Page 74 - 《南京医科大学学报（自然科学版）》2025年第9期

南京医科大学学报（自然科学版） 第45卷第9期
·1286 · Journal of Nanjing Medical University（Natural Sciences） 2025年9月

·临床研究·

心肌最大壁厚和左室流出道解剖学参数评估肥厚型心肌病心

肌纤维化的对照研究

钱芷君，施夏韵，殷 凡，刘王琰，徐 怡，王云飞，朱晓梅 *

南京医科大学第一附属医院放射科，江苏 南京 210029

［摘 要］ 目的：探讨并对比心脏磁共振（cardiac magnetic resonance，CMR）左室舒张末期心肌最大壁厚（maximal wall thick⁃
ness，MWT）和左室流出道（left ventricular outflow tract，LVOT）解剖学参数在评估肥厚型心肌病（hypertrophic cardiomyopathy，
HCM）心肌纤维化中的预测价值，并建立预测模型。方法：回顾性分析77例行CMR检查的HCM患者，测量参数包括部分二尖
瓣前叶长度、总二尖瓣前叶长度、舒张末期和收缩末期的LVOT直径和基底前间隔厚度、左室舒张末期MWT以及心肌延迟强
化百分比（percentage of late gadolinium enhancement，LGE%）等，以LGE%来评估心肌纤维化。利用统计软件随机选择70%的样
本分配至建模组（n=54），通过单因素及多因素回归分析，建立LGE%的预测模型；剩余30%的样本为内部验证组（n=23），所有
患者的超声心动图参数作为外部验证组，评估预测模型的准确性。绘制受试者工作特征曲线，通过计算曲线下面积来确定模
型的预测性能，评估预测模型的灵敏度及特异度。结果：在建模组中，多因素线性回归分析提示，MWT是LGE%的独立预测因
子，线性公式为LGE%=-10.009+0.832×MWT（r=0.466，P < 0.001），而LVOT解剖学参数均与LGE%无线性相关性。在外部验证
组中，超声MWT与CMR MWT呈高度正相关（r=0.856，P < 0.001），且内部验证和外部验证的LGE%预测值均与实际LGE%差异
无统计学意义；当心脏CMR MWT≥30 mm或超声MWT≥25 mm时，其预测LGE%≥15%的准确率分别为82.6%和81.7%。结论：
在评估HCM心肌纤维化时，MWT比LVOT解剖学参数更有预测价值。
［关键词］ 肥厚型心肌病；左室舒张末期心肌最大壁厚；左室流出道；心脏磁共振；延迟强化

［中图分类号］ R542.2；R445.2 ［文献标志码］ A ［文章编号］ 1007⁃4368（2025）09⁃1286⁃08
doi：10.7655/NYDXBNSN250189

A comparative study of maximal wall thickness and anatomical parameters of left
ventricular outflow tract for evaluating myocardial fibrosis in hypertrophic cardiomyopathy

*
QIAN Zhijun，SHI Xiayun，YIN Fan，LIU Wangyan，XU Yi，WANG Yunfei，ZHU Xiaomei
Department of Radiology，the First Affiliated Hospital of Nanjing Medical University，Nanjing 210029，China

［Abstract］ Objective：To comparatively explore the value of left ventricular end ⁃ diastolic maximal wall thickness（MWT）and
anatomical parameters of left ventricular outflow tract（LVOT）for evaluating myocardial fibrosis in hypertrophic cardiomyopathy
（HCM）by cardiac magnetic resonance（CMR）and propose a prediction model. Methods：Seventy⁃seven HCM patients who underwent
CMR examination were retrospectively analyzed. CMR data included partial anterior mitral leaflets length and total anterior mitral
leaflet length. During end⁃diastole and end⁃systole，the diameter of LVOT and the thickness of basal anteroseptum were measured.
Additionally，left ventricular end ⁃ diastolic MWT was collected and the percentage of late gadolinium enhancement（LGE% ）was
analyzed. LGE% was used to assess myocardial fibrosis. Seventy percent of the samples selected randomly by statistical software were
assigned to the modeling group（n=54）for establishing a prediction model of LGE% through univariate and multivariate analysis. The
remaining thirty percent of the samples served as the internal validation group（n=23），and parameters of the echocardiogram of all
patients were used as the external validation group to assess the accuracy of the prediction model. Receiver operating characteristic
curves were plotted，and the predictive efficiency of the prediction model was determined by calculating the area under the curve. The
sensitivity and specificity of the prediction model were also evaluated. Results：In the modeling group，multivariate analysis indicated

［基金项目］ 国家自然科学基金（82302163）；南京医科大学第一附属医院青年学者培养基金（PY2022036）
通信作者（Corresponding author），E⁃mail：xiaomeizhu@njmu.edu.cn（ORCID：0000⁃0002⁃7591⁃9930）
∗