In 1984, the Dallas Criteria were created, which established the first international framework for diagnosing lymphocytic myocarditis by histology [1]. The criteria required the presence of a lymphocytic infiltrate and myocyte injury. If only a mild lymphocytic infiltrate was present, the term “borderline myocarditis” was used. The criteria used hematoxylin & eosin (H&E) stained slides for the diagnosis. In 2013, the European Society of Cardiology (ESC) myocarditis criteria were published [2]. These criteria avoided the difficult question of myocyte injury and instead focused on the presence of immune cells in the biopsy tissue, based on immunohistochemical (IHC) staining patterns. Specifically, these criteria indicated that 14 or more leukocytes per mm2 with at least 7 CD3+ lymphocytes per mm2 were sufficient for the diagnosis of myocarditis. In practice, most centers performed IHC on CD3 alone or with CD68 (for macrophages/histiocytes) to count inflammatory cells in myocardial tissue. A concern of these second criteria became that the threshold for myocarditis was set too low as many non-myocarditis cases could be diagnosed as myocarditis.

In 2021, the Society for Cardiovascular Pathology (SCVP) and the Association for European Cardiovascular Pathology (AECVP) organized two groups tasked with developing new histopathological myocarditis criteria for biopsy and non-biopsy (surgical and autopsy) cardiac specimens, taking advantage of published data that could inform best practices. An important question raised during discussions was how many diffusely infiltrative lymphocytes were present in a non-inflamed heart. While the heart is generally regarded as a non-immune organ, a CD68 stain shows a robust histiocyte network. As well, a CD3 IHC is invariably positive for occasional lymphocytes. Several groups have investigated the presence of lymphocytes in inflamed and non-inflamed hearts. Some of these studies are reported below.

In 1982, Dr. William Edwards led a team that investigated the presence of lymphocytes using H&E stained slides and relying on cell morphology [3]. They counted 20 random 400x fields in 170 biopsy specimens and reported a mean number of less than 5 lymphocytes / high powered field (hpf). The authors were “purposefully conservative” in counting lymphocytes. The patients were biopsied for clinically suspected dilated cardiomyopathy (DCM), myocarditis, dysrhythmia and miscellaneous other conditions.

In 1985, Dr. Bruce McManus's group published a study on 80 patients (adult and pediatric) with failing or irritable hearts [4]. The diagnoses on the samples included “normal,” myocarditis, nondiagnostic findings and other conditions. Among this population, the mean ± standard deviation of a pan T-cell marker (OKT11) was counted as 7.2 ± 9.6 per mm2.

Kühl et al. studied 255 subjects with DCM or other cardiac diseases using T cell IHC. They obtained mean values from counts of 10 or more hpf (x400 / 0.28mm2) per sample [5]. Of 170 patients with DCM, 48 had a mean of >2 lymphocytes/hpf, ranging from 2 to 13.8 T lymphocytes/hpf. This work suggested most patients with DCM had low lymphocyte (<1/hpf) counts, but a subset (28 %) had elevated counts. The Kühl group also investigated 140 biopsies from DCM subjects. Using a digital image analysis system, they demonstrated a median value of 1.11 CD3+ lymphocytes/mm2 in the 53 subjects without cell adhesion molecule (CAM) induction and a median value of 4.56 CD3+ lymphocytes/mm2 in the 87 subjects with CAM induction [6].

Dr. Robert Jennings group used CD5 (Leu1), CD3 (Leu 4), CD4 (Leu3), CD8 (Leu 2) and CD22 (Leu14) IHC to investigate a lymphocytic infiltrate in 96 heart biopsies [7]. Clinical subgroups included idiopathic cardiomyopathy, cardiomyopathy with coronary artery disease, hypertensive cardiomyopathy and miscellaneous conditions. Counts were performed on 200x fields, which were designated as hpf. The results demonstrated 0-3 T-lymphocytes/hpf (x200) in most cases.

The Kawamura group studied 11 patients with DCM and 3 with viral myocarditis using IHC [8]. They reported a mean SD of 8 ± 3 T lymphocytes for the DCM subjects and 23±11 for the myocarditis subjects in a microscopic area of 0.2 mm2. Li et al. studied 10 specimens from DCM and 25 from viral myocarditis, similar to the Kawamura group [9]. They counted randomly selected hpf (x400) and reported 20.4 ± 5.42 CD3+ lymphocytes/0.5 mm2 in DCM cases and 35.44±6.65 in viral myocarditis cases.

Altogether, these studies give a broad range of results. In general, CD3 or lymphocyte counts are low in most heart biopsies, but in a subset of non-myocarditis cases they can be modestly elevated. The studies are difficult to directly compare as they used a range of measures including mean values from multiple locations within a biopsy, different IHC lymphocyte markers, and randomly selected field sizes for their counts.

To generate myocarditis criteria that could be easily employed by pathologists, the SCVP/AECVP tasked group decided that the CD3+ lymphocytes should be counted in the single “busiest” hpf (x400) field. However, the number of CD3+ lymphocytes in a single busy field was not a question specifically addressed by any of these prior reports. Therefore, we designed a protocol to count CD3+ lymphocytes in a single busy hpf in each sample across a range of biopsies performed for reasons other than myocarditis.