Objective: To assess the value of cardiac magnetic resonance (CMR) imaging for predicting adverse left ventricular remodeling in patients with ST-segment elevation myocardial infarction (STEMI).
Methods: We retrospectively analyzed the clinical data and serial CMR (cine and LGE sequences) images of 86 STEMI patients within 1 week and 5 months after percutaneous coronary intervention (PCI), including 25 patients with adverse LV remodeling and 61 without adverse LV remodeling, defined as an increase of left ventricular end-systolic volume (LVESV) over 15% at the second CMR compared to the initial CMR. The CMR images were analyzed for LV volume, infarct characteristics, and global and infarct zone myocardial function. The independent predictors of adverse LV remodeling following STEMI were analyzed using univariate and multivariate Logistic regression methods.
Results: The initial CMR showed no significant differences in LV volume or LV ejection fraction (LVEF) between the two groups, but the infarct mass and microvascular obstructive (MVO) mass were significantly greater in adverse LV remodeling group (P < 0.05). Myocardial injury and cardiac function of the patients recovered over time in both groups. At the second CMR, the patients with adverse LV remodeling showed a significantly lower LVEF, a larger left ventricular end-systolic volume index (LVESVI) and a greater extent of infarct mass (P < 0.001) with lower global peak strains and strain rates in the radial, circumferential, and longitudinal directions (P < 0.05), infarct zone peak strains in the 3 directions, and infarct zone peak radial and circumferential strain rates (P < 0.05). The independent predictors for adverse LV remodeling following STEMI included the extent of infarct mass (AUC=0.793, 95% CI: 0.693-0.873; cut-off value: 30.67%), radial diastolic peak strain rate (AUC=0.645, 95% CI: 0.534-0.745; cut-off value: 0.58%), and RAAS inhibitor (AUC= 0.699, 95% CI: 0.590-0.793).
Conclusion: The extent of infarct mass, peak radial diastolic strain rate, and RAAS inhibitor are independent predictors of adverse LV remodeling following STEMI.
目的: 评估心脏磁共振（CMR）对ST段抬高型心肌梗死（STEMI）后左心室不良重构的价值。
方法: 回顾性分析86例STEMI患者经皮冠状动脉介入术后1周及5个月的临床资料和CMR图像。所有受试者均采集电影和LGE序列。将患者分为左心室不良重构组（n=25）和无左心室不良重构组（n=61）。左心室不良重构定义为第2次CMR检查时左心室收缩末期容积（LVESV）较初始CMR增加15%或更多。CMR分析包括左心室容积、心肌梗死特征、整体和区域心肌功能。采用Logistic回归法分析左心室不良重构的独立预测因素。
结果: 在初始CMR时，两组患者的左心室容积及左心室射血分数（LVEF）差异均无统计学意义（P>0.05），但左心室不良重构组梗死质量百分比（[34.07±10.04）% vs（22.20±11.29）%，P < 0.001]和微血管阻塞（MVO）质量百分比[1.9（0, 3.8）% vs 0（0, 1.06）%，P < 0.001]明显大于无左心室不良重构组。随时间推移，两组患者的心肌损伤和心功能均有恢复。但在第2次CMR时，左心室不良重构患者有更低的LVEF [（42.01± 9.51）% vs（55.23±10.04）%，P < 0.001]、更大的左心室收缩末期容积指数[55.58（43.15，69.91）mL/m2 vs 35.79（26.70，45.04）mL/m2，P < 0.001]和梗死质量百分比（[26.71± 24.51）% vs（17.08± 9.25）%，P < 0.001]。左心室不良重构组整体峰值应变及应变率、梗死区峰值应变、径向和周向峰值应变率更小（均P < 0.05）。多因素分析显示，只有3个因素是左心室不良重构的独立预测因素，其中梗死质量百分比预测左心室不良重构的AUC为0.793（95%CI 0.693~0.873），截断值为30.67%；整体径向舒张期峰值应变率的AUC为0.645（95% CI 0.534~0.745），截断值为0.58%；肾素血管紧张素醛固酮系统（RAAS）抑制剂的AUC为0.699（95% CI 0.590~0.793）。
结论: 在急性期，左心室不良重构组和无左心室不良重构患者的左心室容积、整体及节段的心功能没有显著差异；但随时间推移，左心室不良重构组的左心室容积明显变大，整体及节段的心功能明显更差。梗死质量百分比、径向舒张期峰值应变率及RAAS抑制剂是左心室不良重构的独立预测因子。
Keywords: ST-segment elevation myocardial infarction; adverse left ventricular remodeling; cardiac magnetic resonance; cardiac magnetic resonance feature-tracking; myocardium strain.