S41 Screening for latent TB in HIV: are NICE & BHIVA guidance effective?
HIV infection is the strongest single risk factor for the development of active TB in latently infected individuals. NICE and BHIVA 2011 guidelines both recommend screening for latent TB in HIV positive subjects in the UK. NICE bases this on blood CD4 count; and recommends testing all patients with CD4<500 cells/μl (if CD4<200, this is with an Interferon Gamma Release Assay (IGRA) and tuberculin skin test (TST), and if CD4 200–500, then IGRA ± TST is proposed). BHIVA test-stratify on CD4 count, country of origin and use of antiretroviral therapy (ART). They recommend IGRA in the following populations: all subjects from Sub Saharan Africa, irrespective of CD4 if on ART <2 years; medium TB incidence countries if on ART <2 years and CD4 <500; and low TB incidence countries if on ART <6 months and CD4 <350. To our knowledge, neither strategy has been formally tested in a UK HIV population. Here, using data on all subjects over a 10-year period (2000–2010) within our large HIV service (n=3306), we determine the impact of applying both strategies to detect cases of latent TB infection who subsequently developed active TB during this time period (n=72). Subjects who developed active TB <3 months from HIV diagnosis were excluded from analysis, as it was felt that the time interval was too short for them to be screened for TB. Abstract S41 table 1 illustrates the differences between NICE and BHIVA strategies over the 10-year period. Neither screening method had optimal sensitivity and specificity. Both guidelines' eligible groups were only twice as likely to develop active TB compared to those ineligible for screening. NICE guidelines were poorly specific while the BHIVA strategy missed 42% of cases (largely as they were either UK born or on ART). Using data from our UK HIV population, we find that new TB screening guidance does not appear to discriminate those at risk of active TB.