SCVP myocarditis criteria and definitions

The endomyocardial biopsy criteria comprise 4 grades of myocarditis: no myocarditis, mild myocarditis, moderate myocarditis, and severe myocarditis. Each grade has a specific set of criteria that have to be determined for the correct diagnosis.

• * Single-cell hypereosinophilia; nuclear karyorrhexis/karyolysis; sarcoplasmic membrane scalloping.
• ^† Criteria proposed by the North American contingent
• ^‡ Criteria proposed by the European contingent

A case without myocarditis (grade 0) has no evidence of clusters of lymphocytes (≥ 5) on either an H&E or a CD3 immunohistochemical stain for T lymphocytes.  Further, there are not more than 15 CD3+ cells per 400x  (40x objective) high powered field (hpf).

The diagnosis of mild myocarditis (grade 1) can be made in one of two ways.  The first (1a)  is to identify a single focus of an interstitial cluster of 5 or more lymphocytes noted by a CD3 or H&E stained slide. This cluster of cells should be within the myocardium proper, adjacent to a myocyte and/or myocytes, and not along the endocardium, in areas of loose connective tissue (adipose/fibrosis), scar/dense fibrosis, or within small vessels/capillaries.   Lymphocytes in those areas are not evaluated for myocarditis but can be reported as evidence of chronic inflammation in the biopsy.  The second approach (1b) to make the diagnosis of mild myocarditis, requires the use of CD3 IHC to identify 10 or 15 or more CD3+ cells in a single 400x (40x objective)  hpf within the biopsy material.  (This exact threshold is being finalized, but is controversial.) As noted for lymphocyte clusters, the lymphocytes should be around myocytes and not in the endocardium, dense fibrosis, loose connective tissue, capillaries, or scar.  Ideally, there will be a general pattern of increased CD3+ cells above the baseline throughout the biopsy, but that is not strictly part of the criteria.  Myocyte injury, can manifest as various histologic changes including single-cell hypereosinophilia, nuclear karyorrhexis/karyolysis, and/or sarcoplasmic membrane scalloping.  If present, it should be reported. Myocardial injury is not required for this diagnosis.

Moderate myocarditis (grade 2) is diagnosed by having multifocal myocardial inflammation with or without myocyte injury. There should be two or more clusters of lymphocytes as noted by a CD3 or H&E stained slide, with the definition of a cluster the same as in grade 1.  These clusters of cells should be within the myocardium proper, adjacent to a myocyte and/or myocytes, and not along the endocardium, in areas of loose connective tissue (adipose/fibrosis), scar/dense fibrosis, or within small vessels/capillaries. Myocyte injury, can manifest as various histologic changes including single-cell hypereosinophilia, nuclear karyorrhexis/karyolysis, and/or sarcoplasmic membrane scalloping.  If present, it should be reported. Myocardial injury is not required for this diagnosis. It is expected that diffuse CD3+ lymphocytes will be increased in the setting of multifocal clusters, but there is no criteria based on counting these cells.

Severe myocarditis (grade 3) is diagnosed when there is extensive and heavy predominantly lymphocytic infiltrate with myocyte injury.  A CD3 immunostain is not required to make this diagnosis, but is recommended. For this diagnosis, there should be an extensive, multifocal or diffuse infiltrate.  Myocyte injury should be obvious.  The infiltrate can include macrophages and rare eosinophils.  Giant cells or extensive numbers of eosinophils would suggest different diagnoses (giant cell myocarditis, sarcoidosis and/or eosinophilic diseases).

Below are multiple images of histologic appearances of myocyte injury with explanations of why these images are of different types of myocyte injury. All of the examples depicted below are considered types of myocyte injury as used in the myocarditis consensus criteria.

These 4 images demonstrate myocarditis-induced myocyte injury.  They show nuclear karyorrhexis, sarcoplasmic membrane scalloping, fragmentation of contractile elements, shrunken myocytes, and internal (intramyocytic) lymphocytes.  Additional soft criteria include a dirty background of cellular debris, abundant lymphocytes, and loss of sarcomere visualization in injured cells. Note that most of the injured myocytes are hypereosinophilic on the H&E stain.

Healing Injury

These images demonstrate myocyte injury from a healing, reparative process and not active myocarditis.  While myocyte injury is present (fragmentation of contractile elements) the infiltrate is primarily macrophages and not lymphocytes.

Extensive Infiltrate

Severe myocarditis is denoted by an extensive infiltrate of lymphocytes.  Macrophages and rare eosinophils can be part of the infiltrate, particularly once the myocarditis is established and reparative processes begin. Even at low power, an increase in cellularity can be observed.  CD3+ lymphocytes are extensive, diffuse throughout the tissue, and too numerous to count.

Lymphocyte clusters

Lymphocyte clusters are defined as having 5 or more lymphocytes in near approximation to each other.  They are generally touching or within one to two diameters of a lymphocyte from each other. Clusters should be adjacent to a myocyte or myocytes and are generally discrete with uninflamed regions between clusters, when more than one cluster exists.  Adjacent myocytes can be healthy or injured.  Clusters that appear in the endocardium, in scar, in dense fibrosis, within a vessel or lymphatic or in large areas of loose connective tissue, should not be considered/counted in these criteria.  Clusters can be seen on an H&E stain or on a CD3 immunohistochemical stain.

Diffuse infiltrate

A diffuse infiltrate of 10 or 15 or more CD3+ lymphocytes (this remains an unsettled threshold) in one 400x (40x objective)  hpf is a diagnostic criteria of mild myocarditis (Grade 1). This can be achieved by scanning the slide at medium power and then counting CD3+ cells in one hpf. Cells should be between myocytes and around small vessels.  They should not be counted in the endocardium, within scars, large areas of loose connective tissue, or within vessel and lymphatics.

Areas not to evaluate

Some clusters and diffuse infiltrates of CD3+ cells are not to be evaluated because they are in the endocardium, in a blood vessel or lymphatic, in dense fibrous tissue, in a scar, or in loose connective tissue.  The images below highlight some of these areas to avoid.  On a CD3 stain alone, it can be difficult to discern if the cells are within the myocardium or not.  Thus comparing the CD3 stained area to an equivalent area on an H&E is valuable.

This decision tree is a step-by-step guide to evaluate an endomyocardial biopsy for myocarditis.  Image created using BioRender.

Step one

The first step is to evaluate an H&E slide. Either a florid myocarditis with numerous lymphocytes and myocyte injury is seen or it is not.  If a florid myocarditis pattern is noted, this would be grade 3 severe myocarditis.  One should look for giant cells (either on the H&E or on a CD68 IHC) or numerous eosinophils, as these could represent other forms of myocarditis.  Otherwise, the diagnosis of severe lymphocytic myocarditis can be rendered.  If there is not florid myocarditis, the reviewer advances to step two.

Step two

The remaining H&E stains and a CD3 are reviewed for lymphocyte clusters (5 or more cells around myocytes, as described elsewhere).  Determine if there are one or more clusters.  Clusters should be separated, but can appear on the same biopsy piece or across biopsy pieces.  They can also appear on different levels of the tissue. A CD3 stain may be helpful in confirming if a cluster or clusters are present.  If there are two or more clusters, this is grade 2 moderate lymphocytic myocarditis.  If there is only a single cluster, this is grade 1 mild lymphocytic myocarditis.  In reviewing the H&E slides make note of any myocyte injury. This does not change the diagnostic level, but it should be reported out for the case.  If no lymphocyte clusters are present, advance to step 3.

Step three

The CD3 stain should be evaluated for excessive lymphocytes.  A medium to low power pass can be made of the tissue to identify areas with the most lymphocytes.  Then a single 400x (40x objective) hpf should be used to count lymphocytes within the tissue.  Lymphocytes should be within the myocardium proper, adjacent to myocytes or along small perivascular spaces between myocytes.  They should not be along the endocardium, in scars, within blood vessels, or in large areas of loose connective tissue (adipose/fibrosis). Fifteen or more CD3+ cells within a hpf is grade 1, mild myocarditis.  One can interpret more than one hpf to find the region with the most positive cells. If there are no hpf with 15 or more CD3+ lymphocytes, the diagnosis of grade 0, no myocarditis is given. In the setting of no myocarditis, other diagnoses can be rendered depending on the other findings of the specimen. There is a discussion under the “Biopsy Adequacy” section on different sized high power fields and 40x objectives, which might be useful in thinking about the area being evaluated.

An ideal tissue harvest by endomyocardial biopsy to evaluate for myocarditis is a minimum of five pieces of myocardium (Hauck AJ et al).  Less than 5 pieces decreases sensitivity. Any fewer than three pieces would be considered insufficient/inadequate. At least four levels should be taken on every case with two or more sections per slide for H&E staining.  In addition, a CD3 immunohistochemical stain for T lymphocytes is necessary for any case that does not have severe (grade 3) myocarditis.  The expert panel also recommends a CD68 immunohistochemical stain for macrophages and giant cells (useful for identifying giant cell myocarditis or healing injury) and a Masson Trichrome, Movat Pentachrome or similar to determine the extent of fibrosis (useful for determining the extent of fibrosis which may indicate chronicity).

A useful approach for preparing slides to evaluate a myocarditis case is as follows.

These biopsies have 5 and 4 pieces respectively and would be adequate to evaluate for myocarditis.

These criteria utilize an H&E and CD3 immunohistochemical stain to evaluate the presence of clusters or diffuse infiltrates of T lymphocytes as described. H&E stained slides are useful to determine the cellular features of myocyte injury including single-cell hypereosinophilia; nuclear karyorrhexis/karyolysis; and sarcoplasmic membrane scalloping. The CD3 stain is particularly useful in counting a diffuse infiltrate and can also help visualize lymphocyte clusters.  It is recommended to be performed in all evaluations for myocarditis, but it is considered essential when making the diagnosis of mild and moderate myocarditis (grades 1 and 2), due to the more patchy/heterogeneous patterns of mild and moderate myocarditis. Clusters of inflammatory cells may only be present on H&E stained slides, but not on the CD3 stained slide, which is why both slide types are valuable.  For severe myocarditis, the diagnosis can be made on H&E stained slides without a need for CD3 staining.  In severe myocarditis, a CD68 stain is recommended to help identify giant cells as giant cell myocarditis is a mimicker of severe lymphocytic myocarditis.

These criteria use high powered field (hpf) taken as a single 400x (40x objective and 10x ocular) field.  As different microscopes have slightly different objectives, the area of a single hpf can vary.  This has been discussed. In practical terms, these differences are real but negligible for most modern microscopes.  Below is a general table of field numbers, diameters at 40x and their equivalent area in mm².  The field number is reported on most microscope lenses (FN or OFN).  If your microscope falls outside of this range, you may consider adjusting your counts to be in this general range.

Myocyte injury was a critical aspect of the Dallas criteria to define myocarditis. Conversely, myocyte injury was not a feature of the ESC criteria, in which the presence of a certain number of leukocytes was all that was necessary to define myocarditis. These viewpoints are diametrically opposed. A challenge of using myocyte injury as an essential criteria to define myocarditis is the difficulty in agreement as to what histology defines myocyte injury.  It can be very difficult to discern in many cases and expert cardiovascular pathologists can disagree on its presence or what features to use when evaluating injury.

Therefore, in these new criteria, definitive myocyte injury was chosen to be an essential feature of severe myocarditis only.  In that histologic subtype of myocarditis, it would be expected with the extensive infiltrate of lymphocytes.  The criteria have been written such that myocyte injury could be seen in both mild and moderate myocarditis cases. These types of myocarditis overlap significantly with the ‘borderline myocarditis’ term of the Dallas criteria, where myocyte injury was not seen.  Therefore, it will be of interest to find out how often myocyte injury is reported in these intermediate grades and if the presence of injury with mild or moderate diagnoses of myocarditis portend a worse outcome for the patient.

The presence of myocyte injury in the absence of any inflammation should not be considered as a form of myocarditis.  Other causes such as single cell drop out due to a stress cardiomyopathy, elevated catecholamine levels, septic shock, or other causes should be considered.

The number of clusters used to distinguish between mild and moderate myocarditis is one or more clusters.  Why is this the defining threshold?  Why not 1-2 clusters vs. 3+ clusters?  Or 1-3 clusters vs. 4+ clusters? There is no literature to support any one of these levels.  The expert consensus was that a single cluster may be a random finding in a heart, with little significance.  However, the presence of two or more clusters is less likely to be a random finding and is more likely to be a meaningful discovery.

As all heart biopsies represent a tiny fraction of the heart, one never knows how representative it is of a heterogeneous process which is more frequent in the subepicardial and left ventricular regions, where the biopsies are not typically performed. Is a single cluster the “tip of the iceberg” with much more inflammation in the heart, or is it a random collection of cells, that might be seen infrequently in older individuals?

In an analysis of nearly 100 digitized heart biopsies evaluated for myocarditis, only 3 cases had exactly two cluster of lymphocytes. Therefore, despite the importance of this arbitrary threshold, it is unlikely to be a common decision point. Nonetheless, further evaluation of this threshold is warranted.

A significant amount of effort was spent exploring the need for using hpf or a defined area in mm² for a CD3+ diffuse lymphocytic infiltrate. The benefit of hpf is that anyone on any microscope can use their 40x objective and count the single field.  The challenge of hpf is that different microscopes have slightly different fields of view, such that some hpfs are larger than others.  As a result, there will not be a one-to-one correlation with the use of hpf.

The benefit of a defined area in mm² is that this is an exact area, that should be more reproducible.  The challenge, however, is to both determine what that area is for a given ocular and to consistently use that size.  For example, if a 40x objective ocular’s field number is 25, the area of the view is 0.31mm².  If criteria are set at 0.25mm² (as was debated) then a pathologist should review only 80% of the field to match the 0.25mm² area.  Which part of the field should be ignored?  What biases will be introduced? If the criteria are set to 1mm², then the pathologist with the field number 25 ocular should review and count cells in 3 full fields and 7% of the final field.  Neither of these options is as exact as would be hypothesized with a defined area.

For groups that exclusively use digitized slides, an exact area in mm² can be consistently determined.  However, digital signout is currently the exception to the rule in the evaluation of endomyocardial biopsies for myocarditis.  Therefore, the expert committee thought this was not going to be a widely useful approach.

Therefore, the use of a single 400x (40x objective) hpf field, due to its inherent simplicity, was favored over a defined area in mm².  As practice patterns change, and digital pathology becomes more mainstream, this approach can change.

A second part of this controversy was how much area should be reviewed in which to count a diffuse infiltrate of CD3⁺ cells. It was reasoned that myocarditis is a heterogeneous disease and, as a result, a meaningful lymphocytic infiltrate can be in a small area at a high density, but that infiltrate may not be frequent elsewhere.  Therefore, focusing on a single 400x field would allow the pathologist to identify a significantly elevated number of lymphocytes in a small area.  The approach also reduces the time spent counting cells, which would be increased with more complicated plans such as counting 4 fields and taking the average, or similar schemes.

Although the expert committee initially focused on lymphocyte clusters for making the diagnosis of myocarditis, evidence brought forth showed reasonable examples of an increased lymphocyte infiltrate in which no clusters of 5+ cells were detected.  The question, then, was at what number of lymphocytes over what area would constitute a significant number of cells over baseline? After significant discussion, disagreement, and evaluation of CD3+ staining across a range of specimens, there is still a controversy on what number of  CD3+ cells in a single 400x hpf would be considered mild myocarditis. Debated values have ranged from 25 CD3+ cells/hpf to 10 CD3+ cells/mm².

It was established that CD3+ lymphocytes are rare, but present cells in noninflamed hearts. In general, they range between 0-4 cells/hpf. However, randomly up to 10-12 cells were shown to be in random hpf views of apical left ventrical core tissues removed for heart failure.  There is no perfect threshold, but the consensus group favors a robust threshold to not overdiagnose myocarditis based on spurious collections of lymphocytes, including the counting of some within blood vessels or adjacent soft tissue.

These criteria apply only to examination of ventricular myocardium (autopsy hearts, explanted hearts, apical cores, septal myectomies). The clinical significance of lymphocytic myocarditis in the atria is an area in need of further study and thus atrial appendages are not to be evaluated under these criteria.

Active Myocarditis is defined as:

Myocardial inflammation with myocyte injury that is not explained by another cause (ischemia, trauma, foreign body, amyloid, etc.).

Myocyte injury must be distinct from changes seen in non-inflamed areas and may consist of:

• single-cell hypereosinophilia
• nuclear karryorrhexis/karyolysis
• sarcoplasmic membrane scalloping.

Active Myocarditis should be further qualified by extent, as follows:

Focal: single focus

Diffuse: ≥50% area of a single block involved by confluent myocarditis

Multifocal: more than a single focus but <50% area of a single block involved by confluent myocarditis

**In the absence of myocyte injury, the term “lymphocytic infiltrate of undetermined significance” can be considered.

Any pictures we want unique to large specimens go here.  This can include what is and what is NOT myocarditis.

Tissue should be entirely submitted in a transmural fashion (to visualize epicardium, myocardium, and endocardium on each tissue section).

A minimum of 2 tissue blocks of myocardium should be examined.

Six (6) full-thickness ventricular sections should be examined in 5-6 tissue blocks. A minimum of 1 short-axis slice (taken at the mid-ventricular level) should be saved for additional processing.

Six (6) full-thickness ventricular sections should be examined in 5-6 tissue blocks. A minimum of 1 short-axis slice (taken at the mid-ventricular level) should be saved for additional processing.

With new criteria come new opportunities to determine how they will behave across pathologists and how they will impact patient care.  Here are some questions that are in need of supporting data.

How do biopsy criteria correlate with patient outcome?

Do the different myocarditis levels correlate with patient mortality/transplant? Is myocyte injury a critical indicator of worse patient outcomes? Do criteria correlate with ejection fraction at the time of biopsy? MRI findings?  Blood biomarkers?

How reproducible are criteria levels between pathologists?

What is the amount of inflammation in a “normal” heart?

How do these criteria change the cause of death in autopsy cases?

Viral PCR

A short discussion of viral PCR?

Imaging

A short discussion of MRI and Lake Louise Criteria?

MHC Class II staining

MHC Class II is elevated in endothelial cells due to interferon-gamma effects in myocarditis.  It may, therefore, be a marker of myocarditis, even at a distance from the immune cell infiltrate.  As such, immunohistochemistry for MHC Class II may represent a new biomarker of myocarditis in intermediate cases for which the histology and CD3 staining is inconclusive.

Anti-heart antibodies

A short discussion of auto antibodies?