Background There is a paucity of data on the performance of interferon-gamma release assays (IGRAs) in screening HIV-infected patients for latent tuberculosis infection (LTBI).
Methods Prospective evaluation of two IGRAs [QuantiFERON-TB Gold In-Tube (QuantiFERON) and T-SPOT.TB] for screening HIV-infected patients attending an inner city clinic with a high local prevalence of tuberculosis (TB). For each patient we recorded their gender, ethnicity, country of birth, BCG status and prior treatment for TB, CD4 count, plasma HIV load and receipt of antiretroviral therapy (ART). All patients had both IGRAs (results = positive, negative, or indeterminate) and a chest radiograph.
Results Of 117 patients, 91 (78%) were men, 48 (41%) were white British, 29 (24.8%) black African, 21 (18%) European, 7 (6%) South American and 12 (10%) from other ethnic groups. Individuals born outside the UK had been resident in the UK for a median of 10 years (range 3–45). 91/117 (78%) had previously received BCG and 14/117 (12%) had been treated for TB a median of 9.5 years (range 3–13) previously. Of 106 (90.6%) receiving ART, 105 (99%) had an undetectable plasma HIV load) and the median CD4 count was 550 (range 30–1250). Among those not receiving ART the median (range) plasma HIV load and CD4 count was 21 000 copies/ml (590–160 000) and 520 (250–950) respectively. IGRA results are shown in the Abstract P14 table 1. Overall there was good concordance (92.3%) between the two IGRAs. Among 14 patients previously treated for TB, 10 (71.4%) had previously received BCG; 4/14 (28.6%) had a positive IGRA (3/4 both IGRAs were positive, 1/4 QuantiFERON was positive). Of those previously treated for TB with negative IGRAs (n=10) all had an undetectable HIV viral load, 9/10 had been on ART >2 years and median CD4 count was 670 (range 180–1250). Five other patients with CD4 <200 had negative IGRAs; all had previously received BCG. Mantoux testing, done in 2 of the 5, was negative.
Conclusion Overall there was a high concordance between the two IGRAs. Four of the 12 with positive IGRA results had a past history of TB making interpretation uncertain. The low rate of indeterminate results likely reflects the high CD4 counts among this patient group.