Background: Effective diagnostic methods for detecting latent tuberculosis infection (LTBI) are important for its eradication. A number of studies have evaluated the use of interferon-g-induced protein 10 (IP-10), which is elevated after tuberculosis infection, as a biomarker for LTBI, but conclusive results regarding its effectiveness have not been reported. Objectives: Our objective was to assess the diagnostic value of IP-10 for LTBI. Data sources: We searched the PubMed, Embase, the Cochrane Library and Web of Science databases to find eligible studies. Study eligibility criteria: We included cohort, caseecontrol and cross-sectional studies that evaluated IP-10 in LTBI participants in comparison with tuberculin skin tests (TST) and interferon-g release assays (IGRA). Participants: Individuals with LTBI and uninfected participants. Interventions: IP-10 (index test) compared with TST and IGRA (reference standard) for diagnosing LTBI. Methods: PubMed, Embase, the Cochrane Library, and Web of Science databases were searched up to June 2018. A hierarchical summary receiver operating characteristic (HSROC) model was used to evaluate the pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and HSROC curve for the diagnostic efficiency of IP-10. Results: Twelve studies including 1023 participants and 1122 samples were included. The overall pooled sensitivity was 0.85 (95% CI 0.80e0.88), specificity was 0.89 (95% CI 0.84e0.92), PLR was 7.55 (95% CI 5.20 e10.97), NLR was 0.17 (95% CI 0.13e0.22) and DOR was 44.23 (95% CI 28.86e67.79), indicating a high accuracy for diagnosing LTBI. Based on a meta-regression analysis, high-burden countries, study design, IP-10 method, reference standard and the IP-10 cut-off could not explain the heterogeneity (p >0.05). Conclusions: Our results suggested that IP-10 is a promising biomarker for the diagnosis of LTBI.