Evaluation of the characteristics of the enzyme-linked immunospot assay for diagnosis of active tuberculosis in China.

The purpose of this study was to evaluate the characteristics of the T-SPOT.TB test for the diagnosis of active tuberculosis (ATB) and to distinguish ATB from other diseases using a receiver operating characteristic (ROC) curve. A total of 535 patients with suspected active tuberculosis were enrolled in the study and divided into ATB and nonactive tuberculosis (NATB) groups, as well as pulmonary tuberculosis (PTB) and extrapulmonary tuberculosis (EPTB) subgroups. The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio of the T-SPOT.TB test for the diagnosis of ATB were 84.95%, 85.12%, 82.94%, 86.93%, 5.71, and 0.18, respectively. The median number of spot-forming cells (SFCs) in the ATB group was higher than that in the NATB group (71 versus 1; P < 0.0001). The sensitivities in the PTB and EPTB subgroups were 92.31% and 81.77%. The areas under the curve (AUC) for the diagnosis of ATB using the T-SPOT.TB, early secreted antigenic target 6 (ESAT-6), and culture filtrate protein 10 (CFP-10) were 0.906, 0.884, and 0.877, respectively. A cutoff of 42.5 SFCs for ATB yielded a positive predictive value of 100%. Our study shows that the T-SPOT.TB test is useful for the diagnosis of ATB. Utilizing an ROC curve to select an appropriate cutoff made it possible to discriminate ATB from NATB.