CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

approaches to dampening pathologic inflammation mediated by Th1 immune recovery may improve treatment outcomes in HIV/TB.

1 Lazzaro Spallanzani National Institute for Infectious Diseases, Rome, Italy, 2 Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán, Mexico City, Mexico, 3 Infectious Disease Institute, Kampala, Uganda, 4 Universidad Peruana Cayetano Heredia, Lima, Peru, 5 University College London, London, UK, 6 University of Milan, Milan, Italy, 7 Johns Hopkins University, Baltimore, MD, USA, 8 University of Modena and Reggio Emilia, Modena, Italy Background: Scale up of combination antiretroviral therapy (cART) has a profound impact on the risk of developing tuberculosis (TB) in persons with HIV, both in low and high TB burden countries. Nonetheless, TB remains a major cause of morbidity and mortality in persons with HIV. Knowledge of timing and determinants of TB risk is essential to designing strategies to address this issue. Methods: The study was conducted in four countries with different TB burden (Uganda, Peru ,Mexico, and Italy). We analyzed data of persons enrolled in HIV observational cohorts (one multicenter cohort ICoNA – Italy and 3 single institution cohorts: INNSZ-Mexico, IMTAvH-Peru and IDI-Uganda) from 2006-2016. Cases of TB diagnosed at first presentation (within 3 months of HIV diagnosis/initiation of HIV care) or during follow-up, before and after cART initiation, were considered. Factors associated with the risk of having TB at enrollment were identified by multivariable logistic regression. Incidence rates of TB from enrollment were calculated, and Poisson regression model was used to identify factors associated with the incidence of TB in the study population Results: The analyzed cohort included 24,043 persons of whom 2,455 (10.2%) were diagnosed with TB. TB was diagnosed at first presentation in 1763 (72%), in 260 (11%) at least 3 months after presentation and before cART start, and in 432 (18%) after cART initiation. Proportion of cases diagnosed at first presentation ranged from 69.9% in Uganda to 82% in Mexico. Presentation for HIV care with low CD4 cell count was a strong risk factor for TB in all countries. Preventive therapy was infrequently reported in these patients (<2%). Incidence of TB after cART initiation ranged from 13.3 per 1000 person-years in Uganda to 0.83 in Italy. Incidence declined rapidly during the first year of treatment in all countries. After 12 months of treatment however, it remained higher than the background incidence in each country (Table). Conclusion: Timing of TB diagnosis among persons with HIV was remarkably similar in all four countries despite different TB burdens. More than three- quarters of cases were diagnosed upon presentation to HIV care and were associated with low CD4 cell count. Early HIV diagnosis and immediate initiation of cART may be the most important intervention to further decrease the risk of HIV-associated TB. Additional prevention interventions, such as preventive therapy, may be needed, however, in particular during the first year of cART.

729 HIV TEST YIELD AND REASONS FOR UNKNOWN HIV STATUS AMONG TB PATIENTS: THE KOPANYO STUDY Xiao Jun Wen 1 , Othusitse Fane 2 , Matsiri Ogopotse 2 , Mbatshi Dima 2 , Eleanor Click 1 , Patrick Moonan 1 , John Oeltmann 1 , Rosanna Boyd 1 , Chawangwa Modongo 2 , Christopher Serumola 3 , James Tobias 4 , Alyssa Finlay 1 , Nicola M. Zetola 2 1 CDC, Atlanta, GA, USA, 2 Botswana–UPenn Partnership, Gaborone, Botswana, 3 CDC Botswana, Gaborone, Botswana, 4 Northrop Grumman Corp, Atlanta, GA, USA Background: Botswana is approaching global targets for testing, treatment and viral suppression for HIV. With fewer people living with unknown HIV status, it is more difficult and costly to find and test the remaining few that could benefit from HIV treatment. Based on PEPFAR program reports, the overall yield of HIV testing activities in Botswana was about 5% in 2017. Testing tuberculosis (TB) patients for HIV is a high-yield activity. Our objective was to examine HIV test yield among TB patients never tested previously or with undocumented HIV status and the reasons for unknown HIV status at the start of anti-TB treatment among patients enrolled in a multiyear, population-based TB transmission study in Botswana – the KOPANYO study. Methods: During September 2012 – March 2015, all persons registered with TB in Gaborone and Ghanzi Districts, Botswana were eligible for the study. At enrollment, all TB cases were offered HIV testing in accordance with national guidelines, except those previously documented positive and tested negative within 90 days before enrollment. HIV test results were recorded. The reasons for no results were documented. Results: Among 4331 TB patients enrolled, 14% (623/4331) never tested previously nor had documented HIV status at the start of anti-TB treatment. Of these, 77% (480/623) were tested for HIV during the course of treatment - including 23% (110/480) patients newly diagnosed with HIV. Of all participants, for 3% (143/4331) their HIV status remained unknown to the end of the study. Of these 143 patients, 65% (93) outright refused HIV testing without providing reasons; while 35% (50) did not refuse, but had no test results. The reasons for unknown status among non-refusals included: 29% (16/50) unspecified; 22% (12/50) no test kit was available; 22% (12/50) tested at a different facility with no documented HIV results, and 18% (10/50) agreed to test but deferred for a later time. Conclusion: In this analysis, we found that among the TB patients with unknown HIV status, HIV testing yield was 23%, 4 times higher than other HIV- testing activities in Botswana (5%); however, an unacceptable proportion of HIV status results remained unknown due to patient refusal, or logistical reasons. Further research is needed to understand why patients refuse testing. 730 TIMING AND INCIDENCE OF HIV-ASSOCIATED TUBERCULOSIS: A 4-COUNTRY STUDY Enrico Girardi 1 , Yanink Caro-Vega 2 , Jospeh Musaazi 3 , Gabriela Carriquiry 4 , Alessandro Cozzi-Lepri 5 , Barbara Castelnuovo 3 , Andrea Gori 6 , Yukari C. Manabe 7 , Eduardo Gotuzzo 4 , Antonella D’Arminio Monforte 6 , Brenda Crabtree-Ramírez 2 , Cristina Mussini 8 , for the ICoNA , IMT AvH, INNSZ and IDI Cohorts

Poster Abstracts

731 HIGH INCIDENCE OF TUBERCULOSIS AMONG HIV+ PATIENTS TREATED WITH HAART IN ZAMBIA Simon Mutembo 1 , Jane N. Mutanga 1 , Vincent C. Marconi 2 , Christopher Whalen 3 1 Government of Zambia Ministry of Health, Lusaka, Zambia, 2 Emory University, Atlanta, GA, USA, 3 University of Georgia, Athens, GA, USA Background: Tuberculosis (TB) is the leading cause of morbidity and mortality among Human Immunodeficiency Virus (HIV+) patients. The risk of TB among HIV+ patients on combination Anti-Retroviral Therapy (cART) is heterogenous depending on the timing of cART. However, it is not known whether there are differences in the risk of TB among HIV+ patients accessing cART in rural and urban health settings in sub-Sahara Africa. In urban settings, high TB incidence is sustained by the high HIV prevalence and crowded living conditions. Rural settings have distinct challenges which drive the TB and HIV epidemic. These

CROI 2019 281