HIV and the Heart

Tim P. van de Hoef, MD

Case Study

A 42-year old male patient presents with ongoing fulminant retrosternal chest pain, radiating to the left shoulder and the jaw, with concomitant nausea. The patient awoke with these discerning complaints 21/2 hours ago, and there is no (recent) history of anginal symptoms. His medical history includes human immunodeficiency virus seropositivity since his 24^th, for which he since successfully receives chronic combined antiretroviral therapy with Truvada (tenofovir/emtricitabine), ritonavir, and atazanavir, with undetectable viral loads. In the presence of hypercholesterolemia, further medication includes ezetemibe solotherapy, since statins were not tolerated. The electrocardiogram shows ST-segment elevation in the anterior precordial leads, and the patient is diagnosed with anterior wall acute myocardial infarction. Emergency coronary angiography is performed without delay, revealing a thrombotic occlusion of the left anterior descending coronary artery. Manual thrombus aspiration and subsequent stent implantation is performed successfully. In addition to lifelong aspirin treatment, and, despite revascularization guideline recommendation, the patient is post-treated with clopidogrel instead of prasugrel or ticagrelor for the first year following the acute myocardial infarction.

Introduction

Contemporary antiretroviral therapy (ART) treatment strategies for human immunodeficiency virus (HIV) infection have led to a decrease in acute immunodeficiency syndrome (AIDS)-related (fatal) events (PMID: 16878047). Consequently, life expectancy of HIV-positive patients has improved dramatically, which is now quite comparable to HIV-negative populations (PMID: 20467289, 21587343). Inferably, there has been a shift from opportunistic AIDS-related diseases to non-infectious co-morbidity, including cardiovascular disease (CVD) (PMID: 20588170). The World Health Organization recently estimated that both CVD and HIV will be in the top 3 causes for both global mortality as well as global disability-adjusted life-years in the year 2030, indicating that the interrelation between the two diseases will be an important challenge in global public health in the near future (17132052).

Cardiovascular fatal events comprise 7-15% of total fatal events in HIV-infected patients (PMID: 18645512, 16878047, 20380565). Compared with the non-infected population, HIV-infected patients are at increased risk for CVD (20400883, 17456578), even in HIV-infected patients treated with antiretroviral drugs (PMID: 21811144). Moreover, the age at which CVD surfaces in HIV-infected patients was found to be substantially lower compared with the general non-infected population, suggesting an accelerated atherosclerotic process (PMID:23369416). Several recent reports suggest an important association between HIV-infection, ART, and CVD. Most likely, a combination of factors is involved in the relationship between HIV and the heart, which implicates that the HIV-infected population may benefit from a tailored approach to CVD prevention and management. The attributing factors and their relevance in the management of CVD in HIV-positive patients will be discussed in detail in the next paragraphs.

Factors contributing to HIV-associated cardiovascular disease

The etiology of cardiovascular disease in HIV-positive patients is most likely multifactorial of nature, and includes traditional risk factors for cardiovascular disease, antiretroviral drug effects, as well as factors directly associated with the HIV-infection itself. As such, HIV seropositivity is known to be associated with all stages of atherosclerosis; from the stage of preclinical endothelial dysfunction (PMID: 19177019, 16586393), and the early subclinical stage of atherosclerosis (PMID: 19455012, 19390417, 19595354), to obstructive (epicardial) coronary artery disease (PMID: 19996940).

Traditional cardiovascular risk factors

Traditional risk factors for cardiovascular disease, in particular smoking, dyslipidemia, diabetes mellitus or impaired glucose tolerance, and hypertension, have been shown to be more prevalent in HIV-infected populations, and moreover to confer into an increased risk for CVD compared with the general non-infected population of equivalent age (PMID: 12856222, 17879928, 15903292, 12817187, 18854751, 19372895). Although the increased prevalence of traditional CVD risk factors plays an important role in the increased CVD risk in HIV-infected populations, it is unlikely that traditional risk factors are the sole origin of the increased CVD risk in HIV-positive patients. Indeed, in several large cohort studies, statistical risk-adjustment for traditional CVD risk factors was shown to only partly explain the increase in CVD-risk in HIV-positive patients (PMID: 17456578, 19455012), supporting the hypothesis that antiretroviral drug effects and HIV-infection-related factors play an important role in the added risk. In general, CVD is considered the consequence of a chain of events, initiated by a myriad of risk factors (PMID: 2008853). Intervention anywhere along the chain of events may therefore result in mitigation of the CVD risk, and thereby traditional risk factors constitute an important target for management and prevention of CVD, no less in HIV-infected patients.

Non-modifiable risk factors

Risk factors that do not constitute a modifiable therapeutic target comprise age, sex, and a family history of CVD. With increasing age, structural and hemodynamic changes in the cardiovascular system, oxidative stress, and endothelial dysfunction progress. Moreover, the duration of exposure to pro-atherogenic risk factors accumulates with advancing age. In general, older people have more atherosclerosis that younger people. Thereby, the risk for CVD increases with advancing age (PMID: 9603539). Importantly, age was shown to be a particularly important determinant of CVD risk in HIV-infected patients. In all age subgroups, HIV-infected patients are at increased risk for in particular AMI in comparison with age- and CVD risk factor-matched non-infected controls (PMID: 23459863, 12856222). The risk for CVD becomes clinically pertinent for men at younger ages than for women: for men, this is in their mid-forties, while this is at the time of menopause for women. The origin of this sex-difference in not yet fully elucidated, but is at least in part explained by an earlier onset of traditional CVD risk factors, such as hypertension and dyslipidemia, in men, and supposed hormonal protective effects in pre-menopausal women (PMID: 21587343). A positive family history is a well-known important risk factor for CVD both in men and women, which risk is independent of other concomitant CVD risk factors (PMID: 10723108). Many risk factors are under genetic control (blood pressure, lipid spectra, obesity), but these do not account for the total aggregated risk observed in families (PMID: 7058780), indicating that family history of CVD as a risk factor goes beyond the accumulation of risk from inherited traditional risk factors. It must be noted that while family history is an undisputable risk factor, obviously a large number of modifiable risk factors are found in people with a family history of CVD, adding to the accumulated CVD risk. Although non-modifiable in nature, these risk factors importantly add to the accumulated CVD risk within a specific patient and therefore play an important role in risk stratification and prevention of CVD, in particular in HIV-infected patients.

Modifiable risk factors

Risk factors that comprise modifiable targets for therapeutic strategies include hypertension, dyslipidemia, diabetes, and smoking. Importantly, the prevalence of these traditional risk factors is increased in HIV-positive populations (PMID: 23369416). Compared with the general population, HIV-infected populations in particular have a higher smoking prevalence, and have a diet high in saturated fat. Evidence has accumulated that HIV-infected patients in general have a more atherogenic lipid spectrum, putting them at particular high risk for CVD (PMID: 23369416, 21587343). However, the relative contribution of each traditional risk factor for CVD risk within a patient is comparable in HIV-infected and uninfected populations, irrespective of HIV-infection status (PMID: 21587343). As such, although the risk factor prevalence is increased in HIV-infected populations, it likely does not at itself explain the increased risk of CVD in this population, and, therefore, several HIV-related factors need to be taken into consideration.

Antiretroviral therapy effects

Contemporary ART treatment strategies adhere to a combination of antiretroviral drugs (as such called combined ART (cART)), usually consisting of 2 nucleoside reverse transcriptase inhibitors (NRTIs) in combination with a protease inhibitor (PI), or a non-nucleoside reverse transcriptase inhibitor (NNRTI) in combination with an integrase inhibitor. Several reports have indicated an association between ART treatment and CVD. In particular early studies have reported an increased risk for early acute myocardial infarction (AMI) in patients treated with PIs, which was associated with concomitantly raised cholesterol levels, a known side-effect of PI treatment (PMID: 21587343, 23369416). The associated risk was related to the duration of PI exposure. Nonetheless, the increased risk persisted after statistical correction for lipid concentrations, which implicates an independent direct effect of PI treatment on early occurrence of AMI. Importantly, such increased risk has not been as clearly established for all PIs, and conclusive data remains absent for several ART agents from the PI group. Nonetheless, even in the presence of an increased risk for early AMI, the risk-benefit ratio remains positive towards PI use, due to the unequivocal benefits of cART on survival that outweigh the PI-use associated risk for AMI. For the NRTI group of ART, in particular for abacavir and didanosine, such an association between drugs and CVD is less clearly established, as large cohort studies have yielded conflicting results (PMID: 21587343, 23369416). Nonetheless, there seems to be a consistent reporting of increased AMI risk in HIV-positive populations treated with (c)ART. The complex interrelation of HIV-infection, antiretroviral drugs and concomitant co-morbidities in the presence of traditional CVD risk factors makes it difficult to extract the actual origin of an increased risk for AMI. Nevertheless, an association between ART drugs and AMI must be assumed for all subclasses of ART drugs, although PI use is most consistently found a determinant of early AMI risk.

HIV-infection related factors

Besides the CVD risk conferred by the use of aggressive cART, the HIV-infection itself may add to the accumulated CVD risk in HIV-positive patients. HIV-infection related factors include the persistent inflammatory status and immune dysfunction associated with high viral loads, as well as viral load independent factors.

Inflammation

In general populations, persistent inflammation is associated with an increased risk for CVD (PMID: 19942084, 11877368). Markers of inflammation are persistently elevated in HIV-positive patients (PMID: 2044684), which was shown to be directly associated with (increases in) levels of viral loads (PMID: 19425222, 19954384). Different components of HIV are independently associated with increased markers of endothelial cell activation (sVCAM), systemic inflammation (CCL2, IL-10), and adipose tissue activation (adiponectin), which may actively advance atherosclerosis (PMID: 16890832). Moreover, persistent inflammation in HIV-infected patients is directly associated with mortality in this patient population (PMID: 20581689, 18942885). These findings suggest that inflammation itself may explain (part of) the CVD risk in HIV-positive populations additive to traditional risk factors.

Immune dysfunction and immune activation

Several studies have indicated that immune dysfunction is associated with CVD risk among HIV-infected patients. CD4+ T-cell counts of <500/IL have been associated with an increase in cardiovascular events, independent of traditional CVD risk factors or ART, carrying an additive risk comparable with that of smoking or sub-optimally treated LDL cholesterol levels (PMID: 20597691). CD4+ T-cell counts of <200/IL were associated with an increased risk of AMI, which was comparable in magnitude to the risk imposed by traditional CVD risk factors (PMID: 20827215). Consistently, episodic CD4+ cell count-guided ART is associated with a substantial increase in risk for AMI compared with continuous ART, indicating an important role of viral load and immune dysfunction for the extent of CVD risk (PMID: 17135583). Moreover, several studies have shown a decrease in non-AIDS related events following the start of ART, and a notable increase in CVD events in patients with incomplete immune recovery following start of ART (PMID: 18781885, 18476292). In addition, the risk of AMI was recently shown to be significantly increased in patients with a recent episode of immune dysfunction defined as CD4 count <200 or HIV-RNA count of >500 copies (PMID: 22112603). It can therefore be acknowledged that (residual) compromise of immune function is an important determinant of risk for CVD as well.

Virologic Suppression and the Risk of CVD

The associations discussed in the previous sections may implicate that suppression of HIV replication, and normalization of inflammation and immune function by (c)ART mitigates the associated CVD risk. However, even viral replication suppression with (c)ART does not fully normalize these processes, and even residual levels may result in adverse cardiovascular outcome. Concordantly, it was shown that carotid intima media thickness is increased in all HIV-infected subgroups compared with controls, including HIV-infected patients that maintain an undetectable HIV viral load without the use of (c)ART (PMID: 19390417, 21220772). This finding was independent of prior ART exposure duration, viremia, or advanced immunodeficiency. Moreover, a recent large cohort study of HIV-infected patients compared with age and CVD-risk matched non-infected controls including more than 80.000 subjects indicated that, although mitigated by virologic suppression, the risk for CVD, and AMI in particular, remains present despite virologic suppression (PMID: 23459863).

In summary, the available evidence indicates that CVD in HIV-infected population is indeed a complex multi-factorial process, which etiology goes beyond the role of traditional risk factors, and beyond the effects of antiretroviral medication. HIV-infection related factors, as well as persistent inflammation and immune dysfunction most probably constitute important additional factors resulting in premature atherosclerosis and accelerated ageing even when the viral replication is suppressed.

Prevention of CVD in HIV-infected patients

As a corollary of the aforementioned interplay of CVD risk factors and HIV-infection/–therapy, it may be appreciated that prevention of CVD in HIV-positive patients should be emphasized, especially in the light of the contemporary life expectancy of this patient population. In general, cardiovascular therapy guidelines aimed at prevention of CVD in HIV-patients follow those for the general population, although some HIV-related factors must be taken into consideration (PMID: 18257770). Special emphasis should focus upon the identification and treatment of dyslypidemia, which is frequently present in HIV-infected patients, and which is known to occur in reponse to the start of (c)ART therapy, with a particular association with PIs (PMID: 17460226) and first generation NRTIs (PMID: 23369416). Therefore, assessment of lipid status, especially triglycerides, should be performed prior to the start of cART therapy, at 3-6 months after initiation of therapy, and at least yearly in the absence of abnormalities. A special point of care lies within lipid-lowering therapy in concomitance with cART treatment, because of the possibility of interaction between statins and for example PIs and NNRTIs (PMID: 23369416). The primary mode of interaction is seen through the cytochrome P450 (CYP) pathway. PIs mainly inhibit CYP, and could therefore lead to toxicity of statins as these drugs are most frequently metabolized through particular CYP pathways. NNRTIs are associated with CYP induction, leading to impairment of statin efficacy. Statins not or only mildly associated with CYP interactions include rosuvastatine and pravastatin, and may therefore be considered in HIV-infected patients on cART. More novel statins, such as pitavastatin, without CYP interaction have not yet been evaluated in this setting. Ezetemibe, which is not metabolized through the CYP pathway, may be considered in addition to statin therapy or as stand-alone therapy when statin therapy leads to unendured side effects. Nonetheless, statin therapy should be tailored to the specific situation and side effects or inefficacy of agents may necessitate statin or cART agent switch along the way. Especially when cART leads to a subsequent severe dyslipidemia, switching to another cART combination may be considered (PMID: 23369416). However, no clinical trials have been performed on this subject, and therefore the best approach remains patient-tailored. Moreover, in the light of the complex interplay between all CVD risk factors discussed in the previous paragraphs, it is important to note that the actual role of statin therapy in primary prevention in HIV-infected patients has not been established to date. Nonetheless, the evidence to date indicates a very important role of dyslipidemia in the CVD-risk associated with HIV-infection and –therapy, and optimal medical therapy should be pursued. In addition, smoking cessation should be actively pursued in patients with HIV, because of its high prevalence and associated increase in CVD risk in this patient population, as well as the significant benefit of smoking cessation (PMID: 21251183). The recent smoking cessation guidelines consequently indentified the HIV-infected population as and important target population for smoking cessation therapy, both interms of counseling as well as medical therapeutic strategies (PMID: 18617085). In the presence of hypertension, renin angiotensin system blockers are considered first choice owing to their global protective effects on kidney function, glucose metabolism, and the vasculature in general (PMID: 23369416).

Antiplatelet regimens have not extensively been investigated in HIV-infected populations and it is therefore currently advocated to adhere to general antiplatelet guidelines. In particular aspirin treatment should be initiated as in the general population in the presence of high CVD risk when contraindications are absent (PMID: 22809746). For more aggressive antiplatelet agents, including clopidogrel, prasugrel and ticagrelor, recent investigations have implicated interactions between cART and these potent platelet inhibitors. Ritonavir, a PI-subclass ART, in particular was found to impair prasugrel activity in in vitro experiments (PMID: 21550074). Furthermore, ticagrelor is metabolized by CYP and is contraindicated in patients on PI therapy, in particular in those patients using ritonavir (both ritonavir and ticagrelor are associated with the CYP 3A4/5 pathway) (PMID: 21177984). Although the stent-thrombosis rates in HIV-infected patients after percutaneous coronary intervention and coronary stent implantation have not been found to be substantially higher compared with the general population, it should be borne in mind that evidence on the effectiveness of antiplatelet therapy in terms of platelet activity is very limited, and residual platelet activity levels may well play a role in a particular patient.

Clinical presentation and revascularization

Clinical presentation of advanced CVD in HIV-infected patients involves an equivalently wide spectrum as in the general non-infected population (PMID: 1731952). Nonetheless, as may be appreciated from the previous sections, clinical presentation is predominated by the occurrence of acute coronary syndrome (ACS) (PMID: 12888138, 12739013). Acute coronary syndrome typically occurs in males, at a relatively young age (<50 years of age), in particular in those patients with a prolonged known HIV-infection (>8 years), who remain on (c)ART. The most frequent presentation occurs with an ST-segment elevation myocardial infarction. Nevertheless, presentation with non-ST-segment elevation myocardial infarction and unstable angina occur frequently as well PMID: 1731952, 12739013, 14691419, 12588205). Contrariwise, stable CVD is less frequent. The extent of coronary artery disease in general does not differ between HIV-infected and non-infected populations (12739013). Clinical outcome after revascularization (predominantly for ACS) is similar for HIV-infected and non-infected populations, regardless of whether revascularization is performed percutaneously or surgically (PMID: 16537777, 18288976). HIV-infected patients, however, do remain at increased risk for recurrent ACS during follow-up (PMID: 20965887), which is importantly associated with persistently elevated lipid spectra.

Key Points:

Patients with human immunodeficiency virus infection are at increased risk for cardiovascular disease compared with the general non-infected population. The increased risk originates from an increased pertinence of traditional risk factors for cardiovascular disease, in combination with the detrimental effects of the HIV-infection itself and the adverse (combined) antiretroviral therapy drug effects. Prevention of cardiovascular disease in HIV-positive patients should in particular focus on prevention of dyslipidemia, and smoking, which are both common in HIV-infected patients, and represent important (modifiable) determinants of the cardiovascular disease risk in these patients. Drug interactions in patients treated with (combined) antiretroviral therapy are frequent, especially between antiretroviral therapy and statins or antiplatelet agents such as prasugrel and ticagrelor, which occurs predominantly through the cytochrome P450 pathway. The increased risk for cardiovascular disease predominantly results in an early occurrence of acute myocardial infarction, for which coronary revascularization is associated with good outcome, although recurrent ischemic events are more common in HIV-infected patients than in the general non-infected population.