CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

Background: Chronic inflammation associated with HIV infection can result in increased risk for cardiovascular disease, and the baseline risk of cardiovascular- related mortality is up to 50% higher for HIV+ compared with HIV- persons. Indications for statin therapy are similar among HIV infected and uninfected individuals. Adherence to guideline-based care is critical and has been found deficient across multiple cohorts. Methods: The U.S. Military HIV Natural History Study (NHS, RV168 database) is an ongoing cohort comprised of Department of Defense beneficiaries. We conducted a retrospective cross-sectional analysis of adherence to guidelines for statin therapy, including subjects aged 21-75 whose most recent study visit was between October 2015-September 2016. To determine statin eligibility, we used the American College of Cardiology/American Heart Association 2013 atherosclerotic cardiovascular disease (ASCVD) management guidelines and included subjects who had all necessary data elements for analysis. Results: Selected baseline demographics of the cohort (n=1,066) included median age 47 (Interquartile Range [IQR] 34-55), male (95%), white race (40%), African-American race (45%), smokers (16%) and diabetic (12%). Statin eligibility was noted by having a previous cardiovascular event in 91 (8.5%) subjects, low-density lipoprotein (LDL) levels greater than 190 in 6 (0.6%), qualifying diabetics in 84 (7.9%), and ASCVD 10 year risk >7.5% in 303 (28.4%). In total, 342 (32%) patients met at least 1 criterion for therapy, and of those, 188 (55% of those eligible) were currently prescribed a statin. Among diabetics, 58% of eligible subjects were receiving statin therapy. Individuals receiving statin therapy tended to be older (median age 56 vs 42, p <0.001), white (52% vs 35%, p <0.001), and more likely to currently be on a protease inhibitor (57% vs 32%, p <0.001). Conclusion: We found significant discrepancies between ASCVD guidelines and primary care management of HIV+ persons in the military health system, and racial disparities persisted even in this single-payer network. Despite wide acceptance, poor adherence to the 2013 guidelines remains common in the management of HIV+ persons. This persists despite easy determination of statin eligibility through use of the ASCVD risk calculators. Improved adherence to ASCVD guidelines will be critical to minimize risk of cardiovascular disease in the aging HIV population. Marco Gelpi 1 , Shoaib Afzal 2 , Ashley Roen 3 , Amanda Mocroft 3 , Anne-Mette Lebech 1 , Birgitte Lindegaard 1 , Klaus F. Kofoed 1 , Børge Nordestgaard 2 , Jens D. Lundgren 1 , Susanne D. Nielsen 1 1 Rigshospitalet, Copenhagen, Denmark, 2 Copenhagen University Hospital, Copenhagen, Denmark, 3 University College London, London, UK Background: People living with HIV (PLWH) have increased risk of cardiovascular disease (CVD). Low levels of adiponectin have been linked to atherosclerosis, lipoprotein metabolism disorders, and unfavorable changes in LDL particle size in HIV-negative individuals. High levels of small dense LDL (sdLDL) together with high triglycerides (Tgl) and low HDL define the “atherogenic dyslipidemia” (AD) phenotype, which is a better predictor of CVD events than LDL alone. In this study we aimed to assess possible associations between HIV infection and adiponectin and AD. Methods: 1,099 PLWH from the Copenhagen Co-morbidity in HIV infection (COCOMO) study and 12,161 controls from the Copenhagen General Population Study were recruited. Associations between HIV infection and adiponectin, and AD were explored by uni- and multivariable logistic regression analyses. The model used to assess predictors of low adiponectin was adjusted for HIV infection, age, sex, smoking, BMI, lipid lowering therapy, and physical activity (Model 1). When assessing predictors of AD, low adiponectin was added to Model 1. When assessing HIV-specific predictors of low adiponectin and AD, CD4 nadir, current CD4 count and viral load, duration of HIV infection and cART, and hepatitis C coinfection were added to the models. Key variables are defined in Fig 1. Results: PLWH were younger (50.1 vs 52.2, p <.001), with a higher proportion of males (85.3% vs 81.4%, p .001) compared to uninfected controls. Furthermore, use of lipid lowering therapy was higher in PLWH (14.0% vs 10.6%, p .001). PLWH had lower levels of adiponectin (11.6 vs 12.2 ug/ml, p .014) and higher prevalence of both low adiponectin (2.0% vs 1.1%, p .019) and AD (13.4% vs 8.7%, p <.001) compared to uninfected controls. HIV infection was associated with higher odds of low adiponectin and AD (univariable models: OR 1.81, CI 1.05-2.92 and OR 1.61, CI 1.33-1.95, respectively; adjusted models: OR 2.08, CI 1.16-3.51 and OR 1.67, CI 1.31 -2.12, respectively). Predictors of AD are depicted

in Fig. 1. In PLWH, longer cART duration was associated with low adiponectin and presence of AD (OR per year 1.31, CI 1.10-1.56 and OR 1.09, CI 1.01-1.71, respectively, in adjusted models). Conclusion: HIV infection was associated with higher risk of low adiponectin and AD. Low adiponectin and longer duration of ART were strongly associated with the presence of AD in PLWH. These findings suggest perturbed adiponectin metabolism in PLWH, which is linked with higher risk of AD.

707 NON-CLASSICAL MONOCYTE AND CD4/CD8 RATIO PREDICTS ANKLE BRACHIAL INDEX IN TREATED HIV Dominic Chow , Nattawat Klomjit, Louie M. Gangcuangco, Scott A. Souza, Sean Terada, Lindsay Kohorn, Lishomwa C. Ndhlovu, Cecilia Shikuma University of Hawaii at Manoa, Honolulu, HI, USA Background: HIV-infected individuals on stable antiretroviral therapy are at a heighted risk of peripheral vascular disease (PVD). Chronic inflammation, monocyte (MO) subset and T cell activation have been associated with PVD in the general population. We investigated the relationship between MO, T cell activation, and measures of PVD using ankle-brachial index (ABI). Methods: Cross-sectional analysis of entry data from a cohort study of cardiovascular risk in HIV-infected subjects age > 40 years on stable antiretroviral therapy (ART) > 3 months. ABI was measured following the American Heart Association guidelines and was classified into 3 categories: low (ABI ≤0.90), borderline (ABI 0.91-0.99), normal (ABI 1.00-1.40). Banked PBMCs were phenotyped for MO subsets [classical MO (CD14++CD16-), intermediate (CD14++CD16+), non-classical (CD14low/+CD16++)] and for T cell activation (CD38+HLA-DR+CD8+) using multiparametric flow cytometry. CD4/CD8 ratio was calculated. Linear regression was performed between MO subsets, T cell activation, CD4/CD8 ratio and ABI. Multinomial logistic regression was conducted to determine predictors of ABI categories. Results: Among 160 subjects, median age was 51.0 years and 86%were virally suppressed. There were 5.6%, 22.5% and 71.96% of individuals who had low, borderline and normal ABI, respectively. CD4/CD8 ratio predicted ABI independent of age, gender, hypertension, diabetes, LDL cholesterol and current smoking (β=0.06, p=0.05). Multinomial logistic regression showed that increases in non-classical MO led to an increase in the odds (OR 1.05) of being in the low ABI group compared to the normal group (p = 0.02) and remained statistically significant after adjusting for age, hypertension, diabetes, current smoking, LDL, and CD4 Nadir (p=0.04). Similarly, an increase in non-classical MO led to 3.55 times the odds of being in the borderline ABI group compared to the normal group. No correlation was noted between CD8 T-cell activation and ABI. Conclusion: Higher numbers of non-classical monocytes was associated with low ABI category, while CD4/CD8 ratio predicted ABI. This suggests a potential role of non-classical monocytes and CD4/CD8 ratio in worsening the progression of PVD in HIV patients. 708 MICROBIAL-RELATED METABOLITE TMAO AND CAROTID ARTERY ATHEROSCLEROSIS IN HIV INFECTION Zhilei Shan 1 , David B. Hanna 2 , Clary Clish 3 , Justin M. Scott 3 , Robert Burk 2 , Sabina Haberlen 4 , Sanjiv J. Shah 5 , Joseph B. Margolick 4 , Wendy Post 4 , Howard Hodis 6 , Alan Landay 7 , Kathryn Anastos 2 , Robert C. Kaplan 2 , Qibin Qi 2 1 Harvard University, Cambridge, MA, USA, 2 Albert Einstein College of Medicine, Bronx, NY, USA, 3 Broad Institute of MIT and Harvard, Cambridge, MA, USA, 4 The Johns Hopkins University, Baltimore, MD, USA, 5 Northwestern University, Chicago, IL, USA, 6 University of Southern California, Los Angeles, CA, USA, 7 Rush University Medical Center, Chicago, IL, USA

Poster Abstracts

706 INCREASED RISK OF LOW ADIPONECTIN AND ATHEROGENIC DYSLIPIDEMIA IN HIV INFECTION

CROI 2018 263