CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

factors identified included recent hospitalization (134, 42%), infection (133, 42%), or immobilization/bed rest (78, 25%) within the past 90 days, and current IV drug use (65, 20%). Eighty-seven (27%) had both hospitalization and infection in the past 90 days; 54 (17%) had both immobilization/bed rest and hospitalization. Conclusion: We conducted a robust adjudication process and examined predisposing factors for VTE among PLWH in a large North American cohort. PLWH with VTE were relatively young and most had at least one identified traditional pre-disposing risk factor. In addition, non-traditional risk factors, including IV drug use and recent infection, were common. Almost one-third of patients had detectable viral loads, and almost half were active smokers, suggesting potential modifiable pro-thrombotic risk factors.

odds ratio: 2.2 [95%CI: 1.8;2.9]) p<.001 (Table 1), but not with prolonged APTT (1.7 [95%CI: 0.9;3.0] p=.059). Apart from HIV, higher age, BMI > 25 and current smoking were associated with shortened APTT in adjusted analyses (Table 1). Conclusion: Prevalence of both shortened and prolonged APTT was higher in PLWH than in uninfected controls, and HIV was an independent predictor of shortened APTT. This finding may help explain some of the increased risk of thromboembolic diseases found in PLWH. The mechanisms and etiology should be further investigated in prospective studies with cardiovascular endpoints.

Poster Abstracts

639 DUAL DRUG-MEDIATED MECHANISM FOR NRTI-ASSOCIATED PLATELET HYPERREACTIVITY Kirk A. Taylor 1 , Erica Smyth 1 , Francesca Rauzi 1 , Maddalena Cerrone 2 , Akif A. Khawaja 1 , Brian Gazzard 2 , Mark Nelson 2 , Marta Boffito 2 , Michael Emerson 1 1 Imperial College London, London, UK, 2 Chelsea and Westminster NHS Foundation Trust, London, UK Background: Elevated cardiovascular (CV) risk in people living with HIV may be caused by multiple factors (e.g. lifestyle, HIV-associated inflammation and antiretroviral [ARV] use). Evidence suggests that some ARVs, including abacavir sulphate (ABC), increase CV risk. This is controversial and clinical studies are confounded by HIV status and previous ARV use. ABC, a guanosine analogue, has been suggested to interrupt nitric oxide (NO)–cGMP signalling, but the pharmacological mechanisms linking ARVs with platelets are unclear. Methods: Platelets were isolated from healthy, HIV-negative and ARV naïve volunteers. Aggregation and dynamic granule release were assessed by plate- based aggregometry and flow cytometry in the presence of clinically-relevant levels of ABC, tenofovir alafenamide fumarate (TAF) or tenofovir disoproxil fumarate (TDF). In vivo platelet aggregation was assessed in mice pre-treated with ARVs. Results: ABC, but not TAF or TDF, enhanced peak in vivo aggregation to 65U/Kg thrombin by 7.3±2.2% (P=0.02), whilst the area under the curve for 50µg/Kg collagen- and 0.4mg/Kg ADP-evoked responses were enhanced by 480.1±207.5 (P=0.04) and 160.0±45.8 (P=0.02), respectively. None of the ARVs tested affected in vitro platelet aggregation (P>0.05). Since isolated platelets do not generate NO, the pharmacological impact of ARVs was assessed in the presence of a NO donor, which reduced ADP-evoked aggregation from 48.9±5.2% to 13.0±3.4% (P<0.01). Under these conditions, the active metabolite of ABC (max aggregation: 27.3.1±7.3%, P=0.03), but not TAF/TDF (max aggregation: 13.1±5.4%, P=0.17), interrupted NO-mediated inhibition of platelet activation. Finally, we assessed platelet granule release in the absence of NO and found that ABC uniquely enhanced collagen-evoked alpha and dense granule release by 2.6±0.6- and 1.8±0.3-fold. There was no effect of any ARV on ADP- or thrombin-evoked granule release. Conclusion: Neither TAF nor TDF had any impact on platelet activation, suggesting that these drugs may not pharmacologically interact with platelets during HIV therapy or PrEP. ABC enhanced in vivo and in vitro platelet aggregation in the presence of NO. ABC also elevated platelet granule release in a collagen-specific and NO-independent manner. Together these data suggest

638 SHORTER ACTIVATED PARTIAL THROMBOPLASTIN TIME (APTT) IN PEOPLE LIVING WITH HIV Asbjoern Fink 1 , Andreas D. Knudsen 1 , Rebekka F. Thudium 1 , Jakob Hjorth von Stemann 1 , Shoaib Afzal 2 , Jens D. Lundgren 1 , Thomas Benfield 3 , Sisse R. Ostrowski 1 , Børge Nordestgaard 4 , Susanne D. Nielsen 1 1 Rigshospitalet, Copenhagen, Denmark, 2 Herlev and Gentofte Hospital, Copenhagen, Denmark, 3 Hvidovre Hospital, Hvidovre, Denmark, 4 University of Copenhagen, Copenhagen, Denmark Background: Altered coagulation in people living with HIV (PLWH) including higher d-dimer may contribute to the increased risk of cardiovascular disease (CVD) found in this population. While the extrinsic coagulation pathway has been studied extensively, less is known about the intrinsic pathway in PLWH. The activated partial thromboplastin time (APTT) is a measure of the intrinsic pathway and the overall speed at which blood clots and shortening of the APTT may increase the risk of thromboembolism. We aimed to investigate if the APTT in PLWH is altered compared to uninfected controls (UIC), and if HIV is an independent predictor of shorter APTT. Methods: A total of 985 PLWH from the Copenhagen Co-morbidity in HIV infection study and 2955 uninfected controls (UIC) from the Copenhagen General Population Study matched on age and sex were included in the study. CRP (high sensitivity assay, hsCRP) and APTT were measured in all participants. Shortened and prolonged APTT was defined as below and above the reference interval (25-37 sec.), respectively. We measured height and weight and information about use of tobacco and alcohol was collected using structured questionnaires. We calculated 95% binomial confidence intervals and used logistic regression models adjusted for HIV, age, sex, smoking status, alcohol, diabetes, BMI and hsCRP to investigate the association between HIV and altered APTT. Results: The median APTT was slightly shorter in PLWH than in UIC (27 vs 28 sec., p<.001). A higher proportion of PLWH compared to UIC had both shortened (14% [12;16] vs 8% [7;9] p =<.001) and prolonged APTT (2% [1;3] vs 1% [0.9;2] p =.044). Uni- and multivariable analyses are shown in Table 1. In multivariate analyses, HIV was independently associated with shortened APTT (adjusted

CROI 2019 242