17 Myocarditis in the modern era: analysing the impact of high-sensitivity troponin & C-reactive protein in predicting the extent of myocarditis and LV systolic dysfunction on cardiac magnetic resonance

Background The diagnostic yield of acute myocarditis has improved considerably since the advent of high-sensitivity troponin (hs-Tn) assays and adoption of the 2018 Lake Louise criteria for cardiac magnetic resonance (CMR)-based imaging for myocardial oedema and injury (1). Yet, a seminal 1997 study, predating these technological advances, remains widely cited for highlighting an absence of troponin rise in two-thirds of myocarditis cases (2). We hypothesize that all cases of myocarditis exhibit hs-Tn rise. However, whether such elevations in hs-Tn or inflammatory markers correlate with the extent of myocardial injury and left ventricular systolic dysfunction (LVSD) on CMR remains unclear. To provide insight into this, we examined a contemporary cohort of cases between 2019–2023. Methods Following patient identification through clinical coding, two authors independently conducted case note reviews to verify the diagnosis of myocarditis. CMR findings were stratified into regional (e.g. inferolateral involvement) and global myocarditis (e.g. subepicardial LGE in all segments). This classification aimed to explore potential correlations with LVSD severity, peak hs-Tn and CRP levels. Receiver operating characteristics (ROC) and area under curves (AUC) were calculated to determine any cut-off values for predicting the CMR findings and in turn influence adverse patient outcomes. Results 103 inpatients were clinically coded as 'myocarditis' upon discharge, but subsequent CMR (within 4 weeks) reclassified 23 patients into other causes (figure 1). Analysis of the myocarditis cohort revealed a predominantly male population (76%) with a median age of 53 years, commonly presenting with chest pain without a viral prodrome (figure 2A-D). Elevated hs-Tn levels were observed universally, with a median peak hs-TnT of 610ng/L (IQR 1081.5ng/L) and hs-TnI 3820ng/L (IQR 12416ng/L). In contrast, peak CRP ranged between 2 mg/L (normal) and 600 mg/L. Global myocarditis was detected in 11/80 patients (13.8%) on CMR; however, this finding did not correlate with adverse patient outcomes which included an event rate of all-cause mortality (9.6%) and cardiac re-admissions within 12 months (9.6%, namely chest pains). No arrhythmic deaths were noted. COVID was present in 4% on admission. A CRP cut-off of 140 mg/L predicted the occurrence of global myocarditis with a sensitivity of 72.7% and specificity of 87.0% (AUC 0.74, 95%CI 0.52–0.96, p=0.014) (figure 2E). However, neither hs-Tn or CRP levels correlated with the degree of LVSD (figure 2F) or adverse outcomes. Multivariate logistic regression analysis was also insignificant for predictors, limited by the small sample size. Conclusion Although all cases of myocarditis exhibited raised high-sensitivity troponin, this did not correlate with the degree of LV systolic impairment on CMR. However, whether patients later develop LVSD after myocardial remodeling on follow-up CMR or echocardiogram remains to be determined. Myocarditis with regional LV involvement can have normal CRP, while a CRP cut-off >140 mg/L may predict the occurrence of global myocarditis. Nonetheless, CMR stratification into regional and global was not associated with LVSD or adverse patient outcomes within 12 months. However, a larger sample size will be required to verify this finding, including longer-term follow-up. Conflict of Interest None