Diagnosing lymphocytic myocarditis in non-biopsy specimens remains challenging due to sampling variability, subjective interpretation of histology, and lack of standardized criteria. In 2025, the Society for Cardiovascular Pathology (SCVP) and the Association for European Cardiovascular Pathology (AECVP) proposed the “Seaport criteria” to address these limitations.

To assess the practical applicability of the Seaport criteria in a retrospective cohort of forensic autopsy cases with myocardial lymphocytic-predominant inflammation.

Ninety-three autopsy cases with lymphocytic-predominant myocardial inflammation were re-evaluated according to the Seaport criteria. Death in these cases was attributed to myocarditis (n = 45), unascertained causes (n = 34), or drug toxicity (n = 14). We assessed adherence to the recommended technical requirements, reclassified inflammation as diffuse, multifocal, focal myocarditis, or lymphocytic infiltrates of unknown significance (LIUS), and reviewed the original histological descriptions.

Most cases (82 %) met the minimum technical sampling standards. Pediatric cases were disproportionately non-compliant with sampling requirements, but not due to insufficient myocardium being sampled. Original histological reporting varied substantially in terminology and detail, with myocyte injury inconsistently reported and a lack of sufficient information for grading under the Seaport criteria. Re-classification resulted in 36 cases of diffuse, 27 multifocal, 2 focal myocarditis, and 28 LIUS. The most common diagnosis in cases given myocarditis as the cause of death was diffuse myocarditis (33/45), whereas LIUS was most frequent in the drug-related and unascertained cohorts. Myocyte injury was sometimes difficult to interpret.

The Seaport criteria are feasible to classify lymphocytic myocarditis in autopsy hearts, with potential to standardize histological assessment and therefore improve diagnostic consistency. However, challenges remain in recognizing myocyte injury. Further prospective multicenter validation is recommended.