目的：评价LogisticEuroSCORE与EuroSCOREⅡ风险预测模型对本中心成人单纯二尖瓣置换术后在院死亡率预测的准确性。方法回顾性收集2009年1月至2013年12月497例患者在院接受单纯二尖瓣置换术的临床资料，包括同期行三尖瓣成形、左房减容、左心耳闭合术患者。其中男性186例，女性311例，平均（53．19±11．11）岁。按照第一版的Logistic EuroSCORE模型与新版的EuroSCOREⅡ模型给予评分获得患者的预测死亡率，与患者的实际在院死亡率比较。模型预测的符合程度应用Hosmer－Lemeshow拟和优度检验，而预测的鉴别效度则通过受试者工作特征曲线（ ROC）下面积反映。结果497例患者在院死亡6例，全组实际在院死亡率为1．21％，logistic EuroSCORE及EuroSCOREⅡ预测的在院死亡率分别为2．99％和1．53％。其中EuroSCOREⅡ对全组患者在院死亡率预测符合程度较高，而logistic EuroSCORE明显高估了在院死亡率。 lo⁃gistic EuroSCORE的区分度及校准度差（ P ＝0．673，AUC＝0．621），EuroSCOREⅡ区分度及校准度均较好（ P ＝0．894，AUC＝0．793）。结论 EuroSCOREⅡ对本中心单纯二尖瓣置换手术患者在院死亡风险预测的准确性较高。%Objective To assess the value of Logistic EuroSCORE and EuroSCORE Ⅱ in predicating in-hospital mortality a⁃mong Chinese patients undergoing mitral valve replacement surgery in our center. Methods 497 patients who underwent mitral valve replacement with or without concomitant tricuspid valvuloplasty, left atrium reduction and left auricle closure from January 2009 to De⁃cember 2013 in our department were enrolled in our study. The mean age was 53.19±11.11 years at the time of surgery, and 62.6%were female. They were scored by logistic EuroSCORE and EuroSCOREⅡmodel, and compared with observed in-hospital mortality. Calibration was evaluated by Hosmer-Lemeshow( H-L) test, and discrimination was evaluated by the area under the receiver operating characteristic ( ROC) curve. Results There were 6 in-hospital deaths among the 497 patients and the in-hospital mortality rate was 1.21%, compared with the predicted mortality rate of 2. 99% by Logistic EuroSCORE and 1. 53% by EuroSCOREⅡ. Logistic Eu⁃roSCORE suggested a high prediction, and EuroSCOREⅡsuggested a similar mortality. The discrimination of Logistic EuroSCORE was poor ( P =0.673, AUC=0.621) . EuroSCOREⅡshowed good calibration and discrimination ( P =0.894, AUC=0.793) . Conclusion The EuroSCOREⅡ was suitable in predicting the mortality of mitral valve replacement surgery patients in our department.
展开▼
