探讨扩张型心肌病患者药物治疗后发生心脏结构和功能逆转的预测因素。

回顾性分析2016年8月至2017年4月在徐州市中心医院住院的156例扩张型心肌病患者资料，根据患者治疗后的临床效果和超声心动图结果的不同反应性，将156例扩张型心肌病患者分为左心室逆重构（LVRR）组（27例）和非LVRR组（129例）。记录所有患者的年龄、性别、体质量指数、病程、收缩压、纽约心脏病学会分级、高血压、糖尿病、心房颤动、血管紧张素转化酶抑制剂/血管紧张素Ⅱ受体阻滞剂、β受体阻滞剂、螺内酯、完全性左束支传导阻滞、QRS波时限、N末端脑钠肽前体、左心室射血分数、左心室舒张末期内径、左室收缩末期容积指数、6 min步行试验、服药依从性、6个月再住院率及6个月病死率等资料。采用多因素Logistic回归分析扩张型心肌病患者发生LVRR的相关因素；采用受试者工作特征（ROC）曲线分析QRS波时限、服药依从性对扩张型心肌病患者发生LVRR的预测价值。

LVRR组患者年龄[（54 ± 16）岁vs.（60 ± 12）岁，t=5.899，P=0.021]和QRS波时限[（104 ± 21）ms vs.（122 ± 29）ms，t=10.464，P=0.003]均较非LVRR组显著降低，服药依从性[（6.0 ± 1.0）分vs.（5.3 ± 1.5）分，t=8.962，P=0.002]较非LVRR组显著升高。多因素Logistic回归分析结果显示，QRS波时限[比值比（OR）=0.969，95%置信区间（CI）（0.946，0.993），P=0.011]、服药依从性[OR=1.541，95%CI（1.064，2.230），P=0.022]是心功能恢复正常的独立影响因素。ROC曲线分析结果显示，QRS波时限[曲线下面积（AUC）=0.721，95%CI（0.619，0.824），P < 0.001]、服药依从性[AUC=0.661，95%CI（0.559，0.762），P=0.009]均对扩张型心肌病患者发生LVRR具有预测价值。

QRS波时限及服药依从性是扩张型心肌病患者发生LVRR的重要预测因素。

To explore the predictors for cardiac structural and functional reversal after pharmacological treatment in patients with dilated cardiomyopathy.

A total of 156 patients with dilated cardiomyopathy admitted to Xuzhou Central Hospital from August 2016 to April 2017 were retrospectively analyzed. According to their clinical outcomes and different reactivity of echocardiographic findings after treatment, 156 patients were divided into a left ventricular reverse remodeling (LVRR) group (n=27) and a non-LVRR group (n=129). The age, sex, body mass index, course of disease, systolic blood pressure, New York Heart Association grade, hypertension, diabetes, atrial fibrillation, angiotensin converting enzyme inhibitor/angiotensin receptor blocker, β receptor blocker, spironolactone, complete left bundle branch block, QRS wave time limit, N-terminal pro-brain natriuretic peptide, left ventricular ejection fraction, left ventricular end-diastolic dimension, left ventricular end-systolic volume index, 6-min walk test, medication compliance, 6-month rehospitalization rate and 6-month mortality rate were recorded for all patients. Multivariate Logistic regression was used to analyze factors related to LVRR in patients with dilated cardiomyopathy. Receiver operating characteristic (ROC) curves were used to evaluate predictive value of QRS wave time limit and medication compliance in LVRR.

Compared with the non-LVRR group, the age [(54±16) years vs. (60 ± 12) years, t=5.899, P=0.021] and QRS wave time limit [(104 ± 21) ms vs. (122 ± 29) ms, t=10.464, P=0.003] were significantly lower and the medication compliance [(6.0 ± 1.0) vs. (5.3 ± 1.5), t=8.962, P=0.002] was significantly higher in the LVRR group. The multivariate Logistic regression showed that the QRS wave time limit [odds ratio (OR)=0.969, 95% confidence interval (CI) (0.946, 0.993), P=0.011] and medication compliance [OR=1.541, 95%CI (1.064, 2.230), P=0.022] were independent factors affecting the return to normal cardiac function. The ROC curve showed that the QRS wave time limit [area under the curve (AUC)=0.721, 95%CI (0.619, 0.824), P < 0.001] and medication compliance [AUC=0.661, 95%CI (0.559, 0.762), P < 0.009] both have predictive value for LVRR in patients with dilated cardiomyopathy.

QRS wave time limit and medication compliance are important predictors for LVRR in patients with dilated cardiomyopathy.