A 100 year update on diagnosis of tuberculosis infection

Abstract

Diagnosis and treatment of latent tuberculosis infection (LTBI) is a cornerstone of tuberculosis (TB) control in the developed world. In the last century, the tuberculin skin test (TST) was the only means of diagnosing LTBI. ELISpot and whole-blood ELISA, collectively known as interferon-gamma release assays (IGRAs), are promising new tools. IGRAs are more specific than TST for diagnosis of LTBI as they are not confounded by previous bacille Calmette-Guerin (BCG) vaccination. Assessing IGRA sensitivity in the absence of a gold standard for LTBI is challenging. Studies have therefore used surrogate markers such as active TB and correlation with degree of TB exposure in contact investigations. These studies suggest that sensitivity of ELISpot is higher than TST while whole-blood ELISA has similar sensitivity to TST. Recent longitudinal studies demonstrating the prognostic power of these tests for development of active TB provide definitive evidence that positive IGRA results reflect infection with dormant yet viable bacilli. Is the prognostic power of IGRAs greater than the TST? What are the false-negative rates in immunocompromised individuals with LTBI at high risk of progressing to active TB? IGRAs have been incorporated into national guidelines, although their optimal deployment in diagnostic algorithms is evolving. The health economic benefits of utilizing IGRAs are increasingly recognized, partly because their high specificity avoids unnecessary chemoprophylaxis in BCG-vaccinated persons with false-positive TST results. Current IGRAs are being improved and next-generation tests, with improved sensitivity, could enable the reliable exclusion of LTBI in immunocompromised individuals.