Lian Dan, Liu Yong, Liu Yuan-Hui, Li Hua-Long, Duan Chong-Yang, Yu Dan-Qing
Department of Cardiology, Guangdong Cardiovascular Institute, Guangdong Academy of Medical Sciences, Guangdong General Hospital.
Int Heart J. 2017 Apr 6;58(2):197-204. doi: 10.1536/ihj.16-129. Epub 2017 Mar 21.
To establish a scoring model to predict the risk of contrast-induced nephropathy (CIN) in elderly patients undergoing elective coronary angiography (CAG).A total of 1286 patients aged > 65 years who had undergone elective CAG between August 2009 and February 2013 were enrolled in this study. They were randomly (3:2) assigned to a development (n = 756) or validation dataset (n = 530). Independent predictors of CIN were identified by using logistic regression and were assigned a weighted integer, which was used to establish a score model.CIN incidence in the development set was 6.3%. The risk score model contained 3 variables (with the weighted integer): age > 75 years (1.5), creatinine clearance (CrCl) < 60 mL/minute (1), and congestive heart failure (CHF) (1.5). CIN incidence was 3.1%, 9.1%, and 29.0% in the low-risk group (≤ 1), moderate risk group (1 - 3), and high-risk group (≥ 3), respectively. The risk model demonstrated good prediction value in the development (c-statistic = 0.727) and validation (c-statistic = 0.695) datasets. Compared to the non-CIN group, the CIN group had a significantly higher rate of inhospital major adverse cardiac events (P < 0.01).The risk score model with 3 variables, namely age > 75 years, CrCl < 60 mL/minute, and CHF, is a clinical prediction tool for CIN in elderly patients before elective CAG. CIN is one of the independent risk factors of major adverse cardiac events (MACE).
建立一个评分模型，以预测接受择期冠状动脉造影（CAG）的老年患者发生对比剂肾病（CIN）的风险。本研究纳入了2009年8月至2013年2月期间接受择期CAG的1286例年龄>65岁的患者。他们被随机（3:2）分配到开发数据集（n = 756）或验证数据集（n = 530）。通过逻辑回归确定CIN的独立预测因素，并为其分配一个加权整数，用于建立评分模型。开发集中CIN的发生率为6.3%。风险评分模型包含3个变量（及其加权整数）：年龄>75岁（1.5）、肌酐清除率（CrCl）<60 mL/分钟（1）和充血性心力衰竭（CHF）（1.5）。低风险组（≤1）、中风险组（1 - 3）和高风险组（≥3）的CIN发生率分别为3.1%、9.1%和29.0%。该风险模型在开发数据集（c统计量 = 0.727）和验证数据集（c统计量 = 0.695）中显示出良好的预测价值。与非CIN组相比，CIN组的院内主要不良心脏事件发生率显著更高（P < 0.01）。由年龄>75岁、CrCl < 60 mL/分钟和CHF这3个变量组成的风险评分模型是择期CAG前老年患者CIN的临床预测工具。CIN是主要不良心脏事件（MACE）的独立危险因素之一。
