
Methods: We conducted a targeted metabolomics analysis of plasma samples from 24 pairs of virally suppressed HIV-infected persons from the Duke HIV CVD Cohort, in a case-control format. Cases were defined as HIV-infected persons with documentation of acute myocardial infarction or diagnosis of severe coronary artery disease (CAD) by cardiac catheterization. HIV-infected cases without a history of CAD were matched 1:1 by age, race, sex, mean SBP and CD4 count. Only samples that preceded CAD diagnosis were included. Tandem flow injection mass spectrometry was used to quantify levels of a validated targeted metabolite panel (N=63). To reduce data complexity, we consolidated metabolites with collinear measures into interpretable factors using principal component analysis (PCA). Comparison of factor levels between cases and controls was done by student’s T-test.
Results:
48 HIV-infected persons were included in the analysis. Patient characteristics were: mean age 52.4 ± 8.1; 50% Black, 79% male; mean systolic BP 138mm Hg; mean BMI: 26.5kg/m2 for cases and 26.8 kg/m2 for controls; mean most recent CD4 573 cells/mm3 for cases, 476 cells/mm3 for controls. Based on metabolite levels, 14 factors were derived. The SCDA-derived factor (Ci4-DC/C4-DC, C5-DC, C6-DC, C10-OH/C8-DC, C12-OH/C10-DC, C6:1-DC/C8:1-OH, C8:1-DC acylcarnitines] was one of only two factors that revealed markedly higher plasma levels in cases than controls (p = 0.06).
Conclusion: HIV-infected persons with known major CAD events have significantly higher circulating levels of SCDA prior to their event than HIV-infected persons without known CAD. Although larger studies are ongoing for validation, SCDA may be an important biomarker in optimizing CVD risk stratification for HIV-infected persons.

N. L. Okeke,
None
D. Craig, None
M. Muehlbauer, None
O. Ilkayeva, None
S. Shah, None