CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

769 Aetiology and Outcome of Community-Acquired Pneumonia in HIV-Infected Malawian Adults Stephen J. Aston 1 ; Antonia Ho 2 ; Hannah Jary 1 ; Dean Everett 3 ; Henry Mwandumba 3 ; Robert S. Heyderman 4 ; Mulinda Nyirenda 5 ; Stephen Gordon 6 1 Liverpool Sch of Trop Med, Liverpool, UK; 2 Univ of Liverpool, Liverpool, UK; 3 Malawi Liverpool Wellcome Trust Clinical Rsr Prog, Blantyre, Malawi; 4 Univ Coll London, London, UK; 5 Queen Elizabeth Central Hosp, Blantyre, Malawi; 6 Malawi Liverpool Wellcome Trust Clinical Rsr Proga, Blantyre, Malawi Background: Pneumonia is the commonest reason for adult hospitalisation in Malawi, but there are few data describing its aetiology either locally or in comparable sub-Saharan African settings. We aimed to describe the aetiology of pneumonia amongst hospitalised HIV-infected adults with acute community-acquired pneumonia (CAP) and its relationship to antiretroviral therapy (ART) use and outcome. Methods: Blood, urine, sputum, nasopharyngeal aspirate (NPA) and pleural fluid specimens were collected from adults (≥18 years) with clinically-defined CAP and symptoms for ≤14 days admitted to a large central hospital in Blantyre, Malawi. Bacterial aetiologies were defined by blood culture and additionally for Streptococcus pneumoniae using an immunochromatographic assay for urinary polysaccharide antigen. Mycobacterium tuberculosis was identified by acid-fast bacilli sputum (and pleural fluid) microscopy, mycobacterial culture and Xpert MTB/RIF assay. Influenza, other respiratory viruses and atypical bacterial pathogens were identified using a multiplex RT-PCR assay on NPA. Patients were followed to 30 days. Results: We recruited 459 adults between May 2013 and January 2015 of whom 355 (78%) were HIV-infected (220 (62%) males; median age 35 years (interquartile range (IQR): 30-41); median CD4 count 99 cells/cm 3 (IQR: 44-193). HIV-infection was newly diagnosed in 124 (35%). 189 (83%) of known HIV-infected individuals were on ART. 30-day mortality was 16% (54/342). An organismwas identified in 267 (75%). S. pneumoniae (68 (19%)), Mycobacterium tuberculosis (64 (18%)) and influenza (30 (8.5%)) were the most common organisms. Infection with M. tuberculosis was independently associated with higher mortality (aOR: 2.51; 95% CI: 1.17-5.36). Co-infection, mainly with non-influenza respiratory viruses, was common. 65%, 45% and 50% of patients with S. pneumoniae , M. tuberculosis and influenza, respectively, had at least one co-infection; rates of co-infection did not vary with ART use or CD4 count. Co-infection was not associated with worsened outcome. Conclusions: The major burden of hospitalised pneumonia in Malawi remains in HIV-infected patients with advanced immunosuppression. In addition to S. pneumoniae and influenza, tuberculosis is a major cause of pneumonia, even amongst patients with reported short symptom duration and is associated with increased mortality. Co-infection, typically with respiratory viruses, is common and larger studies are required to better understand prognostic significance. 770 No Increased Risk of HIV Incidence During Pregnancy Chloe A. Teasdale 1 ; Jessica Justman 1 ; Mary-Ann Chiasson 2 ; Kelly Blanchard 3 ; Elaine J. Abrams 4 ; Heidi E. Jones 5 1 ICAP at Columbia Univ, New York, NY, USA; 2 PH Solutions, New York, NY, USA; 3 Ibis Reproductive Hlth, Cambridge, MA, USA; 4 ICAP, Columbia Univ Mailman Sch of PH, New York, NY, USA; 5 City Univ of New York, New York, NY, USA Background: There is inconclusive evidence regarding whether women are at greater risk for acquiring HIV during pregnancy. We examined HIV incidence in women enrolled in a randomized clinical trial during pregnant and non-pregnant person time. Methods: Data came from the Methods for Improving Reproductive Health in Africa (MIRA) study which was conducted in South Africa and Zimbabwe from 2003-2006. Women 18-50 years with at least one follow-up visit within 6 months of enrollment were analyzed. All study visits included HIV and pregnancy testing, as well as self-report of intravaginal practices and sexual risk behaviors. The analysis examined person-time between study visits and estimated HIV incidence using person years (py) at risk during pregnant and non- pregnant periods (non-pregnant periods were classified by self-reported hormonal contraceptive (HC) use). Cox proportional hazards models were fitted using pregnancy status as a time-varying exposure, demographic characteristics as time-fixed covariates and sexual risk behaviors as time-varying covariates (non-pregnant/no HC use was the referent group). Results: 4,549 women from the MIRA study were included in this analysis (92% of all enrolled), 766 (17%) of whom had a lab-confirmed pregnancy. Median follow-up time was 18 months [interquartile range (IQR): 12-24] and median age was 27 years. There were 240 incident HIV cases overall, 16 occurred during pregnant periods. HIV incidence was 3.9/100py overall; during pregnancy, incidence was 3.8/100py, when women were not pregnant incidence was 4.8/100py with injectable HC use, 4.4/100py with no HC and 2.6/100py with oral HC. In unadjusted and adjusted models, pregnancy was not found to increase HIV incidence compared to when women were not pregnant and not using HC (unadjusted hazard ratio (HR) 0.8, 95%CI 0.5-1.3; adjusted 0.7, 95%CI 0.4-1.2). Sex without a condom (HR 1.3; 95%CI 1.0-1.7) and suspecting or knowing a male partner had other sexual partners (HR 1.3, 95%CI 1.0-1.6) were both found to be predictors of HIV incidence in multivariable models. Conclusions: We did not find higher incidence of HIV during pregnancy compared to periods when women were not pregnant however continued high incidence of HIV was observed among pregnant women and is known to increase the risk of mother-to-child transmission of HIV. These findings highlight the need for greater efforts to prevent HIV infection in pregnant women. 771 Effect of Pregnancy on Response to Antiretroviral Therapy Among AfricanWomen Athena P. Kourtis 1 ; JeffreyWiener 1 ; Renee Heffron 2 ; Nelly R. Mugo 3 ; Caroline King 1 ; Kavita Nanda 4 ; Connie M. Celum 2 ; Deborah Donnell 5 ; Jairam R. Lingappa 2 ; Jared M. Baeten 2 1 CDC, Atlanta, GA, USA; 2 Univ of Washington, Seattle, WA, USA; 3 Kenya Med Rsr Inst, Thika, Kenya; 4 FHI 361, Chapel Hill, NC, USA; 5 SCHARP, Fred Hutchinson Cancer Rsr Cntr, Seattle, WA, USA Background: While most recent evidence does not support a role for pregnancy in accelerating HIV disease progression, very little information is available on the effects of incident pregnancy on response to antiretroviral therapy (ART). Hormonal, immune and behavioral changes during pregnancy may influence response to ART. We sought to explore the effects of incident pregnancy (after ART initiation) on virologic, immunologic, and clinical response to ART. Methods: Data were collected from HIV-infected women participating in 3 prospective studies (Partners in Prevention HSV/HIV Transmission Study, Couples Observation Study, and Partners PrEP Study) from seven countries in Africa from 2004 to 2012. Women were included in this analysis if they were ≤ 45 years of age, were started on ART during the study and were not pregnant at ART initiation. Pregnancy was treated as a time-dependent variable to include all pregnancies occurring after ART initiation. Virologic failure was defined as a viral load (VL) greater than 400 copies/ml ≥6 months after ART initiation and viral suppression as viral load <400 copies/ml. Multivariable Cox proportional hazards models assessed the association of pregnancy and time to viral suppression, virologic failure, WHO clinical stage III/IV and death. Linear mixed effects models assessed the association of pregnancy and CD4 count and VL. All analyses were adjusted for confounders, including pre-ART CD4 count and plasma VL. Results: A total of 1041 women were followed, contributing 1097.6 person-years of follow-up. One-hundred ten women became pregnant at least once after ART initiation. Median CD4 count prior to ART initiation was 276 cells/mm 3 (IQR, 209-375); median pre-ART viral load was 17,511 copies/ml (IQR, 2480-69286). Pregnancy was not associated with difference in time to virologic failure (adjusted HR, 0.68, 95% CI, 0.37-1.22), time to viral suppression (adjusted HR, 1.21, 95% CI, 0.82-1.77), time to WHO Clinical stage III or IV (adjusted HR, 0.79, 95% CI, 0.19-3.30) or time to death (adjusted HR, 2.04 (95% CI, 0.25-16.8). Incident pregnancy was associated with an adjusted mean decrease in CD4 count of 47.7 cells/mm 3 (p<0.001) over the course of follow-up, but not with difference in viral load (p=0.06). Conclusions: For HIV-infected women on ART, incident pregnancy does not have an effect on virologic or clinical HIV disease progression. A modest decrease in CD4+ T cell count could be due to physiologic effects of pregnancy.

Poster Abstracts

321

CROI 2016