What is myocyte injury? The histological assessment of lymphocytic myocarditis in autopsy hearts

Myocyte injury is central to the histopathological diagnosis of lymphocytic myocarditis, yet its histological features remain poorly standardised. This study assesses observer agreement in the diagnosis of myocyte injury in lymphocyte-predominant cardiac inflammation and explores which histological features are used in this assessment. Twenty-seven experienced cardiovascular pathologists independently assessed 72 unique photomicrographs of lymphocyte-predominant myocardial inflammation in autopsy hearts. Twenty-four images were scored twice. Intraobserver agreement was assessed using Cohen's kappa and percentage agreement. For interobserver variation, Fleiss' kappa, the percentage of respondents rating a photomicrograph as showing injury, and disagreement index were used. Free-text comments were qualitatively analysed to identify commonly used descriptors in clear-cut and ambiguous cases. Photomicrographs with high and low consensus were reviewed to elicit additional morphological features. Results showed that cardiovascular pathologists are overall internally consistent when assessing myocyte injury (median percentage agreement 83%, median Cohen's kappa 0.644). In contrast, interobserver agreement across the entire image set was only fair (Fleiss' kappa 0.381). Consensus about myocyte injury was near-unanimous in 25 of the 72 images whilst 30 of the 72 yielded very low consensus. This suggests the existence of a large diagnostically challenging 'grey zone' between clear-cut injury and non-injury. Results furthermore suggest that cardiovascular pathologists overall rely on two related concepts when assessing myocyte injury: signs of myocyte degradation and evidence of inflammatory-myocyte interaction. They, however, appear to differ in how these concepts are described, interpreted and applied. Calibration may therefore improve consistency in the histological assessment of lymphocytic myocarditis.