Objective: To explore the value of cardiac magnetic resonance imaging (CMR) in the risk stratification of hypertrophic cardiomyopathy (HCM). Methods: HCM patients who underwent CMR examination in Fuwai Hospital between March 2012 and May 2013 were retrospectively enrolled. Baseline clinical and CMR data were collected and patient follow-up was performed using telephone contact and medical record. The primary composite endpoint was sudden cardiac death (SCD) or and equivalent event. The secondary composite endpoint was all-cause death and heart transplant. Patients were divided into SCD and non-SCD groups. Cox regression was used to explore risk factors of adverse events. Receiver operating characteristic (ROC) curve analysis was used to assess the performance and the optimal cut-off of late gadolinium enhancement percentage (LGE%) for the prediction of endpoints. Kaplan-Meier and log-rank tests were used to compare survival differences between groups. Results: A total of 442 patients were enrolled. Mean age was (48.5±12.4) years and 143(32.4%) were female. At (7.6±2.5) years of follow-up, 30 (6.8%) patients met the primary endpoint including 23 SCD and 7 SCD equivalent events, and 36 (8.1%) patients met the secondary endpoint including 33 all-cause death and 3 heart transplant. In multivariate Cox regression, syncope(HR=4.531, 95%CI 2.033-10.099, P<0.001), LGE% (HR=1.075, 95%CI 1.032-1.120, P=0.001) and left ventricular ejection fraction (LVEF) (HR=0.956, 95%CI 0.923-0.991, P=0.013) were independent risk factors for primary endpoint; Age (HR=1.032, 95%CI 1.001-1.064, P=0.046), atrial fibrillation (HR=2.977, 95%CI 1.446-6.131, P=0.003),LGE% (HR=1.075, 95%CI 1.035-1.116, P<0.001) and LVEF (HR=0.968, 95%CI 0.937-1.000, P=0.047) were independent risk factors for secondary endpoint. ROC curve showed the optimal LGE% cut-offs were 5.1% and 5.8% for the prediction of primary and secondary endpoint, respectively. Patients were further divided into LGE%=0, 0<LGE%<5%, 5%≤LGE%<15% and LGE%≥15% groups. There were significant survival differences between these 4 groups whether for primary endpoint or secondary endpoint (all P<0.001) and the accumulated incidence of primary endpoint was 1.2% (2/161), 2.2% (2/89), 10.5% (16/152) and 25.0% (10/40), respectively. Conclusion: LGE is an independent risk factor for SCD events as well as all-cause death and heart transplant. LGE is of important value in the risk stratification in patients with HCM.
目的： 探讨心脏磁共振（CMR）在肥厚型心肌病（HCM）危险分层中的价值。 方法： 本研究为单中心回顾性研究，连续选取2012年3月至2013年5月于阜外医院接受CMR检查并诊断为HCM的患者，收集患者的临床及CMR资料，通过电话随访和病案记录确认患者生存情况。主要终点为心原性猝死（SCD）或SCD等位事件，次要终点为全因死亡及心脏移植复合终点，根据是否发生主要终点事件分组。采用Cox回归探索不良事件的危险因素，采用受试者工作特征（ROC）曲线评估延迟强化百分比（LGE%）对终点事件的预测效能和最佳临界值，采用Kaplan-Meier法和log-rank检验分析组间生存差异。 结果： 共纳入患者442例，年龄（48.5±12.4）岁，女性143例（32.4%）。随访（7.6±2.5）年，共30例（6.8%）患者发生主要终点事件，包括23例SCD和7例SCD等位事件；共36例（8.1%）患者发生次要终点事件，包括33例死亡，3例心脏移植。多因素Cox回归示，晕厥（HR=4.531，95%CI 2.033~10.099，P<0.001）、LGE%（HR=1.075，95%CI 1.032~1.120，P=0.001）和左心室射血分数（LVEF）（HR=0.956，95%CI 0.923~0.991，P=0.013）是主要终点事件的独立危险因素；年龄（HR=1.032，95%CI 1.001~1.064，P=0.046）、心房颤动（HR=2.977，95%CI 1.446~6.131，P=0.003）、LGE%（HR=1.075，95%CI 1.035~1.116，P<0.001）和LVEF（HR=0.968，95%CI 0.937~1.000，P=0.047）是次要终点事件的独立危险因素。ROC曲线示以5.1%和5.8%为LGE%临界值分别对主要终点和次要终点的预测效能最高。以0、5%和15%为LGE%界值进一步分组的Kaplan-Meier曲线示，LGE%=0、0<LGE%<5%、5%≤LGE%<15%、LGE%≥15% 4组患者的主要和次要终点生存差异均有统计学意义（P均<0.001），其中主要终点事件发生率依次为1.2%（2/161）、2.2%（2/89）、10.5%（16/152）和25.0%（10/40）。 结论： LGE是HCM患者SCD事件以及全因死亡和心脏移植事件的独立危险因素，具有重要的危险分层价值。.