CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

Multivariable linear regression model of the relationship between pericardial fat density and insulin resistance (logHOMA-IR)

Candidate variables were demographics, traditional cardiovascular risk factors, HIV disease factors, measures of total body and regional adiposity, and biomarkers of systemic inflammation and immune activation. Variables were selected by forward and backward selection, with a significance level of p<0.15 for inclusion in the model. Pericardial fat volume was forced into the final model. Parameter estimates are expressed per standard deviation increase for continuous variables or yes vs. no for dichotomous variables. HOMA-IR, Homeostatic model assessment of insulin resistance; HU, Hounsfield Units; HDL, high-density lipoprotein; BMI, body mass index. * = Non-normally distributed variables were log- transformed prior to analyses. Model R2 = 0.4383 Conclusions: Pericardial fat density, a simple measure obtained from non-contrast cardiac CT, is a novel marker of insulin resistance that is independent of pericardial fat volume and measures of total body adiposity. Further studies should examine the utility of this measure to characterize adipose tissue dysfunction in patients with chronic HIV infection 745 The Effect of Physical Activity on Cardiometabolic Health and Inflammation in HIV Sahera Dirajlal-Fargo 1 ; Allison R.Webel 2 ; Bruce Kinley 3 ; Danielle Labbato 3 ;Ying Jiang 4 ; Sara M. Debanne 4 ; Grace A. McComsey 1 1 Rainbow Babies and Children’s Hospital, Case Western Reserve University School of Medicine, Cleveland, OH, US; 2 Frances Payne Bolton School of Nursing Case Western Reserve University, Cleveland, OH, US; 3 University Hospitals Case Medical Center, Cleveland, OH, US; 4 Case Western Reserve University, Cleveland, OH, US Background: In HIV-uninfected populations, physical activity decreases mortality and reduces inflammation. Persistent inflammation is a potential cause for increased co-morbidities in HIV+ adults, yet the evidence examining the effect of physical activity on cardiometabolic health and inflammation in this population is limited. This analysis examines the relationship between physical activity and markers of cardiometabolic health and inflammation. Methods: We conducted a nested study within the SATURN-HIV trial in which 147 HIV+ adults on stable antiretroviral therapy (ART), with HIV-1 RNA< 1,000 copies/mL and LDL-cholesterol <130mg/dL were randomized to 10 mg daily rosuvastatin or placebo. Measures of physical activity, cardiometabolic health, and inflammation were assessed at baseline and 24 weeks later. Spearman correlations were used to explore relationships between physical activity, inflammation and CVD risk markers. Multivariable analyses were conducted to assess associations with physical activity. Results: Median age (Q1, Q3) was 46 (40.4, 52.7) years, 80%were male, 69%were African American and 46% on protease inhibitors. Baseline median physical activity was 49.5 (30.1, 67.9) minutes per week. Physical activity was significantly correlated with several markers of cardiometabolic health and inflammation (all p<0.05) (see table). After adjustment for factors known to affect cardiometabolic health and inflammation, physical exercise remained independently associated with markers of vascular disease (carotid bulb intima-media thickness; β =<-0.01, p=0.03) and endothelial function (brachial hyperemic velocity-time integral; β =<0.01, p=0.04). In addition, physical activity ( β =0.003, p<0.01) was independently associated with insulin resistance (HOMA-IR), even after adjustment for diabetes risk factors, HIV factors, body composition, and inflammation. After 24 weeks, median physical activity was 46.7 (31.4, 64.8) minutes per week and there was no difference between the statin and placebo groups. Changes in physical activity were correlated with changes in insulin resistance ( α =0.19, p=0.03).

Poster Abstracts

Conclusions: Physical activity is independently associated with insulin resistance, vascular disease and endothelial function, and may be a low-risk adjuvant to decreasing co- morbidities in HIV+ adults. Further studies should examine long-term effects of physical activity on markers of cardiometabolic health and inflammation in this population.

458

CROI 2015