Association of Viral Persistence and Atherosclerosis in Adults With Treated HIV Infection

JAMA Netw Open. 2020 Oct 1;3(10):e2018099. doi: 10.1001/jamanetworkopen.2020.18099.


Importance: Persons living with HIV (PLWH) have increased risk for cardiovascular disease, and inflammation is thought to contribute to this excess risk. Production of HIV during otherwise effective antiretroviral therapy (ART) has been associated with inflammation.

Objective: To determine whether higher levels of viral persistence are associated with atherosclerosis as assessed by changes in carotid artery intima-media thickness (IMT) over time.

Design, setting, and participants: In this cohort study, intima-media thickness, a validated marker of atherosclerosis, was assessed over time in a cohort of treated PLWH with viral suppression. Cell-associated HIV DNA and RNA and change in IMT, adjusted for demographics, cardiovascular risk factors, and HIV-related factors, were examined, as well as which factors were associated with viral persistence. One hundred fifty-two PLWH with undetectable viral loads for at least 6 months before study enrollment were recruited from HIV clinics affiliated with 2 hospitals in San Francisco, California, from January 1, 2003, to December 31, 2012. Data were analyzed from February 7, 2018, to May 12, 2020.

Exposures: Cell-associated HIV RNA and DNA were measured using enriched CD4+ T cells from cryopreserved peripheral blood mononuclear cells.

Main outcomes and measures: Carotid IMT was measured at baseline and the last visit, with a mean (SD) follow-up of 4.2 (2.7) years, using high-resolution B mode ultrasonography. The main study outcomes were baseline IMT, annual IMT progression, and incident plaque, defined as a focal region of carotid IMT of greater than 1.5 mm.

Results: The analysis included 152 PLWH (140 [92.1%] male; median age, 48.5 [interquartile range {IQR}, 43.3-53.7] years). Older age, smoking, medications for hypertension, higher low-density lipoprotein levels, and higher interleukin 6 levels were associated with higher baseline mean IMT, whereas cell-associated HIV DNA (estimate, -0.07% [95% CI, -6.1% to 6.4%]; P = .98), and HIV RNA levels (estimate, -0.8% [95% CI, -5.9% to 4.4%]; P = .75) were not. Levels of HIV RNA (0.017 [95% CI, 0.000-0.034] mm/y; P = .047) and HIV DNA (0.022 [95% CI, 0.001-0.044] mm/y; P = .042) were significantly associated with annual carotid artery IMT progression in unadjusted models only. Both HIV RNA (incidence risk ratio [IRR], 3.05 [95% CI, 1.49-6.27] per IQR; P = .002) and HIV DNA (IRR, 3.15 [95% CI, 1.51-6.57] per IQR; P = .002) were significantly associated with incident plaque, which remained significant after adjusting for demographics, cardiovascular risk factors, and HIV-related factors (IRR for HIV RNA, 4.05 [95% CI, 1.44-11.36] per IQR [P = .008]; IRR for HIV DNA, 3.35 [95% CI, 1.22-9.19] per IQR [P = .02]). Higher C-reactive protein levels were associated with higher cell-associated HIV RNA (estimate, 20.7% [95% CI, 0.9%-44.4%] per doubling; P = .04), whereas higher soluble CD14 levels were associated with HIV DNA (estimate, 18.6% [95% CI, 3.5%-35.8%] per 10% increase; P = .01). Higher soluble CD163 levels were associated with a higher HIV RNA:DNA ratio (difference, 63.8% [95% CI, 3.5%-159.4%]; P = .04).

Conclusions and relevance: These findings suggest that measurements of viral persistence in treated HIV disease are independently associated with incident carotid plaque development. The size and transcriptional activity of the HIV reservoir may be important contributors to HIV-associated atherosclerosis.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Atherosclerosis / etiology*
  • Biomarkers*
  • California
  • Carotid Intima-Media Thickness / statistics & numerical data*
  • Cohort Studies
  • Female
  • HIV Infections / complications*
  • HIV Infections / physiopathology*
  • HIV Infections / therapy*
  • Humans
  • Male
  • Middle Aged
  • Viral Load / statistics & numerical data*


  • Biomarkers