Latent Mycobacterium tuberculosis Infection.

To the Editor: In their review article, Getahun et al. (May 28 issue)1 provide a useful overview of issues surrounding latent tuberculosis infection. However, newly published data regarding the performance of the T-SPOT.TB assay call into question the supposition that false conversions and reversions are more common with interferon-γ release assays (IGRAs) than with tuberculin skin tests.2 Specifically, 19,630 serial pairs of more than 16,000 health care workers in 19 U.S. hospitals were evaluated. The mean conversion rate was 0.8%, and the mean reversion rate was 17.6%. In the study by Dorman et al.3 that the authors cite as a basis for concern regarding the operational characteristics of the IGRA, the conversion rate on tuberculin skin testing was 0.9%. Such reversion rates are usually not assessed, since most medical centers do not continue to test once a test is positive. However, reversions on tuberculin skin testing do occur 4 and were seen in approximately 20% of patients with positive results on tuberculin skin testing who were treated for latent tuberculosis.5 Thus, the T-SPOT. TB assay would appear to be similar, if not superior, to tuberculin skin testing for the screening of health care workers. Jeffrey G. Jones, M.D., M.P.H.