at Severance Hospital. The controls with the positive or negative results were selected by matching 1:1 with the case for date of QFT-G IT assay. An indeterminate result was considered when the level in both the tuberculosis antigen and mitogen tube was less than 0.35 IU/ml, or the level in the nil tube was higher than 8.0 IU/ml. Through retrospective review of medical records, age, gender, total lymphocyte count at the QFT-G IT assay, underlying diseases, and medical conditions which can influence the immune function were collected as covariates. The predictive factors for the indeterminate results on QFT-G IT were assessed by multivariate logistic regression models.Results: Age and gender did not show the significant differences between case and control patients (p = 0.105 and 0.906). The percentage of patients with hematologic malignancy (19.4% vs. 11.1%, p = 0.049), chronic renal disease (22.2% vs. 9.7%, p = 0.004), and autoimmune disease (16.0% vs. 6.3%, p = 0.009) in the case group were the significantly higher than those in the control group. Also, the case group had the significantly higher percentage of steroid use (16.0% vs. 8.3%, p = 0.047) and solid organ transplantation (11.1% vs. 2.1%, p = 0.002) than the control group. The total lymphocyte count was the significantly lower in the case group than in the control group (median [interquartile range], 540 [220-1,030] vs. 1,370 [870-1,970], p < 0.001). In the final regression model, clinical factors associated with the indeterminate results were chronic renal disease (odds ratio [OR] 2.26, 95% confidential interval [CI] 1.03-4.94, p = 0.041), autoimmune disease (OR 3.99, 95% CI 1.65-9.66, p = 0.002), and solid organ transplantation (OR 5.06, 95% CI 1.30-19.73, p = 0.020).