Latent Tuberculosis Infection Detection and Clinical Implications in HIV-1 Infected Women

Introduction. Tuberculosis and HIV are major health problems in women during the reproductive years. We designed a series of questions to address the risk of postpartum tuberculosis in HIV-1 infected women and the utility of tuberculosis-specific interferon-gamma (IFN-gamma) release assays (IGRA) during pregnancy/postpartum, using data and specimen from a perinatal cohort of HIV-1 infected women, in Kenya. We determined the risk of postpartum tuberculosis associated with LTBI (or positive IGRA). We determined the sensitivity, specificity and positive predictive value (PPV) of varied cut-offs in baseline IFN-gamma spot count and CD4 count for postpartum tuberculosis. Among women who were IGRA positive during pregnancy, we determined the consistency of a positive response over 1 year postpartum.;Results. Prevalence of LTBI in HIV-1 infected women during pregnancy was 42.7%. LTBI during pregnancy was associated with 4.5-fold increased risk of postpartum tuberculosis (CD4-adjusted Hazard Ratio (aHR): 4.5; 95% confidence interval (CI): 1.1--18.0; p=0.03). IFN-gamma spot count≥6 was associated with PPV of 5.9% and an optimal trade-off in sensitivity (78%) and specificity (55%) compared to lower (>0) and higher (≥10) IFN-gamma cut-offs. In women with CD4 cell count <250 cells/microL, IFN-gamma spot count ≥6 was associated with PPV of 19% and 85% sensitivity and 63% specificity.;Of 18 women with positive IGRAs during pregnancy, 50% retained IGRA positivity over 1 year postpartum. Women with baseline IFN-gamma spot count>8 were more likely to have positive IGRAs postpartum (Odds Ratio: 10.0; 95% CI: 0.9 -- 117.0). Change in IFN-gamma over 3-monthly intervals was 10.2 SFCs/well (95% CI: -1.5 -- 21.8).;Conclusion. Positive IGRAs in HIV-1-infected pregnant women were associated with postpartum active tuberculosis. Magnitude of responses remained consistent in serial testing, with modest increases postpartum. Together, these findings have implications for LTBI detection and IPT strategies in HIV-1 infected pregnant women.