Model Embraced Electromechanical Coupling Time for Estimation of Heart Failure in Patients With Hypertrophic Cardiomyopathy

• Published in: Frontiers in cardiovascular medicine Vol. 9; p. 895035
• Main Authors: Hu, Su, Mi, Lan, Fu, Jianli, Ma, Wangxia, Ni, Jingsong, Zhang, Zhenxia, Li, Botao, Guan, Gongchang, Wang, Junkui, Zhao, Na
• Format: Journal Article
• Language: English
• Published: Frontiers Media S.A 16-06-2022
• Subjects: Cardiovascular Medicine; electromechanical coupling time; heart failure; hypertrophic cardiomyopathy; predictive model; scoring app
• ISSN: 2297-055X; 2297-055X
• DOI: 10.3389/fcvm.2022.895035

ObjectiveThis study aimed to establish a model embraced electromechanical coupling time (EMC-T) and assess the value of the model for the prediction of heart failure (HF) in patients with hypertrophic cardiomyopathy (HCM). Materials and MethodsData on 82 patients with HCM at Shaanxi Provincial People's Hospital between February 2019 and November 2021 were collected and then formed the training dataset (n = 82). Data were used to screen predictors of HF using univariate and multivariate analyses. Predictors were implemented to discover the optimal cut-off value, were incorporated into a model, and shown as a nomogram. The cumulative HF curve was calculated using the Kaplan-Meier method. Additionally, patients with HCM at other hospitals collected from March 2019 to March 2021 formed the validation dataset. The model's performance was confirmed both in training and validation sets. ResultsDuring a median of 22.91 months, 19 (13.38%) patients experienced HF. Cox analysis showed that EMC-T courses in the lateral wall, myoglobin, PR interval, and left atrial volume index were independent predictors of HF in patients with HCM. Five factors were incorporated into the model and shown as a nomogram. Stratification of patients into two risk subgroups by applying risk score (<230.65, ≥230.65) allowed significant distinction between Kaplan-Meier curves for cumulative incidence of HF events. In training dataset, the model had an AUC of 0.948 (95% CI: 0.885-1.000, p < 0.001) and achieved a good C-index of 0.918 (95% CI: 0.867-0.969). In validation dataset, the model had an AUC of 0.991 (95% CI: 0.848-1.000, p < 0.001) and achieved a strong C-index of 0.941 (95% CI: 0.923-1.000). Calibration plots showed high agreement between predicted and observed outcomes in both two datasets. ConclusionWe established and validated a novel model incorporating electromechanical coupling time courses for predicting HF in patients with HCM.