Factors associated with arrhythmia in viral myocarditis and the development of an individualized risk prediction model

WU Ming-xian; LIAO Wei-Lin; YANG Xi

doi:10.3969/j.issn.1672-5301.2026.02.012

Abstract：Objective To explore the factors associated with arrhythmia in viral myocarditis,and to develop an individualized risk prediction model.Methods Medical records from 203 patients diagnosed with viral myocarditis at Dazhou Central Hospital between April 2018 and August 2024 were collected as the derivation cohort.Data from an additional 91 patients treated at the same hospital from September 2024 to November 2025 were collected as an external validation cohort.Based on the occurrence of arrhythmia,patients in the derivation cohort were divided into three groups:grade Ⅰ-Ⅱ arrhythmia,grade Ⅲ-Ⅳ arrhythmia,and no arrhythmia.The propensity score matching was performed among three subgroups,and the baseline characteristics after matching were compared.Multivariable Logistic regression analysis was used to identify factors associated with arrhythmia,which were then used to construct and externally validate a nomogram-based individualized risk prediction model.Results In the derivation cohort,significant differences were observed among the three groups in CRP,D-dimer,cTnI,CK-MB,QRS-T angle,and the proportion of patients with a QRS interval>120 ms(all P<0.05).Logistic regression analysis identified elevated levels of CRP,high D-dimer,high cTnI,high CK-MB,a larger QRS-T angle and QRS interval>120 ms as significant risk factors for arrhythmia(all P<0.05).A nomogram for predicting arrhythmia risk was constructed.Calibration curves showed good agreement between predicted and observed probabilities in both the derivation and validation cohorts,and the Hosmer-Lemeshow test indicated a good model fit(all P>0.05).The AUCs for predicting gradeⅠ-ⅡandⅢ-Ⅳarrhythmia were 0.817(95%CI 0.696-0.905)and 0.821(95%CI 0.700-0.908)in the derivation cohort,and 0.804(95%CI 0.697-0.886)and 0.812(95%CI 0.705-0.893)in the validation cohort,respectively.Decision curve analysis(DCA)demonstrated that the model provided a positive net benefit for predicting both gradeⅠ-ⅡandⅢ-Ⅳarrhythmia across a wide range of threshold probabilities in both cohorts(derivation cohort:0.05-0.97;validation cohort:0.05-0.95).Conclusion CRP,D-dimer,cTnI,CK-MB,QRS-T angle and QRS interval are significant predictors of arrhythmias in viral myocarditis.The individualized risk prediction model developed based on these factors demonstrates good predictive value and clinical applicability.