BACKGROUND: The presence of specific antiglycolipid antibodies in serum and circulating immune complexes (CIC) in children with tuberculosis was detected in order to evaluate their contribution to the value of serodiagnosis of tuberculosis, as has already been shown in adults. METHODS: ELISAs using the three glycolipids LOS, DAT and PGLTb1 were performed in whole serum and immune complexes from 20 children with tuberculous disease or infection, in seven child contacts, and in 26 children with non-tuberculous disease. The contribution of complexed IgG antibody to the diagnostic values was established for each group. RESULTS: The antibody levels in free serum were higher (P < 0.01) in children with tuberculous disease or infection and in contacts than in controls. By contrast, except for PGLTb1, the IgG antibody levels were higher (P < 0.02) in children with tuberculous disease than in the other groups. The highest contribution of IgG antibody against LOS to the predictive values was shown in children with pulmonary tuberculosis (positive predictive value 1,000, negative predictive value 1,000). In paucibacillary tuberculosis (extra-pulmonary and tuberculous infection) and in contacts, the IgG antibody did not contribute to the sensitivity of the serodiagnosis, where the combination of antigens tested in serum increased the diagnostic yield. The very low levels of IgG antibody in these settings may indicate a different B cell response. CONCLUSION: The detection of immune complexes and IgG antibodies against the three glycolipid antigens is useful as a complementary technique for the serodiagnosis of children with a high probability of pulmonary tuberculosis.
BACKGROUND: The presence of specific antiglycolipid antibodies in serum and circulating immune complexes (CIC) in children with tuberculosis was detected in order to evaluate their contribution to the value of serodiagnosis of tuberculosis, as has already been shown in adults. METHODS: ELISAs using the three glycolipids LOS, DAT and PGLTb1 were performed in whole serum and immune complexes from 20 children with tuberculous disease or infection, in seven child contacts, and in 26 children with non-tuberculous disease. The contribution of complexed IgG antibody to the diagnostic values was established for each group. RESULTS: The antibody levels in free serum were higher (P < 0.01) in children with tuberculous disease or infection and in contacts than in controls. By contrast, except for PGLTb1, the IgG antibody levels were higher (P < 0.02) in children with tuberculous disease than in the other groups. The highest contribution of IgG antibody against LOS to the predictive values was shown in children with pulmonary tuberculosis (positive predictive value 1,000, negative predictive value 1,000). In paucibacillary tuberculosis (extra-pulmonary and tuberculous infection) and in contacts, the IgG antibody did not contribute to the sensitivity of the serodiagnosis, where the combination of antigens tested in serum increased the diagnostic yield. The very low levels of IgG antibody in these settings may indicate a different B cell response. CONCLUSION: The detection of immune complexes and IgG antibodies against the three glycolipid antigens is useful as a complementary technique for the serodiagnosis of children with a high probability of pulmonary tuberculosis.