Diagnosis of latent tuberculosis infection with T-SPOT(®).TB in a predominantly immigrant population with rheumatologic disorders.

Abstract

The objective of this study is to compare how likely positive tuberculin skin test (TST) and T-SPOT(®).TB (TSPOT) results predict risk factors for tuberculosis in a predominantly immigrant patient population at risk of latent TB infection (LTBI) and with rheumatologic conditions requiring immunomodulatory therapy (IMT). Prospective study conducted at a referral rheumatology clinic. Inclusion criteria included patients on various IMT, including immunosuppressive drugs that could predispose to TB progression. We studied risk factors associated with LTBI, test results, and tests' agreement. We studied 101 patients. Eighty (79.2 %) were from countries where TB is prevalent and Bacille Calmette-Guérin vaccination is placed routinely. Seventy-four (73.3 %) had rheumatoid arthritis and 92 (90.7 %) were on IMT. Among patients with both TST and TSPOT results, 25 (30.9 %) were TST(+) and 20 (24.7 %) had TSPOT(+) results. Fifteen patients (18.5 %) had TST(+)/TSPOT(+) results, and 51 (63.0 %) had TST(-)/TSPOT(-) results (agreement = 81.5 %; kappa = .54 [95 % CI, .34-.74; P < .001]). Each TSPOT(+) and TST(+) results were independently associated with immigrant status and prior residence in a TB prevalent country after adjustment for immunosuppressive therapy: Adjusted OR(TSPOT+)=6.6 (95 % CI, 1.2-123.3; P = .027); and adjusted OR(TST+)=11.2 (95 % CI, 2.0-209.5; P = .003). Seven out of 10 TST(+)/TSPOT(-) cases had a TST ≥15 mm induration, including three cases with history of TST conversion. TST(+) and TSPOT(+) results predict risk factors associated with LTBI independent of immunosuppressive IMT. Some TST(+)/TSPOT(-) results were unlikely to be false-negatives. The combined use of TST and TSPOT appears to be a reasonable diagnostic strategy to evaluate for LTBI in this population.