QRS时限变化对心脏再同步化治疗后全因死亡及心源性猝死的预测价值

目的: 探索QRS时限（QRS duration， QRSd）的变化（ΔQRSd）对心脏再同步化治疗（cardiac resynchronization therapy，CRT）后全因死亡及心源性猝死的预测价值。方法:纳入2011年3月至2020年12月南京医科大学附属无锡人民医院行CRT的患者并收集其临床资料和12导联体表心电图数据。测量术前和术后QRSd。Cox回归分析筛选全因死亡和心源性猝死的预测因子。以X tile软件分析ΔQRSd（QRSd术后-QRSd术前）对全因死亡和心源性猝死预测的截断值，以Kaplan Meier生存分析判断ΔQRSd与全因死亡和心源性猝死的关系。结果: 共纳入66例患者，中位随访时间72个月，15例死亡，16例发生心源性猝死事件。多因素Cox回归分析提示ΔQRSd是全因死亡的预测因子（HR=1.019，95% CI：1.004～1.035，P=0.014）；ΔQRSd（HR=1.016，95% CI：1.002～1.031，P=0.024）和完全性左束支传导阻滞（HR=0.211，95% CI：0.065～0.682，P=0.009）均是心源性猝死的预测因子。ΔQRSd对全因死亡和心源性猝死预测的截断值均为0 ms。ΔQRSd>0 ms组的术前左室舒张末内径显著高于ΔQRSd≤0 ms组（P<0.001），其术前QRSd显著低于ΔQRSd≤0 ms组（P=0.001）。Kaplan Meier生存分析提示ΔQRSd>0 ms是CRT后全因死亡、心源性猝死、5年生存率及5年心源性猝死率的预测因子（P值分别为<0.001、<0.001、0.01和<0.001）。结论:术后QRSd增宽是CRT患者全因死亡和心源性猝死的独立预测因子。

The predictive value of changes in QRS duration after cardiac resynchronization therapy in all-cause mortality and sudden cardiac death

Objective To evaluate the predictive value of changes in QRS duration （ΔQRSd) after cardiac resynchronization therapy （CRT) in all cause mortality and sudden cardiac death. Methods Patients underwent CRT between March 2011 and December 2020 in Wuxi People′s Hospital Affiliated to Nanjing Medical University were enrolled. The clinical data including 12 lead electrocardiography were obtained. QRS durations derived from pre and post implantation electrocardiography were measured. Predictors for all cause mortality and sudden cardiac death were evaluated using Cox regression analysis. X tile software was used to analyze the cut off of ΔQRSd for predicting all cause mortality and sudden cardiac death. ΔQRSd was assessed in relation to all cause mortality and sudden cardiac death by Kaplan Meier survival analysis. Results Sixty six patients were enrolled with a median follow up time of 72 months, 15 died and 16 suffered from sudden cardiac death. ΔQRSd was a predictor for all cause mortality in multiple Cox regression analysis（HR=1.019， 95% CI1.004-1.035，P=0.014）. Both ΔQRSd（HR=1.016， 95% CI1.002-1.031， P=0.024） and complete left bundle branch block（HR=0.211， 95% CI0.065-0.682， P=0.009） were predictors for sudden cardiac death. The cut off of ΔQRSd for predicting all cause mortality and sudden cardiac death was 0 ms. The pre implantation left ventricular end diastolic diameter in the ΔQRSd> 0 ms group was significantly higher than that in the ΔQRSd ≤ 0 ms group （P<0.001), and the pre implantation QRS duration was significantly lower than that in the ΔQRSd ≤ 0 ms group （P=0.001). Kaplan Meier survival analysis revealed that ΔQRSd> 0 ms was a predictor of all cause mortality, sudden cardiac death, 5 year survival rate, and 5 year sudden cardiac death rate after CRT （P<0.001, <0.001, 0.01 and <0.001, respectively). Conclusion A post implantation widening in QRS duration is an independent predictor of all cause mortality and sudden cardiac death after CRT.