In a recent study published in JAMA Network Open, researchers compared the performance of tuberculin skin test (TST) and interferon-γ release assays (IGRAs) in predicting tuberculosis (TB).
Study: Comparison of Tuberculin Skin Testing and Interferon-γ Release Assays in Predicting Tuberculosis Disease. Image Credit: BLACKDAY/Shutterstock.com
Background
In the United States (US), most TB cases are among asymptomatic subjects with TB infection (TBI), i.e., latent TBI, that could later progress to an active TB.
While studies have shown the effectiveness of TBI treatments in preventing progression to TB, optimal TBI tests are required to identify those at risk. Two types of TB tests have been approved in the US: IGRAs (T-SPOT.TB [TSPOT] and QuantiFERON-TB Gold In-Tube [QFT-GIT]) and TST.
About the study
In the present study, researchers evaluated the ability of the TST and two IGRAs to predict the progression to TB.
Individuals were eligible if they had close contact with a person with TB, were born in, visitors to, or immigrants from a country with high/medium TB incidence, lived with human immunodeficiency virus (HIV), or were members of a local population with TBI prevalence ≥ 25%.
Participants were recruited from 18 TB clinics between July 12, 2012, and May 5, 2017. Excluded subjects were foster children, TBI or TB treatment recipients, and those with anaphylaxis to tuberculin.
Subjects were tested with three tests at the start; they also self-reported their medical history, vaccination status, race/ethnicity, and epidemiologic and demographic risk factors.
All subjects were followed up till November 15, 2020, to examine the development of TB. Possible TB cases were identified based on clinical signs, symptoms, or test results and were verified by a medical review.
Some individuals detected with TB ≤ 30 days post-enrolment were categorized as cases with prevalent TB and were included in sensitivity analyses only.
Fisher exact and chi-squared tests were used to compare demographic and epidemiologic data and medical history between individuals with and without TB. For incident TB cases, the time to TB was calculated.
Sensitivity, specificity, and positive (PPVs) and negative (NPVs) predictive values were estimated. Pairwise comparisons of each test were performed, and PPV ratios were computed.
Several sensitivity analyses were performed by varying the positive test result cutoffs or restricting cases to those with pulmonary TB/culture-confirmed results.
Additionally, the incremental gain of a second test result in predicting TB was calculated. Further, among subjects with ≥ one positive test, the effect of TBI treatment on progression to TB was evaluated.
Findings
Overall, the researchers enrolled 22,020 individuals. Most individuals (82%) were born outside the US. Over 51% were male, 9.6% were close contacts, 11.9% were Hispanic, 9.5% were White, 29.8% were Asian, and 20.8% were Black.
TB was detected in 129 individuals; 87 were prevalent TB cases. Among incident TB cases, 32 had culture-confirmed results, 19 had a smear-positive result, and five had pulmonary TB.
Incident TB was more prevalent among individuals aged < 2 or > 65, close contacts, and those with diabetes or chronic kidney failure.
Participants were followed up for a median of 6.4 years. The median time to TB detection was 2.7 years. QFT-GIT and TSPOT significantly outperformed TST in pairwise comparisons. There were no differences in performance between IGRAs. Sensitivity analyses produced consistent results.
The incremental gain in PPV for positive TSPOT and QFT-GIT results was significant, given a positive TST.
The team found that subjects with ≥ one positive test result were eight times more likely to progress to TB than those without a positive test. Further, treatment completion among those with a positive test was significantly protective in preventing incident TB compared to untreated subjects.
Conclusions
In sum, IGRAs performed superior to the TST in predicting incident TB. These findings were consistent in sensitivity analyses. The study results support using IGRAs for TB prediction even when TST results exist.
Notably, QFT-Plus has replaced QFT-GIT and is currently in use; however, it was unavailable throughout the study. Nonetheless, QFT-GIT is comparable to QFT-Plus.
Furthermore, the findings may not be generalizable as the study was conducted in at-risk individuals.
WHO reports alarming rise in tuberculosis cases in 2023