Article Text

Ischaemic heart disease
Cardiovascular manifestations of HIV infection
  1. Jennifer E Ho,
  2. Priscilla Y Hsue
  1. Cardiology Division, San Francisco General Hospital, University of California, San Francisco, San Francisco, California, USA
  1. Dr Priscilla Y Hsue, San Francisco General Hospital, Cardiology Division, 1001 Potrero Ave, 5G1, UCSF Box 0846, San Francisco, CA 94110, USA; phsue{at}

Statistics from

At the end of 2003, an estimated 1 million people were living with human immunodeficiency virus (HIV) or acquired immunodeficiency syndrome (AIDS) in the USA.w1 Since the advent of highly active antiretroviral therapy (HAART) in 1996, HIV related mortality has decreased dramatically.w2 However, as patients are living longer, chronic health complications such as cardiovascular disease represent an increasing important health issue in this patient population. The aetiology of cardiovascular disease in HIV infection is not well established, and may be related to viral infection itself, the use of HAART, or altered immune responses. This review will focus on cardiovascular manifestations of HIV infection, with particular emphasis on coronary heart disease (CHD) and cardiovascular risk factors.



The first case reports of acute myocardial infarction (MI) in HIV infected patients on HAART were described in 1998.w3 w4 Since then, it has become increasingly clear that individuals with HIV infection are at high risk for cardiovascular events. The relative contributions of HIV infection versus potential adverse effects of HAART to CHD risk, however, remain unclear. Although the data from numerous studies on coronary events in HIV disease are conflicting (table 1), the majority of studies suggest that antiretroviral therapy increases CHD risk in HIV patients.

Table 1 Observational studies of coronary heart disease in HIV infection

The largest study showing no effect of HIV or HAART use on cardiovascular risk is the Veterans Affairs study, which showed no significant increase in cardiovascular or cerebrovascular events in patients treated with HAART or protease inhibitors (PIs) compared with age adjusted US population rates.1 In contrast, the largest study showing an increased incidence rate of MI was the French Hospital Database, where rates of MI were particularly increased in subjects with long term exposure to PIs.w5 Similarly, the Data Collection on Adverse Effects of Anti-HIV Drugs (DAD) study showed an increased incidence rate of MI with PI exposure >6 years, a finding that was partly explained by dyslipidaemia associated with PI use with longer exposure to HAART.3 There was no such association with non-nucleosidase reverse transcriptase inhibitors (NNRTIs).2 Of more recent concern has been the association of abacavir (a nucleosidase reverse transcriptase inhibitor (NRTI), which has also appeared to increase risk of MI in the DAD study (relative rate 1.89, p = 0.0001).4 This finding has also been corroborated in the more recent Strategies for Management of Antiretroviral Therapy (SMART) study, where abacavir was associated with increased CHD events when compared with didanosine.w16

Taken together, these observational studies suggest a 1.5–2 fold increased risk of CHD in HIV infected individuals when compared with uninfected controls, with absolute rates remaining low.5 The discrepancy in findings among observational studies with regard to the impact of PI use is likely due to inherent biases specific to observational studies, such as differences in study design, short follow-up times, lack of non-infected control groups, and unclear duration and specifics of HAART.w17 What appears to indicate an increased risk with long term PI use in observational studies is in contrast to findings from the SMART study. A total of 5472 HIV infected patients were randomised to a strategy of viral suppression (continuous HAART) versus drug conservation (intermittent HAART) and followed for an average of 16 months. Although not powered to examine cardiovascular events as a primary end point, patients assigned to continuous HAART had a decreased risk of fatal or non-fatal cardiovascular disease when compared with those on intermittent HAART (hazard ratio (HR) 1.6, 95% confidence interval (CI) 1.0 to 2.5; p = 0.05).6 Furthermore, the initiation of HAART in treatment naïve patients results in dramatic improvement in endothelial dysfunction,7 suggesting that in the short term, treatment of HIV infection may actually decrease cardiovascular risk. These studies suggest that, although long term treatment with HAART may lead to detrimental cardiovascular effects, there may be beneficial effects of HAART in the short term, providing additional evidence that HIV itself is mechanistically associated with increased CHD risk.

Clinical characteristics of CHD in HIV infected individuals

Clinical presentations of CHD in HIV infection tend to be distinct from CHD due to traditional risk factors. Demographically, HIV patients who develop acute coronary syndrome are more than a decade younger, with a mean age of 50 years, compared with non-infected controls. They are also more likely to be male, to be current smokers and to have low HDL (high density lipoprotein) cholesterol. As might be expected from these clinical features, HIV patients tend to have low TIMI risk scores, and tend to have single vessel rather than multiple vessel coronary artery disease.8 In general, HIV patients hospitalised with acute coronary syndrome have excellent immediate outcomes,w19–21 with successful percutaneous coronary intervention procedures.w20 However, when compared with non-infected controls, HIV patients tend to develop higher rates of future stent related complications. Prior studies comparing outcomes of percutaneous coronary interventions in HIV patients with non-infected controls have demonstrated a higher incidence of restenosis before the era of drug eluting stents (52% vs 14%, p = 0.006),8 and stent related complications requiring target vessel revascularisation (43% vs 11%, p = 0.02).w18 Preliminary data demonstrate that HIV patients treated with drug eluting stents had a 30% rate of major adverse cardiovascular events (Ren et al. Percutaneous coronary intervention in human immunodeficiency virus. Abstract presented at American Heart Association Scientific Sessions 2008, New Orleans, Louisiana, USA). There are no large surgical series of HIV patients who have undergone coronary artery bypass grafting; however, one study in 37 patients after cardiac surgery, of whom 29 had undergone bypass grafting, demonstrated 81% event-free survival at 3 years.w22 In a study performed at Kaiser, 19 HIV patients undergoing cardiothoracic surgery had fewer complications compared to uninfected controls (5.3% vs 26.3%, p = 0.07).w23

Pathogenesis of CHD in HIV infection

Atherosclerosis in HIV patients appears to be pathologically distinct from atherosclerosis in the general population. One autopsy study demonstrated accelerated atherosclerosis in young HIV-1 infected patients, with intermediate features between lesions in common CHD and transplant vasculopathy.w24 In another study, HIV associated atherosclerosis was characterised by diffuse and circumferential vessel involvement with unusual proliferation of smooth muscle cells mixed with abundant elastic fibres, forming endoluminal protrusions.w19 Interestingly, in an autopsy study of young to middle aged patients who died of advanced AIDS, HIV patients had three times greater odds of having a significant stenosis in the coronary arteries, even after adjustment for age and gender.w25

In the absence of HIV infection, chronic inflammation and T cell activation are thought to play a central role in the development of atherosclerosis.w26 The underlying mechanism of early atherosclerosis in HIV disease is not well understood, but similarly may be closely linked to increased vascular inflammation. This may be due to direct viral effects, the use of HAART and associated metabolic changes, or host immune responses. In the HIV population, certain HIV proteins have been shown to activate endothelial cells directly. For example, the HIV envelope protein gp-120 has been linked to higher endothelin-1 concentrations.w27 The potential role of HIV disease in the pathogenesis of early atherosclerosis is supported by the consistent observation that both CD4+ count and viral load influence cardiovascular disease. The CD4+ count nadir predicts subclinical carotid atherosclerosis in our group,9 and a low CD4+ count on HAART has been associated with increased risk of cardiovascular disease.w28–30 A recent study showed that low CD4+ count was independently associated with increased prevalence of carotid lesions.w30 Prior studies have also shown that higher viral loads correlate with endothelial dysfunction as measured by brachial artery flow mediated vasodilation.w31 w32

Several specific pathways of inflammation linking HIV infection to increased cardiovascular risk have been elucidated. Endothelial cells appear to play a central role in the pathogenesis of HIV associated atherosclerosis, as do procoagulant changes, fibrinolytic effects, and increased activation of platelets.w19 Elevations in concentrations of endothelial cell derived markers such as von Willebrand factor antigen have been reported in HIV disease, particularly in patients with a high viral burden or advanced disease.w33 Circulating concentrations of the adhesion molecules intercellular adhesion molecule-1 (ICAM-1) and vascular adhesion molecule-1 (VCAM-1) have also been shown to be elevated in HIV patients compared with non-infected controls, and were directly related to the degree of inflammation as assessed by soluble receptor type 2 for TNF-α (s-TNFR2).w34 Increased rates of inflammation as assessed by cytomegalovirus (CMV) specific T cell responses were associated with higher carotid intima–media thickness.w35 More recently, the SMART study found that untreated HIV infection was associated with high values of interleukin 6 (IL6) and D-dimer, and that these biomarkers were associated with all cause mortality and to a lesser extent cardiovascular disease.w36 IL6 is increased by infection and inflammation, and has been shown to be an independent predictor of mortality in unstable angina.w37 HIV infected individuals also have higher high sensitivity C reactive protein (hsCRP) values and T cell activation compared with uninfected individuals.w35


In addition to a direct viral effect, HIV infection and treatment with HAART may increase cardiovascular risk by influencing other traditional cardiovascular risk factors. A recent study demonstrated increased prevalence of traditional risk factors in HIV infected men without known CHD, leading to higher calculated 10 year Framingham risk scores when compared with non-infected controls (17% vs 11% 10 year risk of ⩾25%).w14

Lipid abnormalities

Changes in lipid profile with HIV infection appear to be twofold: in the early stages of HIV infection before treatment, the predominant changes appear to be hypertriglyceridaemia, low HDL and low LDL (low density lipoprotein) values with predominant small, dense LDL particles when compared with controls.10 w38 After initiation of HAART, however, LDL and total cholesterol concentrations appear to increase, with little change in HDL cholesterol, findings that are particularly associated with the use of PIs.w39 In the Swiss HIV Cohort study, hypercholesterolaemia and hypertriglyceridaemia were 1.7–2.3 times more common among patients on HAART containing PIs when compared with those without PIs.w40 The overall effect of HIV infection is thus toward an atherogenic lipid profile with significant reduction in HDL, and increase in triglycerides, oxidised LDL and small dense LDL.5 The prevalence of hyperlipidaemia in HIV patients is between 28–80% in different studies, with the majority of cases being hypertriglyceridaemia (40–80%).w41

Specific effects of HAART are important to consider. In general, most forms of HAART may increase LDL concentrations. With regard to hypertriglyceridaemia, the PI ritonavir has been associated with the most significant increases in triglycerides, in some cases causing extreme hypertriglyceridaemia exceeding 1000 mg/dl (11.3 mmol/l), although full doses of ritonavir are rarely used today.w42 Increased triglyceride concentrations are also seen with the use of ritonavir–saquinavir or ritonavir–lopinavir combinations. The PI with perhaps the least effect on triglyceride values is the relatively newly developed PI atazanavir.w43

The pathogenesis of lipid abnormalities associated with HIV infection and HAART are not well understood. Proposed mechanisms include increased hepatic lipogenesis, impaired clearance of lipids from the bloodstream, and potential effects of immunologic status.5

Lipodystrophy and metabolic syndrome

HIV associated lipodystrophy is characterised by uniform subcutaneous and peripheral fat loss, with relative preservation or increase in visceral fat, resulting in relative central adiposity, as well as fat accumulation in the neck and dorsocervical region.10 In prospective studies, these abnormalities are clinically evident in 20–35% of patients 12–24 months after starting HAART. Certain combinations of HAART are strongly associated with the development of lipoatrophy, in particular the use of PIs as well as the concomitant use of the two NRTIs, stavudine and didanosine.10 PIs may induce lipoatrophy by inhibiting sterol regulatory enhancer protein 1 (SREBP 1), which activates downstream transcription factors such as peroxisome proliferator activated receptor γ (PPAR-γ), whereas the effects of NRTIs may be due to mitochondrial injury within adipocytes.10 Of note, a recent study showed that use of the newer PI atazanavir for 48 weeks was not associated with abnormal fat redistribution or metabolic disturbances commonly seen in HIV associated lipodystrophy.w44

Lipodystrophy in HIV patients is commonly associated with a constellation of findings characteristic of the metabolic syndrome, including insulin resistance, impaired fasting glucose tolerance, elevated triglycerides, low HDL cholesterol, and hypertension.11 Metabolic syndrome appears to be highly prevalent among HIV patients. In a recent cross-sectional study of 710 HIV patients, 17% had metabolic syndrome, a finding which was independently associated with the use of stavudine and lopinavir/ritonavir.w45 Progression to metabolic syndrome is substantial in the first 3 years after initiation of HAART, and incident metabolic syndrome was associated with increased risk of CHD in a study of 88 HAART naïve patients who started on treatment (HR 2.73, 95% CI 1.07 to 6.96; p = 0.036).w46

Other risk factors

In addition to the above metabolic changes, other traditional cardiovascular risk factors also appear quite prevalent in the HIV infected population. Smoking is quite common among HIV patients, and in a French cohort comparing HIV patients aged 35–44 years on HAART to a population based cohort, the estimated attributable risks due to smoking were 65% for men, and 25% for women, respectively.w47 Similarly, the DAD study investigators found that smoking was the most powerful predictor of CHD in patients with HIV aside from an existing history of CHD.2

The prevalence of diabetes mellitus and hypertension also appear quite elevated when compared with the general population.w14 w48


Carotid artery intima–media thickness

In HIV uninfected patients, carotid artery intima–media thickness (IMT) as assessed with B mode ultrasound has been strongly correlated with coronary atherosclerosis, and is directly associated with increased risk of MI and stroke in older patients without known cardiovascular disease.w49 w50 There have been numerous studies using carotid IMT to assess the presence of subclinical atherosclerosis in HIV patients, of which the largest have been summarised in table 2.

Table 2 Carotid intima–media thickness in HIV patients: major studies

The effect of HAART and PI use in particular on cardiovascular risk is no more clear in carotid IMT studies as it was in observational studies, and studies demonstrating a correlation between carotid IMT and clinical outcomes are lacking in the HIV infected population. However, IMT studies have provided further insight into specific factors associated with worse cardiovascular risk. For example, our group showed that HIV infected patients had a higher baseline IMT with more rapid progression of IMT when compared with controls at 1 year. Predictors of progression included age, Latino race, and nadir CD4 count ⩽200 cells/μl.9 In another study, CMV specific T cell responses were independently associated with higher IMT, suggesting that accelerated atherosclerosis may be mediated by heightened CMV induced responses in HIV patients.w35 Kaplan et al showed that a low CD4 count was independently associated with increased prevalence of carotid lesions, suggesting that beyond traditional risk factors, CD4 count was the most robust risk factor for subclinical atherosclerosis in HIV patients.w30 In general, the impact of HIV and HAART on subclinical atherosclerosis is still incompletely understood; however, most studies appear to demonstrate premature atherosclerosis in the HIV infected population.

One of the current limitations of carotid IMT measurement is the lack of uniform methodology. Whereas some studies examine the common carotid,w52 others measure IMT at the carotid bifurcation region.w35 w51 Others still report the presence of carotid plaque within the imaged segment distinct from the underlying IMT measurement at a specified location.w30

Brachial artery flow mediated dilation

Endothelial dysfunction is thought to play a central role in the development and progression of atherosclerosis, and in non-HIV infected patients has been shown to predict future cardiovascular events.w60 w61 The hallmark of endothelial dysfunction is impaired endothelium dependent vasodilation, which can be non-invasively assessed using brachial artery flow mediated vasodilation (FMD).w62 Whereas carotid IMT is thought to reflect long term exposure to atherogenic factors, brachial artery FMD is a measure of current vascular function and short term exposures.

It is known that HIV infected patients have impaired endothelial function as assessed by FMD when compared with non-infected controls.w32 The mechanism of endothelial dysfunction in HIV disease is unclear. Previous observational studies have shown worse endothelial dysfunction with higher viral load,w31 w32 and others have demonstrated improved endothelial function in HAART treated compared with untreated individuals.w63 w64

However, the relative contribution of HAART to endothelial dysfunction in HIV infected individuals is complex. Data from numerous studies examining endothelial function and the use of HAART, in particular PI based regimens, are conflicting. Some studies suggest worse endothelial dysfunction with the use of PIs,12 w63 w64 while others do not support this phenomenon.w32 The potential effect of PIs may be specific to certain medications, as studies in non-infected controls demonstrated no change in brachial artery FMD on lopinavir–ritonavir,w65 but significantly decreased FMD after indinavir use.w66 w67 Adverse effects of PIs may be mediated via atherogenic lipid changes,12 although AZT and AZT + indinavir dramatically reduced endothelium dependent vasodilatation in an animal model, without changes in cholesterol.w68 Most recently, initiation of HAART in treatment naïve patients resulted in dramatic improvement in endothelial function both at 4 and 24 weeks of therapy, irrespective of whether a PI was used or not.7

Coronary artery calcium scoring

The quantification of coronary artery calcium (CAC) by electron beam computed tomography is a non-invasive marker of atherosclerosis that has been shown to predict coronary death and non-fatal MI in the non-infected population.w69 There are few studies examining CAC scores in HIV infected individuals. One study demonstrated elevated CAC >100 in 8.6% of men and 6.0% of women in a relatively young cohort, although no direct comparison was made to HIV uninfected controls.w56 More recently, data from the Multicentre AIDS Cohort Study showed that the prevalence of CAC was marginally increased among long term HAART users; however, the extent of CAC was actually reduced among HAART users compared with uninfected controls.w70


Currently, the treatment of CHD in HIV infected individuals should largely be guided by existing recommendations in uninfected patients, as clinical studies in HIV patients with CHD are limited. However, there are two aspects particular to HIV infected patients that deserve mention: the potential role of HAART with regard to cardiovascular disease, and the treatment of hyperlipidaemia in HIV disease, for which separate recommendations have been devised.


Recent evidence suggests that the risk of non-AIDS related mortality may exceed the risk of AIDS related mortality in individuals with CD4 counts >200 cells/μl.w71 Of particular concern are increased rates of coronary events and mortality compared with non-infected controls.3 9 w9 Despite existing evidence that long term HAART may have adverse effects, it is becoming clearer that uncontrolled HIV replication leads not only to increased cardiovascular risk but also other non-AIDS complications, as was shown in the SMART study.6 w29 w71 Current recommendations by the International AIDS Society USA Panel guidelines thus support HAART initiation for asymptomatic individuals at CD4 count <350 cells/μl, with “individualised” therapy at CD4+ count ⩾350 cells/μl.13 Whether earlier initiation of HAART at higher CD4+ counts in the course of HIV disease improves cardiovascular risk is not known; however, guidelines do support earlier initiation of HAART in the setting of high cardiovascular risk, in addition to other high risk clinical features (high viral loads >100 000 copies/ml, rapidly declining CD4 count >100/μl per year, active hepatitis B or C infections, or the presence of HIV associated nephropathy).14

The initial choice of HAART regimen is primarily targeted at viral suppression, although metabolic profiles of drugs should be considered in patients at high cardiovascular risk. While traditional HAART may be constrained by HIV resistance and medication tolerability, novel antiretroviral agents such as integrase inhibitors and viral entry inhibitors may in the future provide better options with regard to cardiovascular side effect profiles.15


The Infectious Disease Society of America (IDSA) and Adult AIDS Clinical Trials Group (AACTG) have developed specific guidelines for the evaluation and management of HAART related hyperlipidaemia.13 These recommendations are largely based on National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) guidelines, and advocate adjusting individual cholesterol targets to the underlying cardiovascular risk based on the Framingham predicted 10 year risk.w72 A general algorithm of treatment is summarised in fig 1.

Figure 1

Treatment of hyperlipidaemia in HIV infected individuals. HAART, highly active antiretroviral therapy; NCEP ATP III, National Cholesterol Education Program Adult Treatment Panel III; TG, triglyceride.

Currently there is no evidence to treat hyperlipidaemia in HIV patients to goals that are different from those for HIV uninfected patients. However, it is important to consider specific drug–drug interactions when initiating lipid lowering therapy in this patient population. Both PIs and NNRTIs can affect cytochrome P450 isoforms. In general, all PIs inhibit CYP3A4, with the highest level of inhibition with ritonavir, followed by indinavir, nelfinavir, amprenavir, and saquinavir. Delavirdine, an NNRTI, is also an inhibitor of CYP3A4, whereas nevirapine and efavirenz result in induction of the enzyme. Both simvastatin and lovastatin concentrations increase dramatically in the setting of PI use and have led to rhabdomyolysis,w73 thus are contraindicated in this setting. Atorvastatin concentrations appear to be increased to a lesser degree, and atorvastatin may be used at lower doses in HIV patients. Since pravastatin is not metabolised by CYP3A4, it is a first line agent for lowering LDL in HIV patients. Similarly, fluvastatin is metabolised by CYP2C9, and can be used as a second line agent.w73 Rosuvastatin has minimal P450 metabolism, although concentrations appear to be increased when used in combination with atazanavir/ritonavir and lopinavir/ritonavir, limiting doses to 10 mg in that setting.w74 w75

Current recommendations for the treatment of hyperlipidaemia in HIV patients include lifestyle modifications, and dietary and exercise interventions have been shown to decrease total cholesterol values by 11–25% in the HIV population. When lipid lowering therapy is indicated, pravastatin or atorvastatin are first line therapies for the treatment of elevated LDL cholesterol in patients taking any PI or delarvidine (fig 1). The recommended starting doses are pravastatin 20–40 mg daily and atorvastatin 10 mg daily. Fluvastatin 20–40 mg daily may be considered an alternative second line agent. The treatment of hypertriglyceridaemia is mainly achieved with fibrates (gemfibrozil 600 mg twice a day or micronised fenofibrate 54–160 mg daily) for triglyceride values exceeding 500 mg/dl (5.6 mmol/l). Niacin can cause insulin resistance, and is therefore not recommended as first line therapy with concurrent PI use or the presence of lipodystrophy. In general, bile sequestering resins are not recommended for use in HIV patients.13 Ezetemibe appears safe and effective when added to maximally tolerated doses of lipid lowering therapy,w76 and has modest lipid lowering activity when used alone in HIV patients.w77

What remains unknown is whether HIV should be considered as a separate cardiovascular risk factor, potentially lowering LDL thresholds for starting therapy. Furthermore, given potential anti-inflammatory and other pleiotropic actions, it is unclear whether statins will have additional benefits beyond lipid lowering in this patient population.

Modification of other risk factors

Clearly, traditional cardiovascular risk factors are enriched in the HIV population, such that modification of underlying risk factors is essential to the clinical care of HIV patients. Smoking cessation, lifestyle changes in diet and exercise, and the appropriate treatment of hypertension and diabetes mellitus are all measures which should be pursued aggressively in this population.


Although several multivariate models of calculating probability of CHD are used in the general population, none have been validated in the HIV infected population. The most widely used screening tool is the Framingham risk score, which appears to underestimate CHD risk in HIV patients who are smokers.16 HIV specific risk prediction models of CHD have been proposed, and one such tool from the DAD study group incorporated PI exposure with traditional risk factors and appeared to be reasonably accurate in preliminary studies.16

There have been several screening guidelines for cardiovascular risk factors in HIV patients. The Infectious Diseases Society of America HIV Medicine Association (IDSA/HIVMA) recommendations include fasting lipid profiles before and within 4–6 weeks after starting HAART, fasting glucose values before and during HAART, and routine measurements of body weight and changes in body shape.w78

The sensitivity and specificity of exercise or pharmacologic stress testing are not established in HIV disease, and utilisation currently follows guidelines for the general population. Similarly, the role of surrogate markers of CHD such as carotid IMT, inflammatory biomarkers such as hsCRP, adiponectin, and apolipoprotein B100 are also unclear in HIV disease, and further work is needed to clarify their role in the early detection of CHD in HIV patients.17


HIV associated pulmonary arterial hypertension

The prevalence of pulmonary arterial hypertension (PAH) in HIV infected patients is several thousand times that of the general population. Older studies had shown an incidence of 0.5%, which appears to have remained constant since the advent of the HAART era.w79 w80 However, recently we have shown that the prevalence of asymptomatic elevations in pulmonary arterial systolic pressures (PASP) as assessed by echocardiography may in fact be much higher, with a PASP >30 mm Hg in 35.2% of HIV patients, compared with 7.7% of non-infected controls.w81 This association between HIV and PAH is largely independent of secondary causes of PAH, with HIV being the sole risk factor for PAH in 82% of HIV infected patients.w82 It has also been shown that HIV infected patients with PAH have higher mortality rates and a more rapidly progressive disease course when compared with those without HIV, with a median survival rate of 6 months.w83

The pathogenesis of HIV associated PAH has not been clearly defined. Certain HIV proteins have been shown to activate endothelial cells indirectly, such as the envelope glycoprotein-120, which has been linked to higher endothelin-1 values.w27 w84 Endothelin-1 in turn is a potent vasoconstrictor and may play a central role in the pathogenesis of PAH. Increased markers of inflammation such as vascular endothelial growth factor A, platelet derived growth factor, and interleukin 1 and 6 have also been demonstrated in HIV associated PAH.w85 In addition, autoimmunity may play a role,w86 and a genetic predisposition has also been suggested.w87 Co-infection with human herpesvirus 8, on the other hand, did not appear to be associated with elevated PASP.w81

The treatment of HIV associated PAH including the role of HAART remains unclear. Few studies have examined the role of pulmonary vasodilators: one uncontrolled study showed improvement in clinical measures of heart failure and haemodynamics with bosentan, suggesting that endothelin may play a role in the pathogenesis.w88 The role of prostacyclin analogues in HIV associated PAH is limited to several small prospective studies demonstrating haemodynamic improvement.w89

Cardiovascular manifestations of HIV: key points

  • The prevalence of HIV infection in older individuals is increasing.

  • HIV appears to increase the risk of coronary heart disease (CHD) by 1.5–2 fold in HIV infected individuals when compared with uninfected controls.

  • Although long term treatment with highly active antiretroviral therapy (HAART) may lead to detrimental cardiovascular effects, there may be beneficial effects of HAART in the short term, providing evidence that HIV itself is mechanistically associated with increased CHD risk.

  • The identification and treatment of cardiovascular risk factors, particularly smoking and hyperlipidaemia, should be considered in all patients with HIV infection.

HIV related left ventricular dysfunction and myocarditis

The incidence of HIV related dilated cardiomyopathy before HAART was 15.9 per 1000 person-years, and has decreased significantly after introduction of HAART.w90 The aetiology of HIV related cardiomyopathy is likely multifactorial, and may be due to direct infection of myocardial cells by HIV-1 virions, immune activation, or co-infection with other viruses such as coxsackievirus B3 and CMV, as well as nutritional deficiencies, autoimmune factors (increased anti-α myosin antibodies), and HAART toxicity (zidovudine).w91 One myocardial biopsy series before the HAART era demonstrated that pathology is consistent with myocarditis in more than half of patients.w92

The treatment of HIV related cardiomyopathy is unclear, and usual treatment for heart failure with afterload reduction appears reasonable. What remains unknown is the role of inflammation and immune response in the disease. One study in children with HIV and left ventricular dilation showed improvement in left ventricular contractility in those with higher endogenous IgG values or after treatment with intravenous immunoglobulin, suggesting that myocardial impairment may be immunologically mediated.w93 The prognosis of HIV related cardiomyopathy appears to be worse than other non-ischaemic cardiomyopathies, and one study in children with vertical HIV transmission demonstrated that even mild left ventricular dysfunction was associated with increased overall mortality.w94 Further data will emerge from the ongoing HIV-Heart study, which is examining the prevalence and natural history of myocardial dysfunction in HIV infected adults.w95

Cerebrovascular disease

Although up to 40% of AIDS patients appeared to have neurologic complications, most of these were related to encephalopathy or infectious causes, although both ischaemic stroke and intracranial haemorrhage have been described in HIV patients.w96 In one cohort, the prevalence of cerebrovascular disease was 1.9% (transient ischaemic attack 0.8%, stroke 1.2%), with an annual incidence rate of 216/100 000. The prevalence appeared to be higher in later stages of infection and poorer immunologic state, although HAART did not change the risk of cerebrovascular events in two studies.1 w97 In contrast, DAD study results showed an incidence rate of 5.7 events per 1000 person-years for the combined end point of cardiovascular and cerebrovascular disease events, with an increased risk with longer HAART exposure (relative risk per year of exposure 1.26, 95% CI 1.14 to 1.38; p<0.0001).w98

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College of Physicians (London) and the European Board for Accreditation in Cardiology—you need to answer the accompanying multiple choice questions (MCQs). To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to:

Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into Heart with their journal’s username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.


Pericardial effusion is the most common cardiac manifestation of HIV infection with a prevalence up to 20%.w99 Although most patients are asymptomatic and effusions are generally small,w100 the presence of a pericardial effusion appears to be an independent predictor of mortality and poor prognosis.w101 The risk of bacterial endocarditis in HIV infected patients is similar to cohorts with similar risk behaviours, and the diagnosis and management is the same as in uninfected patients.18 Cardiac malignancies are quite rare in HIV patients, and include Kaposi’s sarcoma and malignant lymphoma.w91 HIV patients also have been noted to have prolonged QTc intervals, a finding which may be associated with myocarditis, cardiomyopathy, and autonomic neuropathy.w91


According to the Centers for Disease Control and Prevention (CDC), HIV patients who are age 50 years and older will account for half of all HIV/AIDS cases in the USA by the year 2015. Although similar data are not available for European countries, it is known that within western and central Europe, the UK together with France, Italy, and Spain continue to have a large HIV epidemic.w102 In addition to typical HIV related complications, this patient population will be facing consequences of other chronic diseases, including atherosclerosis. We believe that HIV infection and HAART may accelerate typical problems associated with aging, particularly cardiovascular disease via chronic persistent inflammation. Since cardiovascular disease is so prevalent in the aging population, any process that accelerates this process further will likely lead to significant clinical and public health problems. In the future, prevention strategies to decrease cardiovascular risk in the HIV infected population will be important from both a clinical as well as a public health perspective.


  1. 1.
  2. 2.
  3. ▸ An observational study demonstrating an increase in cardiovascular events with the long term use of PIs.

  4. 3.
  5. 4.
  6. 5.
  7. ▸ A summary statement outlining the problem of cardiovascular disease in HIV infected individuals.

  8. 6.
  9. ▸ Randomised clinical trial demonstrating that a strategy of continuous HAART was associated with improved outcomes compared with intermittent therapy.

  10. 7.
  11. ▸ A prospective clinical trial demonstrating a notable improvement of endothelial function after the acute initiation of antiretroviral therapy in treatment naïve individuals.

  12. 8.
  13. 9.
  14. 10.
  15. ▸ A comprehensive review of cardiovascular risk factors, with a focus on metabolic changes in HIV infected individuals.

  16. 11.
  17. ▸ A comprehensive review of cardiovascular disease in HIV patients.

  18. 12.
  19. 13.
  20. ▸ Guideline statement regarding the treatment of hyperlipidaemia in HIV infected individuals.

  21. 14.
  22. ▸ Contemporary guideline statement regarding the treatment of HIV infection in adults.

  23. 15.
  24. ▸ Summary statement from a panel of experts regarding prevention strategies of cardiovascular disease in HIV infected patients.

  25. 16.
  26. ▸ Summary statement from a panel of experts on cardiovascular risk prediction in HIV infected individuals.

  27. 17.
  28. ▸ Summary statement from a panel of experts on cardiovascular screening in HIV infected individuals.

  29. 18.
  30. ▸ Contemporary review of cardiovascular disease in HIV infection.

View Abstract


  • Additional references are published online only at

  • Competing interests: In compliance with EBAC/EACCME guidelines, all authors participating in Education in Heart have disclosed potential conflicts of interest that might cause a bias in the article. The authors have no competing interests.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.