Negative effect of ferritin levels on T-SPOT.TB, an enzyme-linked immunospot based interferon-gamma release assay using Mycobacterium tuberculosis specific antigens, in the smear-positive active pulmonary tuberculosis.

Background: T-SPOT.TB (TSPOT), is a major diagnostic tests for Mycobacterium tuberculosis (MTB) infection due to excellent sensitivity. Though the sensitivity of the test is less than expected, factors related to false-negative of the assay have not been elucidated. Purpose: To evaluate sensitivity of TSPOT in microbiologically confirmed pulmonary tuberculosis (PTB) subjects, and identify factors related to false-negative (FN) of the assay. Method: From Oct 2013 to Sep 2016, TSPOT assays were conducted using newly diagnosed microbiologically confirmed PTB patient’s blood samples at the time of admission for isolation and treatment. Demographic and clinical characters were obtained from participants. Result: A total of 145 patient were enrolled. Twenty-five (17.2%) of 145 showed FN. One (0.7%) subject was invalid and eight (5.5%) were boarder line. The sensitivity of TSPOT was 76.6％ (111 cases). In comparison with positive group, FN showed significant higher ferritin (FRT) level (p = 0.028) and lower hemoglobin (p = 0.005), serum albumin (p = 0.06) and PC well spots (p = 0.043). Among them, FRT was most significantly associated with FN by multivariable logistic regression analysis (p = 0.0073). The utility of FRT to predict FN was assessed with a receiver-operating characteristic analysis. The maximum Youden9s index was at 321 ng/mL with a sensitivity of 63.2 % and a specificity of 70.5 %. Conclusions: TSPOT must be carefully interpreted in cases with high FRT. Further validation studies in other cohorts, and exploring underlying mechanisms is also required.