGRESS, 4., 1994, São Paulo. Proceedings. p.3-21.
- DESHPANDE, B.R.; SARDESHPANDE, P.D.; SANE, C.R.; PUROHIT, B.L. Tuberculous endometritis in a buffalo cow. Indian Vet.J., v.43, n.4, p.288-290, 1966.
- DOBBELAER, R.; ·O'REILLY, L.M.; GENICOT, A.; HAAGSMA, J. The potency ofbovine PPD tuberculins in guinea-pigs and in tuberculous cattle. J.Biol.Standard., v.11, p.213-20, 1983.
- FERNANDES, M.C. Métodos escolhidos de técnica microscópica. Rio de Janeiro: Imprensa Nacional. 1943. 611p.
- FRANCIS, J.; SEILER, R.J.; WILKIE, I.W.; O'BOYLE, D.; LUMSDEN, M.J.; FROST, A.J. The sensitivity and specificity of various tuberculin test using bovine PPD and other tuberculins. Vet. Rec., v.4, p.420-425, 1978.
- HAAGSMA, J. Bovine tuberculosis. Off. Intern. Épizooties. Manual Amendment 2, 1995. 11 p.
- HERMANS, P.W.M.; SOOLINGEN, D.; DALE, J.W.; SCHUITEMA, A.R.J.; McADAM, R.A.; CATTY, D.; EMBDEN, J.D.A. Insertion element IS986 from Mycobacterium tuberculosis: a useful tool for diagnosis and epidemiology of tuberculosis. J.Clin.Microb., v.28, n.9, p.2051-2058, 1990.
- KANTOR, I.N. Bacteriologia de la tuberculosis humana y animal. OPAS/OMS. Nota técnica n.11, 1979. 63p.
- KANTOR, LN.; ODEÓN, A.C.; STEFFAN, P.E.; AZA, M.J.; MADRID, C.R.; MARCHEVSKY, N. Sensitivity of the cervical and the caudal fold tuberculin tests with Mycobacterium bovis in infected cattle of Argentina. Rev.Sci.Tech:Off. Intern. Épizooties, v.3, n.1, p.137-150, 1984.
- KANTOR, I.N. & RITACCO, V. Bovine tuberculosis in Latin America and Caribbean: current status, control and eradication programs. Vet.Microbiol., v.40, n.1/2, p.5-14, 1994.
- KLEEBERG, H.K. Tuberculosis humana de origen bovino y salud pública. Rev.Sci.Tech. Off. lntern. Épizooties, v.3, n.1, p.55-76, 1984
- LÁU, H.D. Important economic diseases in buffaloes. In: WORLD BUFFALO CONGRESS, 4., 1994, São Paulo. Proceedings. p.209-220.
- MANDAL, P.C. & IYER, K.R. Tuberculous mastitis in a buffalo. Indian J.Anim.Scie., v.41, n.3, p.148-150, 1971.
- MANDAL, P.C. & SINGH, B. Tuberculous metritis in a buffalo (Bos bubalus). Indian J.Anim.Hlth, p.121-122, 1975.
- MOHAN, R.N. Diseases and parasites of buffaloes. Vet. Buli., v.38, n.10, p.647-659, 1968.
- NAIN, S.P.S. & KAUSHIK, R.K. Examination of tuberculin-positive animais. Indian J.Anim.Scie., v.54, n.10, p.877-878, 1985.
- NEILL, S.D.; POLLOCK, J.M.; BRYSON, D.B.; HANNA, J. Pathogenesis of Mycobacterium bovis infection in cattle. Vet.Microbiol., v.40, n.1/2, p.41-52, 1994.
- NICOLETTI, P. An update on selected diseases which affect buffaloes. In: WORLD BUFFALO CONGRESS, 4., 1994, São Paulo. Proceedings. p.221-228.
- O'REILLY, L.M. & DABORN, C.J. The epidemiology of Mycobacterium bovis infections in animais and man: a review. Tub.Lung Disease, v.76 , p.1-46, 1995. Suplemento 1.
- ORGANIZACIÓN PANAMERICANA DE LA SALUD. Plan de acción para la erradicación de la tuberculosis bovina en las Americas. OPAS/OMS. HPV/TB. n.113, 1992. 41p.
- PORTUGAL, M.A.S.C.; GIORGI, W.; SIQUEIRA, P.A. Ocorrência de tuberculose em bubalinos (Bubalus bubalis var. bubalis Linneus, 1758) no Estado de São Paulo. Arq.Inst.Biol., S.Paulo, v.4, n.38, p.231-238, 1971.
- RATLEDGE, C. & STANFORD, J. The biology of the mycobacteria. London: Academic Press. 1983, v.2, 554p.
- ROSWURM, J.D. & KONYA, L.D. The comparative-cervical tuberculin testas an aid to diagnosing bovine tuberculosis. In: ANNUAL MEETING OF UNITED STATES ANIMAL HEALTH ASSOCIATION, 77., 1973, Saint Louis. Proceedings. p.368-389.
- SAUBOLLE, M.A. Nontuberculous mycobacteria as agents of human disease in the United States. Clin.Microb. Newsletter, v.11, n.15, p.113-117, 1989.
- SIEGEL, S. Estatística não-paramétrica para as ciências do comportamento. São Paulo: McGraw-Hill. 1975. 350p.
- VACCARO, A.; CAPUANO, G.; DAMIANO, N. Ricerche sulla prevalenza della tuberculosi nell'allevamento bufalino della province di caserta e salermo. In: CONGRESO INTERNATIOij L SULL' ALLEMAMENTO BUFALINO NÉL MONDO. Atti del Convegno, 2., 1982, Caserta, p.191.
- VALE, W.G. Water buffalo world uptake. Prospects of buffalo production in Latin America. In: WORLD BUFFALO CONGRESS, 4., 1994, São Paulo. Proceedings. p.75-87.
- VANNAMAYYA, P.R.; SHARMA, A.K.; PARAI, T.P.; PALIWAL, O.P.; PARIHAR, N.S. Evaluation of tuberculin and johnin tests with pathological lesions in buffaloes. Indian J.Anim.Scie., v.3, n.57, p.189-190, 1987.
- VIEIRA, S. Introdução à bioestatística. Rio de Janeiro: Campus. 1942. 293p.
- WORLD HELTH ORGANIZATION Report of the who meeting on zoonotic tuberculosis (Mycobacterium bovis). WHO/CDS/VPH. Monograph Series n.130, 1993. 27p.
