Lipid Impact on HIV Heart Disease and Antiretroviral Impact on Lipids

April 20, 2017

 1  |  2  |  3  |  Next > 

Abstract: Research implicates abnormal lipids in incident cardiovascular disease among people with HIV infection. Avoiding high total cholesterol could prevent 43% of myocardial infarctions, according to findings in one large HIV cohort. Studies also tie dyslipidemia to waning kidney function, declining cognition, sensory neuropathy, and erectile dysfunction in HIV-positive people. Abundant research links antiretroviral therapy to higher rates of dyslipidemia, but this association is evolving as less lipid-toxic protease inhibitors, nonnucleosides, and integrase inhibitors see wider use. Besides antiretrovirals, more than a dozen other medications can inflate lipids, including several antihypertensives, antidiabetics, and antiinflammatories often taken by people with HIV infection.

Strong evidence linking abnormal lipids to cardiovascular disease and mortality in the general population dates to the 1970s1 and grew deeper and wider in succeeding decades.2-4 Research also shows that lowering cholesterol has a clinical impact: Analysis of 10 prospective cohorts, three international studies, and 28 randomized trials figured that cutting total cholesterol about 10% lowered the risk of incident ischemic heart disease approximately 54% in men 40 years old.5 The impact on women was similar though based on slimmer data.

Impact of Lipids on Cardiovascular Disease With HIV

Smaller studies in HIV populations also tie unruly lipids to heart disease. Three studies found links between incident cardiovascular disease and high total cholesterol, high triglycerides, small dense low-density lipoprotein cholesterol (LDL-C), or apolipoprotein B.6-8

A case-control study matched 52 HIV-positive people who had incident cardiovascular disease from 1995 through 2009 to 104 HIV-positive people with no cardiovascular disease.6 Prevalence of dyslipidemia (use of lipid-lowering drugs or any abnormal lipid) was significantly higher in cases than controls (87% versus 72%, P = 0.05), and cases had significantly higher levels of total cholesterol (212 versus 188 mg/dL, P = 0.002) and LDL-C (130 versus 116 mg/dL, P = 0.04) in the 4 months before the cardiovascular event. Multivariable analysis identified high total cholesterol 4 months before the event as an independent predictor of the cardiovascular diagnosis (P = 0.0005).


In another case-control analysis, Swiss HIV Cohort Study (SHCS) investigators evaluated the impact of two highly atherogenic particles on coronary event risk from April 2000 through July 2008: small dense LDL-C and apolipoprotein B, which is the main protein in LDL-C.7 This study involved 98 antiretroviral-treated cases who had a first coronary event and 393 antiretroviral-treated controls matched for age, gender, and smoking status. Logistic regression models determined that every 1 mg/dL higher small dense LDL-C level raised the odds of a coronary event 7% (adjusted odds ratio [aOR] 1.07, 95% confidence interval [CI] 1.01 to 1.12), while every 10 mg/dL higher apolipoprotein B level upped the odds 17% (aOR 1.17, 95% CI 1.06 to 1.28). In both the LDL-C model and the apolipoprotein B model, longer duration of protease inhibitor (PI) therapy did not affect coronary event chances, but higher viral load and lower nadir CD4 count did.

A study of HIV-positive people in DAD cohorts isolated high triglycerides as an independent predictor of myocardial infarction (MI).8 The analysis involved 33,308 DAD participants who had 580 MIs from 1999 to 2008. An unadjusted analysis determined that every doubling of triglyceride level boosted the MI risk by two thirds (relative risk [RR] 1.67, 95% CI 1.54 to 1.80). After adjustment for latest total cholesterol and high-density lipoprotein cholesterol (HDL-C), every twice higher triglyceride tally raised MI risk by one third (aRR 1.33, 95% CI 1.21 to 1.45), meaning total cholesterol and HDL-C also contributed to MI risk. Even after additional adjustment for other cardiovascular risk factors, HIV factors, and treatment factors, every twice-higher triglyceride level independently boosted MI risk 11% (aRR 1.11, 95% CI 1.01 to 1.23).

Researchers studying 29,515 HIV-positive people in 7 U.S./Canadian NA-ACCORD cohorts determined that avoiding high total cholesterol would prevent 43% of MIs in this group.9 The NA-ACCORD team figured population-attributable fractions (PAFs) for various risk factors, defining those fractions as "the proportion of MIs that could be avoided in HIV-infected adults if all were unexposed to the modifiable risk factor of interest" and other risk factors remained unchanged. PAFs combine two metrics, prevalence of the risk factor in cohort members who had an MI and the adjusted hazard ratio for that risk factor. Elevated total cholesterol (>240 mg/dL) had the highest PAF, 43%, meaning this group would avoid 43% of MIs if no one ever had high total cholesterol (Figure 1). Avoiding hypertension would prevent 41% of MIs, according to this analysis, and never smoking would prevent 38%. But avoiding a CD4 below 200 cells/mm3 would avert only 10% of MIs, and avoiding a viral load at or above 400 copies/mL would avert only 6%.

Figure 1. Classic cardiovascular risk factors had a bigger impact in preventing myocardial infarction (MI) than did HIV-related risk factors in an analysis of population-attributable fraction (PAF) in a large HIV population. PAF is the "proportion of MIs that could be avoided in HIV-infected adults if all were unexposed to the modifiable risk factor of interest," if all other variables are held constant.9 (TC, total cholesterol; HTN; hypertension; CD4 count <200 cells/mm3; HIV RNA >400 copies/mL.)

(Illustration from Servier PowerPoint Image Bank.)

MIs Averted by Avoiding Key Risk Factors

Figure 1. MIs Averted by Avoiding Key Risk Factors

No randomized trials or long-term epidemiologic studies show that lowering harmful LDL-C, total cholesterol, or triglycerides with antilipid drugs prevents cardiovascular disease in people with HIV. In the general population, according to the National Lipid Association, evidence abounds that "reducing elevated levels of atherogenic cholesterol will lower atherosclerotic cardiovascular disease risk in proportion to the extent that atherogenic cholesterol is reduced."10 A Cholesterol Treatment Trialists' meta-analysis found that every 1 mmol/L (38.7 mg/dL) lower LDL-C achieved with statin therapy yielded a 22% drop in relative risk of cardiovascular events.11 A large trial in people with HIV is under way to confirm that this correlation holds true in people with HIV.

Lipids, Kidneys, Cognition, Neuropathy, and Sex

Other research in people with HIV ties out-of-line lipids to falling renal function, cognitive decline, sensory neuropathy, and even erectile dysfunction (Figure 2).

Figure 2. Research in people with HIV links abnormal lipids to cardiovascular disease, falling renal function, cognitive decline, sensory neuropathy, and perhaps erectile dysfunction.

(Illustration from Servier PowerPoint Image Bank.)

MIs Averted by Avoiding Key Risk Factors

Figure 2. Research in people with HIV links abnormal lipids to cardiovascular disease, falling renal function, cognitive decline, sensory neuropathy, and perhaps erectile dysfunction.

Two recent studies linked dyslipidemia to waning kidney function in people with HIV. A Multicenter AIDS Cohort Study (MACS) analysis involved 365 gay/bisexual men who started antiretroviral therapy (ART) between the end of 1995 and September 30, 2011.12 In adjusted models 3 years after ART began, men who gained more than 50 mg/dL in total cholesterol averaged a 2.6-mL/min yearly drop in estimated glomerular filtration rate (eGFR) (P < 0.001). In contrast, men whose total cholesterol changed less than 50 mg/dL had an average 1.4-mL/min yearly drop in eGFR (P < 0.001). Five years after ART began, the group that gained more than 50 mg/dL in total cholesterol had an average 2.4-mL/min yearly drop in eGFR (P = 0.008), compared with an average 0.1-mL/min yearly decline in men with a smaller total cholesterol change (P = 0.594).

A retrospective study in Japan tracked eGFR over 6 years in 661 people with HIV.13 The group averaged an annual eGFR decline of 2.01 mL/min, more than 6 times higher than the drop in age-matched controls without HIV. Dividing non-HDL-C levels into quartiles, the researchers used multivariate logistic regression to determine that the distribution of eGFR declines increased significantly across non-HDL-C quartiles (P = 0.0359). A non-HDL-C level above the cohort's median value boosted the odds of decreased eGFR more than 75% (aOR 1.77, 95% CI 1.07 to 3.00).

The MACS investigators suggest "dyslipidemia may promote renal injury through damage to glomerular capillary endothelial and mesangial cells as well as podocytes, resulting in glomerulosclerosis and interstitial fibrosis."12,14 At the same time, chronic kidney disease is an acknowledged risk factor for dyslipidemia,10 so causality can run in both directions. MACS researchers published two recent studies linking dyslipidemia to cognitive decline in men with HIV.15,16 A 1996-2010 analysis compared 273 HIV-positive men with 516 sociodemographically matched HIV-negative men, 81% of them white and 89% with more than a high-school education.15 Median age stood at 51 years, and all HIV-positive men had a baseline viral load below 400 copies/mL. Mixed-effects models tied higher total cholesterol and LDL-C to faster cognitive decline (P < 0.01), while higher HDL-C and statin use slowed that decline (P = 0.02 for both).

A study of 364 MACS men with HIV found that severity of HIV-associated neurocognitive disorder (HAND) remained stable through 4 years of follow-up in 77% of men, deteriorated in 13%, and improved in 10%.16 Logistic regression determined that a self-reported hypercholesterolemia diagnosis nearly tripled the odds of HAND progression (aOR 2.8, 95% CI 1.3 to 5.9, P = 0.01). That finding led the MACS team to suggest that lipid-mediated cerebrovascular disease could contribute to cognitive impairment in people with HIV.

High triglycerides are linked to sensory neuropathy in people with diabetes, and a study in 436 people with HIV made the same association.17 Among these people with a median age of 52 years, 75% were taking ART and 27% had clinically defined sensory neuropathy. Multivariable logistic regression figured that triglycerides in the highest group tertile versus the lowest (>244 versus <142 mg/dL) nearly tripled the odds of neuropathy (aOR 2.7, 95% CI 1.4 to 5.5).

Finally, HIV clinicians may want to tell men who complain of erectile dysfunction that abnormal lipids could contribute to this problem.18 The study involved 109 HIV-positive men who completed the International Index of Erectile Function (IIEF) survey at a Mexico City HIV clinic. The group averaged 39.9 years in age, had been diagnosed with HIV for an average 92.7 months, and had taking ART for an average 56.4 months. Almost two thirds, 65%, had IIEF-determined erectile dysfunction, and 33% had dyslipidemia (triglycerides >200 mg/dL, total cholesterol >240 mg/dL, or both). Dyslipidemia emerged as the sole variable linked to erectile dysfunction in univariate or multivariate analysis (adjusted hazard ratio [aHR] 3.75, 95% CI 1.25 to 11.28, P = 0.02). The authors note that dyslipidemia alters vascular relaxation, which can compromise blood flow and thus lead to erectile dysfunction.

 1  |  2  |  3  |  Next > 

Related Stories

Dyslipidemia With HIV: High Prevalence, Multiple Risks
Some Keys to Dyslipidemia Care in People With HIV
Screening for and Managing Dyslipidemia in People With HIV

This article was provided by The Center for AIDS Information & Advocacy. It is a part of the publication Research Initiative/Treatment Action!. Visit CFA's website to find out more about their activities and publications.

No comments have been made.

Add Your Comment:
(Please note: Your name and comment will be public, and may even show up in
Internet search results. Be careful when providing personal information! Before
adding your comment, please read's Comment Policy.)

Your Name:

Your Location:

(ex: San Francisco, CA)

Your Comment:

Characters remaining:


The content on this page is free of advertiser influence and was produced by our editorial team. See our content and advertising policies.