Heart Failure among People with HIV: Evolving Risks, Mechanisms, and Preventive Considerations

Abstract

Purpose

People with HIV (PHIV) with access to modern antiretroviral therapy (ART) face a two-fold increased risk of heart failure as compared to non-HIV-infected individuals. The purpose of this review is to consider evolving risks, mechanisms, and preventive considerations pertaining to heart failure among PHIV.

Recent Findings

While unchecked HIV/AIDS has been documented to precipitate heart failure characterized by overtly reduced cardiac contractile function, ART-treated HIV may be associated with either heart failure with reduced ejection fraction (HFrEF) or with heart failure with preserved ejection fraction (HFpEF). In HFpEF, a “stiff” left ventricle cannot adequately relax in diastole – a condition known as diastolic dysfunction. Diastolic dysfunction, in turn, may result from processes including myocardial fibrosis (triggered by hypertension and/or immune activation/inflammation) and/or myocardial steatosis (triggered by metabolic dysregulation). Notably, hypertension, systemic immune activation, and metabolic dysregulation are all common conditions among even those PHIV who are well-treated with ART. Of clinical consequence, HFpEF is uniquely intransigent to conventional medical therapies and portends high morbidity and mortality. However, diastolic dysfunction is reversible – as are contributing processes of myocardial fibrosis and myocardial steatosis.

Summary

Our challenges in preserving myocardial health among PHIV are two-fold. First, we must continue work to realize UNAIDS 90–90-90 goals. This achievement will reduce AIDS-related mortality, including cardiovascular deaths from AIDS-associated heart failure. Second, we must work to elucidate the detailed mechanisms continuing to predispose ART-treated PHIV to heart failure and particularly HFpEF. Such efforts will enable the development and implementation of targeted preventive strategies.

Introduction

Early in the course of the HIV epidemic, front-line physician-scientists noted diverse presentations of profound myocardial and pericardial disease among patients with AIDS¹. In this pre-antiretroviral therapy (ART) era, AIDS-associated heart failure was typically characterized by overt reductions in cardiac contractile function, with or without ventricular chamber dilation¹. Several factors were noted to underlie such heart failure presentations. First, direct myocardial infiltration by HIV² - with or without concomitant infiltration by opportunistic viruses, parasites, and/or bacteria¹. Second, a robust autoimmune reaction in the myocardial structural space, likely triggered by infection³. Third, pericardial disease, particularly among those patients also battling tuberculosis, Kaposi’s sarcoma, or lymphoma¹. With the introduction of early ART, including the nucleoside reverse transcriptase inhibitor (NRTI) zidovudine (AZT), the contributions of co-infections to heart failure risk abated. At once, it became apparent that select antiretroviral therapeutics exerted cardiotoxic effects⁴.

Today, PHIV with access to modern ART continue to face significant health threats from heart failure. Heart failure is a disease of aging, whereby the general-population prevalence increases steadily across successive age deciles⁵. Adjusting for age, PHIV with access to ART confront an approximately two-fold increased risk of heart failure^6,7,8,9,10. Thus, as the global population of PHIV ages^11,12,13, heart failure prevalence in this group of 37 million individuals may be anticipated to spike. Of further concern, heart failure outcomes in the general population are poor¹⁴ and worse still among PHIV: Indeed, the 5-year mortality rate among PHIV diagnosed with heart failure approaches 50%¹⁵.

Our challenges in preserving myocardial health among PHIV are thus two-fold. First, we must continue work to realize UNAIDS 90–90-90 goals. This achievement will reduce AIDS-related mortality, including cardiovascular deaths from dilated cardiomyopathy. Second, we must remain cognizant of the ongoing threat heart failure continues to pose to aging ART-treated PHIV. Our attention should center on elucidating mechanisms fueling heart failure risk among ART-treated PHIV, including persistent systemic immune activation and metabolic dysregulation. Enhanced understanding of factors contributing to increased heart failure risk and associated adverse outcomes in the aging HIV population will enable the development and implementation of targeted preventive strategies. Given the poor prognosis of heart failure among PHIV, primacy is on prevention. In this context, the present review focuses on heart failure risks, mechanisms, and preventive considerations relevant to ART-treated PHIV.

Heart Failure Risks among Contemporary Cohorts of PHIV

A systematic review and meta-analysis of cardiac dysfunction among PHIV – inclusive of studies across time and place – highlights the manner in which population-specific heart failure risks are evolving¹⁶: Erqou et al. selected 54 studies conducted in diverse regions (Africa, Asia, Europe, and North America) and published anywhere between 1988 to 2017¹⁶. Analyzing data from 125,382 PHIV (82% men), Erqou et al. determined a pooled heart failure prevalence of 6.5% (4.4%, 9.6%). This observed heart failure prevalence among PHIV was surprisingly high, given the relatively low average age of the cohort (47 years). Of note, among those PHIV studied, only 77% were on ART and a significant proportion had untreated, uncontrolled HIV/AIDS¹⁶.

Importantly, key North American studies analyzing incident heart failure by HIV status in contemporary cohorts consistently suggest an increased relative risk of heart failure among PHIV with access to ART^6,7,8,9. Butt et al. examined data from an all-male cohort of US Veterans without prior cardiovascular disease (CVD), including 2,391 PHIV and 6,095 controls followed from 2000 through 2007⁶. Median age in both groups was 48 years. Age- and race-adjusted heart failure incidence per 1,000 person-years (based on diagnosis codes) was 7.21 among PHIV (95% confidence interval, CI, 6.90–7.34) and 4.82 (95% CI 4.72–4.91) among controls. Overall, the hazard ratio for heart failure among individuals with vs. without HIV was 1.81 (95% CI 1.39–2.36), after adjustment for traditional heart failure risk factors⁶. Of note, this hazard ratio was not adjusted for history of depression, renal dysfunction, or liver dysfunction – all parameters subsequently associated with incident heart failure in this cohort^17,18,19. Feinstein et al. explored data from a predominantly-male US cohort including 4,640 PHIV (83% men) and 4,250 matched control subjects (80% men) – all without prior heart failure – followed from 2000 to 2016⁷. In this study, the hazard ratio for incident physician-adjudicated heart failure among PHIV vs. controls was 2.10 (95% CI 1.38–3.21) after adjustment for traditional heart failure risk factors. Compared with the VA Cohort, this cohort was slightly younger (average age 40–41 years)⁷. Notably, in both cohorts, only approximately half of PHIV evidenced viral loads <500 copies/ml^6,7. In the study by Butt et al., the increased risk of heart failure among PHIV was only significant among those with uncontrolled viremia. By contrast, in the study by Feinstein et al., higher viral load (and lower CD4⁺ T cell count) tracked with heart failure risk among PHIV, but even those PHIV with suppressed viremia saw increased heart failure risk as compared with controls^6,7. Finally, the US Partners Healthcare Database heart failure study - focused on 1,388 women with HIV (WHIV) and 13,781 matched control subjects without prior heart failure - demonstrated a nearly four-fold increased incidence of physician-adjudicated heart failure among WHIV⁹. In this study, led by Janjua et al., the average age of both cohorts was 59 years. Among the WHIV, 92% were on ART, though again, only approximately 50% had achieved viral suppression⁹.

Most recently, Yen et al. published a large study analyzing nationally representative data on heart failure among PHIV from the Taiwan Centers for Disease Control and Prevention HIV Surveillance System, coupled with data on heart failure among age and sex-matched controls without HIV from the Taiwan National Health Insurance Research Database¹⁰. Overall 24,153 PHIV (94% men; 72% on ART) and 96,612 control subjects (94% men) without known heart failure were followed from 2003–2014. In this cohort, the average age was only 33 years. Nevertheless, PHIV demonstrated a 1.5-fold increased incidence of heart failure (non-adjudicated). The hazard ratio for heart failure among individuals with vs. without HIV was 1.52 (95% CI 1.27 to 1.82) after adjustment for traditional heart failure risk factors. Moreover, the time to incident heart failure was significantly shorter among PHIV vs. controls (P<0.001)¹⁰.

Taken together, studies from North America and Asia reveal an approximately 2-fold increased risk of heart failure among contemporary cohorts of PHIV with access to ART. Within these cohorts, even when ART has been prescribed, suboptimal viral control is common and appears to be associated with augmented risk.

Outcomes among Contemporary Cohorts of PHIV with Heart Failure

Soberingly, multiple studies across regions suggest heart failure outcomes are worse among PHIV vs. non-HIV-infected individuals. In the general population, a diagnosis of heart failure is associated with recurrent hospitalization, decreased quality of life, and high rates of mortality within 5 years¹⁴. Analysis of data from the predominantly-male US Veterans Cohort suggests that among PHIV with heart failure, 5-year mortality rates approached 50%¹⁵. Analogously, Janjua et al. showed through the US Partners Healthcare Database heart failure study that among women with heart failure, HIV positivity conferred an increased risk of all-cause mortality, cardiovascular mortality, and heart failure hospitalization⁹. Indeed, the hazard ratio for recurrent heart failure hospitalization among WHIV vs. non-HIV-infected women was 2.58 (95% CI 1.55–4.29) after adjustment for traditional heart failure risk factors⁹. Finally, through the Sub Saharan Africa Survey of Heart Failure (THESUS HF) study, Sliwa et al. analogously confirmed worse outcomes among PHIV with heart failure^20,21. This prospective multi-center study recruited 1,006 patients with acute heart failure (51% women) from 9 countries in Sub-Saharan Africa between 2007–2010 and followed these patients for 6 months. Within this cohort, hypertension prevalence was a striking 56% while HIV prevalence was 7%. Analyses by Sliwa et al. illustrated that in this group, HIV status conferred an increased risk of all-cause mortality and 60-day hospital readmission^20,21.

A recent mixed-sex study of heart failure outcomes among US PHIV vs. controls suggested worse outcomes only among those PHIV with unchecked viremia²². In this study, Alvi et al. studied all individuals admitted with heart failure to an urban academic medical center (Montefiore) in 2011 and followed for 2 years²². The recruited cohort included 374 PHIV (53% men) and 1,934 non-HIV infected individuals (55% men). Groups were similar with respect to demographics (average age 60 years) and traditional CV risk factors. In follow up, the broad group of PHIV demonstrated increased rates of all-cause mortality, CV mortality, and heart-failure re-admission. Amongst this group, 92% were on ART, but a smaller percentage evidenced suppressed viral load. In sub-analyses, the aforementioned trends held up only for the subgroup with unchecked viremia (viral load >500 copies/ml) or markedly reduced immune function (CD4⁺ T cell count <200 cell/mm³). Of potential clinical relevance, additional analyses performed within the cohort of PHIV showed that use of ritonavir-boosted protease inhibitors (PIs) was associated with a 2-fold increased risk of cardiovascular mortality and 30-day heart failure readmission²³. This finding ran in contrast to early work implicating older NRTI’s in mitochondrial injury and attendant cardiotoxicity^24,25,26.

What Clinicians Caring for PHIV Need to Know about Heart Failure Subtypes

Heart failure is subclassified based on left ventricular ejection fraction (EF), which is the percentage of blood leaving the left ventricle each time it contracts²⁷. Patients with clinical heart failure may thus have heart failure with reduced ejection fraction (HFrEF; EF <40%), heart failure with preserved ejection fraction (HFpEF; EF ≥50%), or heart failure with borderline ejection fraction (EF 40–49%). Several experts believe HFrEF and HFpEF constitute distinct disease processes, whereby HFpEF may not represent a stage en route to HFrEF²⁷. Indeed, differences abound in HFrEF vs. HFpEF etiology, triggering risk factors, and ensuing pathophysiology.

HFrEF tends to be caused by disorders which affect myocardial contractile function and/or afterload (the pressure against which the heart pumps to circulate blood)²⁸. Examples include acute myocardial infarction leading to focal myocardial fibrosis, infectious/autoimmune myocarditis, toxin-induced cardiomyopathy, and valvulopathy. Thus, parameters predisposing to HFrEF include traditional metabolic risk factors (hypertension, dyslipidemia, dysglycemia, cigarette smoking) as well as infectious risk factors and toxic exposures^28,29. With the aforementioned etiologies, myocyte loss and associated cardiac remodeling may ensue, possibly leading to chamber dilation. From a pathophysiologic perspective, by affecting contractile function and/or afterload, etiopathologic causes of HFrEF reduce the stroke volume (blood volume pumped from ventricle with each cardiac cycle) and, in turn, the cardiac output (a product of stroke volume and heart rate). Thus, HFrEF patients experience reduced forward flow of blood during systole and a secondary back-up of high pressures²⁸.

HFpEF, by contrast, is typically caused by processes which incite the myocardium to stiffen, resulting in inadequate relaxation during diastole (or “diastolic dysfunction”)³⁰. Examples include diffuse myocardial fibrosis and myocardial steatosis. Risk factors for HFpEF include select traditional metabolic risk factors such as hypertension and dysglycemia, as well as female sex and advanced chronological age^29,30. Myocyte loss and ensuing chamber dilation are less common, but increased wall thickness is characteristic. By affecting relaxation during diastole, etiopathologic causes of HFpEF result in unacceptably highly filling pressures required to achieve adequate preload and maintain cardiac output. Increased left ventricular filling pressures, in turn, lead to a back-up of high pressures. Moreover, while patients with HFpEF appear to have grossly normal systolic function at rest, these patients often cannot adequately augment cardiac output in response to exercise³⁰.

Differences notwithstanding, HFrEF and HFpEF have select features in common²⁷. For example, a shared clinical presentation of heart failure characterized by a combination of exertional dyspnea (related to reduced forward flow), pulmonary edema (related to left sided pressure back-up), and possibly hepatic congestion/peripheral edema (related to right-sided pressure back-up). Though systolic dysfunction defines HFrEF and diastolic dysfunction remains a hallmark of HFpEF, subtle forms of both types of dysfunction may co-exist in patients with either of these diseases. Moreover, both diseases are characterized by chronotropic incompetence (inability to adequately augment heart rate) and vascular dysfunction (e.g. impaired endothelial vasorelaxation, leading to vascular stiffness). Finally, both forms of heart failure may trigger analogous compensatory responses including activation of the sympathetic nervous system and renin-angiotensin-aldosterone system (RAAS) and release of natriuretic peptides. In the short term, these compensatory processes may help maintain cardiac output, but in the long run, these responses exacerbate ventricular remodeling and worsening symptomatology²⁷.

As HFrEF and HFpEF may differ with respect to underlying etiology and pathophysiology, it is plausible that these two diseases may respond differently to medical therapies. Among patients with HFrEF, several pharmacologic interventions significantly reduce the hazard ratio for heart failure hospitalization or death²⁷. Thus, there is a standard paradigm for treating patients with HFrEF: RAAS blockade, beta blockade, and possibly symptomatic treatment with diuresis or afterload reduction³¹. In contrast, among patients with HFpEF, analogous interventions have not been shown to reduce the hazard ratio for heart failure hospitalization or death³⁰. Thus, the astute clinician must try to identify early HFpEF phenotypes and intervene on relevant risk factors in the hopes of forestalling the transition to overt HFpEF – a morbid condition intransigent to medical therapy.

Clinical and Pre-Clinical Heart Failure Phenotypes among PHIV: Evolving Patterns

Erqou’s meta-analysis of cardiac dysfunction as assessed through multiple studies across time and place highlights ways in which heart failure risk is evolving among PHIV¹⁶. In this meta-analysis, a shift towards a lower prevalence of systolic dysfunction among PHIV was noted in more recent studies and studies conducted in regions characterized by widespread ART access. Overall, the pooled prevalence of left ventricular systolic dysfunction (EF<50% or fractional shortening <26%) was 12.3% (95% CI 6.4%−19.7%). Consonant with these observations are juxtaposed findings on clinical heart failure phenotypes among PHIV gleaned from the South African Heart of Soweto Study and from three US studies. Through the Heart of Soweto study, Sliwa et al. characterized the presentation of 518 PHIV (62% women, average age 39 years, 54% on ART) presenting to a major urban hospital between 2006–2008 with acute CVD³². Among all CVD presentations, a significant proportion were caused by dilated cardiomyopathy (38%) or pericarditis (25%). No cases of HFpEF were noted³². By contrast, studies of predominantly-male cohorts of US PHIV with heart failure (US Veterans Cohort; US Montefiore Cohort) suggest a near-even split of HFrEF vs. HFpEF^33,22. Moreover, in Janjua’s analysis of US WHIV with heart failure (Partners Healthcare Cohort), a striking shift towards HFpEF was noted (71% HFpEF vs. 29% HFrEF)⁹. Notably, the average age in the US cohorts was significantly higher than that in the South African cohort and ART uptake was higher^33,22,9. By inference, it appears that among PHIV, ART access, chronologic age, and women’s sex may represent risk factors for HFpEF. Thus, as ART access expands and the global population of PHIV ages, we may expect to see a surge in HFpEF cases – particularly in Sub-Saharan Africa. In Sub-Saharan Africa, where the majority of all adults with HIV reside³⁴, approximately 50% of PHIV are women³⁴ and among both WHIV and MHIV, hypertension is highly prevalent^35,36,37.

Among PHIV, while the prevalence of overt systolic dysfunction appears to be declining over time, rates of diastolic dysfunction – a pre-clinical phenotype, which progresses at a rate of 2%/year to symptomatic heart failure³⁸ – may be on the rise, Erqou’s meta-analysis incorporating studies old and new suggested the population prevalence of diastolic dysfunction (grades 1–3) among PHIV was 29.3% (22.6% to 36.5%)¹⁶. By contrast, general-population studies suggest the prevalence of diastolic dysfunction runs closer to 10%³⁹. Perhaps not surprisingly, then, in Erqou’s meta-analysis, the relative risk of all grades of diastolic dysfunction among studied PHIV vs. controls was 3 (95% CI 1.8–5.1)¹⁶. Notably, a meta-analysis of cardiac dysfunction among PHIV by Cerrato et al. including 11 studies (1 from North Africa, 1 from Asia, 6 from Europe, and 3 from North America) all published between 2004–2011 suggested an even higher prevalence of diastolic dysfunction among PHIV⁴⁰. In this study, among 2,242 PHIV (%men not reported, median age 42 years, 98% ART, 74% undetectable viral load), the prevalence of diastolic dysfunction was 43.38% (95% CI 31.73–55.03). By contrast, the prevalence of overt systolic dysfunction (impaired LVEF) was 8.33% (95% CI 2.20–14.25). In multivariable analysis, factors associated with diastolic dysfunction included older age (OR = 2.50 per 10 years increase; 95% CI: 1.70–3.60) and hypertension (OR = 2.30; 95% CI: 1.20–4.50). By contrast, factors associated with systolic dysfunction included history of myocardial infarction 15.90 (95% CI: 1.94–329.00) and active cigarette smoking 1.70 (95% CI: 1.03–2.77)⁴⁰.

Two Processes which may be Expected to Contribute to Diastolic Dysfunction among ART-treated PHIV are Myocardial Fibrosis and Myocardial Steatosis

Myocardial fibrosis and myocardial steatosis are two pathologic processes which may be expected to contribute to diastolic dysfunction among contemporary cohorts of ART-treated PHIV. In a healthy state, cardiomyocytes occupy most of the myocardial structural space. The remaining myocardial structural space is filled by the cardiac interstitium, including collagen produced by myofibroblasts^41,42. Further, in a health state, cardiomyocytes contain miniscule amounts of triglyceride – on the order of 0.4%−0.6%⁴³. Myocardial fibrosis is a pathologic process in which the myocardial tissue collagen volume fraction is increased⁴². The distribution of myocardial fibrosis may be diffuse (reactive interstitial fibrosis) or focal (replacement fibrosis or scarring). Major drivers of diffuse myocardial fibrosis include aging, hypertension, systemic/myocardial inflammation/immune activation, and micro-ischemia caused by suboptimal stress-induced augmentation of myocardial blood flow. Focal myocardial fibrosis, by contrast, is most frequently caused by acute myocardial infarction⁴². Myocardial steatosis is a pathologic process characterized by increased deposition of lipids (predominantly triglycerides) within cardiomyocytes – increased intramyocardial triglyceride content. Major drivers of myocardial steatosis include age, obesity, abnormal lipid metabolism, dysglycemia, and endothelial dysfunction⁴³. Both fibrosis and steatosis contribute to myocardial stiffness - impairing relaxation and resulting in diastolic dysfunction^42,43. Over the past decade, general-population studies have revealed that fibrosis relates closely to diastolic dysfunction, HFpEF, and cardiovascular mortality^44,45,46. Moreover, studies in populations with diabetes suggest a close inverse relationship between steatosis and diastolic function^47,48. Importantly, among populations without HIV, both myocardial fibrosis and myocardial steatosis have been shown to be modifiable with therapy tailored to the pathophysiologic driver – e.g. antihypertensive therapy geared toward reducing fibrosis or therapy with peroxisome proliferator-activated receptor gamma (PPAR-γ) agonism aimed toward mitigating steatosis^49,50,51,52.

Among ART-treated PHIV, systemic immune activation may drive myocardial fibrosis while ongoing metabolic dysregulation may predispose to myocardial steatosis. Systemic immune activation persists even among those individuals with HIV whose virus is completely suppressed by combined antiretroviral therapy^53,54,55. ART, even when administered early in the course of the disease, dampens select indices of systemic immune activation but many indices of systemic immune activation remain elevated⁵⁶. In this regard, low-level viral replication, viral co-infection, enhanced microbial translocation may be culprit⁵⁷. Further, a significant proportion of ART-treated PHIV experience metabolic dysregulation, which may be driven by HIV infection itself, HIV-associated immune activation, HIV-associated hormonal perturbations including relative growth hormone deficiency⁵⁸ and RAAS activation⁵⁹, and/or off-target effects of select antiretroviral therapeutics^24,25,26. Forms of metabolic dysregulation demonstrated by PHIV are varied. While select early ART regimens tended to evoke overt lipodystrophy (peripheral lipoatrophy with or without central lipohypertrophy)⁶⁰, many modern regimens are better-tolerated. Nevertheless, ART initiation continues to be associated with weight gain⁶¹, accumulation of excess adiposity⁶², and ectopic fat deposition^63,64 – all processes associated with development of traditional cardiometabolic risk factors.

Of interest, theoretically, systemic immune activation may predispose to myocardial fibrosis either by prompting inflammation in the myocardial structural space (inflammatory mechanism) or by contributing to arterial inflammation^65,66 and downstream coronary microvascular dysfunction⁶⁷ (inflammatory-ischemic mechanism). Meanwhile, myocardial steatosis may trigger inflammation in the myocardial structural space, feeding into in situ fibrosis (metabolic-inflammatory mechanism). General-population studies have also highlighted relationships between coronary microvascular dysfunction and myocardial steatosis⁶⁸, but cause/effect determinations remain elusive. In at least one cardiac MRI/MRS study among PHIV, myocardial fibrosis has been shown to correlate with myocardial steatosis⁶⁹ and in two such studies, myocardial steatosis has been shown to correlate with diastolic dysfunction^70,71. Notably, no cardiac MRI/MRS study among PHIV has identified a relationship between myocardial fibrosis and diastolic dysfunction, although such a relationship has been convincingly demonstrated in general-population studies⁴⁴. Moreover, general-population studies have linked myocardial fibrosis with ensuing heart failure and cardiovascular death^45,46. Additional work is needed to examine whether the increased burden of diffuse myocardial fibrosis among asymptomatic PHIV with access to ART may help to explain heightened risks of diastolic dysfunction⁴⁰, heart failure^{6,33,7,8,9,10}, pulmonary hypertension⁷², and cardiac dysrhythmia/sudden cardiac death⁷³.

Insights on the Development of Pre-Heart Failure Phenotypes Among Contemporary Cohorts of PHIV – Comparative Analysis of Population-Specific Physiology Studies Employing Cardiac Magnetic Resonance Imaging and/or Spectroscopy

Cardiac magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) technologies enable simultaneous, detailed characterization of myocardial structure and function^{74,75,76,77,78,79}. A series of cardiac MRI/MRS studies in contemporary cohorts of asymptomatic individuals with vs. without HIV have contributed to elucidating pathophysiologic mechanisms underlying HIV-associated myocardial fibrosis/steatosis and diastolic dysfunction.

Two European studies employing cardiac MRI/MRS to study a predominantly male cohort revealed PHIV (vs. controls) have an increased prevalence of patchy focal fibrosis, suggestive of prior myocarditis, as well as diffuse fibrosis and steatosis^80,81,82. Further, these studies showed evidence of diastolic dysfunction among PHIV^80,81,82. Specifically, Holloway et al. enrolled 90 PHIV on ART (76% men) and 39 controls (67% men) without prior CVD from the United Kingdom between 2011 and 2012⁸⁰. Groups were well-matched on age (mid 40’s) and BMI (24–25 kg/m²), although PHIV tended to have higher levels of circulating triglycerides and fasting glucose. Among PHIV, the authors noted an increased prevalence of focal patchy myocardial fibrosis (MRI: late gadolinium enhancement), an increased burden of diffuse myocardial fibrosis (MRI: T1 mapping), and apparent myocardial steatosis (MRS: intramyocardial triglyceride content 0.53 vs. 0.36% P=0.003). Further, PHIV evidenced signs of diastolic function (MRI: circumferential diastolic strain rate) as well as subtle systolic dysfunction (MRI: circumferential systolic strain rate). In multivariable modeling, HIV status remained an independent predictor of myocardial steatosis as well as both diastolic dysfunction and subtle systolic dysfunction⁸⁰. Through follow-up of the same cohort (augmented by the addition of ART-naïve PHIV), Ntusi et al. observed among PHIV frequent pericardial effusions and suggestive evidence of increased myocardial edema/inflammation⁸¹. Separately, a German study by Luetkens et al. centering on 28 asymptomatic ART-treated PHIV (79% men) vs. controls (68% men) similarly suggested among PHIV possible myocardial edema/inflammation, as well as reduced arterial distensibility⁸².

Building on these findings, two North American studies elucidated important functional consequences of myocardial steatosis among PHIV. Nelson et al. enrolled from the US 27 asymptomatic ART-treated men with HIV (MHIV) and 22 controls between 2012–2013⁷⁰. The authors demonstrated among PHIV marked myocardial steatosis (MRS: intramyocardial triglyceride content) and reduced diastolic function. Among PHIV, myocardial steatosis related directly to triglyceride levels and ART exposure and inversely to diastolic function⁷⁰. Thiara et al. enrolled from the US 95 asymptomatic PHIV (75% men; 93% on ART) and 30 matched controls (73% men) between 2010–2013. In this cohort, PHIV were slightly older than controls (49 years vs. 46 years), and both groups were borderline overweight (BMI 28–30 kg/m²)⁶⁹. While PHIV exhibited increased rates of smoking, diabetes, and lipid-lowering medication use, Framingham Risk Scores (FRS) were comparable between groups. Among PHIV, myocardial steatosis was noted (MRS: intramyocardial triglyceride content 1.14 vs. 0.58%, P=0.04), as was an increased burden of diffuse myocardial fibrosis (MRI: extracellular volume). Further, in this group, a positive correlation was discerned between the degree of steatosis and fibrosis. In multivariable modeling, women’s sex and visceral adiposity independently predicted steatosis among PHIV while women’s sex, hypertension, and HIV status predicted fibrosis among the whole group. Of note, among PHIV, neither HIV-specific parameters (viral load, CD4⁺ T cell count), nor measured indices of systemic immune activation related to fibrosis or steatosis. By contrast, relationships between levels of systemic immune marker monocyte chemoattractant protein-1 (MCP-1) and measures of subclinical systolic dysfunction were identified⁶⁹.

Most recently, a North American study focused on WHIV (vs. non-HIV-infected women) revealed diffuse myocardial fibrosis in relation to heightened systemic immune activation⁸³ as well as myocardial steatosis in relation to metabolic dysregulation⁷¹. In this study, 20 asymptomatic ART-treated women with HIV (WHIV) and 14 control subjects were recruited from the US between 2016–2017⁸³. Groups were well-matched on age (52–53 years) and BMI (32 kg/m²). As compared with non-HIV infected women, WHIV evidenced increased myocardial fibrosis (MRI: extracellular volume) and decreased diastolic function (MRI: circumferential diastolic strain rate). Additionally, novel systemic immune indices relevant to HIV-associated myocardial fibrosis and/or diastolic dysfunction were identified. Specifically, among WHIV, increased levels of the monocyte activation marker sCD163 related to myocardial fibrosis, while heightened expression of the cell-surface receptor CCR2 on inflammatory monocytes (CD14+CD16+) related both to myocardial fibrosis and diastolic dysfunction⁸³. Of interest, the monocyte-expressed honing receptor, CCR2, is known to promote cell-specific transmigration into target tissues among PHIV⁸⁴. It remains possible, then, that among WHIV, primed CD14+CD16+ monocytes more robustly hone to the myocardial structural space, where they may transform into inflammatory macrophages and engender a local fibrotic response. Within the same US cohort, Toribio et al. demonstrated that WHIV (vs. non-HIV-infected women) exhibited marked myocardial steatosis reflected in a three-fold increase in intramyocardial triglyceride content⁷¹. Moreover, among women, myocardial steatosis was noted to relate to metabolic and hormonal perturbations⁷¹.

Additional insights are anticipated from three ongoing cardiac MRI ± MRS studies being conducted among diverse cohorts of PHIV. Underway and fully enrolled, the NHLBI-funded CHART study (Characterizing HIV-related Diastolic Dysfunction) is assessing myocardial structure and function among ~200 asymptomatic US PHIV through application of cardiac MRI and dynamic ECHO⁸⁵. Detailed immunophenotyping as well as proteomic/metabolomic assessments will permit for sophisticated correlational analyses. Meanwhile, Baker, Ntsekhe et al. are applying cardiac MRI to assess subclinical myocardial pathology among asymptomatic PHIV from South Africa, as well as region-specific parameters fueling this pathology. Finally, Neilan, Zanni et al. are leading a study applying cardiac MRI/MRS to determine whether statin therapy (vs. placebo) forestalls the progression of myocardial fibrosis and myocardial steatosis among 129 asymptomatic PHIV recruited from the US and South Africa.

Preventive Considerations – Including Knowledge Gaps and Future Directions

How can we apply what we know about mechanisms underlying heart failure risk among ART-treated PHIV to conceptualize and critique possible preventive strategies? The first strategy to consider is immediate ART. Ample evidence supports the notion that unchecked HIV/AIDS and associated conditions adversely affect the heart muscle¹. Further, the START study confirmed that immediate ART upon HIV diagnosis reduces all-cause mortality⁸⁶. Thus, immediate ART, as recommended by the WHO, may help protect myocardial health among PHIV. However, we must augment our understanding as to which antiretroviral therapeutic regimens exert fewest cardiotoxic effects and work to expand access to these. Second, we must identify and target region-specific risk factors for myocardial damage. Such risk factors may include co-infections (e.g. tuberculosis), nutritional deficiencies, and/or toxic exposures^87,88,89,90. Third, we must target behavioral and traditional metabolic risk factors for myocardial disease including sedentary lifestyle, cigarette smoking, excess alcohol use, cocaine use, obesity, hypertension, dyslipidemia, and dysglycemia. Behavioral and traditional metabolic risk factors feed into several general myocardial injury mechanisms which are not specific to PHIV (ischemic heart disease, coronary microvascular disease, myocardial toxicity, valvular disease, and arrhythmia)⁹¹ but which may be synergistically activated by HIV-specific mechanisms. Fourth, we must determine the specific mechanisms by which systemic immune activation and metabolic dysregulation predispose to myocardial fibrosis and steatosis among PHIV and identify safe, targeted strategies to forestall these processes. Future research in the field will need to be attentive to the influence of underlying genetics, as well as sex, gender identity, race/ethnicity, and region-specific risk factors. Finally, educational outreach emphasizing heart failure risks among ART-treated PHIV will facilitate early detection of pre-heart failure phenotypes (e.g. exercise-induced dyspnea) and obviate diagnostic overshadowing (i.e. false attribution of heart-failure suggestive symptomatology to HIV itself or secondary infection). Though heart failure portends a poor prognosis, particularly among PHIV, general-population studies highlight the promise of efforts geared toward heart failure prevention⁹².