Units

Keywords

Link

Contact

---

Citation

Type

Journal

Publication Date

Pages

Publisher

Brief description/objective

Interferon-gamma release assays (IGRAs) are promising candidates
to replace the tuberculin skin test (TST) in screening for latent tuberculosis infection (LTBI). However, there are some limitations: IGRAs are, like the TST, only an indirect marker for Mycobacterium tuberculosis (MTB) infection and sensitivity for diagnosing active TB is sub-optimal, with false-negative results. This article gives an overview of the performance of the two commercially available IGRAs in adult HIV-positive individuals. We will focus on sensitivity and indeterminate rates among HIV-positive patients and the potential influence of CD4+ cell count. Among the studies published using commercial IGRAs, we found that the sensitivity of IGRAs in HIV co-infected individuals with or without an indeterminate result is impaired compared with HIV-negative patients with TB. Indeterminate rates are higher in HIV-positive individuals and the number of CD4+ cells is significantly associated with the rate of indeterminate results, whereas the effect of CD4+ cell numbers on actual sensitivity is less pronounced. From our findings, both IGRAs are likely to give very high rates of indeterminate and probably unreliable results when CD4+ cell count is <100 cells/μl. However, the tests seem to give reliable results at a CD4+ cell count >200 cells/μl, but this needs to be confirmed in studies including a sufficient number of HIV-positive individuals with low CD4+ cell numbers to allow meaningful statistical calculations. IGRAs have a high negative predictive value in a low endemic but not in a high
endemic setting. The high proportion of indeterminate results underlines the importance of positive control as a marker of anergy, helping to exclude false-negative results, which is not an option for the TST. The positivity rate increased by 10% when indeterminate results were excluded and therefore we strongly recommend full openness on the proportion and use of indeterminate results when reporting the performance and accuracy of IGRA.