Abstract: Background　The risk prediction scoring systems for acute coronary syndrome （ACS） have been extensively used for assessing the long-term prognosis of patients with acute non-ST-segment elevation myocardial infarction （NSTEMI）. Both STEMI and NSTEMI belong to ACS，but NSTEMI shows an increasing percentage in ACS，with a possible trend of exceeding the percentage of STEMI. Moreover，they have significant differences in incidence，mortality risk during the acute phase，and long-term prognosis. However，there is no literature about tools for effectively predicting long-term prognosis of patients with NSTEMI. Objective　To identify the risk factors for long-term major adverse cardiovascular events（MACEs） in discharged patients with NSTEMI，and based on this，to develop a predictive scoring system. Methods　Three hundred and twenty-four patients with an admission diagnosis of NSTEMI were recruited from Hebei General Hospital from January 2017 to August 2018. Clinical data were obtained，including age，sex，Killip class，systolic and diastolic blood pressure，heart rate，cardiac function at admission，past medical history（history of old myocardial infarction，PCI，coronary artery bypass grafting，thrombolysis，hypertension，diabetes，hyperlipidemia，chronic kidney disease，old cerebral infarction，cerebral hemorrhage，anemia），smoking，drinking，incidence of in-hospital heart failure，in-hospital medication，GRACE score，left ventricular ejection fraction，hemoglobin，white blood count，lymphocytes，platelet count，alanine aminotransferase，aspartate transaminase，total protein，globulin，albumin，serum creatinine，glomerular filtration rate，electrolytes，blood lipids，and treatment strategies. All cases were followed up till 2020-09-01，during which the incidence of MACEs and all-cause death were recorded. The risk factors of out-of-hospital MACEs were identified，and used to construct a predictive scoring system. Results　（1） Compared with patients without MACEs，those with MACEs had older mean age，higher mean higher prevalence of grade 2-3 hypertension，in-hospital betamethasone treatment，in-hospital heart failure，and non-revascularization treatment，lower prevalence of smoking，as well as lower mean levels of glomerular filtration rate，low-density lipoprotein，lymphocyte count，and hemoglobin（P<0.05）. （2） Multivariate Logistic regression analysis showed lymphocyte count〔OR=0.621，95%CI（0.408，0.946），P=0.026〕，hemoglobin〔OR=0.983，95%CI（0.969，0.996），P=0.013〕，non-revascularization 〔OR=2.030，95%CI（1.153，3.573），P=0.014〕，and the cut-off value of total protein〔OR=2.412，95%CI（1.244，4.675），P=0.009〕were associated with out-of-hospital MACEs in NSTEMI patients.（3） A regression equation based on these risk factors of MACEs was developed：logit（P）=3.98-0.476X1-0.018X2+0.88X3+0.708X4（X1-X4 stand for lymphocyte count，hemoglobin，total protein，and non-revascularization，respectively）. Combined with the risk coefficient β of each factor，a scoring system with a total score of 21 points and a predictive threshold of 9 points was constructed，with an AUC value of 0.704. Conclusion　Long-term MACEs were common in patients with acute NSTEMI，which may be associated with lymphocyte count，hemoglobin，total protein and non-revascularization. Our predictive scoring system for long-term MACEs with a total score of 21 points and a predictive threshold of 9 points，and has good predictive capability.

Key words: Non-ST elevated myocardial infarction, Cardiovascular diseases, Adverse cardiovascular events, Risk factors, Forecasting

摘要： 背景　对于急性非ST段抬高型心肌梗死（NSTEMI）患者远期预后预测目前应用的是急性冠脉综合征（ACS）患者的评分，其在ACS中占比逐年升高，有超越急性ST段抬高型心肌梗死（STEMI）的趋势，二者虽同属ACS，但其发病率、急性期死亡风险及远期预后存在明显差异。然而截止到目前，尚无有关针对于NSTEMI患者远期预测的有效工具的文献。目的　分析NSTEMI患者院外不良心血管事件（MACE）发生的危险因素并构建预测评分系统。方法　收集2017年1月至2018年8月于河北省人民医院入院诊断为NSTEMI的患者324例为研究对象。收集患者一般资料，包括年龄、性别、Killip分级、入院时收缩压和舒张压、心率、心功能、既往病史〔陈旧性心肌梗死、经皮冠状动脉介入治疗（PCI）史、冠状动脉旁路移植术（CABG）史、溶栓史、高血压、糖尿病、高脂血症、慢性肾功能不全、陈旧性脑梗死、脑出血、贫血〕、吸烟、饮酒。记录患者院内是否心力衰竭、院内用药、GRACE评分、射血分数、血红蛋白、白细胞计数、淋巴细胞计数、血小板计数、丙氨酸氨基转氨酶、天冬氨酸氨基转移酶、总蛋白、球蛋白、白蛋白、肌酐、肾小球滤过率、离子、血脂以及治疗策略选择情况。对患者进行随访至2020-09-01，观察MACE及全因死亡发生情况。根据患者随访期间是否发生MACE分为MACE组和非MACE组。分析NSTEMI患者院外MACE发生的危险因素并构建预测评分系统。结果　（1）与非MACE组相比，MACE组患者年龄、高血压≥2级比例、院内应用倍特类比例、总蛋白、球蛋白、院内心力衰竭比例、未血管重建比例均较高，吸烟比例、肾小球滤过率、低密度脂蛋白、淋巴细胞计数、血红蛋白均较低（P<0.05）。（2）多因素Logistic回归分析显示，淋巴细胞计数〔OR=0.621，95%CI（0.408，0.946），P=0.026〕，血红蛋白〔OR=0.983，95%CI（0.969，0.996），P=0.013〕，未血运重建〔OR=2.030，95%CI（1.153，3.573），P=0.014〕，总蛋白〔OR=2.412，95%CI（1.244，4.675），P=0.009〕是NSTEMI患者发生院外MACE的影响因素。（3）进一步建立危险因素的回归方程：logit（P）=3.98-0.476X1-0.018X2+0.88X3+0.708X4（注：X1表示淋巴细胞计数，X2表示血红蛋白，X3表示总蛋白，X4表示未血运重建）。结合各因素的危险系数β，最终建立了总分值为21分，预测截断值为9分的评分系统；ROC曲线下面积为0.704。结论　NSTEMI患者远期发生MACE较为常见，淋巴细胞计数、血红蛋白、总蛋白、未血运重建是其影响因素。本研究构建了总分值21分，预测截断值9分的远期MACE发生预测评分系统，该评分系统具有较好的预测效能。

关键词: 非ST段抬高型心肌梗死, 心血管疾病, 不良心血管事件, 危险因素, 预测