CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

726 HIGH TB INFECTION RATE IN CHILDREN & YOUNG ADULTS IN RURAL UGANDA IN THE SEARCH TRIAL Carina Marquez 1 , Atukunda Mucunguzi 2 , Gabriel Chamie 1 , Dalsone Kwariisima 3 , Theodore Ruel 1 , Laura B. Balzer 4 , Tamara D. Clark 1 , Edwin D. Charlebois 1 , Maya L. Petersen 5 , Diane V. Havlir 1 1 Univ of California San Francisco, San Francisco, CA, USA, 2 Infectious Diseases Rsr Collab, Kampala, Uganda, 3 Makerere Univ Joint AIDS Prog, Kampala, Uganda, 4 Harvard Univ, Boston, MA, USA, 5 Univ of California Berkeley, Berkeley, CA, USA Background: To end TB, age-specific prevention strategies are needed; however, data on TB risk factors in rural East African children and young adults are limited. We estimated TB transmission and characterized predictors of TB infection in children and young adults in rural Uganda. Methods: In the SEARCH Study (NCT:01864603) we performed a tuberculin skin test (TST) survey, from 2015-2016, in a sample of residents ≥5 years in 8 rural communities in Eastern Uganda. Households were randomly sampled, and enriched for those with an HIV-infected adult. TB infection was defined as a positive TST, an induration ≥10mm for HIV-uninfected persons and ≥5mm for HIV infected persons. The annual risk of TB infection (ARTI) was calculated as 1-(1-prevlance)^(1/mean age +0.5). Risk factors for prevalent TB infection in children (5-14 years) and young adults (15-24 years) were assessed using multivariable logistic regression. All models were adjusted for community, BCG vaccination, and living in a household with an HIV-infected adult. Results: 2,093 children and 953 young adults completed a TST. In this sample, the ARTI among 5-24 years olds was 1.2%; the prevalence of TB infection was 9% among children and 23% among young adults, and the prevalence of HIV was 1.2% in children and 1.9% in young adults. Predictors of TST positivity in children were age (aOR:1.1, 95%CI:1.0-1.2), household contact with TB (aOR:2.7, 95%CI:1.3-5.4), lowest wealth quintile (aOR:1.7; 95%CI:1.1-2.6) and a trend towards an association with living away from home for school (aOR:1.7, p=0.11). Among mother and child dyads that both had TSTs placed, maternal TB infection predicted TB infection in children (aOR:2.0, 95%CI:1.3-3.2). Predictors of TB infection in young adults were age (aOR:1.1, 95%CI:1.1-1.2), household contact with TB (aOR:4.0, 95% CI:1.5-10.7), and being a student (aOR:1.6, 95%CI:1.1-2.2). 5% of 5-24 year olds with a positive TST had a known household TB contact. HIV infection, mother’s HIV status, and living in a household with an HIV-infected adult, were not associated with TB infection in children or young adults. Conclusion: Our data suggest TB transmission is high in rural Uganda, and nearly a quarter of young adults in our sample are already infected with TB. Only 5% of 5-24 year olds with TB infection had a known household contact, suggesting undiagnosed household contacts or community and school-based contacts drive ongoing TB infections in children and young adults. 727 THE HIV-ASSOCIATED BURDEN OF RECURRENT TB DISEASE IN CAPE TOWN, SOUTH AFRICA Sabine Hermans 1 , Nesbert Zinyakatira 2 , Judy Caldwell 3 , Frank Cobelens 4 , Andrew Boulle 2 , Robin Wood 2 1 Amsterdam Inst for Global Hlth and Development, Amsterdam, Netherlands, 2 Univ of Cape Town, Cape Town, South Africa, 3 City Hlth, Cape Town, South Africa, 4 Univ of Amsterdam, Amsterdam Zuidoost, Netherlands Background: Retreatment tuberculosis (TB) disease is common in high-prevalence settings such as Cape Town. We recently estimated that one third of TB patients go on to develop another episode of disease and risk of recurrence increases with every subsequent episode. The impact of HIV on this burden of disease is unclear. We determined the rates of recurrent TB stratified by HIV status as well as the HIV-associated population attributable risk fraction (PAF) over a period of 12 years. Methods: All recorded TB episodes in the Cape Town metropolitan area between 2003-2015 were linked to individuals by deterministic linkage of personal identifiers. We created a virtual cohort of individuals who had their first episode notified in Cape Town (excluding those who were reported to have had TB treatment before, but whose first episode was not identified by the matching algorithm). We estimated the rate of recurrent TB disease stratified by HIV status and by the number of previous episodes. In case of HIV seroconversion between episodes we split the accrued person-time in half. We calculated the rate ratio of recurrent TB disease associated with HIV which we utilized to calculate the PAF of HIV infection in recurrent TB disease. Results: A total of 287,003 TB episodes were included which represented 245,495 individuals; 16% had two or more episodes of TB. Rates of recurrent TB increased by subsequent episode (Figure 1). HIV-positive rates were higher than HIV-negative rates until episode 5: the rate ratio of recurrent TB disease associated with HIV decreased by subsequent episode: from 1.65 at the second episode to 0.86 at the sixth. The proportion of retreatment disease attributable to HIV in this population increased by subsequent episode: from 42% to 48% in the second to the sixth episode, respectively. Figure 1. Rates of recurrent TB by number of episodes, stratified by HIV status at start of current episode (with 95% confidence intervals). Conclusion: We found a very high rate of TB disease recurrence in both HIV-negative and HIV-positive TB patients, with less than half of retreatment TB attributable to HIV. These findings suggest that the HIV epidemic does not explain the high burden of retreatment TB in Cape Town, and therefore that high antiretroviral coverage will not be sufficient to curb it. It is more likely explained by a high annual risk of TB infection in combination with an increased risk of infection or progression to disease associated with previous TB treatment. 728 HIV TESTING UPTAKE AMONG HOUSEHOLD CONTACTS OF MDR-TB INDEX CASES IN 8 COUNTRIES Valarie S. Opollo 1 , Xingye Wu 2 , Richard Lando 3 , Susan Swindells 4 , Amita Gupta 5 , Anneke Hesseling 6 , Rodney Dawson 7 , Michael D. Hughes 8 , N. Sarita Shah 8 , for the A5300/I2003 StudyTeam 1 Kenya Med Rsr Inst, Kisumu, Nyanza, Kenya, 2 Harvard Univ, Boston, MA, USA, 3 Kenya Med Rsr Inst, Kisumu, Kenya, 4 Univ of Nebraska, Omaha, NE, USA, 5 The Johns Hopkins Univ, Baltimore, MD, USA, 6 Stellenbosch Univ, Tygerberg, South Africa, 7 Univ of Cape Town, Mowbray, South Africa, 8 CDC, Atlanta, GA, USA Background: HIV co-infection rates among MDR-TB cases vary globally, and are associated with higher morbidity and mortality. Household contacts (HHC) of MDR-TB/HIV co-infected cases are at high risk for both HIV and TB infection. However, uptake of HIV testing among HHC is understudied. As part of a cross-sectional feasibility study for a randomized trial of preventive therapy for HHC of MDR-TB index cases (IC), we evaluated factors associated with HIV test uptake among HHC. Methods: Adult IC with at least one HHC were eligible. A HHC was defined as living in the same dwelling and sharing housekeeping arrangements with an IC in the 6 months before the IC started MDR-TB treatment. All adult and child HHC were offered HIV testing if never tested or last tested HIV-negative >1year prior to study entry. HIV testing was done using standardized algorithms. Logistic regression for clustered data was used to evaluate associations. Results: From 10/2015–5/2016, 1007 HHC of 284 IC were enrolled from 16 sites in 8 countries (Botswana-1 Brazil-1, Haiti-1, India-2, Kenya-1, Peru-2, South Africa-7 and Thailand- 1). Among the 284 IC, 102 (36%) were HIV-infected, 156 (55%) were HIV-uninfected, and 26 (9%) had unknown status. HIV status was known for 225 (22%) HHC: 39 (4%) were HIV- positive, 186 (18%) were HIV-negative. HIV testing was offered to 770 (98%) of the 782 remaining HHC, of whom 545 (71%) agreed to testing; 535 (98%) were tested, and 26 (5%) were HIV-positive. Testing uptake varied by site (median 86%; p<0.001); 4 sites had 100% uptake, but 5 sites had <50% uptake. Uptake was 74% for females versus 67% for males, and was lower in children 2–4y (51%), 5–12y (56%) and 13–17y (63%), compared to ˂ 2y (77%) and adults ˃ 18y (78%). Of the 225 HHC who declined testing, 119 (53%) gave a reason; common reasons were perception of low risk (23%), not wanting repeat testing (9%), not ready (5%), not enough time (3%), fear of disclosure (3%). The proportion of HHC of HIV-infected IC versus HIV-uninfected IC agreeing to HIV testing was similar (68% versus 67%, P=0.87), but the proportion testing positive differed (8% versus 2%, P=0.008). Of the 225 HHC who declined testing, 71 (32%) were contacts to an HIV-infected IC. Conclusion: HIV testing uptake varied considerably among sites and was lower in children and adolescents compared to infants and adults. Addressing participant perceptions of HIV risk may increase HIV test uptake, with particular emphasis among HHC of HIV-positive IC given their higher risk of HIV infection. 729 HIV CONTINUUM AND EXPEDITED TB DIAGNOSIS IN TB/HIV COINFECTED PATIENTS IN BOTSWANA Diane Gu 1 , Sanghyuk S Shin 1 , Chawangwa Modongo 2 , Cynthia Caiphus 3 , Othusitse Fane 2 , Matsiri Ogopotse 2 , Mbatshi Dima 2 , Nicola M Zetola 4

Poster and Themed Discussion Abstracts

CROI 2017 317