230 POOR CONCORDANCE BETWEEN DIAGNOSTIC TESTS FOR LATENT TUBERCULOSIS INFECTION INCLUDING INTERFERON-γ RELEASE ASSAYS IN HIV+ PERSONS.

Purpose Two new interferon-γ release assays (IGRA) T-SPOT.TB (TSPOT) and QuantiFERON-TB Gold in Tube (QFT-G) that employ tuberculosis (TB)-specific antigens are now available for the diagnosis of latent tuberculosis infection (LTBI). Few data are available on these tests in immunocompromised individuals. Our study was carried out to determine the utility of these tests in HIV+ patients. Methods Patients were enrolled at two clinics in Atlanta from 9/05 to 6/06. Patients had blood drawn for TSPOT and QFT-G and a tuberculin skin test (TST) placed. Medical and laboratory data were also collected. Results 338 HIV+ patients were enrolled; 65% were male, and 85% African American. Median CD4 count was 318/μL; mean viral load was 40,095 copies/mL. Eight percent were foreign born, and 7% had a history of BCG. Overall, 44 patients (13%) had a positive test for TST, QFT-G, or TSPOT. Six (2% [95% CI 0.8-3.8]) persons had a positive TST; 10 (3% [95% CI 1.6-5.4]) a positive QFT-G; and 32 (10% [95% CI 6.8-13.1]) a positive TSPOT. There were significant differences in the number of persons who tested positive for TSPOT compared with TST and TSPOT compared with QFT-G (McNemar9s test p p Conclusions Poor concordance was noted between the three diagnostic tests. Indeterminate test results with IGRA were significantly more likely to occur among those with lower CD4 counts. Additional studies among HIV+ persons with high TB prevalence are needed to further assess the utility of these new IFN-γ tests among HIV+ persons.