Background: Chronic inflammation contributes to increased cardiovascular (CV) risk among HIV-infected individuals with low plasma viral RNA levels. Hyperemic velocity (HV), a measure of microvascular function, is the stimulus for brachial artery flow-mediated vasodilation (FMD), a measure of macrovascular function. HV is more predictive of CV events than FMD in the general population. The contribution of HIV persistence and traditional CV risk factors to microvascular and macrovascular dysfunction has not previously been evaluated. Methodology: We studied 71 HIV-infected individuals. Endothelial function was assessed using FMD. HV was measured as peak velocity-time integral (VTI) of the first complete velocity envelope after cuff release. D-dimer, hs-CRP, IL-6, sCD14, sCD163, and TNF-α were also measured. HIV persistence was determined using ultrasensitive plasma HIV RNA (lower limit of detection <0.3 copies/ mL, single copy assay), total proviral DNA in PBMCs (copies/ million CD4+ T-cells), and cell-associated RNA in PBMCs using TMA assay (signal/cutoff per million CD4+ T-cells). Multivariate Poisson regression was used to determine predictors of HV and FMD. Results: Most participants were male (96%) with a median age of 51 years. Approximately 30% had traditional CV risk factors. Based on conventional assays, 72% were treated and suppressed while 28% were untreated elite controllers. The median CD4+ T-cell count was 554 (IQR 314-741). The median plasma RNA using single copy assay was 0.8 copies/mL; 63% had <0.3 copies/mL plasma RNA. The median cell-associated RNA level was 5.6 (IQR 3.8-6.8) and the median proviral DNA level was 1.75 (IQR 1.1-2.5) log10 S/Co per million CD4+ T-cells, respectively. The median HV was 85 cm (IQR 57.8-100.7) and the median FMD was 4.8% (IQR 2.9-5.7). Worsened HV was independently associated with lower CD4+ T-cell count (RR 1.04 per 100 CD4+ T-cells, p<.001). In contrast, lower eGFR (RR 1.72, p=0.002), smoking (RR 1.25, p=0.039), and higher cell-associated RNA (RR 0.95, p=0.045) were associated with impaired FMD after adjustment for age and gender. Inflammatory and coagulation markers were not significantly associated with FMD or HV. Among treated and suppressed individuals only, lower eGFR and smoking remained significantly associated with impaired FMD while the association with cell-associated RNA weakened. Conclusions: Among HIV-infected subjects with low plasma RNA levels, a low CD4+ T-cell count was associated with impaired HV, while traditional risk factors and cell-associated HIV RNA levels were associated with worsened FMD. Our findings suggest that HIV may preferentially affect the microvasculature leading to endothelial dysfunction. Therapies targeted at further reduction of HIV persistence may also decrease CV risk.