Evaluation of a rapid immunochromatographic test for the serologic diagnosis of tuberculosis in Italy

Abstract

To determine specificity, sensitivity and predictive values of a rapid immunochromatographic assay (ICT tuberculosis) for the diagnosis of tuberculosis (TB) in an Italian clinical setting, and to identify tentative new guidance for the interpretation of test results. The ICT tuberculosis test is an immunochromatographic test based on the detection of IgG antibodies directed against five highly purified antigens secreted by Mycobacterium tuberculosis during active growth. Sera from 60 patients with active pulmonary (48 sputum smear-positive and six sputum smear-negative cases) and extrapulmonary (six cases) TB were obtained. Personal, anamnestic and clinical data were investigated and recorded for each patient. The control groups comprised 156 subjects: 40 healthy individuals, half of them Mycobacterium bovis BCG-vaccinated, and 116 patients with mycobacterial diseases other than TB (five cases), with nonmycobacterial lung diseases (30 cases), with nonmycobacterial nonlung diseases (30 cases), with nonmycobacterial diseases and rheumatoid factors positivity (30 cases), and with asymptomatic HIV infection (21 cases). For 21 individuals the test was simultaneously performed with both serum and whole blood sample. Each positive result of the ICT test was reported with regard to the number (1-4), position (A, B, C, D) and color intensity (+ to ++++) of the evidenced lines in order to assess the quality of the antibody response. The overall sensitivity and specificity were 56.7% and 90.4%, respectively. The sensitivity for pulmonary TB patients was 61.1% (66.7% for smear-positive and 16.7% for smear-negative cases) and 16.7% for extrapulmonary TB patients. The difference between ICT results in pulmonary TB patients and control subjects was statistically significant (P < 0.0001). The analysis of the positive ICT tests revealed that samples with strong color intensity (>/=++) and specific antibodies bound to antigens immobilized on line D were significantly more frequent in TB patients than in controls (P = 0.001 and P= 0.027, respectively). ICT test results with the presence of at least three visible lines were more often observed in the TB patients than in controls, although not reaching statistical significance (P = 0.052). No difference was observed between the results of the ICT test performed both on serum and whole blood sample. The ICT tuberculosis test was confirmed to be rapid and easy to perform without requiring special equipment, both on serum and whole blood sample. Our data, in accordance with those obtained in a previous study conducted in extra-European countries, confirmed higher sensitivities for the smear-positive TB patients than for the smear-negative TB patients, and for pulmonary TB patients than for the extrapulmonary TB patients. Data obtained on the quality of antibody response in the ICT positive samples, might be used to improve the performance of the test.