CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

636LB GUT MICROBIOTA, TRYPTOPHAN CATABOLISM AND ATHEROSCLEROSIS IN HIV INFECTION Qibin Qi 1 , Simin Hua 1 , David B. Hanna 1 , Clary Clish 2 , Jason Lazar 3 , Wendy Post 4 , Kathryn Anastos 1 , Robert Burk 1 , Robert C. Kaplan 1

1 Albert Einstein Coll of Med, Bronx, NY, USA, 2 Broad Inst of MIT and Harvard, Cambridge, MA, USA, 3 SUNY Downstate Med Cntr, Brooklyn, NY, USA, 4 Johns Hopkins Univ, Baltimore, MD, USA Background: Gut microbiota alteration and disturbed tryptophan catabolism have been observed in HIV infection. Yet, the relationship of tryptophan catabolismwith HIV- related cardiovascular disease (CVD) risk remains unknown. Methods: Plasma metabolomics profiling (including 155 soluble metabolites of known identity) was performed in 407 women (294 HIV+, 113 HIV-) in the Women’s Interagency HIV Study (WIHS) and 339 men (229 HIV+, 110 HIV-) in the Multicenter AIDS Cohort Study who underwent repeated B-mode carotid artery ultrasound imaging in 2004─2013. We examined associations of plasma tryptophan, kynurenic acid, a metabolite of tryptophan catabolism, and tryptophan-to-kynurenic acid (Kyn/Trp) ratio, a measure of tryptophan catabolism, with incident carotid plaque (focal intima-media thickness >1.5 mm) over 7 years (all participants without carotid plaque at baseline). High-resolution bacterial community profiling was performed in fecal samples in 49 WIHS women. Results: Over 7 years, 114 participants developed incident carotid plaque. Plasma tryptophan was significantly associated with decreased risk of incident carotid plaque (RR 0.75 [95% CI 0.64-0.88] per SD), while kynurenic acid (RR 1.34 [1.08-1.65] per SD) and Kyn/Trp ratio (RR 1.41 [1.22-1.64] per SD), were significantly associated with increased risk of incident carotid plaque (Figure 1A), after correction for multiple testing on 155 metabolites. Plasma tryptophan was significantly lower (P<0.001), while Kyn/Trp ratio was significantly higher (P=0.01) in HIV+ vs. HIV- persons (Figure 1B). Of note, tryptophan and related metabolites correlated with HIV viral load and specific inflammation/immune activation markers (e.g., sCD14, T-cell activation surface markers) rather than global inflammation markers or traditional CVD risk factors (Figure 1C). In addition, we identified that gut microbes enriched in Proteobacteria phylumwere significantly correlated with decreased plasma tryptophan and increased plasma kynurenic acid and Kyn/Trp ratio, after correction for multiple testing on 193 taxa. Conclusion: Our data indicate that disturbed tryptophan catabolism and related metabolites, correlating with HIV infection parameters, specific inflammation/immune activation markers, and gut microbiota dysbiosis, may play a role in the progression of HIV-related atherosclerosis. Our study suggests a potential modifiable target for the prevention and intervention of CVD in people with HIV, though further investigations are needed. 637 INTEGRATING CARDIOVASCULAR DISEASE RISK-FACTOR SCREENING IN HIV SERVICES IN SWAZILAND Miriam Rabkin 1 , Anton M. Palma 1 , Margaret L. McNairy 1 , Samkelo Simelane 2 , Averie B. Gachuhi 1 , Raymond A. Bitchong 3 , Harriet Nuwagaba-Biribonwoha 1 , Pido Bongomin 2 , Velephi Okello 4 , Wafaa M. El-Sadr 1 1 ICAP at Columbia Univ, New York, NY, USA, 2 ICAP Swaziland, Mbabane, Swaziland, 3 Raleigh Fitkin Memorial Hosp, Manzini, Swaziland, 4 Swaziland Ministry of Hlth, Mbabane, Swaziland Background: Cardiovascular disease risk factors (CVDRF) are prevalent among people living with HIV (PLWH), yet routine screening for CVDRF in HIV programs in sub-Saharan Africa is uncommon. We assessed the feasibility and acceptability of CVDRF screening of adults on antiretroviral therapy (ART) at a large HIV clinic in Swaziland. Methods: PLWH ≥40 years on ART were screened via blood pressure (BP) measurement, assessment of tobacco use, and point-of-care (POC) testing for HbA1c and total cholesterol (TC). Hypertension (HTN) was defined as SBP > 140 mmHg, DBP > 90 mmHg, and/or current use of anti-HTN medication. Diabetes mellitus (DM) was defined as HbA1c > 6.5% and/or current use of DMmedication. High cholesterol (HC) was defined as TC > 6.2 mmol/L. WHO/ISH risk stratification was used to assess 10-yr CVD risk. Screened and unscreened patients (pts) were observed using external-observer continuous observation time-motion methodology. We compared total visit time and time spent on HIV services among screened/unscreened pts using Wilcoxon rank-sum tests. We tracked the number of pts screened per week and explanations for variation in screening volume using a run chart. A subset of screened pts participated in exit interviews. Results: 1,826 ART pts were screened for CVDRF over a period of 42 weeks, of whom 1,063 (58%) were women. 532 (29%) had one CVDRF and 140 (8%) had ≥ 2 CVDRF. 407 pts (22%) had HTN, 167(9%) were smokers, 136 (7%) had HC, and 116 (6%) had DM. 36 pts (2%) had ≥ 10% 10-year CVD risk. Time-motion data from the visits of 172 pts (50 unscreened, 122 screened) showed that visit length for screened pts was significantly higher than for unscreened pts (median 15 min vs. 4 min, p <0.01) with no difference in the time spent providing HIV services (p = 0.57). The most time-consuming elements were POC testing (median 10 min, range 4-20) and BP measurement (median 2 min, range 0-3). On average, 43 pts were screened per week; screening volume varied markedly, and was limited by challenges related to staffing, space, systems, and supplies (Figure 1). 126 pts participated in exit interviews; all described the screening process as satisfactory and indicated that they would recommend it to others. Conclusion: While screening for CVDRF added substantial time to visits, with implications for staffing and wait time, routine screening of PLWH for CVDRF was feasible and appreciated by pts. It was also high-yield, identifying CVDRF in 37% of PLWH screened.

Poster and Themed Discussion Abstracts

CROI 2017 272