Posting #4 - Gold standard in LTBI testing for TB in people who are HIV+?

The tuberculin skin test (TST) has been used historically to test individuals for latent TB infection (LTBI) and has a sensitivity and specificity of approximately 90% in immunocompetent people. Since the discovery of Human Immunodeficiency Virus (HIV), the global TB picture has changed considerably, including the way we test for LTBI. In the context of an HIV infection, the TST has proven to have a much lower sensitivity and specificity than with immunocompetent people. The reduced reliability of the TST in HIV patients has led to a search for a test that will perform better with those patients. Interferon-γ release assays (IGRA), including QuantiFERON-TB Gold (QFT-G) have been studied as a possible replacement for the TST due to the limitations of the TST in testing the general population, particularly in areas where the bacillus Calmette-Guerin (BCG) vaccine is being used. QFT-G testing in the general population has compared favorably to TST (Fietta, et al., 2003; Detjen, et al., 2007), so studies were undertaken to determine if QFT-G would overcome the reliability problems of the TST in HIV positive patients.

The question of reliability for testing of LTBI has to include the discussion of what is considered reliable. When talking about reliability, many times it isn't reliability that is being discussed, but validity. Reliability, strictly speaking, means that a test gets the same result every time, while validity means that the test is accurate. In this case, it is the relative validity, or accuracy, of the tests that is in question. Usually, a test for a particular condition is found that has a relatively high sensitivity and specificity. That test is then considered to be the gold standard for that particular condition. Other tests can be compared to the gold standard to determine their relative accuracy (Michel, Mouillet, and Salmi, 2006). Unfortunately, in the case of LTBI, there is no gold standard. Since there is no way to confirm the presence of TB until it progresses into a clinical illness, there isn't any way to determine the accuracy of LTBI tests. This is particularly true of LTBI in HIV positive patients where anergy reduces the response of the immune system to the TST. So the challenge in this case is to determine the accuracy of both tests without knowing the true status of the LTBI patients who are being tested. To try to resolve this issue, two different strategies were employed. One was to try both tests on patients were had confirmed TB disease and find how many tested positive. This would give us the sensitivity for both tests. The other strategy was to test people who had known risk factors, compare the results of both tests with each other and against a predicted value of positive results. One other factor was important to take into consideration in regards to test strategies. The efficacy of testing is related to the prevalence of the infection in the population being tested. If TB is not endemic in the population, the positive predictive value of the test (the chance that you actually have the condition if you test positive) will be low, even if the sensitivity and specificity are high. This effects the cost effectiveness of the test.

Four studies were selected to compare the sensitivity of QFT-G versus TST in people with active tuberculosis (Raby et al., 2008; Vincenti et al., 2007; Syed Ahamed Kabeer et al, 2009; Rangaka et al., 2007). Three studies were found that compared the results of QFT-G with TST where subjects did not active cases of TB, but had risk factors associated with TB infection (Luetkemeyer et al., 2007; Raby et al., 2008; Jones et al., 2007).

The results from the studies were mixed overall. In those among high prevalence groups, either all with active TB or in high LTBI settings, two studies showed that QFT-G was significantly more sensitive and two showed no significant difference. More information as to the methodology of the studies would be necessary to determine the validity of each of these studies, but on face value, there is some evidence that QFT-G is more sensitive in high prevalence settings. For the studies in low prevalence settings, the results uniformly showed that there was no significance between the two tests. No studies were found that compared the specificity of the two tests.

A couple of statistics from the studies did stand out. Both tests generally performed poorly in testing for TB in HIV negative subjects. In only two studies out of four, results between HIV positive and HIV negative subjects was equivalent (QFT-G in Jones et al., 2007; Rangaka et al., 2007) and in one study, both of those results were significantly less than was reported for the TST results for HIV negative subjects. Also, as the CD4 count dropped below 100, the QFT-G indeterminate results increased significantly. The significance of this result is that QFT-G indeterminate results correlate to an anergy response while the TST doesn't distinguish between a negative and an anergy response. This may be helpful in determining true negative results, although further research needs to be done in this area.

Finally, a gold standard has yet to be established. Neither the QFT-G or TST can be considered a gold standard, especially in regards to testing of LTBI in HIV positive people. Valid reasons exist to use the QFT-G instead of a TST, though. Concerns about a patient not showing up for a 2nd visit to get a reading of their TST result or testing in areas where the BCG vaccine is widely used are among those reasons. The wide variation in data shows that further work needs to be done to determine the factors in HIV positive patients that hamper validity in TB testing.

Reference

Detjen, A. K., Keil, T., Roll, S., Hauer, B., Mauch, H., Wahn, U., et al. (2007). Interferon-gamma release assays improve the diagnosis of tuberculosis and nontuberculous mycobacterial disease in children in a country with a low incidence of tuberculosis. Clinical Infectious Disease, 45, 322.

Fietta, A., Meloni, F., Cascina, A., Morosini, M., Marena, C., Troupioti, P., et al. (2003). Comparison of a whole-blood interferon-gamma assay and tuberculin skin testing in patients with active tuberculosis and individuals at high or low risk of Mycobacterium tuberculosis infection. American Journal of Infectious Control, 31, 347-53.

Jones, S., de Gijsel, D., Wallach, F.R,, Gurtman, A.C., Shi, Q., Sacks, H. (2007). Utility of QuantiFERON-TB Gold in-tube testing for latent TB infection in HIV-infected individuals. International Journal of Tuberculosis Lung Disease, 11, 1190-5.

Luetkemeyer, A.F., Charlebois, E. D., Flores, L. L., Bangsberg, D. R., Deeks, S. G., Martin, J.N., et al. (2007). Comparison of an interferon-gamma release assay with tuberculin skin testing in HIV-infected individuals. American Journal of Respiratory Critical Care Medicine, 175, 737-42.

Michel, P., Mouillet, E., Salmi, L. R., (2006). Comparison of Medical Subject Headings and standard terminology regarding performance of diagnostic tests. Journal of the Medical Library Association, 94, 221-223.

Raby, E., Moyo, M., Devendra, A., Banda, J., De Hass, P. Ayles, H., et al. (2008). The effects of HIV on the sensitivity of a whole blood IFN-gamma release assay in Zambian adults with active tuberculosis. PLoS One, 3, e2489.

Rangaka, M. X., Diwakar, L., Seldon, R., van Cutsem, G., Meintjes, G. A., Morroni C, et al. (2007). Clinical, immunological, and epidemiological importance of antituberculosis T cell responses in HIV-infected Africans. Clinical Infectious Disease, 44, 1639-46.

Rangaka, M. X., Wilkinson, K. A., Seldon, R., Van Cutsem, G., Meintjes, G. A., Morroni, C., et al. (2007). Effect of HIV-1 Infection on T-Cell–based and Skin Test Detection of Tuberculosis Infection. American Journal of Respiratory Critical Care Medicine, 175, 514-520.

Syed Ahamed Kabeer, B., Sikhamani, R., Swaminathan, S., Perumal, V., Paramasivam, P., Raja, A. (2009). Role of interferon gamma release assay in active TB diagnosis among HIV infected individuals. PLoS One, 4 , e5718.

Vincenti, D., Carrara, S., Butera, O., Bizzoni, F., Casetti, R., Girardi, E., et al. (2007). Response to region of difference 1 (RD1) epitopes in human immunodeficiency virus (HIV)-infected individuals enrolled with suspected active tuberculosis: a pilot study. Clinical Experimental Immunology, 150, 91-8.