Histofarcin test for the diagnosis of epizootic lymphangitis in Ethiopia: development, optimisation, and validation in the field.

Histofarcin, a skin test antigen for the diagnosis of epizootic lymphangitis, was locally produced from the mycelial form of Histoplasma capsulatum var. farciminosum (HCF) in disease-endemic districts of Ethiopia and tested for its application in the field between April 2002 and May 2003. The test was evaluated using 108 mules, 84 in endemic and 24 in disease-free districts. Microscopic and mycological examinations of clinical lesions were used as the "gold standard" for the validation of the test. The concentration of histofarcin that caused an optimum reaction was 0.2-0.4 mg/mL in a 0.1 mL dose and this was attained 24-48 h post-injection. The sensitivity and specificity of the histofarcin test were 90.3% (CI = 73.1, 97.5) and 69% (95%, CI = 48.1, 84.9%) in disease-endemic districts. On the other hand, specificity was 100% (CI=94.8, 100) in disease-free districts. Positive and negative predictive values of the histofarcin test were 77.78% (95% CI = 60.4, 89.3) and 85.71% (95% CI = 62.6, 96.2), respectively. A marginal substantial agreement (kappa = 0.61, P = 0.0000) was observed between the clinical status and the result of the histofarcin test. A large proportion (31%) of 'false positives' was recorded in endemic districts, which could be due to the pre-clinical stage of the disease. The latter ended in lower specificity and positive predictive value of the test since the true infection status of a 'false positive' could not be known on the basis of clinical features. Therefore, standard test validation procedures including slaughtering and isolation of HCF is required. After proper validation, we conclude that the histofarcin test could play a significant role in detecting early infection, and differentiating of EL from glanders, ulcerative lymphangitis, and sporotrichosis.