急性心肌梗死患者发生无复流现象的危险评分

目的：建立一种预测急性心肌梗死（AMI）患者经皮冠脉介入诊疗（PCI）术中发生无复流风险的评分方法。方法收集接受介入治疗的 AMI 患者的无复流相关因素的临床资料，分为评分模型组和验证组，采用二项多变量 Logistic 回归研究的方法，筛选独立危险因素，建立无复流风险评分系统，对发生风险进行分层，并对评分系统真实性及可靠性进行评价。结果模型组多变量 Logistic 回归分析显示女性（OR ＝0.587，P ＝0.019）、心功能 Killip 分级≥2级（OR ＝3.656， P ＜0.01）、TIMI 血流分级≤2（OR ＝0.774，P ＝0.013）、PCI 术前血栓负荷＞4分（OR ＝2.629，P＜0.01）、发病至球囊扩张病变时间＞6 h （OR ＝1.485，P ＝0.083）为 AMI 患者 PCI 中发生无复流的独立危险因素。危险分层：评分＜2分为低危，2～5分为中危，＞5分为高危。模型组受试者工作曲线下面积（AUC）为0.716（95％CI：0.671～0.761）。采用 Hosmer-Lemeshow 拟合优度检验评价得出：χ2＝1.027，P ＝0.994，提示预测值与实际值差异无统计学意义。验证组二项 Logistic 回归分析的 AUC ＝0.891（95％CI：0.822～0.959）。分别比较验证组 killip 分级、血栓负荷、评分、危险分层的受试者工作曲线，显示无复流评分的 AUC 最大（AUC ＝0.851，95％ CI：0.776～0.927）。结论根据基于 AMI 患者临床和造影特征建立的简单无复流风险评分系统，可用于对 PCI中无复流发生风险进行分层，为临床提供了一个方便快捷预测无复流发生的工具。

Development and validation of no-reflow phenomenon risk score in patients with acute myocardial infarction

Objective To establish a bedside available risk scoring system of no-reflow in the acute stage of STEMI.Methods Data from STEMI patients treated with PCI divided into model group and validation group were analyzed.Multivariable binary logistic regression analysis was used to identify independent no-reflow predictors of the model group.Finally,a score according to the odds ratio on logistic regression analysis was designed,and then risk stratification was established,and no-reflow high-risk patients with myocardial infarction were selected.The authenticity and reliability of the logistic regression courses were validated using receiver operator characteristic curve (ROC)and Hosmer-Lemeshow goodness-of-fit.Results Multivariate logistic regression analysis demonstrated that female (OR =0.587,P =0.019),Killip class of myocardial infarction≥2 (OR =3.656,P <0.01),TIMI flow ≤2 before primary PCI (OR =0.774,P =0.013),thrombus burden score ≥4 on baseline angiography (OR =2.629,P <0.01),pain to balloon time ≥ 6 h (OR =1.485,P =0.083)were independent correlate predictors of no-reflow phenomenon in the STEMI after PCI.The risk score system demonstrated a good risk prediction in the model group with AUC of 0.716 (95%CI:0.671 -0.761)based on ROC analysis.There was no significant discrepancy between multivariate logistic regression analysis and Hosmer-Lemeshow goodness-of-fit (χ2 =1.027,P =0.994).In risk stratification,total value <2 was assigned into low risk level,and 2-5 was put into the medium risk level,and >5 was arranged into high risk level.The risk score system demonstrated a good risk prediction in the validation group with AUC of 0.891 (95%CI:0.822 -0.959)based on ROC analysis.ROC analysis in the validation group was applied to Killip class,thrombus burden,score and risk stratification in the validation group ,and the no-reflow score was more accurate,with a larger area under the curve (AUC = 0.851,95% CI:0.776 -0.927 ).Conclusions Establishment of no-reflow scoring system with STEMI patients undergoing PCI was benefit to select high risk patients with no-reflow.