Version May 2024
INTRODUCTION — For patients who have access to antiretroviral therapy (ART), the overall incidence of AIDS and death related to HIV infection has decreased dramatically [1]. Prior to 1996, the annual mortality among individuals with HIV exceeded 20 percent; this rate declined to <2 percent a decade later with the availability of effective treatment [2]. However, following the introduction of ART, new concerns arose about drug toxicities, including body fat maldistribution and metabolic abnormalities (eg, dyslipidemia, diabetes mellitus), and their potential association with cardiovascular disease [3-6].
This topic addresses the epidemiology of cardiovascular morbidity and mortality and cardiovascular risk factors in the setting of HIV infection and treatment with a focus on ischemic heart disease. The incidence of subclinical atherosclerosis and the pathogenesis of cardiovascular disease in patients with HIV are discussed elsewhere. (See "Pathogenesis and biomarkers of cardiovascular disease in patients with HIV".)
Management of dyslipidemia in patients with HIV, as well as HIV lipodystrophy, and HIV-associated cardiac complications, such as pericarditis, myocarditis, pulmonary hypertension, and cardiomyopathy, are also discussed separately. (See "Management of cardiovascular risk (including dyslipidemia) in patients with HIV" and "Cardiac and vascular disease in patients with HIV" and "Epidemiology, clinical manifestations, and diagnosis of HIV-associated lipodystrophy".)
EPIDEMIOLOGY — With more effective and widespread treatment of HIV in resource-rich settings, morbidity and mortality from non-AIDS-related events have surpassed those from AIDS-related events [7-9]. In particular, cardiovascular disease has emerged as an important cause of death in patients with HIV relative to the decreasing incidence of opportunistic disease. Several lines of evidence, from modeling of calculated cardiovascular risk to clinical studies evaluating such hard endpoints as stroke, myocardial infarctions (MIs), and sudden cardiac death have cumulatively supported this finding [10-16]. The data evaluating the rate of cardiovascular disease in patients with HIV compared with uninfected populations and its association with antiretroviral therapy (ART) and HIV disease state are discussed below.
Incidence compared with uninfected populations — Several studies have analyzed large clinical databases and cohorts, mainly in the United States, Canada, and Europe, but also in more resource-limited settings, to compare the incidence of cardiovascular disease in patients with and without HIV [17-27]. Although some of these studies are limited by low number of events, short follow-up, and incomplete assessments of other cardiac risk factors, they consistently report a 1.5- to 2-fold increase in the rate of cardiovascular events in individuals with HIV compared with control populations. As an example, in a systematic review of such studies that comprised nearly 800,000 people with HIV with 3.4 million person-years of follow-up, the incidence of cardiovascular disease was 62 events per 10,000 person-years, and the risk ratio compared with people without HIV was 2.16 (1.79 for myocardial infarction and 2.56 for stroke) [27]. From 1990 to 2015, the fraction of cardiovascular disease attributable to HIV infection increased from 0.36 to 0.92 percent; the highest attributable fractions were in sub-Saharan Africa, where HIV was estimated to account for more than 15 percent of the cardiovascular disease burden.
In the United States, one of the largest of these cohort studies evaluated California state-sponsored health insurance claims data, which included 28,513 patients with HIV and 3,054,696 patients without HIV [18]. The incidence of coronary heart disease (CHD) (including acute MI, other ischemic heart disease, and coronary atherosclerosis) in patients between the ages of 18 and 24 years was low overall but increased in those infected with HIV compared with the uninfected (relative risk [RR] 6.76, 95% CI 3.36–13.58 for men and 2.47, 95% CI 1.23–4.95, for women). The relative risk of CHD was the most increased in patients with HIV over the age of 45 years compared with uninfected populations.
Other analyses have found that the risk of MI alone is elevated in patients with HIV across a wide range of ages. In a cohort of 27,350 age-matched predominantly male veterans with HIV and 55,109 without HIV followed for a median of six years, HIV infection was associated with a greater risk of acute MI overall, even after adjusting for Framingham risk factors and other comorbidities (adjusted hazard ratio [HR] 1.48, 95% CI 1.27-1.72) [24]. When the cohort was analyzed by 10-year age groups, HIV infection was associated with a greater MI risk among patients aged 40 to 69 years but not among younger or older patients. Similarly, in a cohort from Kaiser Permanente Northern California, the age-adjusted rate of hospitalization for MI was 4.3 versus 2.9 events per 1000 person-years in men with and without HIV, respectively [17]. However, a subsequent update from the Kaiser study demonstrated a decline in rates of MI among the HIV group leading to a rate similar to that in the group without HIV [28]. A slightly higher overall MI rate was observed from two tertiary care hospitals in Massachusetts, with 11 versus 7 acute MI events per 1000 person-years in patients with and without HIV, respectively [19]. Even when adjusted for age, gender, race, hypertension, diabetes mellitus, and cholesterol, HIV infection remained associated with a significantly higher incidence of MI (RR 1.75, 95% CI 1.51–2.02). Among women, the relative risk for MI compared with controls was further increased. Similar findings were reported from cohorts in France and Denmark [20,21].
Effect of antiretroviral therapy — The degrees to which HIV infection itself, traditional cardiovascular risk factors, and ART each contribute to the elevated risk of cardiovascular disease in the population with HIV remain an active area of investigation. However, the preponderance of data suggests individual ART agents might differ in their contributions to the risk of cardiac events. Most studies that show an association between ART and cardiovascular events implicate older protease inhibitor (PI) use; however, this does not appear to be a class effect. (See 'Protease inhibitors' below.)
The Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) Study Group prospectively studied the incidence of MI in 23,468 patients with HIV residing in Australia, Europe, and the United States [29,30]. The overall incidence was low at 3.5 events per 1000 person-years, but the incidence increased with cumulative exposure to ART (age-adjusted RR 1.26, 95% CI 1.12–1.41, for each additional year of exposure). The association between ART and MI persisted in models that controlled for total cholesterol and triglyceride levels, suggesting that ART may negatively affect cardiovascular risk independent of its effects on lipid profile. (See 'Effect of antiretroviral therapy on lipid levels' below.)
Some, but not all, other subgroup analyses of retrospective database and cohort studies similarly demonstrate that the relative risk of cardiovascular disease compared with patients without HIV was greater in patients with HIV on ART than those not on ART [18,21,31]. One notable exception is the United States Veterans Study that retrospectively evaluated 36,766 veterans with HIV for risk of cardiovascular or cerebrovascular events [32]. Between 1995 and 2001, the overall mortality decreased by more than 75 percent. The rates of hospital admissions related to cardiovascular or cerebrovascular disease remained constant over time and were comparable between ART-treated and untreated patients. These findings persisted in a follow-up study that extended the evaluation period from 1993 to 2003 and included 41,213 patients [33]. Compared with no therapy, there was no significant difference in the rate of serious cardiovascular events after 24, 28, and 72 months of exposure to ART. The reasons for the discrepancy between this and other studies of cardiovascular risk with ART are unknown, although the veteran population may represent a group with higher cardiovascular risk at baseline.
Nevertheless, several studies suggest that higher CD4 cell counts and lower HIV RNA levels are associated with decreased MI risk [34-38], and in one trial, interruption of ART led to increased cardiovascular events [39], suggesting that treating HIV with ART clearly has beneficial effects on MI risk. (See 'Effect of antiretroviral therapy interruption' below and 'Effects of CD4 cell count and viral load' below.)
The full extent of cardiovascular risk in treated patients with HIV will become more discernible as many of the initial cohorts continue to follow patients over time. Since the population with HIV is relatively young and actual cardiovascular events are infrequent, many investigators have used surrogate measures of coronary artery disease, such as intima media thickening of the carotid and coronary arteries, as indirect endpoints. The effect of ART on these markers of cardiovascular disease is discussed elsewhere. (See "Pathogenesis and biomarkers of cardiovascular disease in patients with HIV", section on 'Subclinical atherosclerosis'.)
Protease inhibitors — Some, but not all, studies show an association between PIs and cardiovascular events [17,22,29,40,41]. In the studies that show a positive association, the PIs evaluated included agents such as lopinavir-ritonavir and indinavir, which are no longer considered preferred first-line agents in the United States (although lopinavir-ritonavir is still used in resource-limited settings). Furthermore, specific evaluation of individual PIs suggests that the increased risk of cardiovascular disease is not a class effect. More commonly used PIs (atazanavir and darunavir, each combined with low-dose ritonavir) may not have the increased risk of cardiovascular disease seen with other agents from this class [40].
Abacavir — Controversy persists over the potential association between abacavir and myocardial infarction despite considerable research on this topic. A consistent finding among those studies that do demonstrate an association is that the risk appears greatest among those with a greater burden of traditional risk factors, leading most experts to recommend avoiding abacavir in patients with high baseline risk. (See "Selecting antiretroviral regimens for treatment-naive persons with HIV-1: Patients with comorbid conditions", section on 'Cardiovascular disease'.)
Various studies, large observational cohorts, and analyses of treatment trials have suggested an increased cardiovascular risk with abacavir [21,42-46]. As an example, in an analysis of the D:A:D cohort evaluating 941 MI events over 367,559 person-years, there was an increased risk of MI among those receiving abacavir at the time (0.47 versus 0.21 per 100 person-years among those not receiving abacavir; adjusted risk ratio 1.98, 95% CI 1.72-2.29) [47]. Similarly, an analysis of North American cohorts that evaluated incident MIs between 2001 and 2013 among patients initiating ART identified an increased risk of both type 1 (spontaneous) and type 2 (secondary to an ischemic imbalance) MI with abacavir use within the past six months, even after adjusting for excess cardiovascular risk factors among abacavir users [48].
However, not all studies consistently show risk of MI with abacavir use [49-54]. A pooled analysis evaluated the results of 52 industry-sponsored studies in the drug manufacturer's database of HIV trials, which included 9502 patients who received abacavir and 4672 who took other antiretroviral medications [50]. Baseline demographics and clinical factors, including lipid and glucose values, were similar between both groups. Rates of MI were comparable among patients exposed or not to an abacavir-containing regimen.
The mechanism of action for the potential association remains uncertain [44]. Studies evaluating changes in inflammatory or coagulopathic biomarkers upon initiation or discontinuation of abacavir similarly have demonstrated conflicting results [55,56]. However, there is a growing body of evidence to suggest that a possible reversible impact of abacavir on platelet function may play an important role [44,57-59]. (See "Pathogenesis and biomarkers of cardiovascular disease in patients with HIV", section on 'Pathogenesis of vascular disease'.)
Effect of antiretroviral therapy interruption — Some studies suggest that longer duration of ART use, and particularly of some PIs, is associated with an increased risk of cardiovascular events. However, in the SMART study, a large international study of CD4 cell count-guided ART interruption, the 2720 patients randomly assigned to the drug interruption arm had an increased risk of major cardiovascular events compared with the 2752 patients in the continuous treatment arm [39]. After approximately 3700 person-years of follow-up (median 16 months), there were 79 fatal and nonfatal MIs; the event rate in the interruption group was 1.3 per 100 person-years compared with 0.8 per 100 person-years in the continuous therapy group.
Thus, despite some concerns that certain ART agents may be associated with cardiovascular risks, the discontinuation may result in an even greater risk of disease. One possible explanation for this finding is that suppression of HIV itself may be cardioprotective, possibly by reducing proinflammatory cytokines (eg, interleukin-6) that may play a role in arterial inflammation [29,60]. Additionally, discontinuation of ART may impact the lipid profile, leading to a reduction in high-density lipoprotein (HDL) cholesterol, a finding that was associated with MI in a case control analysis of SMART study participants [61]. (See "Pathogenesis and biomarkers of cardiovascular disease in patients with HIV", section on 'Inflammation'.)
Effects of CD4 cell count and viral load — Data on the effect of CD4 cell count and viral load on cardiovascular morbidity are not entirely consistent, although the bulk of evidence suggests that lower CD4 cell counts and higher HIV viral loads are associated with greater cardiovascular risk [34-38]. In a nested case-control study within the French Hospital Database on HIV, 289 individuals with HIV who had a prospectively recorded first-time MI were age-, gender-, and site-matched to 884 controls with HIV and no history of MI [35]. The following findings were noted:
●Cases had a lower mean CD4 cell count nadir than controls (135 versus 177 cells/mm3), but current CD4 counts were similar between the two groups.
●Lower nadir CD4 cell count and higher plasma viral load were associated with a small but statistically significant increased rate of MI, independent of exposure to ART and presence of traditional risk factors.
Furthermore, in a retrospective analysis of 3068 patients with HIV in the Netherlands, 57 patients experienced a cardiovascular event (MI, stroke, need for coronary artery intervention) during the nearly 11,000 patient years of follow-up [34]. Those patients with poor immunological recovery (ie, CD4 cell counts remained <200 cells/mm3 two years following initiation of ART) had higher estimated five-year cardiovascular event rates (4.7 percent) than those with better CD4 cell count improvements (2 to 2.6 percent). This associated cardiovascular risk was attenuated after adjusting for variables such as age, gender, and prior history of an event. Similarly, in a large cohort study of over 27,000 individuals with HIV, each 300-cell increase in current CD4 count was associated with a 25 percent decrease in MI risk, and higher HIV-1 RNA levels also predicted MI risk [38].
By contrast, in the large D:A:D cohort study, there was no association between nadir CD4 cell count (lowest recorded value) or peak HIV viral load and incidence of MI [29].
Other studies have suggested an association between low CD4 to CD8 ratio and cardiovascular risk [62,63].
HIV controllers — Studies have suggested that people who control HIV viremia without ART have higher levels of inflammation than both individuals whose HIV is suppressed on ART and individuals without HIV [64]. This observation has raised concern that HIV controllers may have higher risk for cardiovascular disease, but clinical evidence has not supported this. As an example, in an analysis of the Multicenter AIDS Cohort and the Women’s Interagency HIV Study, there were no differences in the burden of carotid plaque and measures of carotid intima media thickness among HIV controllers or long-term nonprogressors (individuals who maintain a CD4 cell count >500 cells/microL without ART) compared with people with HIV viremia [65]. These findings suggest that the inflammatory changes seen with immunologic control of viremia may not translate into a higher risk of cardiovascular disease. Further studies of these unique individuals are needed to further evaluate this issue.
TRADITIONAL RISK FACTORS — Overall, the classic cardiovascular risk factors of dyslipidemia, hypertension, diabetes, and smoking are common among populations with HIV, although the frequency of these comorbidities is not sufficient to explain the overall increased incidence of cardiovascular disease observed in the setting of HIV infection.
The presence of major risk factors for cardiovascular disease were evaluated among 1455 women and 931 men participating in two United States-based cohorts of patients with HIV (the Women's Interagency HIV Study [WIHS] and the Multicenter AIDS Cohort Study [MACS]) and compared with controls without HIV [66]. The 10-year risk of developing coronary heart disease (CHD) was estimated using the Framingham risk score equations (see "Cardiovascular disease risk assessment for primary prevention: Risk calculators"). This cross-sectional study demonstrated the following results:
●Compared with controls without HIV, individuals with HIV more frequently had low high-density lipoprotein (HDL) cholesterol and elevated triglycerides.
●Among men with HIV, 2 percent had moderate and 17 percent had high predicted CHD risk, in contrast with men without HIV (5 and 11 percent, respectively).
●Among women with HIV, 2 percent had moderate and 12 percent had high predicted CHD risk, similar to women without HIV.
●The risk of CHD was significantly lower in patients who were treatment-naïve compared with those patients taking protease inhibitors (PIs; odds ratio [OR] 0.57).
●Predictors of elevated risk also included low-income status and elevated body mass index (BMI).
A high prevalence of modifiable risk factors was present in the study participants. For example, 40 percent were current smokers. In addition, >40 percent of men with HIV and >60 percent of women with HIV met criteria for being overweight or obese. However, rates of obesity were less than in individuals without HIV.
A similar study in France had comparable findings. To calculate a predicted risk for myocardial infarction (MI) and CHD, a cohort of individuals with HIV using PIs (APROCO) was compared with a cohort of individuals without HIV (WHO MONICA) [67]. Compared with the 1038 age-matched individuals without HIV, the 274 patients with HIV were more likely to be thin, have elevated triglyceride levels, and low HDL cholesterol levels. The individuals with HIV were also more likely to smoke. The predicted five-year risk of CHD was greater among men and women with HIV compared with the WHO MONICA cohort (relative risk [RR] 1.20 and 1.59 for men and women, respectively). Of note, only 51 women with HIV were included in this study. More recent US data in patients with HIV have demonstrated continued upward trends in the prevalence of diabetes mellitus, hypertension, and hypercholesterolemia [68].
Dyslipidemia — Studies have consistently shown a high prevalence of dyslipidemia among patients with HIV, with and without antiretroviral therapy (ART). Abnormalities of lipid metabolism were originally reported in patients with advanced AIDS during the era prior to the introduction of combination ART [69,70]. Patients with HIV had lower HDL cholesterol and low-density lipoprotein (LDL) cholesterol levels, followed by higher plasma triglyceride levels prior to developing AIDS [71]. The degree of viremia may correlate to the amount of triglyceridemia [71]. In one prospective study of 50 men from MACS, significant declines in mean serum Total cholesterol, HDL cholesterol, and LDL cholesterol were observed for a mean of eight years after HIV infection when compared with pre-seroconversion levels [72]. Subsequent initiation of ART was associated with increases in total cholesterol and LDL cholesterol but with little change in HDL cholesterol.
It is not understood why HIV infection causes a shift in the lipid profile. One postulated mechanism links HIV inhibition of cholesterol efflux from human macrophages [73]. Accumulation of lipids within macrophages occurs via HIV impairment of the ATP-binding cassette transporter A1 (ABCA1)-dependent cholesterol efflux.
Effect of antiretroviral therapy on lipid levels — ART contributes to lipid abnormalities in patients with HIV [74,75]. There has been great interest in defining which antiretroviral agents may be the greatest contributors to dyslipidemia. It is also prudent to distinguish between agents within each class as important differences have emerged with newer agents. The impact of drugs on lipid profiles must be analyzed carefully since increases of certain fractions (such total cholesterol, LDL cholesterol, and triglycerides) are considered unfavorable, whereas increases in the HDL cholesterol fraction are considered cardioprotective.
Protease inhibitors — Dyslipidemia, including hypercholesterolemia and hypertriglyceridemia, develops with the use of some protease inhibitors (PIs), although the effect will vary with the individual PI [76]. Overall, the recommended PI regimens that require a lower dose of ritonavir for boosting (eg, darunavir) tend to have less unfavorable effects on the lipid profile [77].
Ritonavir — Hypertriglyceridemia is particularly common and may be severe with ritonavir [78-81]. Even the low dose of ritonavir that is used for "pharmacokinetic boosting" of concurrently used PIs has been associated with lipid abnormalities, although these are less pronounced in regimens that minimize the dose of ritonavir.
Darunavir-ritonavir — In the pooled subgroup analysis of the clinical trials of boosted darunavir (POWER 1 and POWER 2), 15 percent of patients developed elevated triglyceride levels (>8.4 mmol/L) compared with 7 percent in the comparator PI arms [82]. However, in a head-to-head comparison of darunavir-ritonavir (800 mg/100 mg every day) with lopinavir-ritonavir dosed either once or twice daily (ie, the ARTEMIS trial), darunavir-ritonavir patients had smaller median increases in triglycerides and total cholesterol than lopinavir-ritonavir patients [83]. Lipid levels remained below National Cholesterol Education Program cutoffs for darunavir-ritonavir-treated patients during 96 weeks of follow-up.
Data suggest that darunavir and atazanavir have clinically comparable effects on lipids, with similar small increases in triglycerides, total cholesterol, and LDL levels after 48 to 96 weeks of therapy [77,84,85].
Nevertheless, darunavir-ritonavir does not appear to be as lipid neutral as raltegravir [77]. (See 'Integrase inhibitors' below.)
Atazanavir-ritonavir — Atazanavir-ritonavir is associated with more favorable lipid profile compared with older-generation ritonavir-boosted PIs [86]. In the CASTLE study, atazanavir-ritonavir compared favorably with lopinavir-ritonavir in terms of virologic suppression with a backbone of tenofovir-emtricitabine among treatment-naïve patients [87]. At 48 weeks, the estimated differences in lipid and lipoprotein levels between the two arms (percent change from baseline) were: -9.5 (total cholesterol), -11.6 (non-HDL cholesterol), and -25.2 (triglycerides), all statistically significant and favoring atazanavir-ritonavir versus lopinavir-ritonavir.
Atazanavir was also associated with declines in cholesterol and triglyceride levels in treatment-experienced patients who switched to an atazanavir-containing regimen [88-90]. (See "Management of cardiovascular risk (including dyslipidemia) in patients with HIV", section on 'Treatment-experienced patients'.)
Nevertheless, atazanavir-ritonavir does not appear to be as lipid neutral as raltegravir [77]. (See 'Integrase inhibitors' below.)
Lopinavir-ritonavir — The addition of lopinavir 400 mg twice daily to ritonavir 100 mg twice daily in adults without HIV did not further exacerbate triglyceride and LDL levels but did cause additional increases of total cholesterol and HDL cholesterol without affecting the total cholesterol/HDL cholesterol ratio. [80].
Nonnucleoside reverse transcriptase inhibitors — Nonnucleoside reverse transcriptase inhibitor (NNRTI) use is associated with an increase in LDL cholesterol and total cholesterol levels compared with the lipid profiles of treatment-naïve individuals. However, this increase is counterbalanced by an increase in HDL cholesterol during treatment with NNRTI therapy [76]. The favorable overall lipid profile with NNRTI therapy is particularly pronounced with rilpivirine [91-95] and nevirapine (which is rarely used) [96].
Rilpivirine compares favorably with efavirenz with regards to effect on lipid levels [95]. As an example, in two large trials of antiretroviral-naïve subjects with HIV initiating an NNRTI-based regimen, total cholesterol, HDL cholesterol, LDL cholesterol, and triglyceride levels increased significantly more among patients randomly assigned to receive efavirenz compared with rilpivirine [93,94]. However, there was no difference in changes of the total cholesterol/HDL cholesterol ratio between the two treatments due to the greater rises in HDL cholesterol among those receiving efavirenz.
Etravirine was comparable to placebo in terms of any induced lipid abnormalities in the DUET 1 and 2 trials [97,98].
Doravirine, an investigational NNRTI, has a favorable lipid profile compared with both efavirenz and darunavir. As an example, in a randomized clinical trial of treatment-naïve patients, doravirine resulted in a more favorable mean change in LDL-cholesterol from baseline to week 48 (-4.5 mg/dL) compared with darunavir plus ritonavir (+9.9 mg/dL, with a mean difference -14.6 mg/dL, 95% CI -18.2 to -11.1) [99].
Nucleoside reverse transcriptase inhibitors — First-line nucleoside reverse transcriptase inhibitors (NRTIs), such as tenofovir and emtricitabine, do not have an adverse effect on lipid profiles. In an observational cohort of 2267 patients who started their first ART regimen, tenofovir, in combination with either emtricitabine or lamivudine, was associated with lower levels of total cholesterol, LDL cholesterol, and triglycerides compared with other NRTIs; however, there was no associated benefit on HDL cholesterol levels [100]. Similarly, in an open-label trial of 311 patients with viral suppression on a regimen of abacavir-lamivudine with a boosted PI, modestly more favorable changes in the total cholesterol (median change of -21 versus -3 mg/dL) and LDL cholesterol (median change of -7 versus 2 mg/dL) levels were observed among those randomly assigned to switch to a tenofovir-emtricitabine backbone compared with those assigned to continue the regimen [101]. Tenofovir plus emtricitabine appears to have a modest lipid-lowering effect compared with placebo [102].
The newer formulation of tenofovir, tenofovir alafenamide (TAF) has different lipid effects compared with tenofovir disoproxil fumarate (TDF). In a trial of treatment-naïve patients initiating ART, the single tablet regimen of elvitegravir-cobicistat-TAF-emtricitabine was associated with mildly but statically significantly greater increases in total cholesterol, LDL, and HDL compared with elvitegravir-cobicistat-TDF-emtricitabine, but the ratio of total cholesterol to HDL ratio remained unchanged for both regimens [103,104]. The clinical significance of this lipid difference is not yet known.
Integrase inhibitors — Overall, all agents in the integrase inhibitor class are associated with favorable lipid profiles.
A multicenter, randomized, double-blind study of raltegravir compared with efavirenz, both combined with tenofovir-lamivudine, was performed in 198 treatment-naïve patients with HIV [105]. Lipid profiles were basically unchanged in patients randomly assigned to the raltegravir arm. Over a longer period of follow-up (240 weeks), among those who were not on lipid-lowering therapy at entry, fewer raltegravir recipients initiated lipid-lowering therapy compared with efavirenz recipients (9 versus 34 percent) [106]. Increases in fasting triglycerides, total cholesterol, LDL cholesterol, and HDL cholesterol levels from baseline were significantly greater at week 240 in the efavirenz recipients compared with those who received raltegravir.
Raltegravir also appears to have favorable lipid effects compared with PIs. In a trial of 1800 treatment-naïve patients randomly assigned to a regimen containing ritonavir-boosted atazanavir, ritonavir-boosted darunavir, or raltegravir, total and LDL cholesterol and triglyceride levels all increased over 144 weeks with the PIs but not with raltegravir [77]. Furthermore, switching from a boosted PI to a raltegravir-based regimen has been associated with decreases in total cholesterol, non-HDL cholesterol, and triglycerides in several randomized controlled trials [107,108].
Elvitegravir, which is available as a component of a once-daily fixed dose combination that also includes cobicistat, emtricitabine, and tenofovir (EVG/COBI/FTC/TDF), has a lipid profile comparable to efavirenz and atazanavir [109]. The change from baseline in total cholesterol was greater in EVG/COBI/TDF/FTC compared with ATV/r-containing regimens, whereas triglyceride increases were greater in the ATV/r arm. There were no differences in the total cholesterol/HDL cholesterol ratio between the two treatments. By contrast, when compared with an efavirenz-based regimen, mean changes in total cholesterol, HDL cholesterol, and LDL cholesterol were less with EVG/COBI/TDF/FTC. The total cholesterol/HDL cholesterol ratio remained the same with both regimens.
Dolutegravir, a once-daily integrase inhibitor, resulted in no significant increases in lipid levels in treatment-naïve patients, similar to raltegravir in a head-to-head comparison [110]. A subsequent trial demonstrated reductions in total cholesterol, LDL cholesterol, and triglycerides in virologically controlled individuals who switched from a ritonavir-boosted, protease inhibitor-based regimen to a dolutegravir-based regimen [111].
Bictegravir, available in a fixed-dose combination with tenofovir alafenamide and emtricitabine, also has a favorable lipid profile when compared directly with dolutegravir-containing initial ART [112].
Insulin resistance and diabetes mellitus — Exposure to older ART agents was associated with insulin resistance and an increased incidence of diabetes mellitus [113,114]. In the Multicenter AIDS Cohort Study (MACS), for example, the rate of incident diabetes was 4.7 cases per 100 person-years among men with HIV using ART compared with 1.4 cases per 100 person-years among men without HIV during the four-year period of observation [113]. These metabolic abnormalities can develop independent of changes in body composition [115]. Other cohorts have reported incident diabetes rates among treated patients with HIV ranging from 4.4 per 1000 person years to 5 per 100 person-years [116-118], whereas other studies have not consistently demonstrated an increased risk of diabetes mellitus with treated HIV infection. These studies have differed in definition of diabetes mellitus, ART regimens commonly used, their ability of adjust for key confounding factors, and underlying prevalence of obesity.
Antiretroviral agents from several classes have been implicated in insulin resistance. In vitro models, animal models, studies of HIV-negative volunteers given PIs, and clinical trials of PIs have all demonstrated insulin resistance with these agents [119]. One possible explanation for the association is the observation that PIs can direct down-regulation of the glucose transporter isoform GLUT4, the major transporter of glucose into fat cells and cardiac and skeletal muscle [114]. Several studies have also noted an association between cumulative exposure to NRTIs and greater insulin resistance and diabetes risk compared with no NRTI exposure [120-122].
Most of these studies evaluated patients using older antiretroviral agents, such as stavudine and indinavir, which are no longer commonly used in resource-rich settings. It is unclear whether there is a significant risk of diabetes with newer antiretroviral drugs.
Hypertension — Patients with HIV may have higher rates of hypertension compared with those without HIV [19,123,124]. In a study that analyzed billing codes from two large hospitals in Massachusetts, the diagnosis of hypertension was recorded more frequently among patients with HIV (21 versus 16 percent in those without HIV infection) [19]. This higher prevalence of hypertension may be related to ART, as illustrated by findings of MACS, which followed 5578 men with HIV over 19 years and observed a significantly higher systolic blood pressure in those using ART for greater than five years [123]. This increase was not noted if ART was used for less than two years. However, an association between specific ART agents and the risk of incident hypertension has not been identified [125]. An association between hypertension and higher waist-hip ratio among patients with HIV has also been suggested [124,126].
Cigarette smoking — Smoking rates among individuals with HIV are considerably higher than those seen in the general population. As an example, in a survey of over 4000 adults with HIV in the United States, 42 percent were current smokers and 20 percent were former smokers, in contrast to 21 and 22 percent of the general population [127]. Other studies from various countries have reported rates of current or former smoking among patients with HIV ranging from 57 to 72 percent [19,29,128].
As in the general population, smoking is a major risk factor for cardiovascular disease in individuals with HIV and is associated with increased mortality [128-130]. One Danish cohort study suggested that the increase in cardiovascular risk associated with smoking was greater for individuals with HIV than for controls without HIV [129]. In another cohort study of 2921 patients with HIV and free access to ART who did not use injection drugs, current compared with never smoking was associated with increased mortality from both cardiovascular and all causes (RRs 4.3, 95% CI 1.4-13.1, and 4.4, 95% CI 3.0-6.7, respectively) [128]. This translated into a loss of 12.3 years of life among smokers with HIV compared with nonsmokers with HIV, more than double the 5.1 years of life lost attributable to HIV alone.
RISK MODIFIERS
Lipodystrophy syndrome — The lipodystrophy syndrome refers to abnormal fat redistribution with lipoatrophy and/or lipohypertrophy that can occur in patients with HIV on antiretroviral therapy (ART), which is sometimes associated with metabolic syndrome. The exact prevalence of this phenomenon is unclear, and there may be different risks for fat loss versus accumulation, suggesting discrete processes. Nevertheless, although patients with HIV and abnormal fat redistribution may not have overt obesity, it is often associated with significant metabolic abnormalities, including dyslipidemia and insulin resistance, and thus increased cardiovascular risk. As an example, 10-year coronary heart disease (CHD) risk estimates were compared in 91 patients with HIV and patients without HIV from the Framingham Offspring Study who were matched for age, gender, and body mass index (BMI) [131]. The risk estimates were significantly higher in patients with HIV and evidence of fat redistribution. However, when these patients were matched to controls based on waist-to-hip ratio measurements, the CHD risk was similar. Furthermore, patients with HIV and without evidence of fat redistribution did not have an increased CHD risk compared with controls. These data suggest that increased cardiovascular risk in patients with HIV may be associated with fat redistribution itself.
The epidemiology of lipodystrophy and its association with dyslipidemia and insulin resistance are discussed elsewhere. (See "Epidemiology, clinical manifestations, and diagnosis of HIV-associated lipodystrophy".)
Metabolic syndrome — Not surprisingly, since treatment of HIV infection is associated with central fat accumulation, dyslipidemia, and insulin resistance, studies that evaluate populations with HIV for the presence of metabolic syndrome have found high rates [132-134] (see "Metabolic syndrome (insulin resistance syndrome or syndrome X)", section on 'Definition'). In the INITIO trial, a study of 881 patients with HIV who initiated ART, the prevalence of metabolic syndrome at baseline was approximately 8 percent [116]. During three years of follow-up, the incidence of metabolic syndrome was 8 to 12 cases per 100 person-years, depending on the criteria used. Both baseline and incident metabolic syndrome were associated with the development of diabetes.
Hepatitis C virus infection — The effect of coinfection with hepatitis C virus (HCV) on cardiovascular disease risk in patients with HIV is uncertain, with studies yielding conflicting results. Several studies in men have suggested a greater risk of cardiovascular disease in the setting of HCV infection [135-137]. In the largest of these, the incidences of acute myocardial infarction (MI) and cerebrovascular events were greater among 6136 veterans with HIV-HCV coinfection (4.2 and 12.5 events per 1000 person-years, respectively) compared with those observed among 13,288 veterans with only HIV (3.4 and 11.1 events per 1000 person-years, respectively) [136]. However, when adjusted for diabetes mellitus, hypertension, age, and duration of ART, the increased risk of MI among those with coinfection was not statistically significant. Additionally, in an analysis of data prospectively collected from 32,395 individuals with HIV in the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study, the risk for development of MI was similar for patients with HIV-HCV coinfection compared to patients with HIV monoinfection after adjusting for potential confounders [138].
Controlling for potential confounders is important as HCV infection does appear to be associated with various factors that affect cardiovascular risk. There are a few studies that suggest that increased cholesterol levels may be blunted in patients coinfected with HCV [136,138-141]:
●In a retrospective study of 357 patients with HIV and 115 patients with HIV-HCV coinfection taking ART, the mean changes in cholesterol were significantly lower in patients with coinfection than in patients with HIV alone [140]. Eight percent of patients with HIV had to initiate lipid-lowering agents, while none of the patients with HIV-HCV coinfection started treatment.
●Another retrospective study evaluated the incidence and risk factors associated with the development of lipid abnormalities in 282 patients initiating ART [141]. No differences were noted between patients who were exposed to nonnucleoside reverse transcriptase inhibitors (NNRTIs) versus protease inhibitors (PIs); however, a protective effect against developing hypercholesterolemia was seen among patients who were HCV-infected.
The mechanism for these observations is unclear but it is likely to be related to impaired cholesterol synthesis in patients with HCV coinfection and may underestimate cardiovascular risk assessment. Nevertheless, any benefit to the lipid profile may be offset by modulation of other risk factors. As an example, in the large study of over 19,000 veterans with HIV described above, those with HCV coinfection were less likely to have abnormal lipid profiles but more likely to be smokers and to have hypertension and diabetes [136]. A small study suggested that effective treatment for HCV was associated with improvement in the levels of biomarkers associated with CVD risk [142].
Host genetic factors — As in the general population, certain genetic markers have been associated with a greater risk of cardiovascular events among patients with HIV, independent of traditional risk factors [143]. The clinical utility of testing for these polymorphisms to improve cardiovascular risk stratification has not yet been established. (See "Overview of possible risk factors for cardiovascular disease", section on 'Genetic markers'.)
Substance use — Cardiovascular disease has been associated with cocaine and marijuana use in patients with HIV.
Cocaine has been linked to coronary calcification [144]. In a study of 165 African American patients with HIV, significant coronary stenosis (≥50 percent) was detected in 15 percent of participants [145]. Those patients with long-term cocaine use (>15 years) had the highest prevalence of stenosis (42 percent).
Marijuana use has been linked to cardiovascular disease in patients with HIV. In a prospective cohort study of 558 men with HIV between the ages of 40 and 60 years, long-term heavy marijuana use (defined as daily or weekly use reported at ≥50 percent of biannual visits) was independently associated with cardiovascular events (odds ratio [OR] 2.5, 95% CI 1.3-5.1) [146]. Given the observational study design, these findings may be the result of confounding, although the analysis attempted to control for age, tobacco smoking, HIV viral load, and traditional cardiovascular risk factors.
SUMMARY AND RECOMMENDATIONS
●With more effective and widespread treatment of HIV, the overall incidence of AIDS or death related to infection by HIV has decreased dramatically. However, as patients are living longer, cardiovascular disease has emerged as an important cause of death in patients with HIV. (See 'Introduction' above.)
●The incidence of coronary artery disease, specifically myocardial infarction (MI), in patients with HIV is low overall, but it is approximately 1.5 times higher than that seen in patients without HIV matched for age and gender. (See 'Incidence compared with uninfected populations' above.)
●The degrees to which HIV infection itself, traditional cardiovascular risk factors, and antiretroviral therapy (ART) each contribute to the increased risk of cardiovascular disease in the population with HIV are unknown. Protease inhibitors recommended as first-line therapy do not appear to increase the risk of cardiovascular disease. Some data suggest an increased risk of MI with abacavir (a nucleoside reverse transcriptase inhibitor [NRTI]) use, although this finding has not been consistent across studies. (See 'Effect of antiretroviral therapy' above.)
●Despite possible contributions of some ART drugs on cardiovascular risk, discontinuation of ART is associated with an even greater risk of cardiovascular events, suggesting a protective effect of suppression of HIV replication. (See 'Effect of antiretroviral therapy interruption' above.)
●Patients with HIV tend to have lower high-density lipoprotein (HDL) cholesterol and low-density lipoprotein (LDL) cholesterol levels and greater triglyceride levels, independent of any exposure to ART. Protease inhibitor (PI) regimens that require a lower dose of ritonavir for boosting (eg, darunavir) tend to have less unfavorable effects on the lipid profile. Treatment with nonnucleoside reverse transcriptase inhibitors (NNRTIs) is associated with an increase in the HDL cholesterol. Rilpivirine is associated with a more favorable lipid profile than efavirenz. Overall, integrase inhibitors are also associated with a favorable lipid profile. (See 'Dyslipidemia' above.)
●There is a high prevalence of other traditional cardiovascular risk factors, such as hypertension and cigarette smoking, among this patient population. (See 'Hypertension' above and 'Cigarette smoking' above.)
●Some studies have demonstrated that coinfection with hepatitis C virus (HCV) may be associated with a lower risk of developing hypercholesterolemia during ART, although it does not appear that this reduces overall cardiovascular risk. In fact, higher rates of cardiovascular disease have been reported in patients with HCV coinfection. (See 'Hepatitis C virus infection' above.)
ACKNOWLEDGMENT — UpToDate gratefully acknowledges John G Bartlett, MD (deceased), who contributed as Section Editor on earlier versions of this topic and was a founding Editor-in-Chief for UpToDate in Infectious Diseases.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
