© 2008 American Thoracic Society doi: 10.1164/rccm.200802-251ED
Predicting TuberculosisDoes the IGRA Tell the Tale?
Duke University Medical Center
McGill University The tuberculin skin test (TST) has been used to diagnose infection with Mycobacterium tuberculosis for close to 100 years. This longevity reflects the TST's low cost and ease of administration, as well as the numerous longitudinal studies correlating the size of the TST reaction with future risk of active tuberculosis (TB) (1). These studies demonstrated that persons with positive tests had increased risk of developing active TB in the future, whereas persons with negative tests had low risk. Furthermore, placebo-controlled trials have demonstrated that antituberculous drug treatment of subjects with positive TSTs reduced their risk of developing active TB (2), whereas treatment of persons with negative tests had no benefit, even if they were immunocompromised (3). This experience is the basis for authoritative recommendations to identify and treat a positive TST, de facto considered equivalent to latent TB infection. The TST has several limitations, particularly poor specificity because of cross-reactivity with the antigens of the bacillus Calmette-Guérin (BCG) vaccine, as well as many of the nontuberculous mycobacteria (4). The injection of antigens can stimulate anamnestic immune recall, so that repeated administration can lead to a larger reaction ("boosting"). Finally, patients must return for a second visit 48–72 hours after TST placement for reading.
In recent years, IFN- Despite the theoretical advantages of IGRAs, there has been little evidence that a positive IGRA result is predictive of future risk of active TB. This type of evidence is crucial, as the only gold standard for a test of latent TB is the future development of active disease. This makes the study by Diel and colleagues (8) in this issue of the Journal (pp. 1164–1170) very important. The authors prospectively evaluated 601 immunocompetent close contacts of infectious TB cases in Germany—a low-incidence country. Of those tested, 243 (40.4%) were TST positive and 66 (11.0%) were positive using an IGRA (the QuantiFERON Gold In-Tube). Of the 601 contacts, 278 (46.3%) had a history of prior BCG vaccination. The rate of positive TST was significantly higher among BCG-vaccinated contacts (67.3% positive) than among unvaccinated contacts (17.3% positive), whereas the IGRA test did not appear to be affected by prior BCG vaccination (11.5% positive among vaccinated persons, 10.5% positive among unvaccinated persons). Isoniazid treatment was offered only to contacts with positive IGRA tests. Contacts were then followed for a mean of 103.3 weeks. Among contacts who were not offered or who refused isoniazid, 5 of 219 (2.3%) with positive TSTs developed active TB, compared with 6 of 41 (14.6%) with positive IGRA tests (P < 0.003). Notably, all six contacts who subsequently developed active disease had a positive IGRA result, whereas only five of the six had a positive TST. The authors concluded that the IGRA test was at least as sensitive as the TST for identifying contacts at high risk for progression to active TB, and identified a significantly smaller group of persons than the TST. These findings have important implications for TB control, as a test equally sensitive to the TST and more specific could allow identification of persons at highest risk to develop TB disease. There are several important limitations relating to this study that need to be considered. Of the six contacts diagnosed with active TB, only two were culture confirmed and the others were clinically diagnosed—these might have been misclassified. Although the difference in rates of progression to active TB between the TST-positive and IGRA-positive subjects was statistically significant, the total number of six cases of active TB is still small. As a result, the 95% confidence interval for the estimate of sensitivity of IGRA in predicting active TB ranges between 54 and 100%, and for TST ranges from 36 to 100%. The population studied was immunocompetent, so these results cannot be extrapolated to contact investigations involving persons with the human immunodeficiency virus or other immunocompromising conditions. The results should also not be generalized to persons who are being screened for remote latent TB, as the performance of IGRAs among persons with remote TB infection is less clear (9). Finally, every study that has compared IGRA tests to each other (QuantiFERON Gold/Gold In-Tube to T-SPOT.TB) has found a significant proportion of subjects with discordant test results, the significance of which is unknown (10). Perhaps the most important reason to be cautious in interpreting these results is the findings of the only other similar study that has been published to date. In the Gambia, household contacts with positive TSTs had the same risk of progression to active TB during 2 years of follow-up (1.7%) as contacts with a positive IGRA (1.7%) (11). In that study, each test (skin test or IGRA) identified persons who progressed to active TB but had a negative result with the other test. It is a rare editorial that does not end with a call for further studies, and we will not break with tradition. The results of Diel and colleagues are very promising, as they suggest the potential for IGRAs to enable programs to target the highest-risk patients for latent TB treatment. These findings do not call, they shout, for further prospective studies to better define the ability of IGRAs, relative to the TST, to predict future risk of active TB. FOOTNOTES Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
release assays (IGRAs) have been proposed as alternatives to the TST. IGRAs examine the in vitro IFN-
