Inflammation and Heart Failure: Therapeutic or Diagnostic Opportunity?∗
The role of inflammation in the pathogenesis and progression of heart failure has important therapeutic and diagnostic implications. If inflammation is a direct cause of heart failure, then treatments targeting the immune response may be beneficial; however, if inflammation is primarily a marker of disease, then immune modulatory treatment is unlikely to work, but one might be able target patients with documented inflammation to identify and treat unrecognized left ventricular dysfunction.
The association of inflammation and heart failure has been recognized after many studies demonstrating that proinflammatory biomarkers are elevated in patients with a variety of cardiomyopathies (1–3). The levels of markers of inflammation have correlated with prognosis and severity of disease. These markers include proinflammatory cytokines such as tumor necrosis factor (TNF)-α, interleukin-1, interleukin-6, and galectin-3 (1). Importantly, the association between heart failure and markers of inflammation is observed for both reduced and preserved ejection fraction cohorts (1).
Additional support for the hypothesis that inflammation is a direct cause of heart failure comes from evidence of viral infection in patients with heart failure and the associated immune response. Patients with dilated cardiomyopathy have a detectable viral genome in their myocardium in >60% of cases (1). In addition, persistent virus in patients with myocarditis has been associated with persistent left ventricular dysfunction but not with the development of chronic heart failure (1). Complicating the interpretation is that although most markers of immune activation are associated with negative cardiac effects, certain CD4+ regulatory T cells appear to be cardioprotective, preventing progression of heart failure in a pressure-overload model (3). Thus, our understanding of the balance of immune-mediated injury and repair is limited, particularly for the chronic phase of myocardial injury.
The study by Mantel et al. (4) in this issue of the Journal provides evidence that inflammation is part of a direct causal pathway of heart failure. Prior studies have indicated that the prevalence and incidence of heart failure are increased in patients with rheumatoid arthritis; however, it has been unclear whether this is just the result of more hypertension and ischemic heart disease, both of which are increased in patients with rheumatoid arthritis, or whether the immune disorder directly causes myocardial dysfunction. The authors addressed this question using data from the Swedish rheumatology registry and matched control subjects from the Swedish total population registry, which tracks health data on all Swedish residents. At the time of the diagnosis of rheumatoid arthritis, heart failure was no more common than in the general population; however, during follow-up, heart failure occurred more frequently in the rheumatoid arthritis group (4.1 per 1,000 person-years) than in control subjects (3.2 per 1,000 person-years), and this corresponded to a highly significant hazard ratio (HR) of 1.22. Importantly, the increased risk was similar for those with nonischemic heart failure (HR: 1.22) and ischemic heart failure (HR: 1.27). The risk was strongest for those positive for rheumatoid factor (HR: 1.45 vs. 0.97 for those negative for rheumatoid factor). This finding that a chronic immune disorder can lead directly to nonischemic heart failure suggests that treatment of chronic inflammation might prevent or slow the progression of heart failure for many patients with or without immune disorders.
Unfortunately, the results of randomized trials targeting inflammation have been disappointing (1,2). CORONA (Controlled Rosuvastatin Multinational Trial in Heart Failure) found a significant reduction in high-sensitivity C-reactive protein but no mortality reduction in older patients with heart failure treated with rosuvastatin. Similarly, the GISSI-HF (Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico Heart Failure) trial did not find a reduction in mortality or hospitalization for heart failure with rosuvastatin treatment. Although initial reports suggested that inhibition of TNF-α with etanercept might improve cardiac performance, 2 larger trials were discontinued because of futility (1). A third trial using an alternative anti-TNF chimeric monoclonal antibody (infliximab) was stopped because of harm in the intervention arm. Placebo-controlled trials of intravenous immunoglobulin and immunomodulation of cytokines have been similarly unsuccessful (1). In a meta-analysis of patients with acute myocarditis, corticosteroids appeared to improve left ventricular ejection fraction but had no impact on survival (5).
If treatment of inflammation is not yet useful for improving heart failure outcomes, can we improve diagnosis? Improved survival is the primary motivation for detecting heart failure in its earliest stages. Asymptomatic patients with reduced left ventricular ejection fraction (<40%, stage B heart failure) have been shown to have improved outcomes with angiotensin-converting enzyme inhibitors and β-blockers, and their use is recommended by clinical guidelines. In addition to those who are asymptomatic, certain patients with rheumatoid arthritis and other chronic immune disorders may be too limited in activity level to describe symptoms of heart failure despite impaired cardiac function. In addition, some patients may describe the nonspecific symptoms of heart failure such as fatigue and shortness of breath but have these symptoms attributed to their immune disorder.
The potential for benefit in those with unrecognized heart failure raises the question of whether one should screen patients in some way. Possibilities for such screening include nonspecific markers of left ventricular dysfunction such as B-type natriuretic peptide (BNP) or N-terminal pro-BNP or a limited echocardiogram examination. Prior work has suggested that screening with such tests could be cost-effective if the prevalence of unrecognized left ventricular systolic dysfunction were at least 1% (6). Prior studies from the United States and Europe suggest this value may range from 0.75% (women) to 2.5% (older men), and for every patient with recognized heart failure and systolic dysfunction, there is a person with unrecognized or asymptomatic systolic dysfunction. In the present study by Mantel et al. (4), the prevalence of recognized heart failure was 2.5%, which was 77% higher than in the general population. Thus, it is likely that there is at least a 1% prevalence of asymptomatic left ventricular dysfunction in patients with rheumatoid arthritis, which makes this group a reasonable target for screening. Indeed, a small study of patients with rheumatoid arthritis and no known cardiac disease who were screened with BNP found a 1% prevalence of systolic dysfunction (7).
In summary, although effective treatments for heart failure targeting inflammation are at least years away, an immediate clinical implication of the association of rheumatoid arthritis and heart failure is the potential for improved diagnosis of left ventricular dysfunction.
1. : "Chronic heart failure and inflammation: what do we really know?". Circ Res 2016; 119: 159.
2. : "Inflammation in heart failure: the holy grail?". Hypertension 2016; 68: 27.
3. : "The role of inflammation and cell death in the pathogenesis, progression and treatment of heart failure". Heart Fail Rev 2016; 2: 169.
4. : "Association between rheumatoid arthritis and risk of ischemic and nonischemic heart failure". J Am Coll Cardiol 2017; 69: 1275.
5. : "Corticosteroids for viral myocarditis". Cochrane Database Syst Rev 2013; 10: CD004471.
6. : "Cost-effectiveness of screening with B-type natriuretic peptide to identify patients with reduced left ventricular ejection fraction". J Am Coll Cardiol 2004; 43: 1019.
7. : "Use of B-type natriuretic peptide as a screening tool for left ventricular diastolic dysfunction in rheumatoid arthritis patients without clinical cardiovascular disease". Arthritis Care Res (Hoboken) 2011; 63: 729.
Dr. Heidenreich has reported that he has no relationships relevant to the contents of this paper to disclose.