CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

be appropriate for a young African population. Logistic regression analysis was used to investigate if HIV was associated with abnormal arterial stiffness. Results: cfPWV was available in 853 participants, of whom 365 (42.8%) were HIV-positive. HIV-positive participants were older, mainly women and had fewer cardiovascular risk factors than HIV-negative participants. Median CD4 count was 491 (IQR 335-680), and 77.7%were on antiretroviral treatment. The prevalence of abnormal arterial stiffness was 32.3% in the HIV-positive group, in contrast to 24.6% in the HIV-negative group, and this was significantly different after adjustment for gender and age (p = 0.02). HIV was independently associated with abnormal stiffness following regression analysis adjusted for age, gender, hypertension, BMI, DM and dyslipidemia (OR 1.64, 95% CI 1.13- 2.39). Conclusion: Abnormal arterial stiffness is more common in HIV-positive individuals than in HIV-negative individuals despite a lower burden of cardiovascular risk factors in the HIV-positive group. HIV is independently related to abnormal arterial stiffness. More research is needed to identify HIV- related factors that contribute to arterial stiffness in order to develop targeted prevention and treatment strategies. Awaiting these results we emphasize the need for screening and treatment of well-known cardiovascular risk factors.

2020 (Figure1). Among people living with HIV, the predicted 2017 prevalence of hypertension is 46%. In the total 2017 adult population, a projected 52% has either hypertension, HIV, or both. Conclusion: The model projects a continued increase of a substantial dual burden of HIV and hypertension both at the population and individual level. In the absence of more recent data on the co-burden of disease, these estimates of prevalent and incident cases approximate the dynamics of both conditions and may be used to address the gap in data.

Poster Abstracts

718 CORONARY ARTERY CALCIFICATION IN VIROLOGICALLY SUPPRESSED AGING HIV-INFECTED THAIS Pairoj Chattranukulchai 1 , Monravee Tumkosit 1 , Sarawut Siwamogsatham 1 , Tanakorn Apornpong 2 , Sivaporn Gatechompol 2 , Akarin Hiransuthikul 2 , Win M. Han 2 , Stephen J. Kerr 2 , Opass Putcharoen 1 , Smonporn Boonyaratavej 1 , Kiat Ruxrungtham 1 , Praphan Phanuphak 2 , Anchalee Avihingsanon 2 1 Chulalongkorn University, Bangkok, Thailand, 2 HIV–NAT, Thai Red Cross AIDS Research Centre, Bangkok, Thailand Background: Coronary artery calcification (CAC) is a known surrogate marker for coronary atherosclerosis, and is significantly related to future major cardiovascular events (MACE) and mortality. HIV-infected patients are at risk of cardiovascular disease (CVDs) but there is scarce data in Asian adults. We determined the prevalence and factors associated with subclinical atherosclerosis by CAC among HIV-infected adults from Thailand. Methods: HIV infected subjects aged >50 years who received HIV care in a prospective long term cohort at HIV-NAT, Thailand were enrolled. Subjects with a history of MACE were excluded. CAC was measured by cardiac multidetector row computed tomography (MDCT). All MDCT scans were read by an experienced radiologist blinded from patient care. Subclinical atherosclerosis was defined as CAC >0. Liver fibrosis was assessed by fibroscan. Results: 316 subjects (60.8%male, median age 54 years, 13.3% /21.5% current/former smokers, 97% virally suppressed) were enrolled. The median duration of ART was 16 years and 38%were on boosted PIs. Median overall CAC was 35.4 (IQR 6.9-130.1) and 46.8% had CAC>0. The CAC score category frequencies were 1-10 (minimal CAD:13.6%), 11-99(mild: 19.3%), 100- 399(moderate: 8.5%) and >400 (severe: 5.4%). Compared to CAC=0, CAC score >0 group had significantly higher traditional risk factors (older, male sex, higher ASCVD risk score (9.0% vs 4.1%), diabetes (25.7% vs 8.6%), hypertension (52.7 % vs 28.6%), BMI>25 kg/m2, waist/hip ratio, fasting glucose, fasting glucose >100 mg/dl, low HDL, blood pressure > 130/85 mmHg), NNRTI use and higher liver fibrosis scores. In a multivariate regression model, older age (aOR 1.07, 95%CI 1.01-1.14, p=0.02), male sex (aOR 4.03, 95%CI 1.88-8.64, p<0.001), hypertension (aOR 1.79, 95%CI 1.01-3.18, p=0.046), and fibroscan >7.2 KPa (aOR 2.15,95% CI 1.14-4.05), p=0.02) were independently associated with the CAC score >0 after adjusting for smoking history, diabetes, current ART regimen, statin use and BMI. Other HIV related factors such as CD4 count and ART duration, were not associated with CAC score>0. Conclusion: Despite low prevalence of current/former smokers, subclinical atherosclerosis among well suppressed HIV-infected Aging Thais was relatively high (46.8%). In addition to traditional CVD risk factors, liver fibrosis was significantly associated with subclinical atherosclerosis. Strategies to prevent future MACE such as statins, are warranted.

717 HYPERTENSION AND HIV AS COMORBIDITES IN SOUTH AFRICA: MODELING THE DUAL BURDEN Brianna Osetinsky 1 , Omar Galarraga 1 , Mark Lurie 1 , Stephen McGarvey 1 , Till Bärnighausen 2 , Sake de Vlas 3 , Jan A. Hontelez 3 1 Brown University, Providence, RI, USA, 2 Heidelberg University, Heidelberg, Germany, 3 Erasmus University Medical Center, Rotterdam, Netherlands Background: Non-communicable diseases (NCDs) are a significant and growing source of morbidity and mortality among HIV positive people in sub-Saharan Africa. Calls for greater research in healthcare priorities and direct investments in the treatment of the co-epidemics of NCDs and HIV are stymied by the lack of viable population level estimates of hypertension and the co- burden of disease. In this study we demonstrated that we can effectively model the prevalence and incidence of hypertension, as well as HIV and hypertension comorbidity among the adult population of KwaZulu-Natal (KZN), South Africa. Methods: We incorporated microsimulation of hypertension to the established agent-based Sexually Transmitted Diseases Simulation Model (STDSIM). Hypertension was modeled as a stochastic distribution calibrated to fit age- and sex-specific prevalence using Africa Health Research Institute’s HIV and hypertension surveillance data of adults over 18 years in KZN from 2003 and 2010, with 6,751 and 15,343 respondents respectively. It was parameterized as the initial development of hypertension along the life course of an individual as drawn from a convex combination of densities for the baseline, calibrated to the 2003 prevalence, and a growth factor calibrated from the 2010 data. Hypertension was added to the STDSIMmodel calibrated for HIV prevalence and incidence in KZN, and we conducted a sensitivity analysis of the fit for HIV and hypertension as comorbidities. Results: The model recreates the adult population increase of hypertension from 21% in 2003 to 33% in 2010. We predict a 2017 prevalence of 42%, with 595 new cases per 10,000 people. The prevalence of HIV and hypertension as comorbidities from the surveillance estimates is replicated with an increase from 3.5% in 2003 to 8% in 2010. The predicted 2017 prevalence of both as comorbidities is 10%, with 150 new cases per 10,000 people. There are substantial temporal increases in these conditions from the 2003 baseline to

CROI 2018 267