心脏结构及功能指标的联合轨迹与慢性心衰预后的关联

目的  探讨心脏结构及功能指标纵向变化轨迹群对慢性心力衰竭(chronic heart failure, CHF)预后存在的潜在价值。方法  纳入2012年1月-2019年1月在山西省两所三级甲等医院住院并符合标准的277例患者，利用多次住院的心脏结构及功能指标，建立基于群组的多轨迹模型，识别具有相似轨迹变化的患者群体。使用Cox比例风险回归模型分析轨迹群对患者死亡风险的影响。结果  共识别轨迹指标变化各异、基线特征分布存在差异的3组轨迹群组。其中轨迹群1中患者男女均衡，纽约心脏病协会(New York Heart Association, NYHA)心功能分级集中于Ⅱ~Ⅲ级，射血分数(ejection fraction, EF)整体较稳定并高于另外两组，不稳定型心绞痛的患者较多；轨迹群2中男性、陈旧性心肌梗死患者占比高，EF稳定下降；轨迹群3中男性、NYHA心功能分级为Ⅳ级的患者比例高，EF呈现波动下降状态，患缺血性心肌病、瓣膜性心脏病和房颤的患者多。Cox比例风险回归模型结果表明轨迹群组在单独纳入、校正基线特征和校正共病影响三种情况下均为CHF患者发生死亡的影响因素。结论  基于心脏结构及功能指标纵向变化识别的轨迹群能够对CHF患者的死亡风险进行分层，有助于帮助识别高危患者。

Association between combined trajectories of cardiac structure and function indicators and prog-nosis of chronic heart failure

  Objective  To explore the potential value of the longitudinal trajectory groups of cardiac structure and function indicators for the prognosis of patients with chronic heart failure.  Methods  A total of 277 patients who were hospitalized in two tertiary hospitals in Shanxi Province from January 2012 to January 2019 and met the criteria were included. The cardiac structure and function indicators of multiple hospitalizations were used to establish the group-based multi-trajectory model to identify patient groups with similar trajectory changes. The Cox proportional hazards regression model was used to analyze the influence of trajectory group on the risk of death of patients.  Results  Three trajectory groups with different changes in trajectory indicators and baseline feature distributions were identified. Among them, patients in trajectory group 1 were balanced between men and women, New York Heart Association (NYHA) functional class was concentrated in grades Ⅱ-Ⅲ, the overall ejection fraction (EF) was relatively stable, which was higher than the other two groups, and more patients had unstable angina. The proportion of men and patients with old myocardial infarction in trajectory group 2 were high, and the EF was steadily decreasing. Trajectory group 3 had a high proportion of males and NYHA grade IV patients, EF presented a fluctuated decline state, ischemic cardiomyopathy, valvular heart disease and atrial fibrillation were more common. The results of the Cox proportional hazards regression model showed that the trajectory group were an influencing factor for the death of patients with chronic heart failure under the three conditions of individual inclusion, adjustment of baseline characteristics, and adjustment of comorbidity.  Conclusion  The trajectory groups identified based on the longitudinal changes of cardiac structure and function indicators can stratify the death risk of patients with chronic heart failure and help identify high-risk patients.