Objective and setting: We evaluated a rapid-format antibody card test and the tuberculin skin test for diagnosis of active tuberculosis (TB) in high (Cairo, Egypt) and low (St. Louis, USA) prevalence areas.
Design: Prospective study of hospitalized TB patients and controls with other chest diseases.
Results: Test performance varied significantly in the two study sites. The antibody test detected 87% of 71 smear-positive pulmonary TB cases (86% of smear-negative pulmonary cases and 48% of TB meningitis cases) in Egypt; specificity was 82%. The tuberculin test was highly sensitive in Egypt in subjects with pulmonary TB (100%) but not in those with meningitis (23%); specificity was 70%. The sensitivity and specificity of the antibody test in St. Louis were 29% and 79%, respectively; 50% of St. Louis TB cases and 15% of controls had positive tuberculin tests.
Conclusions: This convenient antibody card test may have value for diagnosis of patients suspected of having TB in high prevalence areas like Egypt. However, the specificity of the test is too low for it to be useful as a screening test. Our results suggest that neither the antibody test nor the tuberculin test have much diagnostic utility in low prevalence settings like St. Louis.