CROI 2016 Abstract eBook
Abstract Listing
Poster Abstracts
adjusting for age, CD4 cell count and ART status, the incidence of TB was significantly reduced during pregnancy compared to pre-conception (IRR, 0.46; 95% CI, 0.25-0.84) and was lowest in women on ART in pregnancy (IRR, 0.38; p=0.021) and increased slightly postpartum (IRR, 1.50; p=0.417). Conclusions: TB incidence appears reduced in HIV+ women during pregnancy compared to the pre-conception period but the absolute incidence remains high in this setting. While intensified case detection interventions in PMTCT services warrant consideration, expanded access to ART in pregnancy under Option B+may further reduce the burden of TB in pregnant and postpartumwomen. 758 Pregnancy Intensifies the IFN-gamma Suppression of HIV in TB-Infected IndianWomen Jyoti S. Mathad 1 ; Ramesh Bhosale 2 ; Usha Balasubramanian 2 ; Prasad Deshpande 2 ; Savita Kanade 2 ;Vidya Mave 3 ; Amita Gupta 4 1 Weill Cornell Med Coll, New York, NY, USA; 2 Byramjee Jeejeebhoy Med Coll, Pune, India; 3 BJ Med Coll, Pune, India; 4 Johns Hopkins Univ Sch of Med, Baltimore, MD, USA Background: Tuberculosis (TB) is more likely to occur immediately postpartum than at any other time in a woman’s life. Low interferon gamma (IFN-g) has been associated with increased risk of TB. The aim of this study was to determine if immune changes in pregnancy and postpartum affect the IFN-g response to M. tuberculosis (MTB) antigens. Methods: Screening for latent TB with an IFN-g release assay (QuantiFERON ® Gold in tube, QGIT) was performed during 2 nd /3 rd trimester of pregnancy, at delivery and 3 months postpartum in HIV-infected and -uninfected pregnant women at a government hospital in Pune, India. A subset enrolled during pregnancy was followed longitudinally with testing repeated at delivery and postpartum. Sociodemographic and medical history data were also collected. Concentrations of IFN-g were compared using the Wilcoxon ranksum test. Logistic regression was performed to assess predictors of maximal IFN-g response (defined as 10 IU/mL) to MTB antigens. Results: Valid QGIT results were available for 310 women during pregnancy, 426 women at delivery and 209 postpartumwomen. Among those with a positive QGIT, HIV-infected women produced a lower median IFN-g than HIV-uninfected women at all time points (median 2.7 vs 3.2 IU/mL in pregnancy, 1.8 vs 1.9 IU/mL at delivery). This was statistically significant postpartum (2.62 in HIV-infected vs 6.02 IU/mL in HIV-uninfected, p=0.01). The 21 HIV-uninfected and 55 HIV-infected women in the longitudinal cohort showed the same trend (see Figure 1). HIV-infected women trended towards a significant decrease in IFN-g between pregnancy and delivery (p=0.07). HIV-uninfected women had a significant increase in IFN-g production between delivery and postpartum (p=0.0002) while HIV-infected women did not (p=0.19). Having a maximal IFN-g response to MTB antigens was associated with not having HIV (1.6, CI 0.9-2.9, p=0.09) and being postpartum (OR 3.5, CI 2-6.2, p<0.05). Conclusions: IFN-g reached a nadir for both HIV-infected and –uninfected women at delivery; HIV-uninfected women had a significant increase in IFN-g production postpartum while HIV-infected women did not. Lower IFN-g production during pregnancy, especially in those with HIV, may facilitate progression to active TB, though symptoms may not appear until postpartum, when IFN-g production increases again. TB diagnostics development should take these dynamic immune changes during pregnancy into account.
Poster Abstracts
759
Evaluation of Provider-Initiated Cryptococcal Antigen Screening, South Africa Nicky Longley 1 ; SnigdhaVallabhaneni 2 ; Mariette Smith 3 ; Rachel Smith 2 ; Meg Osler 3 ; Nicola Kelly 3 ; Anna Cross 3 ; Andrew Boulle 4 ; Graeme Meintjes 3 ; Nelesh Govender 5 1 St George’s Univ of London, London, UK; 2 CDC, Atlanta, GA, USA; 3 Univ of Cape Town, Cape Town, South Africa; 4 Cntr for Infectious Disease Epi and Rsr, Cape Town, South Africa; 5 Natl Inst of Communicable Diseases, Johannesburg, South Africa Background: Cryptococcal antigen (CrAg) screening can identify persons at risk for disseminated cryptococcal disease (DCD); pre-emptive treatment can prevent disease progression. In August 2012, the Western Cape Province (WC) implemented provider-initiated CrAg screening among antiretroviral therapy (ART)-naïve patients with CD4 count <100 cells/µl followed by fluconazole treatment for asymptomatic CrAg-positive patients. We evaluated the implementation and effectiveness of this program. Methods: We used National Health Laboratory Service data for CD4 count and CrAg test results, provincial HIV program data for ART status, nationwide DCD surveillance program data for patient outcome, and medical records of CrAg-positive patients for fluconazole prescriptions. We assessed the proportion of eligible patients screened for CrAg, and treated pre-emptively, and the prevalence of CrAg positivity among those screened during September 1, 2012 –August 31, 2013. We compared incidence of DCD among eligible patients who were screened vs. not screened. Results: Of 4,395 eligible patients, 26.6% (n=1170) were screened. The proportion of patients screened increased from 15.9% in September 2012 to 36.6% in August 2013. The prevalence of CrAg positivity was 2.1% (24/1170). Treatment data were available for 13/24 CrAg-positive patients, 9 of whomwere treated with any fluconazole. There were 9 (0.8%) incident cases of DCD among the 1170 patients who were screened for CrAg, all of which occurred among those who screened CrAg-negative but had a substantial delay in starting ART vs. 49 (1.5%) incident cases among the 3225 eligible patients who were not screened (p=0.07). Median time between CD4 count and DCD was 103 days (range: 35–497) in the screened group vs 155 days (range: 17–589) in the group not screened. A significantly higher proportion of those who were CrAg screened were also prescribed ART (72.9%, 853/1170) compared with those were not screened (48.3%, 1556/3225) (p<0.001). Conclusions: The penetrance of provider-initiated CrAg screening in the WC was poor. Although not statistically significant, there were half as many DCD cases in the CrAg- screened group compared with those not screened. CrAg screening along with prompt ART initiation can reduce the DCD burden, but needs to be implemented well. Laboratory- based reflex screening, where CrAg testing is automatically performed on CD4 test remnant blood, would enhance screening, and should be considered where possible.
316
CROI 2016
Made with FlippingBook - Online catalogs