Um rebanho bovino naturalmente infectado por tuberculose foi analisado através de diferentes métodos diagnósticos. Um teste intradérmico simples (TIC) identificou 21 animais como positivos. Após 90 dias deste resultado, um teste intradérmico comparativo (TIC) foi aplicado nos 21 animais positivos ao TIS, além de outros 29 animais com resultados prévios negativos escolhidos aleatoriamente. De todos estes animais (50), foram coletadas amostras de leite e secreção nasal para isolamento e identificação de microrganismos por cultura e PCR; amostras de sangue de cada um dos animais foram coletadas para exames de ELISA: produção de Interferon-gama (IFN) e pesquisa de anticorpos frente aos antígenos MPB70 e MPB83. Tais amostras sanguíneas foram coletadas em três diferentes momentos: no dia da execução do TIC e nos dias dia 7 e dia 21 após a execução do TIC. Os animais que foram positivos a este teste foram abatidos; exames de identificação do agente, tais como cultivo e PCR foram realizados post-mortem para confirmação da doença. Baseado na análise Kappa, IFN apresentou resultados estatisticamente comparáveis aos resultados de isolamento e identificação bacteriana por cultura e PCR, além do TIC ao longo de todo o experimento. No entanto, TIC, ELISA e IFN não foram estatisticamente comparáveis. Tais resultados sugeriram que nenhum dos atuais métodos de diagnóstico para tuberculose possibilitou a identificação de todos os animais infectados. Por este motivo, uma estratégia mais abrangente deveria incluir métodos de diagnóstico que pudessem identificar a resposta imune celular e humoral, uma vez que animais de um mesmo rebanho poderiam se encontrar em diferentes estágios da infecção.

A herd infected naturally with tuberculosis was investigated by different diagnostic methods. Ninety days after a screening test that identified 21 cows as skin test positive, a Comparative Intradermal Tuberculin Test (CITT) was performed in those 21 cows and in 29 other randomly selected skin test negative cows. Milk samples and nasal swabs were collected prior to the CITT for bacteriological culture and PCR, while blood samples were collected for IFN release and antibody responses to MPB70 and MPB83, at three time points post tuberculin injection. Animals positive by CITT were slaughtered and disease confirmation undertaken. Based on the Kappa test, IFN was comparable to the standard tests (culture, PCR and CITT) at all three sampling points. Results from both antibody ELISAs were similar but were not comparable to the standard tests. T-test analysis of the CITT, IFN and ELISAs demonstrated that their performances were not correlated. There is increasing recognition that individually, available diagnostic tests do not detect all infected cattle. Therefore, a comprehensive strategy for the diagnosis of bovine TB should include test results for the detection of both cellular and humoral immune responses where there may be animals at different stages of infection.

In order to evaluate the predictive value of acid fast bacilii (AFB) smear for the diagnosis of Mycobacterium tuberculosis in respiratory specimens in a setting with a high prevalence of Aids and an unknown prevalence of nontuberculous mycobacteria (NTM), we retrospectively examined specimens cultured for mycobacteria between 1 September 1993 and 30 September 1994 and medical records of patients with positive culture in a General Hospital, Aids reference in Rio de Janeiro, Brazil. Seventy three per cent (1517/2077) of samples were respiratory specimens and mycobacteria were recovered from 20.6% (313/1517) of these. M. tuberculosis was identified in 94.2% (295/313) and NTM in 5.8% (18/313). The yield of positive AFB smear and of positive culture was 6.1% (93/1517) and 20.6% (313/1517), respectively. The positive predictive value (PPV) of AFB for M. tuberculosis was 98.4% in expectorated sputum and 96.4% in bronchoalveolar lavage. Forty four percent (130/295) of specimens with positive culture for M. tuberculosis and 66.7% (12/18) for NTM were from patients HIV positive. The conclusion was that in our study population, the PPV of AFB for M. tuberculosis in respiratory specimens was high and the prevalence of NTM was low despite the high prevalence of HIV positive.