We would like to study a group like this with gallium-67 scintigraphy. We have a good experience with reumatic, myocarditis and acute heart transplantation rejection patients and Ga-67 is a good marker for active infflamation.

It depends which kind of diagnostic criteria you take. In my opinion, if you compared with the EMB criteria, the diagnostic yields of cardiac MRI should be high. On the other hand, current diagnostic criteria of CMR for myocarditis in JACC white paper is very strict. even there are some abnormal findings in patients such as pericardial effusion or regional wall motion abnormalities, these signs can not be used as CMR diagnostic evidence for myocarditis.

I don't think that myocarditis is overdiagnosed. Based on lots of data, we have to assume that more than 50% of all systemic viral diseases (remember the last severe cold or flu you had?) come with myocardial involvement. Up to about 8 of 100 people have a severe (with necrosis) myocarditis in their lifetime, without knowing about it.

Early enhancement may stay positive after myocarditis, so it should be used in combination with the other CMR consensus markers for myocarditis ("Lake Louise Criteria"; Myocarditis White Paper; Friedrich et al. JACC. 2009;53:1475–1487.).

In fact, compared with other clinical markers, the limitation of CMR is rather UNDER- not overcalling, since the sensitivity and negative predictive value are only 67% and 69%, respectively. For Early enhancement this would be 74% and 70%.

We have had many cases with non-specific symptoms, sometimes combined with ventricular ectopy, which was associated with edema (assessed by T2-weighted CMR) and early enhancement.

It looks like the myocardium is not much different to other tissue (muscle ache in influenza is very common), yet may be more sensitive to viral injury because it cannot go into "resting" (as we all can do when we have inflamed muscles. Animal models have shown that exercise during inflammation may cause a DCM phenotype.

Persisting early enhancement indicates a worse prognosis (Wagner et al., MAGMA 2003;16:17-20 and Mavrogeni et al., European Journal of Heart Failure 2011;13:830–837.). While we have to get a more robust image quality (or use T1 mapping in the future instead of signal intensity analysis), the concept of visualizing hyperemia in myocardial inflammation is solid.

I think we are identifying real disease. Luckily, myocarditis is usually a benign course and thus, the value for most patients is in its explanation for non-specific symptoms, while in patients with heart failure it is the best gatekeeper to decide on endomyocardial biopsies.

I agree with Friederich's suggestions: in our experience we use strict diagnostic parameters which are: chest pain, ST elevation, troponin I increase, evidence of increase signal at STIR images + enhancement with non-ischemic pattern at LGE. Using these criteria we have a lot of "new entries" showing normal coronary angiography and mild depression of regional wall motion at echocardiography with fast recovery of normal regional wall motion. At follow-up ,in the vast majority of patients STIR are normal with persistent LGE. The prognostic meaning of LGE in these patients, with normal regional wall motion, is still unclear and a prospective multicenter trial should be organized to give a final answer to this question.

I agree with both Alberto Roghi and Friederich. However, it also depends on several accessory parameters (expertise of the observers, image quality, timing from onset,..). Surely Lake Louise criteria are fine and for skilled operators the overcalling is definitely not the case. It is my personal experience that more inexperienced operators tend to overcall a lot on MR...

In my opinion, CMR helps in cases of suspected myocarditis plus reduced lv function. In those cases we need 1) support our diagnosis of inflammatory myocardial damage in contrast to other reasons for impaired function. And 2) we need to get a thorough follow up if lv function does not improve under medical management. As pointed out above, both can be well accomplished by MRI and thus help select patients, that we would have to take to EMB if damage persists or LV function does not improve. After all, CMR can not distinguish chronic, virus negative from virus positive forms of myocarditis.

On the other hand, in patients with suspected myocarditis plus normal LV function, there is not really a target/need for specific treatment. Thus, I find it difficult to make out a diagnostic benefit for these patients.

CMR for acute myocarditis is the most useful non-invasive test, particularly in the scenario of patients with chest symptoms, positive serum troponin, and angiographicallly normal coronary arteries.

Increased signal on T2w imaging with sub-epicardial LGE is highly suggestive. We have had less experience with the early-enhancement ratio for its incremental benefit to T2w+LGE.

However... an interesting and common scenario are patients with NO acute symptoms who are undergoing CMR for other reasons, and have sub-epicardial LGE in the posterolateral segments.... How common is "old" or "healed" myocarditis, what are the differentials and how can we further confirm this putative diagnosis?

Lets put things into perspective...before we had myocarditis diagnosed with EMB and Dallas criteria with random biopsies that were positive in up to 1/3rd of patients with suspect symptoms. This ended up in very heterogeneous groups of patients, most of which probably didn't have anything to do with viral myocarditis and in consequence very controversial findings.

CMR has brought a lot of order and answers in all of this. It allows to have a more homogeneous group of true "acute viral myocarditis" patients. A lot of the significant work on this topic has been done by the Stuttgart group (Sechtem and Mahrholdt) with an institution strategy to perform systematic EMBs on the right and left LV if possible guided by DE-CMR and excellent Virology expertise. They have also recently shown the very significant predictive value of DE on acute myocarditis patients' outcome.

In a very well written update on acute myocarditis in JACC, Kindermann puts CMR at the center of diagnostic workout, first line with 6 month follow-up and EMB second line if patients evolve towards complications and DCM (roughly 10 to 20% of patients...)

In routine clinical practice, what is robust and easily assessed: LV function, volumes and DE.

T2w imaging, very sensitive to artifacts and early Gd enhancement are good for research in a strict setting, but too variable or too complex and time consuming to be used on an everyday practice. Maybe as mentioned earlier by M Friedrich, T2 mapping and T1 mapping are going to increase diagnostic performance...but this the future might tell us.

As always for any diagnostic tool, CMR is not perfect, and you will always have to combine MR findings with clinical context and eventually biomarkers.

Also, never forget to exclude acute coronary syndromes in the emergency setting, and in this perspective coronary CT with an additional low dose cardiac delayed enhanced acquisition is another very interesting tool to use on first line basis.

Regarding LGE in myocarditis, it is important to keep in mind that it doesn't differentiate acute from chronic injury (may be a 30 year old scar...) and its' sensitivity is highly variable depending on patient selection (in some less than 30%!). So at least T2 should be added. The vast majority of comparative data indicate that LGE just isn't sufficient in suspected acute myocarditis.

The initial paper by Mahrholdt et al. actually was a mixed group of presumably acute, chronic and even healed myocarditis, generally more severe cases. Therefore their data, while important, are not applicable to more ordinary cases.

I agree with M. Friedrich on the fact that it doesn't cost much to add T2w sequences in the diagnostic workup of myocarditis patients. And indeed it gives information on the acute aspect or even "active" aspect of things. Abdel Aty's work on this in JACC is also very interesting. However, when T2w is positive...

"Ordinary" cases of myocarditis....well it might be interesting to give a definition of these cases, since a significant number of patients are either asymptomatic or diagnosed at later stages of the disease.

That's where a larger group/cohort of patients diagnosed in a multicenter setting and with a homogeneous diagnostic based on MR at the center (but not only) would be interesting, to understand truly what is an "ordinary" myocarditis patient and what's his prognosis.

Good points, Nathan.

I think our hearts have benign colds too and as long as you listen to your Grammy and stay at home if you have a fever, that's not dangerous (and we may not even need to know). But for my taste many of those people out there with palpitations and "well, there are a few PVCs, so what" type of arrhythmia are probably working on a "non-ordinary" myocarditis and indeed we don't now much about what makes this different from your "heart cold".

Of note, there is a registry for CMR in myocarditis, which is currently being expanded under the lead of Andrew Howarth and Yoko Mikami at the Stephenson CMR Centre in Calgary. Hopefully the registry will not only help getting a better understanding of the criteria, but als help validating and streamlining the protocol. Furthermore, the registry can then serve as the platform for a prospective muti-centre trial.

That should help...

In my experience, the appearance of myocarditis in CMR is very inhomogeeous. Some cases show early enhancement, only slight alterations on T2-weighted images, small pericardial effusion and hardly late enhancement. Other cases exhibit extensive edema and late enhancement as well as significant effusion. Extensive alterations on CMR do not automatically mean a significantly restricted left-ventricular function. It seems to be curious.

At our hospital, unclear cases are sometimes proved by biopsy. I add CMR images of an interesting case. Everybody can think of an appropriate diagnosis. Next week, I will add the diagnosis.

My humble opinion is "yes", since we are looking for LGE, we could include non-ischemic patients without certain symptoms.