Myocarditis

No population-based epidemiological study has comprehensively documented the range of clinical presentations of acute and chronic myocarditis, likely because of 2 factors: the protean presentation of patients continues to challenge healthcare professionals in their differential diagnostic efforts, and the lack of reliable noninvasive tests reduces the yield conclusive diagnostics or accurate prognostic value. The gold standard for diagnosing suspected myocarditis remains the EMB; however, improved molecular and histological tests using tissue samples reveal great variability, partly because of a variety of factors including interobserver variability, tissue quality, stage of presentation, causes, and disease severity. Similarly, myocardial inflammation and injury may be diffuse or focal, involving any of the cardiac chamber walls to greater or lesser degrees concurrently, and at various stages of disease.

Differential diagnosis of myocarditis is a great concern because of the heterogeneous nature of the clinicopathologic scenarios.^35a,b For example, Pieroni et al³⁶ has recently demonstrated that patients with right ventricular myocarditis are indistinguishable from arrhythmogenic right ventricular cardiomyopathy on the basis of clinical features, presence, and severity of structural and functional right ventricle abnormalities. Other study reported similar observations that myocarditis involving the right ventricle can mimic arrhythmogenic right ventricle cardiomyopathy.³⁷ Myocarditis may also present similarly to ischemic heart disease with chest pain, abnormalities on electrocardiograms, and elevated cardiac biomarkers. Furthermore, myocarditis may seem similar to other causes of HF, including cardiac amyloidosis and hypertrophic cardiomyopathy.^38,39 In all, thorough consideration must be taken when firmly diagnosing suspected myocarditis. Although the real prevalence of myocarditis is likely underestimated, its presentation, features, and course remain a challenge to be resolved.

The majority of individuals with acute myocarditis who present with acute DCM have a surprisingly mild disease that may resolve within a few days. Frequently, these individuals are experiencing hypersensitivity myocarditis in the context of peripheral eosinophilia and fever, often with recent initiation of new medication(s). Yet, patients who present with giant cell myocarditis and eosinophilia myocarditis are at high risk of downstream difficulties. These patients may also present with thymoma,^40,41 autoimmune disorders,⁴² ventricular tachyarrhythmia, or advanced heart block. Moreover, for patients who fail to respond positively to standard supportive care, cardiac sarcoidosis should be considered in patients with DCM, chronic HF, or ventricular arrhythmias.

Many cases of myocarditis likely go undetected because of asymptomatic presentations or nonspecific symptoms. For example, a recent study suggests that smallpox vaccination results in ≈1:200 patients having elevated troponin levels, whereas the prevalence of myocarditis has been estimated at ≈1:5500.⁴³ Creatine kinase was previously a preferred biomarker for myocarditis; however, its sensitivity in late diagnosis is poor because studies show that creatine kinase decreases steadily back to basal levels by ≈36 hours.^44,45 Elevated levels of soluble ST2, a truncated soluble receptor of interleukin-33, have also recently been demonstrated to be correlated with HF and DCM cases and have been gaining popularity for detecting HF.⁴⁶ However, measuring levels of biomarkers in patients presenting with suspected myocarditis remains low in dependability but may still aid in confirming a diagnosis of myocarditis.^47,48 Other nonspecific serological markers of inflammation including leukocyte count and C-reactive protein may be elevated in suspected myocarditis. However, nonelevated levels fail to rule out an inflammatory response in the myocardium,⁴⁹ suggesting that serological tests for myocarditis remain a poorly developed method of diagnosis, with relatively low prognostic value.

One approach to distinguishing between patients presenting with acute myocardial infarction or myocarditis is by obtaining an ECG. Despite the low sensitivity of electrocardiograms, it is widely used as an initial screening tool.⁵⁰ Electrocardiograms of confirmed myocarditis patients may exhibit nonspecific T waves and ST-segment changes including ST-segment elevation.⁵¹ Recently, a study by Ukena et al⁵² investigated the prognostic value of electrocardiography in patients suspected to have myocarditis and found that a prolonged QRS duration is a significant independent predictor for cardiac death or heart transplantation. In cases of acute myocarditis, electrocardiograms may show sinus tachycardia. In general, changes in electrocardiograms suggest signs of myocardial infarction with ST-segment elevation. Electrocardiograms provide a convenient tool for risk stratification and initial screening but provide weak diagnostic value.

Echocardiography is a useful measurement tool in the diagnostic assessment of suspected myocarditis and to rule out other causes of HF. Echocardiography investigates cardiac chamber sizes, wall thicknesses, and systolic and diastolic functions, and thus does not provide direct evidence of myocarditis. However, patients with fulminant myocarditis often lack cardiac dilation and exhibit increased septal thickness, whereas patients with acute myocarditis may exhibit a spherical-shaped ventricle that remodels to a more elliptical shape over the period of a few months with normal wall thickness.^6,53 In addition, patients with fulminant myocarditis exhibit substantial recovery in ventricular function by 6 months when compared with patients with acute myocarditis.⁶

Speckle tracking has recently been a useful technique in diagnosing patients with acute myocarditis. A recent study by Hsiao et al⁵⁴ demonstrated that LV strain and strain rate may be useful measurements for diagnosis of suspected myocarditis patients, even those with preserved LV ejection fraction (LVEF). In addition, speckle tracking is useful for predicting deterioration and overall event-free survival. In general, strain rate and global strain values were decreased in myocarditis patients when compared with healthy individuals.

To date, cardiac magnetic resonance imaging (cMRI) is mostly used to detect significantly decreased EFs and abnormalities in wall motion. Contrast enhancement (CE) CMR is a more sensitive technique of cMRI and can detect areas of myocardial damage in patients with acute myocarditis. Studies evaluating the diagnostic value of cMRI demonstrate that late gadolinium enhancement imaging is a superior method in diagnosing myocarditis because of the improved difference in signal intensity between myocardial regions with and without gadolinium-diethylenetriaminepentacetate.⁵⁵ Moreover, cMRI sensitivity and specificity are as high as 100% and 90%, respectively, when compared with histology confirmed samples identified by CE-CMR.⁵⁶ In patients with acute myocarditis, areas of CE are often located in the lateral wall originating from the epicardial quartile, even if the pattern of myocardial injury is influenced by viral infection.⁵⁷ This technique also rules out ischemic causes because CE in ischemic infarction will include subendocardial layers of the myocardium. From our observations and understanding, eosinophilic myocarditis is the only form of myocarditis with striking involvement of the endocardium.

A study conducted by Mahrholdt et al⁵⁸ demonstrated that late gadolinium enhancement imaging of confirmed virus-induced myocarditis exhibited either a rim-like pattern in the septal wall or a subepicardial distribution in the free LV lateral wall that recovered within months of detection in the case of PVB19, whereas patients coinfected with both PVB19 and HHV6 progressed toward chronic HF. cMRI can also provide indirect evidence of viral involvement because PVB19 patients exhibit CE most frequently in the lateral wall, whereas HHV6 exhibits CE in the midwall of the interventricular septum.⁵⁸ The caveat for cMRI is that it lacks the ability to determine the magnitude of myocardial inflammation and depends heavily on the clinical context. Moreover, the high sensitivity of cMRI is only found in cases of acute myocarditis because diagnostic features of chronic myocarditis resolve after several weeks to months. However, because patients presenting with suspected chronic myocarditis often lack viremia, chronic myocarditis can be inferred when chronic HF or myocardial fibrosis can be excluded. cMRI is a reliable method of detecting myocardial fibrosis in DCM. A study by De Cobelli et al⁵⁹ was able to use CE to identify areas of myocardial inflammation in 70% of biopsy-proven chronic myocarditis. To date, available data on the diagnostic accuracy of cMRI are limited although CE-CMR is becoming useful in obtaining accurate and reliable preliminary information in a noninvasive manner.^59a Because of the ability of CE-CMR to locate areas of inflammation, it can serve as a guidance for EMB sampling when necessary,^59b thus enhancing the diagnostic accuracy of biopsies for further testing using conventional histopathology and molecular polymerase chain reaction techniques for cellular infiltrates and viral presence.^59c

EMB positivity in acute myocarditis patients is defined as lymphocytic infiltration in association with myocyte necrosis/death (Dallas Criteria). Patients exhibiting borderline myocarditis are defined as those with lymphocytic infiltration in the absence of myocyte necrosis.⁶⁰ EMB is currently the best method to differentiate between giant cell myocarditis and sarcoidosis. For example, giant cell myocarditis exhibits an abundance in eosinophils, the lack of sarcoidosis or granulomas, and never affects the pericardium. In addition, eosinophilic myocarditis is best identified by myocyte necrosis accompanied by the presence of intracavitary thrombi containing eosinophils observed in the lumen of intramyocardial coronary vessels. Also, an observational study investigating patients with progressive HF and EMB-confirmed lymphocytic myocardial infiltration showed that the positive detection of enterovirus genomes in ventricular myocardium correlated with worsened outcomes.⁶¹ A study by Frustaci et al⁶¹ treated 41 patients, presenting with progressive HF, with immunosuppressive therapy (prednisone and azathioprine) for 6 months. Of 41 patients, 21 exhibited improved LVEFs, whereas 20 failed to respond. Of these 20 patients who failed to respond to immunosuppressive therapy, 17 patients were detected positive for enteroviral genomes, whereas only 3 patients who responded to treatments tested positive for hepatitis C virus genomes. Consistent with this observation, Kühl et al²⁷ demonstrated that viral persistence in the myocardium is associated with compromised LV function, whereas patients with viral clearance have improved LV function. In contrast, there are many examples of patients who tested positive for viral genomes including HHV6 and PVB19, yet had noted absence of infiltrating cellular immune responses in the myocardium and did not satisfy the Dallas criteria for myocarditis.^62–65

It is important to note that the use of EMB varies widely based on the physician and clinical presentation of the patients. For example, the European Society of Cardiology recommends the use of EMB as early as possible for histology,² immunohistochemistry, and viral polymerase chain reaction analysis. Whereas the American College of Cardiology/American Heart Association recommends EMB based on 3 classes of comprehensive clinical presentations, risks, and measurements using noninvasive techniques.⁶⁶ In brief, the use of EMB by the European Society of Cardiology and American College of Cardiology/American Heart Association both aim to clearly define myocarditis with the least amount of risk to the patient, but vary in the timing of EMB sampling. It is worth noting that the sensitivity of the EMB for detection of myocarditis is rather low.^66a

Immunohistochemistry on EMB samples has been gaining popularity in characterizing myocarditis. Caforio et al¹⁹ have demonstrated that immunohistochemistry yields increased sensitivity and shows that more than half of previous cases of borderline myocarditis diagnosis would be missed by hematoxylin-eosin alone. Moreover, Kindermann et al⁶³ show that 69 of 181 (38%) EMB samples are positive for myocarditis using the Dallas criteria, whereas 91 of the 181 (50%) had positive staining for CD3, CD68 (macrophages), and human leukocyte antigen class II antibodies. Currently, endomyocardial biopsies provide relatively high diagnostic and prognostic value in suspected myocarditis patients with low complication rates of <1%.^63,67

Evolving molecular technologies have led to the characterization of miRNA profiles among acute, chronic, and fulminant myocarditis, including a relationship to the severity of myocardial damage. For example, miR-208b and miR-499 are upregulated upon myocardial damage and can be detected in plasma of myocarditis patients.⁶⁸ Furthermore, although miR-208b is upregulated in myocarditis patients, simultaneous upregulation of both miR-208b and miR-499 is only observed in fulminant virus-induced myocarditis.⁶⁸ Other miRNAs demonstrated to be altered in acute myocarditis patients include miR-155, miR-21, miR-146b, miR-511, and miR-212.^69–71 Currently, no miRNAs have been found to be specific for patients with chronic myocarditis. The altered expression of specific miRNAs during the pathological progression of myocarditis makes miRNAs interesting future targets for treating and diagnosing suspected myocarditis patients with greater accuracy.

There is considerable evidence that reflects an important role for the direct effect of viral infection on cardiomyocytes in the development of viral myocarditis. Early research has shown that CVB3 infection is sufficient to induce myocardial injury in severe combined immune deficient mice in which mature T and B lymphocytes are absent.⁷⁶ It was also demonstrated that cardiac-specific expression of a replication-restricted CVB3 genome in mice is able to induce a DCM phenotype in the absence of an apparent host immune response.⁷⁷ Other evidence of the contribution of viral direct damage on the myocardium is provided by the findings that elimination of the coxsackievirus-adenovirus receptor (CAR) specifically in the adult mouse hearts completely blocks viral infection of myocardial cells and virus-induced inflammation in the myocardium.⁷⁸ These CVB3-infected mice exhibit improved cardiac function, whereas other tissues (eg, pancreas and spleen) that also express CAR are CVB3 positive.⁷⁸ This study suggests that viral infection of cardiomyocytes and consequent productive viral replication within the myocardium is required for the development of myocarditis. CAR, a transmembrane protein in the family of adhesion molecules, localized predominantly at the intercalated disc and the cell–cell junctions of the AV node in the adult heart is required for virus entry into different cell types.^79,80 Studies from independent research groups using CAR-deficient mice have revealed an important role of the CAR in embryonic development and the maintenance of normal AV conduction and cardiac function.^79–81 The observation of increased expression of CAR in the heart of patients with DCM and relatively high levels of CAR in young adult heart may partially explain the increased susceptibility of young hearts to myocarditis.⁸² In addition to CAR, viral entry or virulence of CVB3 is also determined by decay-accelerating factor (CD55) coreceptor in certain cell types.⁸³ For instance, CVB3 binding with decay-accelerating factor on the apical surface of epithelial cells promotes the interaction between virus and CAR, which is localized to the tight junction; normally inaccessible to virus.⁸⁴ The mechanisms underlying direct myocardial injury have been extensively studied. Enteroviruses encode 2 proteinases, 2A and 3C, which are crucial in the completion of the full viral lifecycle by processing viral polyprotein into individual structural and nonstructural proteins. These viral proteinases also contribute to the cellular and molecular phenotype of myocarditis by targeted cleavage of host proteins essential for the regulation of protein translation, transcription, cardiac contraction, signaling transduction, and host structural proteins.^85,86 The list of protein substrates is growing rapidly (Table 2). Recently, a proteomic approach allowed for global screening of viral proteinase targets.¹²³ Cardiac-specific expression of viral proteinase 2A is sufficient to induce a DCM phenotype.¹²⁴ Notably, dystrophin and dysferlin deficiency confers increased susceptibility to CVB3 infection of the heart by enhancing viral propagation to adjacent cardiomyocytes and via disrupting membrane repair function.^99,125 It has been suggested that these proteins may serve as genetic predisposing factors for viral myocarditis. Wong et al¹⁰⁶ also revealed that serum response factor, a muscle-enriched transcriptional factor controlling the expression of many cardiac contractile and regulatory genes, is cleaved through the activity of viral proteinase 2A, partially contributing to virus-induced cardiac dysfunction. Virus can produce direct cytotoxicity in the heart through targeting host protein translational machinery^111–113 and inducing apoptosis by direct processing of caspases.¹²⁶ Increasing evidence reveals that aberrant accumulation of protein aggregates plays an important role in the development of human heart diseases.^127–129 Misfolded protein aggregates are frequently detected in common heart diseases and are thought to be highly toxic to cardiomyocytes.^130,131 Abnormal accumulation of ubiquitin conjugates that occur in CVB3-infected cardiomyocytes has also been demonstrated, suggesting that protein homeostasis is disrupted in CVB3-infected hearts.^86,132,133 In addition, cleavage of autophagic adaptor proteins sequestosome/p62 and neighbor of BRCA1 gene protein, as well as RNA-binding protein transactive response DNA-binding protein-43, has also been documented after CVB3 infection, contributing to the increased accumulation of protein aggregates.^96,110,119 Together, the loss of cellular homeostasis, translation/transcription shutoff, contractile dysfunction, and apoptosis as a result of viral proteinase activity may contribute to direct damage of infected cells (Figure 1).

Analogous to many other viruses, CVB3 has evolved processes to subvert or manipulate the host cellular machinery to ensure productive viral infection. For example, virus uses the host signaling and miRNA system to enhance viral replication, contributing to the pathogenesis of myocarditis and the progression to end-stage DCM. Multiple cellular signaling pathways are known to be activated upon viral infection and play a critical role in regulating viral infectivity at various steps of viral lifecycle, including viral entry, replication, release, and evasion from host immune responses.⁷⁵ Coyne and Bergelson⁸⁴ demonstrated that viral entry into epithelial cells through tight junctions is mediated by Fyn and Abl kinases. Studies have also revealed that the activation of the tyrosine protein kinase p56^lck, a member of the Src family kinases, the extracellular signal-regulated kinase 1/2 mitogen-activated protein kinase, and protein kinase B signaling as a result of CVB3 infection of immune and myocardial cells is required for efficient viral replication and is associated with host susceptibility to viral myocarditis.^134–138 In addition, virus-induced p38 mitogen-activated protein kinase and glycogen synthase kinase-3β activation induces increased cell death and consequent viral progeny release.^139,140 Administration of a p38 inhibitor to CVB3-infected mice reduces viral replication, attenuates myocardial damage, and improves cardiac function, suggesting a potential value for CVB3-induced myocarditis.¹⁴¹ It is notable that these signaling pathways do not work independently; indeed, CVB3 can subvert the host cell signaling networks cooperatively to ensure optimal viral infection.^142,143 A statistical model has been built to systematically analyze and predict the intracellular signaling responses to CVB3 infection.¹⁴³

The interplay between CVB3 and host miRNAs also plays a pivotal role in viral pathogenesis (Figure 2). The miRNAs comprise a large, novel class of noncoding RNAs (20–24 nucleotides) that regulate gene expression by targeting messenger RNAs for degradation and translational repression.^144,145 Ho et al¹⁴⁶ conducted the first investigation on the role of miRNAs in CVB3 replication, where upregulation of miR-141 upon CVB3 infection is partially responsible for the shut-off of host protein translation via targeting the cap-binding protein eukaryotic initiation factor 4E. Following this study, several other miRNAs were identified as induced after CVB3 infection. For instance, miR-203 is upregulated upon CVB3 infection through the activation of the protein kinase C/transcription factor AP-1 pathway, and in turn, enhances CVB3 replication by targeting zinc finger protein-148 to promote cell survival during early infectious stages.¹⁴⁷ Similarly, miR-126, which is upregulated during infection through a sprout-related EVH1 domain containing 1-extracellular signal-regulated kinase 1/2 positive feedback loop, facilitates viral propagation and contributes to viral cytopathogenicity by suppressing LRP6 and WRCH1 genes.¹⁴⁸ In addition to negative regulation of gene expression profiles, miRNAs can also positively affect the translation of their targets.¹⁴⁹ For example, it was shown that miR-10a* promotes CVB3 biosynthesis by targeting the 3-dimensional coding sequence of CVB3 genomic RNA.¹⁵⁰ Moreover, certain miRNAs were found to contribute to the development of viral myocarditis without evidence of altering viral replication dynamics. One example is miR-21, whose change in expression levels during CVB3 infection is controversial because some studies indicated its upregulation,^70,151 whereas others demonstrate downregulation.¹⁵² Nevertheless, it was found that a high level of miR-21 does not influence CVB3 replication but leads to disrupted myocyte interaction by suppressing and disorientating critical components of intercalated disks.¹⁵³ Another example is provided by miR-1, which is induced in CVB3-infected mouse hearts.¹⁵⁴ Upregulation of miR-1 leads to repression of its target gene GJA1 (connexin-43), suggesting a possible mechanism by which CVB3 impairs the function of cellular gap junctions and triggers arrhythmias.¹⁵⁴ Although most miRNAs upregulated in CVB3 infection facilitate viral replication and pathogenesis, some can also act as host defensive components against viral infection. MiR-221 and miR-222, 2 miRNAs significantly elevated during CVB3-induced viral myocarditis, however, directly target several genes benefiting viral replication and inflammation, including genes encoding c-ets-1, interferon regulatory factor 2, bcl-2-interacting mediator of cell death, thymocyte selection-associated high mobility group box, bcl-2 modifying factor, and chemokine (C-X-C motif) ligand 12. Inhibition of miR-211/miR-222 augments viral load and strongly aggravates CVB3-induced cardiac injury, whereas a high level of these 2 miRNAs reduces viral replication.¹⁵⁵ In brief, accumulated evidence indicates that CVB3-induced differential expression of miRNAs plays a crucial role in modulating host and viral gene expression, and in turn directly or indirectly promotes CVB3 pathogenesis.

Viral infection of the heart triggers the activation of the host antiviral immune response, which is characterized by the infiltration of natural killer cells and macrophages, followed by virus-specific T lymphocytes. The immune response functions as a double-edged sword: initial activation is beneficial to the host by limiting viral spread; however, a persistent and excessive immune response conveys harmful consequences contributing to the progression of myocarditis and DCM. The balance between the antiviral influences and the deleterious effects on cardiac function is an important determinant of the severity of myocarditis and the ultimate progression to DCM.

The innate immune response is essential for host defense early during an infection. It is mainly mediated through cytokines, including interleukins, tumor necrosis factor-α, and interferons (IFNs). The pattern recognition receptors, including the toll-like receptor (TLR), retinoic-acid inducible gene-I–like receptor, nucleotide-binding oligomerization domain-like receptor, and C-type lectin receptor, play a key role in the innate immune response by detecting specific pathogen-associated molecular patterns that are primarily present in invading microbes.¹⁵⁶ These receptors can recognize the CVB3 genome and trigger an intracellular signaling cascade leading to the production of type I IFNs and proinflammatory cytokines.⁸⁵ Type I IFNs, including IFN-α and IFN-β, serves as important antiviral mediators to inhibit CVB3 replication.^157–159 It was reported that certain TLR3 polymorphisms are associated with increased susceptibility to enterovirus-myocarditis and DCM in patients.¹⁶⁰ Animal studies using TLR3- and TLR4-deficient mice demonstrate a protective function for both TLR3 and TLR4 in CVB3-induced myocarditis and DCM.^161–163 Moreover, animal studies suggest that sex differences in TLR signaling play an important role in differential susceptibility to CVB3-induced myocarditis in men and women.^21,164,165 The antiviral significance of IFNs has also been observed at the level of their gene targets. For example, IFN-inducible protein 10 is upregulated in viral-infected myocardium, and overexpression of this gene specifically in cardiomyocytes decreases virus titers leading to improved cardiac function.¹⁶⁶ Furthermore, ablation of IFN stimulated gene 15 results in amplified virus yields and increased severity of myocarditis and mortality.¹⁶⁷ In addition to their function in extracellular matrix remodeling,^168–171 a critical process involved in the progression of myocarditis to DCM, matrix metalloproteinases also serve as important modulators of the antiviral immune response. A recent study has revealed a novel role for matrix metalloproteinase-12 in innate immunity via mediating the secretion of IFN-α by transcriptional regulation of inhibitor of kappa B alpha, alpha (IκBα).¹⁷² The absence of matrix metalloproteinase-12 in mice was demonstrated to prevent the release of IFN-α from the cytoplasm and increase the severity of viral myocarditis.¹⁷²

To dampen the host antiviral innate immunity, viruses have developed sophisticated mechanisms. Viral proteinase 2A antagonizes the type I IFN signaling by cleaving retinoic-acid inducible gene-I–like receptor melanoma differentiation-associated protein 5 and mitochondrial antiviral signaling proteins.⁸⁷ It was also reported that the mRNA level of the suppressor of cytokine signaling proteins, which inhibit the innate immunity by suppressing the janus kinase/signal transducers and activators of transcription signaling, is upregulated in response to CVB3 infection to counteract the host defense mechanism.^173,174

The initial beneficial role of the immune response can become detrimental to the myocardium if it is not properly regulated. For example, sustained production of the proinflammatory cytokines, such as tumor necrosis factor-α^175–178 and interleukin-1,^178,179 after viral infection can exacerbate cardiac damage, leading to impaired cardiac function. In addition, the balance between regulatory T (Treg) and interleukin-17–producing (Th17) cells has also been suggested to play an important role in the pathogenesis of viral myocarditis.¹⁸⁰ Treg inhibits proinflammatory responses, resulting in reduced immunopathology, whereas Th17-mediated immunity induces enhancement of immunopathology.¹⁸⁰

The adaptive immunity responses begin later in the acute and subacute phases of myocarditis. Lymphocyte infiltration can aid with the clearance of virus-infected cardiomyocytes, but it may also cause myocardial damage. The latter effects are highlighted by the early evidence that depletion of T lymphocytes decreases mortality rate and reduces cardiac inflammation and injury after CVB3 infection.¹⁸¹ It was subsequently shown that the absence of both CD4⁺ and CD8⁺ T-lymphocytes attenuates the pathological phenotype of myocarditis and significantly reduces mortality.¹⁸² Molecular mimicry between viral and host antigens has also been proposed to be a mechanism responsible for autoimmune-mediated destruction of the myocardium.¹⁸³ After viral infection, exposure of cardiac antigens, such as cardiac myosin that cross-react with coxsackieviral antigens, to the immune system elicits an autoimmune response, leading to the damage of cardiomyocytes through production of autoantibodies and autoreactive immune cells.¹⁸³ The miRNAs also play a role in the unfavorable inflammatory response to viral infection of the heart. It was reported that miR-155 is strongly induced on CVB3 infection and is colocalized with infiltrated inflammatory cells in virus-infected mouse heart and human myocarditis.¹⁵¹ Inhibition of miR-155 results in attenuated inflammatory infiltration, improved cardiac function, and reduced mortality in a mouse model of viral myocarditis.¹⁵¹

Early application of antiviral agents is a potential therapeutic avenue to halt the development of viral myocarditis. Several antiviral agents have shown favorable effects in clinical trials. IFNs, the most commonly used antiviral agent, have been assessed in several circumstances as a treatment for myocarditis. In a Phase II clinical study, IFN-β was administered to 22 patients presenting with persistent LV dysfunction coinciding with enterovirus or adenovirus infection. The outcome was positive demonstrating complete viral clearance indicated by untraceable viral genomes in all patients and improved LV function in 15 patients.¹⁸⁵ In another instance, IFN-α2a was used in the treatment of 2 patients with confirmed myocardial enterovirus infection with outcomes similar to IFN-β treatment.¹⁸⁶

In addition to the agents mentioned above, more promising antiviral strategies are being tested in laboratories. In nonhuman models, more data have been gathered indicating the potential efficacy of IFN in the treatment of myocarditis.¹⁸⁷ Chemical compounds have also been demonstrated to inhibit myocarditis-associated viruses, but their direct mechanistic and clinical evidence is lacking.^188–190 Administration of natural products such as astragaloside IV can significantly decrease CVB3 titers by upregulating IFN-γ as well as reducing mononuclear cell infiltration and cardiomyocyte injury in mice.¹⁹¹ Chen et al¹⁹² showed that astragaloside IV significantly decreases the fibrosis of the heart muscle tissue and increases the murine survival rate caused by the inhibition of the TGFβ 1-Smad signaling in DCM.

The clinical application of antiviral agents is limited by the nonspecific targeting, unclear mechanisms, and side effects. The development of nucleic acid–based antiviral agents has introduced more specific strategies against viral infection, which allows one to target the viral genome directly. Antisense oligodeoxynucleotides were found to inhibit gene expression by binding specific sequences of mRNA and through cleavage of target RNA strands in a DNA–RNA duplex by the cellular RNase H.¹⁹³ Thus, antisense oligodeoxynucleotides designed to target viral gene sequences can be used to inhibit viral gene expression and inhibit viral replication. An early study designed and tested 7 antisense oligodeoxynucleotides that target various regions of CVB3 genome RNA; all of which demonstrated strong antiviral activity in both cultured mammalian cells and a mouse myocarditis model.¹⁹⁴ RNA interference is an another tool that can be used to diminish the infection of viruses. Previously, artificially designed small interfering RNAs targeting several regions of CVB3 transcripts have been shown to decrease cell death in cell culture models.^194–196 In addition, host cellular genes involved in viral replication can also be used as potential therapeutic targets of small noncoding RNAs. For example, small interfering RNAs targeting components in critical pathways for viral pathogenesis, including the ubiquitin-proteasome pathway, the lysosome pathway, and the autophagy pathway, result in significant inhibition of CVB3 replication and myocarditis.^{86,96,133,197,198} However, such data are obtained from studies using cultured mammalian cells, and the therapeutic role of these small noncoding RNAs in myocarditis should be further verified in vivo. Fewer studies have substantiated the effects of small noncoding RNAs in the treatment of myocarditis. The limitation of using small noncoding RNAs is the lack of a specific mode of delivery to target viral-infected cells. An attempt to address this issue is the use of artificial miRNAs that specifically target CVB3 genomes.¹⁹⁹ These studies indicate a potential trend in the targeted delivery of antiviral drugs although its data are limited.

Preventing viral entry is an another strategy to confine viral spread and attenuate viral myocarditis. A previous study used the treatment of cell cultures with a modified CAR or decay-accelerating factor containing the human IgG1 fragment crystallization domain on the C-terminal region of the molecule, resulting in a dimeric antibody-like molecule. These recombinant CAR-fragment crystallization and decay-accelerating factor-fragment crystallization proteins competitively bind to soluble viral particles and aid in viral-specific activation of the innate and adaptive immunity.^200–203 Simultaneous administration of soluble CAR-fragment crystallization and small interfering RNA targeting CVB3 genome exerts synergistic antiviral activity in the treatment of a persistently infected cardiac cell line in vitro.²⁰⁴ However, the utility of recombinant proteins in disease treatment needs more optimization to avoid unexpected side effects. Currently, there is an abundance of animal data on antiviral strategies in myocarditis and a lack of evidence in human trials.

The acute phase of myocarditis characterized by intense pathogen infection and myocyte necrosis and apoptosis lasts only for a few days. After the acute phase, the further development of myocarditis in the subacute phase is thought to be predominantly a result of exaggerated autoimmune responses.²² Thus, the application of immunosuppressive drugs seems plausible for the treatment of myocarditis. In a clinical trial, administration of immunosuppressant drug azathioprine and prednisone improved the LVEF and New York Heart Association functional class of patients with myocarditis and symptoms for >6 months.²⁰⁵ Similar results were reported in another trial of patients with chronic nonviral myocarditis and symptomatic HF, in which both LVEF and New York Heart Association functional class were also improved after treatment with azathioprine and prednisone. Despite the favorable outcomes of immunosuppressant use for chronic myocarditis, most attempts to alleviate acute myocarditis using anti-inflammatory drugs or immunosuppressants have failed, with little positive or even adverse effects.¹⁸ Similarly, adiponectin mediates cardioprotection and favors positive outcome in patients with nonviral inflammatory cardiomyopathy.^206,207 But in a recent study, administration of adiponectin in CVB3-induced acute myocarditis mouse models promoted the development of myocarditis by suppression of TLR-dependent innate immune responses.²⁰⁸

Several other immune modulators have been tested in experimental animal models or cell cultures and exhibited inspiring therapeutic potentials. Nonsteroidal anti-inflammatory drugs are considered effective anti-inflammatory therapeutics and have been applied in the treatment of pericarditis.²⁰⁹ Nonsteroidal anti-inflammatory drugs have shown increased myocardial inflammation and mortality in murine models with acute viral myocarditis, however, should not be considered as an antiviral therapy.^210,211 Intravenous immunoglobulin is an another type of immunomodulatory agent that also shows antiviral effects in experimental models; however, data remain controversial.^16,212

Agents modulating specific genes can also be used to relieve exaggerated autoimmune responses in myocarditis. For instance, tissue inhibitor of metalloproteinase 1–specific small interfering RNA or polyclonal antisera ameliorates CVB3-induced myocarditis by enhancing the matrix metalloproteinase–dependent migration of inflammatory cells at sites of infection and thereby anatomically focusing the adaptive immune response.²¹³

In all, treatment with immunomodulatory drugs seems beneficial for nonviral myocarditis or the subacute stage of viral myocarditis, but for the early stage of acute infectious myocarditis, such treatment will bring little or possibly detrimental effects. Thus, it is rather crucial to detect pathogen-induced myocarditis with sensitive techniques and to treat at early stages to improve detrimental outcomes.

Because antiviral therapy benefits little in the subacute stages of myocarditis and the effects of immunomodulatory drugs on infectious myocarditis are still controversial, therapy to relieve HF is still a major strategy used in the treatment of patients with myocarditis. HF drugs used for myocarditis include angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, β-blockers, aldosterone antagonists, and calcium-channel blockers.²² In different animal models with viral myocarditis or autoimmune myocarditis, angiotensin-converting enzyme inhibitors, captopril, and angiotensin II receptor blockers, losartan and olmesartan, significantly reduce inflammation, necrosis, and fibrosis of the myocardium and improve LV function.^214–218 β-blockers generally improve LV function, relieve HF, and increase survival of patients with inflammatory cardiomyopathy,²² but different types of β-blockers show variable impact on myocarditis. Carvedilol shows a cardioprotective function in rats with autoimmune myocarditis by suppressing inflammatory cytokines and oxidants,¹¹⁵ whereas metoprolol exerts adverse effects on CVB3-induced myocarditis leading to increased inflammation and mortality in the murine model.²¹⁹ Aldosterone antagonists were reported to reduce hospital admission for worsening HF and to increase survival of patients with acute or chronic HF.²²⁰ Calcium-channel blocker amlodipine shows a protective effect against myocardial injury in mice with congestive HF induced by viral myocarditis through blunting of the overproduction of nitric monoxide.²²¹ However, administration of calcium-channel blockers is not generally recommended in the management of acute HF.²²⁰

Mechanical circulatory supports for heart function also aids in the recovery of myocarditis. Although such supports cannot cure myocarditis itself, the devices allow avoidance of fatality and increase the survival longevity of patients, especially in those with cardiogenic shock. A myocarditis and acute cardiomyopathy 2 (Intervention in Myocarditis and Acute Cardiomyopathy [IMAC] 2) study showed that myocarditis was the strongest predictor of bridge to recovery in recent onset nonischemic cardiomyopathy patients requiring the support of LV assist devices. Furthermore, female sex was associated with even greater likelihood of bridge to recovery in myocarditis patients on LV assist device support.²²² As for fulminant myocarditis, a ventricular assist device is associated with complete recovery of myocardial function.²²³ Moreover, external pulsatile mechanical ventricular assistance also improves survival in patients with acute myocarditis.²²⁴ If myocarditis leads to end-stage HF, cardiac transplantation is the ultimate measure that should be considered.⁴⁸

Although different strategies for the therapy of myocarditis have been introduced, most are still yet to be verified in the research laboratory and are far from clinical application. As well, most clinical assessments performed on myocarditis treatment are still related to symptoms and depend much on the clinical presentation. Development of new monitoring and therapeutic techniques, such as molecular diagnoses, nucleic acid–based antiviral drug development, and targeted drug delivery, will greatly enhance the therapy for myocarditis in the future. An exhaustive list of current treatments for viral and nonviral-associated myocarditis is included in Table 3.