CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

388 Cardiovascular Risk Factors and the Rate of Change in Brain Structure in HIV Disease James T. Becker 1 ; Mikhail Popov 1 ; Lawrence Kingsley 1 ; Francine Barrington 2 ; Andrew J. Levine 3 ; Eileen Martin 4 ; Eric N. Miller 3 ; Cynthia Munro 5 ; Ann Ragin 6 ; Ned Sacktor 5 ; for the NeuropsychologyWorking Group of the Multicenter AIDS Cohort Study 1 Univ of Pittsburgh, Pittsburgh, PA, USA; 2 Johns Hopkins Univ, Baltimore, MD, USA; 3 Univ of California Los Angeles, Los Angeles, CA, USA; 4 Rush Univ Med Cntr, Chicago, IL, USA; 5 Johns Hopkins Univ Sch of Med, Baltimore, MD, USA; 6 Northwestern Univ, Chicago, IL, USA Background: Since the introduction of cART, factors not directly associated with HIV disease are now having a significant impact on brain structure and function. In particular, cardiovascular disease (CVD) is known to affect brain structure and performance on cognitive tests. The purpose of this analysis was to identify associations between CVD risk factors and change in brain structure over time. Methods: 122 men (56% HIV-infected) participating in the Multicenter AIDS Cohort Study (age: 55.8 + 4.9) underwent two brain MRI scans 3.2 years apart (T1 weighted magnetization-prepared rapid gradient-echo sequence). Gray matter, white matter, and CSF volumes (expressed as percent of total intracranial volume) were measured using FMRIB’s Automated Segmentation Tool, and an annualized rate of change was calculated. Correlation analyses identified hypertension, diabetes, urine protein/creatinine ratio (as an index of small vessel disease) and age as having significant associations with the outcome variables. Lipid levels and illicit drug use were not linked to rate of change. Linear regression models evaluated the associations between change in the three tissue class volumes and the significant predictor variables (from the unadjusted analysis), as well as CD4+ cell counts and plasma HIV RNA viral load (among infected men only). Results: After adjusting for age, race, and cohort of entry, HIV serostatus was unrelated to rate of change in any of the three brain tissue compartments. The rate of expansion of the CSF compartment was significantly faster among individuals with diabetes; no other CVD-related variable predicted brain volume change. Among the HIV-infected men only, nadir CD4+ cell count was associated with changes in gray and white matter volumes. CSF expansion was significantly greater among individuals in the most recent enrollment cohort, and among individuals with diabetes. Conclusions: In the cART era, CVD risk factors are at least as important as HIV-associated factors in predicting rate of change of brain tissue compartments. Finally, among HIV- infected individuals, factors associated with HIV treatment history are likely critical for predicting subsequent brain health. 389 NewMethod Measuring Intracranial Vessel Caliber Reveals Arterial Remodeling In HAND Paba M. De Alwis 1 ; Cristah Artrip 2 ;Winston Liu 1 ; Bryan Smith 1 ; Avindra Nath 1 ; Daniel S. Reich 1 ; Govind Nair 1 1 NIH, Bethesda, MD, USA; 2 Vanderbilt Univ, Nashville, TN, USA Background: In vivo MRI measurements of cerebrovascular caliber could offer insight into the pathology and spread of HIV in the CNS. Previous postmortem studies have shown a decrease in blood-vessel wall thickness in HIV infected individuals, indicating possible arterial remodeling. 1 Here, we measured the vessel caliber in HIV patients in vivo using high-resolution T2* weighted MRI and compared it to that frommatched control subjects. Methods: T2* weighted MRI was acquired at 0.55 mm isotropic resolution on a Philips 3T scanner with 8-channel head coil in 13 subjects with HIV-associated neurocognitive disorders (HAND, mean and standard deviation of age 49±9 years, 62%male), 43 HIV without HAND (HIV, 51±8 years, 56%male) and 23 socioeconomically matched controls without HIV (Control, 44±11 years, 57%male). Cross-sectional areas (averaged values from right and left arteries) of the A1 segment of anterior cerebral arteries and M1 branch of middle cerebral arteries were measured using a MATLAB program developed in house. One-way ANOVAs and two-tailed t-tests were used to assess group differences, and results are reported as mean ± standard deviation. Results: HAND group had significantly higher M1 cross-sectional areas (8.8±2.2 mm 2 ) than either the Control (7.7±1.6 mm 2 ) or HIV (7.5±1.5 mm 2 ; ANOVA p<<0.01) groups. However, such differences were not seen in A1 segment among the three cohorts (ANOVA p=0.9) (Figure). The technique also detected a difference between the M1 and A1 segment areas (7.8±1.7 mm 2 and 6.9±2.2 mm 2 respectively, p<<0.01, paired t-test). Men had consistently larger M1 areas than women in Control (8.1±1.6 mm 2 and 7.0±1.2 mm 2 respectively, p<0.05) and in HIV (8.0±1.6 mm 2 and 7.0±1.0 mm 2 , p<<0.01) groups. There were no observed correlations between any of the cross-sectional areas and age. Conclusions: We found that HIV patients with HAND had larger MCA cross-sectional areas than did HIV subjects without HAND or control subjects, which could be consistent with the reports of significantly thinner vessel walls seen in HIV patients post-mortem. The differences in vessel size between men and women and between the M1 and A1 branches show the ability of this method to detect subtle differences in arterial caliber. Another strength of this method is its ability to characterize vascular caliber in-vivo, which should allow for multiple longitudinal measurements over time. Reference: 1 Gutierrez et. al. Neuropathology 2013; 33, 256–263

Poster Abstracts

390

Isolated Cingulate Correlates With Reduced Executive Function in HIV Xiong Jiang 1 ; Manya Magnus 2 ; Chenglong Liu 1 ; Rebecca Barasky 2 ; CuiweiWang 1 ; MichaelW. Plankey 1 ; Seble Kassaye 3 ; David J. Moore 4 ; Ron Ellis 4 ; MaryYoung 1 1 Georgetown Univ Med Cntr, Washington, DC, USA; 2 George Washington Univ, Washington, DC, USA; 3 Georgetown Univ, Washington, DC, USA; 4 Univ of California San Diego, San Diego, CA, USA Background: Executive dysfunction is a common neurocognitive deficit among persons living with HIV (HIV+), and has been attributed to damage in fronto-striatal circuits. The specific neural mechanisms underlying executive dysfunction, however, remain to be elucidated. A recent study with a task-switching paradigm revealed that HIV+ older adults adapt less quickly to changing task demands than HIV- controls, and the behavioral impairments correlated with disrupted brain activation in the anterior cingulate cortex (ACC)

147

CROI 2016