Rationale: Silicosis is a well-recognized risk factor for tuberculosis (TB).
Objectives: To compare T-Spot.TB with tuberculin skin test (TST) in predicting the development of TB.
Methods: Male patients with silicosis without clinical suspicion of active TB, past history of TB, and treatment for latent TB infection (LTBI) were offered both T-Spot.TB and TST in the Pneumoconiosis Clinic of Hong Kong from 2004 to 2008, and followed prospectively until September 30, 2009, for development of TB.
Measurements and main results: Active TB and culture- or histology-confirmed TB developed in 17 (5.5%) and 14 (4.5%) of 308 recruited subjects at an annual rate of 2,247 and 1,851 per 100,000 person-years, respectively. Active TB occurred in 7.4% (15 of 204) and 1.9% (2 of 104) of T-Spot.TB-positive and -negative subjects, respectively, whereas the corresponding figures for TST (cutoff 10 mm) were 6.4% (13 of 203) and 3.9% (4 of 205), respectively. A positive T-Spot.TB test significantly predicted the subsequent development of active TB (relative risk, 4.50; 95% confidence interval, 1.03-19.68) and culture- or histology-confirmed TB (relative risk, 7.80; 95% confidence interval, 1.02-59.63). Consistent results were obtained after exclusion of subjects treated for LTBI and adjustment for potential confounders. TST did not significantly predict the development of active TB or culture- or histology-confirmed TB, irrespective of the cutoff values with or without exclusion of subjects treated for LTBI. Culture filtrate protein 10 spot count, but not early secretary antigenic target 6 spot count, was significantly associated with subsequent TB development.
Conclusions: T-Spot.TB performs better than TST in the targeted screening of LTBI among patients with silicosis.