CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

639

Stroke in HIV-Infected Patients in the Combination Antiretroviral Therapy Era Juan Berenguer 1 ; Alejandro Alvaro-Meca 2 ; Asuncion Diaz 3 ; Dariela Micheloud 1 ; Salvador Resino 3 1 Hosp General Universitario Gregorio Marañón, Madrid, Spain; 2 Univ Rey Juan Carlos, Alcorcón, Spain; 3 Inst de Salud Carlos III, Madrid, Spain Background: Both HIV and HCV infections have been associated with increased risk of stroke. We estimated incidence and mortality rates of stroke (hemorrhagic or ischemic) in HIV-infected (HIV+) patients (Pts.) in the combination antiretroviral therapy (cART) era, with particular attention to HIV/HCV-coinfected (HIV/HCV) Pts Methods: We reviewed the computerized data from patients in the Spanish Minimum Basic Data Set (MBDS), that includes information from Pts. discharged in almost 300 hospitals. Pts. were identified according to the following ICD-9-CM codes: HIV infection (042 or V08) with or without HCV infection (070.44, 070.54, 070.7x, or V02.62) with hemorrhagic (h) stroke (430-432) or ischemic (i) stroke (433-437). HBV infection (070.2x, 070.3x, or V02.61) was an exclusion criterion. Pts. were classified as HIV-monoinfected (HIV-Mono) or HIV/HCV. We estimated rates (events per 10,000 patient-years) in the period 1999-2011; time interval that was broken down into three periods: 1 st (1997-1999), 2 nd (2000-2003) and 3 rd (2004-2011). For the calculation or rates, the numerator was the number of events within each period. The denominator was the number of patient-years at risk within each period, for this purpose we estimated the number of HIV+, HIV/HCV, and HIV-Mono Pts. in each period. Results: h-stroke rates: In the 1 st period rates of h-stroke were higher for HIV-Mono Pts. than for HIV/HCV Pts. From the 1 st to the 2 nd period, rates of h-stroke decreased in HIV-Mono Pts. (from 16.0 to 5.5; P <0.001) and increased in HIV/HCV Pts. (from 1.3 to 7.6; P <0.001). In the 3 rd period, rates of h-stroke were higher for HIV/HCV Pts. than for HIV-Mono Pts. ( Figure 1A ). i-stroke rates: Similar trends were found for i-stroke. Rates decreased significantly from the 1 st to the 3 rd period in HIV-Mono Pts. (from 27.7 to 16.4; P <0.001), and increased significantly in HIV/HCV Pts. (from 1.8 to 12.6; P <0.001) ( Figure 1B ). Mortality rates: mortality rates for both h-stroke and i-stroke were higher for HIV-Mono Pts. than for HIV/HCV Pts. in the 1 st period; however, this trend was reversed by the 3 rd period ( Figure 1C & 1D ). The adjusted likelihood of death for h-stroke in the 3 rd period was higher for HIV/HCV Pts. than for HIV-Mono Pts. Conclusions: In the cART era, incidence and mortality rates of stroke decreased in HIV-Mono Pts. but increased steadily in HIV/HCV Pts.

Poster Abstracts

640 Incidental Carotid Plaque in HIV is AssociatedWith Subsequent Cerebrovascular Events Sumbal Janjua ; Pedro Staziaki; RichardTakx; Orla Hennessy; Michael Lu; MarkellaV. Zanni; Steven Grinspoon; Udo Hoffmann;Tomas G. Neilan Massachusetts General Hosp, Boston, MA, USA

Background: Coronary atherosclerotic plaque is increased in HIV-infected individuals and associated with subsequent cardiovascular events. Cerebrovascular (CV) events including stroke and transient ischemic attack (TIA) are also increased in HIV; however there are no data characterizing the prevalence, characteristics and prognostic associations of incidental carotid plaque in HIV. Methods: From a registry, we identified all HIV-infected individuals free of known CV disease who underwent a contrast neck CT from 2005 to 2014. Data collection, including CV and HIV-specific risk factors and image analysis, were performed by blinded independent teams. Image variables included the presence of carotid plaque and non-calcified plaque (NCP). The outcome of interest was a CV event (stroke, TIA) defined by ICD code and independently adjudicated. Association between plaque and events was determined using Cox proportional hazard models and compared with propensity-matched (age, gender, indication for CT, DM, HTN, HLD, smoking) HIV-uninfected controls. Results: 248 HIV-infected individuals free of prior CV disease (43+9 years, 24% female, 10% DM, 10% HTN, 33% smokers, 15% on statins, mean LDL 89+38mg/dl) were compared to 118 matched HIV-uninfected controls. The median duration of HIV was 16 yrs (10-21) and mean nadir CD4 count was 120 cells/mm 3 . At time of CT, the mean CD4 count was 308 cells/mm 3 , 79%were on ART, 20%were co-infected with HCV and 51% had an undetectable viral load (VL). On CT, HIV-infected individuals (vs. controls) had a higher prevalence of any carotid plaque (41% vs. 25%, P=0.005) and NCP (56% vs. 29%, P=0.03). Longer duration of HIV and ART were associated with increased plaque, while lower VL was associated with decreased plaque. Over a median follow-up of 3.1 yrs, 28 events occurred in HIV-infected individuals, rate of 4%/yr, as compared to 1%/yr in uninfected controls (P=0.013). Within HIV, the presence of carotid plaque (adjusted HR: 3.5, 1.5-8, P=0.002) and NCP (adjusted HR: 2.7, 1.3-5.8, P=0.01) were associated with an increased risk of subsequent

261

CROI 2016