ObjectivesImmune checkpoint inhibitor (ICI)‐associated myocarditis, particularly severe ICI‐associated myocarditis, has a high mortality rate. However, the predictive value of electrocardiogram (ECG) remains unclear. The present study aimed to evaluate the predictive value of clinical and electrocardiographic parameters for severe myocarditis.MethodsClinical and electrocardiographic data of 73 cancer patients with ICI‐associated myocarditis were retrospectively collected. The severity of ICI‐associated myocarditis was graded using the NCCN guidelines for managing immunotherapy‐related toxicities. Myocarditis grades 1–2 and grades 3–4 were classified as mild and severe myocarditis, respectively. Logistic regression analysis was performed to analyze the predictive value of each parameter in predicting severe myocarditis.ResultsAmong the 73 patients with myocarditis, 20 (27.4%) patients had severe myocarditis. Compared with mild myocarditis group, sinus tachycardia (p = 0.001), QRS duration ≥110 ms (p = 0.001), prolonged QTc interval (p < 0.001), and bundle branch block (p = 0.007) at the time of myocarditis were more common in the severe myocarditis group. Logistic regression analysis revealed that sinus tachycardia (p = 0.028) and QTc interval prolongation (p = 0.007) were predictors of severe myocarditis. Whereas the predictive value of other electrocardiographic parameters was weak. Concurrent targeted therapy didn't increase the risk of severe myocarditis. A high NT‐proBNP level was associated with severe myocarditis.ConclusionsECG at the onset of myocarditis manifested as sinus tachycardia and prolonged QTc interval predicted a high risk of severe myocarditis. Early detection of ECG abnormalities may faciliate early detection of severe ICI‐associated myocarditis.