Original versus 2018 Lake Louise Criteria for Acute Myocarditis Diagnosis: Old versus New

See also the article by Luetkens et al in this issue.

Matthias Gutberlet, MD, PhD, EBCR, is professor of radiology and chairman of the department of diagnostic and interventional radiology at the Heart Center Leipzig-University Leipzig since 2007. He started his career in cardiology and radiology at the Charité, Berlin in 1995. His main research interests are noninvasive cardiovascular imaging. From 2014 to 2017, he was the president of the European Society of Cardiovascular Radiology (ESCR). He had the privilege to be part of the International Consensus Group on Cardiovascular Magnetic Resonance in Myocarditis and coauthor of the 2009 LLC I and 2018 LLC II.

Christian Lücke, MD, EBCR, is a consultant in the department of diagnostic and interventional radiology at the Heart Center Leipzig, Germany. Originally from Hameln, Germany, he received his medical degree in 2005 in Leipzig. His special interest is tissue characterization by multiparametric imaging for the diagnosis of myocarditis and myocardial fibrosis in cardiomyopathies. He was responsible for the imaging in the MyoRacer trial. He is author and coauthor of several peer-reviewed articles in the field and is a reviewer for European Radiology, European Journal of Radiology, and Herz.

Introduction

Luetkens and coauthors report in this very interesting study (1) on an initial use of the updated version of the Lake Louise criteria II (LLC II) from 2018 (2) for the cardiac MRI diagnosis of acute myocarditis. They compared the results by using the original LLC I (3) from 2009 with the LLC II from 2018 in 40 patients clinically suspected of having acute myocarditis and 26 control subjects. They concluded that, especially because of the higher sensitivity of multiparametric cardiac MRI, the 2018 LLC should be implemented into routine diagnostic protocols.

The clinical need for reliable noninvasive diagnostic tools in patients suspected of having myocarditis cannot be ignored, when looking at the main indications in large cardiac MRI registries, such as the Cardiac MR/CT Registry of the European Society of Cardiovascular Radiology (ESCR) (https://www.mrct-registry.org/), with per date almost 150 000 submitted cardiac MRI cases. The indication “suspected/known myocarditis” accounts for almost 30% of all nonstress cardiac MRI examinations (4) according to the ESCR Registry. If you then add the indication “suspected/known cardiomyopathy,” which includes in its differential diagnosis a diagnosis of myocarditis or chronic myocarditis, respectively, you´ll end up with almost 50% of all nonstress cardiac MRI examinations (4).

The original LLC I is based mainly on three cardiac MRI sequences: (a) a T1-weighted spin-echo sequence before and after contrast material administration to calculate the so-called early gadolinium enhancement ratio (3) or global relative enhancement (5), which should represent the hyperemic aspect of myocardial inflammation; (b) a T2-weighted fat-suppressed (eg, a T2 short inversion time inversion-recovery) sequence to depict myocardial edema and to calculate the T2 signal intensity ratio; and (c) an inversion recovery sequence to look at the late gadolinium enhancement (LGE) patterns, typically occurring subepicardially.

The “two-out-of-three” approach (3) was proposed, meaning that if two or more of the three sequences demonstrated global or focal pathologic results, the diagnosis of acute myocarditis according to the LLC I could be made.

It needs to be borne in mind that the calculations and published (3,5) cutoffs from the first two sequences were originally only validated for the use of the integrated cardiac MRI body coil, but not for the more commonly used dedicated surface coils, which was also the case in the study from Luetkens et al. Furthermore, the published cutoffs might differ when using other scanners or different sequences and were originally validated only clinically in patients suspected of having acute myocarditis. Thus, it was recommended to create site-specific normal values. In addition, the acquisition, the evaluation, and therefore the final calculation are prone to artifacts.

Therefore, the experts and authors of 2018 LLC II replaced the calculation of the early gadolinium enhancement ratio (3) or global relative enhancement (5), and this sequence is not explicitly recommended anymore (2).

However, despite its limitations, the LLC I were extensively used within the last decade, provided reasonable results, especially in infarctlike acute myocarditis (6), and are still of importance in experienced hands. Furthermore, the use of LLC I also helped to improve the diagnosis of subclinical acute cellular rejection after heart transplantation (7) and provided new insights on inflammation in Takotsubo cardiomyopathy (8).

Nevertheless, in the meantime, multiparametric imaging with T1 mapping, T2 mapping, and extracellular volume calculations became commercially available for most sites and allowed the pixelwise calculations of absolute values for tissue characterization. Despite all inherent limitations, which these mapping techniques also have in common with the parameters used in LLC I, they provide for the first time absolute values and are therefore supposed to provide much more robust, reliable, and reproducible (9) values. Its usefulness, feasibility, and benefits have been proven in all kinds of cardiomyopathies, myocarditis, and storage diseases, especially in diffuse disease by several groups within the last decade, including Luetkens et al.

Therefore, it was more than consequent to update the 10-year-old 2009 LLC I with 2018 LLC II, and to include the new T1 and T2 mapping techniques, but also try to simplify the whole cardiac MRI protocol for diagnosing acute myocarditis. The cardiac MRI diagnosis of acute myocarditis can be made according to 2018 LLC II if T1-based and T2-based imaging shows pathologic findings (2). This is equivalent with a revaluation of T2-based imaging, which is most sensitive to increased water content and edema, and therefore one of the most robust indicators of acute and/or ongoing inflammation, whereas T1-based imaging like T1 mapping before and after contrast material administration, as well as LGE, can show pathologic findings even in “healed” myocarditis; this is because they are not only sensitive to myocardial edema, but also to diffuse fibrosis and scar formation, and do not represent the acuity of the disease as well as T2 mapping techniques do.

Therefore, it is not surprising that Luetkens et al could demonstrate in their cohort of patients with clinically obvious acute myocarditis that not only T1 mapping alone with an area under the curve (AUC) of 0.92, but also T2 mapping alone (AUC, 0.91), provided comparable results to the combined approach of LLC II (AUC, 0.92).

However, the most promising results provided T2 mapping in chronic, or perhaps better referred to as “ongoing” or “reactivated,” myocardial inflammation, which was demonstrated by Lurz et al in the MyoRacer trial (10). In clinical routine, cardiac MRI helps to confirm the suspected clinical diagnosis of acute myocarditis. The real clinical challenge, but also a challenge for cardiac MRI, is patients suspected of having chronic myocarditis.

Nevertheless, we are still at the beginning of the use of these new LLC II. Future studies using the updated 2018 LLC should also be applied to a group of patients with more chronic disease assuming, that like in the study on patients with acute myocarditis by Luetkens et al, it will provide much more reliable results than LLC I.

However, we believe that at least these future studies on patients suspected of having chronic myocarditis need to use endomyocardial biopsy (EBM) as the standard of reference, despite all the limitations of EBM in focal disease. This limitation of EBM might be less important in chronic forms of myocarditis. The simple clinical evaluation, which was the basis of most cardiac MRI studies in acute myocarditis like in the study of Luetkens et al, is certainly not sufficient enough in this clinically very important group of patients.

Finally, we congratulate the authors for their excellent first results and definitely agree with them that the 2018 LLC should be implemented into routine diagnostic protocols with all the mentioned precautions.