SETTING: The applicability of serodiagnosis of tuberculosis using Mycobacterium tuberculosis-complex-specific antigens in a Tanzanian population with high prevalence of HIV. OBJECTIVE: This study was performed to evaluate the usefulness, sensitivity and specificity of serology using M. tuberculosis-specific antigens in the diagnosis of tuberculosis in patients with and without HIV co-infection. DESIGN: Patients with proven pulmonary and extrapulmonary tuberculosis at a major referral centre in Tanzania were enrolled in the study. The control group consisted of patients without a history of previous tuberculosis admitted to the trauma ward and of healthy volunteers. Sera were analysed by an enzyme linked immunoassay (ELISA) using two M. tuberculosis specific proteins as antigen: the 38 kDa protein [3T] and a 17 kDa protein. In addition was recorded presence or absence of BCG scar and tuberculin sensitivity and the sera were tested for HIV and analysed for beta-2-microglobulin content. RESULT: Sensitivity and specificity were markedly reduced in tuberculosis patients with HIV co-infection compared to patients without this disease (73% and 70% versus 52% and 50% respectively). CONCLUSION: Serology for diagnosis of tuberculosis is not feasible in an HIV endemic region.
SETTING: The applicability of serodiagnosis of tuberculosis using Mycobacterium tuberculosis-complex-specific antigens in a Tanzanian population with high prevalence of HIV. OBJECTIVE: This study was performed to evaluate the usefulness, sensitivity and specificity of serology using M. tuberculosis-specific antigens in the diagnosis of tuberculosis in patients with and without HIV co-infection. DESIGN:Patients with proven pulmonary and extrapulmonary tuberculosis at a major referral centre in Tanzania were enrolled in the study. The control group consisted of patients without a history of previous tuberculosis admitted to the trauma ward and of healthy volunteers. Sera were analysed by an enzyme linked immunoassay (ELISA) using two M. tuberculosis specific proteins as antigen: the 38 kDa protein [3T] and a 17 kDa protein. In addition was recorded presence or absence of BCG scar and tuberculin sensitivity and the sera were tested for HIV and analysed for beta-2-microglobulin content. RESULT: Sensitivity and specificity were markedly reduced in tuberculosis patients with HIV co-infection compared to patients without this disease (73% and 70% versus 52% and 50% respectively). CONCLUSION: Serology for diagnosis of tuberculosis is not feasible in an HIV endemic region.
Authors: R J Wilkinson; K Hasløv; R Rappuoli; F Giovannoni; P R Narayanan; C R Desai; H M Vordermeier; J Paulsen; G Pasvol; J Ivanyi; M Singh Journal: J Clin Microbiol Date: 1997-03 Impact factor: 5.948
Authors: Krishna K Singh; Yuxin Dong; Sai A Patibandla; David N McMurray; Vijay K Arora; Suman Laal Journal: Infect Immun Date: 2005-08 Impact factor: 3.441