Predictors of Outcome in Patients with Suspected Myocarditis

1. Kindermann I, Kindermann M, Kandolf R, Klingel K, Bultmann B, Muller T, Lindinger A, Bohm M,,  Predictors of outcome in patients with suspected myocarditis.,  Circulation,  118 (6) 639-48. View in PubMed

Over a 13-year period, the authors performed heart biopsies on patients with "suspected myocarditis." Suspicion was apparently based on left ventricular (LV) dysfunction, a febrile episode some time in the prior 6 months, absence of coronary disease, and "necrosis markers." It is not clear whether all patients had all factors. Patients underwent endocardial biopsies that were analyzed for viral genome, histology, and immunohistochemistry [primarily cell surface markers to identify cell type and human leukocyte antigen (HLA)]. Clinical and biopsy results were then correlated with outcomes using uni- and multivariate analysis.

Univariate significant correlations included New York Heart Association (NYHA) functional class, LV size, ejection fraction, LV end-diastolic pressure (LVEDP), beta blocker therapy (inverse), and positive immunohistochemistry. Multivariate significant correlations included NYHA functional class, beta blocker therapy (inverse), and positive immunohistochemistry. Important factors that had no correlation included the presence or absence of viral genome in the myocardial tissue and myocardial histology based on a published criteria used for the diagnosis of myocarditis (Dallas criteria). Also interesting was the absence of correlation between the presence or absence of viral genome and immunohistochemistry.

The authors conclude that "For patients with suspected myocarditis, advanced New York Heart Association functional class, immunohistological signs of inflammation, and the lack of beta blocker therapy, but not histology (Dallas criteria) or viral genome detection are related to poor outcome." This reviewer has a slightly different take on these data. 

Clinical Implication/Application

Many years ago (in a galaxy far, far away), I wrote a book chapter entitled "Immunosuppressive therapy in myocarditis: Acute viral versus chronic idiopathic myocarditis."[1] This was an attempt to distinguish between the well-defined syndrome of viral myocarditis that presents with fever, pericarditis, minimal, if any, left ventricular dilatation, myocardial edema with or without a fall in ejection fraction and an ill-defined process that, clinically, was dilated cardiomyopathy, but in some patients, had inflammatory cells present in small amounts in myocardial tissue. Most of us in the field at that time (1988) were convinced that most, if not all, patients with idiopathic dilated cardiomyopathy had this syndrome as a consequence of a prior viral infection, and some of these patients had ongoing inflammation. The implication was that if you could correct the inflammation, you could arrest or improve the myopathic process. In fact, we were in the process of planning a multicenter trial under the direction of Jay Mason to test this hypothesis.[2] It is important to emphasize that the principal diagnostic criteria of myocardial inflammation could not be agreed upon by any of the pathologists active in the field; and hence, a conference of pathologists was held in Dallas, in essence to “agree to agree” on what they would call “myocarditis” for the purposes of this upcoming clinical trial. The “Dallas criteria” is a consensus that has never been subjected to any form of proof of validity. This background is particularly important given the title of the current manuscript in review "suspected myocarditis."

Let's first look at the patient population. The only clinical factor that distinguishes the cohort from patients with idiopathic dilated cardiomyopathy was a history of a febrile illness of any sort within 6 months of their presentation. Even if one accepts the Dallas criteria as diagnostic for myocarditis, only 38% had this feature.

Next, what predicts outcome in this patient population? NYHA functional class, ejection fraction, cardiac size, filling pressures, and beta blocker therapy. Again, there is very little to distinguish this patient population from run-of-the-mill idiopathic dilated cardiomyopathy.

One, and probably the only, noteworthy finding of this study was the association between positive immunohistochemistry and outcome. Although one could interpret this finding as evidence of myocarditis, these same cell types are present in any form of cell necrosis and clean-up. It would not be surprising if ongoing nonapoptotic cell death in patients with dilated cardiomyopathy were a marker for poor outcome. It would be interesting to know how many of these patients with positive immunohistochemistry were on intermittent or continuous inotropic therapy, a popular treatment during the time many of these patients were recruited.

What about other so-called features of myocarditis, such as viral persistence? It is well established that the mere presence of viral genome does not establish pathogenesis.[3,4] Furthermore, not only did the presence of viral genome not predict outcome, there was no correlation between the presence or absence of viral genome and one of the more powerful predictors, positive immunohistochemistry.

Over the past 10 years, thanks largely to the mapping of the human genome, we now know that many cases of even apparent sporadic cases of idiopathic cardiomyopathy have a genetic basis, and the point mutations have been identified. It is more than likely that most cases of idiopathic cardiomyopathy will have an identified molecular basis. The consistent interpretation of the negative results for the Myocarditis Treatment trial [2] is that immunosuppression is not effective in chronic myocarditis. A more plausible interpretation is that immunosuppression is not effective therapy for dilated cardiomyopathy. With all of the inconclusive or negative data, despite millions of dollars and hours of research, it is time for those wedded to the viral hypothesis to move on. I have. So, no Virginia, there is no chronic myocarditis. 

1. Hosenpud JD. Immunosuppressive therapy in myocarditis: Acute viral versus chronic idiopathic myocarditis. In: Schultheis HP, ed. New concepts in viral heart disease, Berlin: Springer-Verlag, 1988:393-401.
2. Mason JW, O'Connell JB, Herskowitz A, et al. A clinical trial of immunosuppressive therapy for myocarditis. The Myocarditis Treatment Trial Investigators.[see comment] New Engl J Med 1995;333(5):269-75.
3. Keren A. Invited for debate: Is virus persistence a determinant for disease progression? Ernst Schering Res Found Workshop. 2006;(55):55-61.
4. Pagano JS, Blaser M, Buendia MA, et al. Infectious agents and cancer: Criteria for a causal relation. Semin Cancer Biol. 2004;14(6):453-71.