Background & Objectives: Cellular resistance to tuberculosis (TB) infection depends on signals from mycobacterium tuberculosis-specific T lymphocytes activating mycobacterial killing mechanisms in infected macrophages. The network of soluble factors, or cytokines, responsible for these cellular communications has been progressively unraveled over the last 3 decades; however, new regulatory and effectors cytokines continue to be discovered. Infection of a host with a pathogen first result in activation of cells of the innate immune response including effectors molecules including cytokines such as TNF-a & IFN-g. The aim of this study was to investigate the frequency of TNF-a , IFN-g alleles, evaluating serum concentration of TNF-a and IFN-g and relationship of between susceptibility to TB and TNF-a and IFN- g gene variations. Material & Methods: In this prospective case-control study, 93 patients with smear positive tuberculosis selected from Masih Daneshvari Hospital, Tehran, Iran. They were matched with 103 controls without any history TB. Genotype of 5 regions of TNF-a and 1 region of IFN-g were distinguished by PCR-RFLP method, and level of serum concentration between case & control groups were evaluated by ELISA method. Data were analyzed with Mann-Whitney U & earning the cut off analyzed with ROC curve for ELISA method and the results of PCR-RFLP method were analyzed by SPSS, Fisher exact and X2. Results: In PCR-RFLP method, the results showed a significant difference at TNF-308 and TNF-857 between two groups of control and patient (P<0.05). In ELISA method, a significant difference in IFN-g was observed between the groups of control and patient (P<0.05). Also a cut off point as a serologic marker between the positive and negative states, for rapid TB examination about IFN-g was found; the cut off for IFN-g was 0.19.Conclusion: Mutation in TNF-308 and TNF-857 regions were identified significantly, but it was not observed in other TNF-a and IFN-g regions. In this study, ELISA method suggest that mycobacterial pathogens are frequently associated with production of cytokines such as IFN-g and serologic tests considering the cut off patients, help us to detect TB in cases that we want to earn results rapidly.