Abstract: Objective  To evaluate the value of T-SPOT.TB assay in the diagnosis of tuberculosis (TB). Methods  Five hundred and eighty-seven cases of hospitalized patients (128 TB cases including 103 cases with active TB and 25 cases with old TB; 459 non-TB cases including 241 cases with immune impairment and 218 cases without immune damage) from 2011 July to 2012 April in Affiliated Hospital of Inner Mongolia Medical University were analyzed retrospectively in their age, gender, course of disease, clinical manifestations, the results of T-SPOT.TB assay, pathological examination, purified protein derivative (PPD) skin test, acid-fast bacilli smear and anti-TB antibody. The data were analyzed with software SPSS 16.0. The measurement data were compared with Wilcoxon rank test, and the count data were analyzed using chi square test. P<0.05 was considered as significant difference.  Results  Of 187 T-SPOT.TB-positive cases, 106 cases were diagnosed as TB, the positive predictive value was 56.68% (106/187). Of 400 T-SPOT.TB-negative cases, 378 cases were diagnosed as non-TB diseases, the negative predictive value was 94.50% (378/400). Of 587 cases, 128 cases were diagnosed as TB, the sensitivity of T-SPOT.TB assay was 82.81% (106/128), which was higher than that of PPD (35.71%, 30/84), acid-fast bacilli smear (8.74%, 9/103) and anti-TB antibody (14.06%, 9/64). The sensitivities of T-SPOT.TB in active TB and old TB were 93.20% (96/103) and 40.00% (10/25), respectively. Of 459 cases with non-TB diseases, the specificities of T-SPOT.TB in immune impairment group and non immune damage group were 71.37% (172/241) and 94.50% (206/218). The sensitivity of T-SPOT.TB in tuberculous pleurisy and peritonitis was 100.00% (37/37). The median count of T cell spots in active TB and old TB were 502/10⁶ PBMCs and 430/10⁶ PBMCs, respectively, four quartile range (P₂₅, P₇₅) were 217/10⁶ PBMCs, 1287/10⁶ PBMCs and 140/10⁶ PBMCs, 1303/10⁶ PBMCs, respectively. There were no significant difference between the 2 groups using the Wilcoxon rank test (U=429.5, P=0.585). The median count of T cell spots in pulmonary TB and extrapulmonary TB were 456/10⁶ PBMCs and 528/10⁶ PBMCs, respectively, four quartile range (P₂₅, P₇₅) were 264/10⁶ PBMCs, 950/10⁶ PBMCs and 186/10⁶ PBMCs, 1244/10⁶ PBMCs, respectively. There were no significant difference between the 2 groups (U=1083.0, P=0.871).  Conclusion  T-SPOT.TB assay has higher specificity in the low-risk population, while lower specifi-city in the high-risk population with latent tuberculosis infection, higher sensitivity in the patients with tuberculous effusion of serous cavity compared with PPD, acid-fast bacilli smear and antibody. But it cannot identify active TB with inactive TB.

Key words: Tuberculosis/diagnosis, Interferon-gamma release tests