Vlietstra R E, Frye R L, Kronmal R A, Sim D A, Tristani F E, Killip T
Circulation. 1980 Aug;62(2):254-61. doi: 10.1161/01.cir.62.2.254.
The findings for 14 risk variables were correlated with the results of coronary arteriography in 8807 patients registered in the interinstitutional Coronary Artery Surgery Study (CASS). Discriminant-function analysis revealed that age, sex, cigarette smoking and the level of blood cholesterol best distinguished between the groups with (6688 patients) and without (2119 patients) coronary artery disease. A family history of coronary artery disease and the presence of hypertension or diabetes were of addition, but less, discriminating value. The relative risk for coronary artery disease in patients with the combination of cigarette smoking and an elevated cholesterol level was high (> 4) in females age 55 years or younger and in males age 35 years or younger. Few females age 45 years or younger (seven of 97) had coronary artery disease when both of these risk factors were absent. In spite of these correlations, only limited gains accrued from the use of discriminant-function analysis in correctly allocating patients into disease and nondisease groups. This indicates that, while certain factors are significantly correlated with coronary arteriographic findings, their value for predicting the presence of coronary artery disease is limited.
对14个风险变量的研究结果与机构间冠状动脉手术研究（CASS）登记的8807例患者的冠状动脉造影结果进行了相关性分析。判别函数分析显示，年龄、性别、吸烟和血胆固醇水平最能区分患有冠状动脉疾病（6688例患者）和未患冠状动脉疾病（2119例患者）的两组人群。冠状动脉疾病家族史以及高血压或糖尿病的存在也有一定的判别价值，但相对较小。在55岁及以下的女性和35岁及以下的男性中，吸烟和胆固醇水平升高同时存在的患者患冠状动脉疾病的相对风险较高（>4）。当这两个风险因素都不存在时，45岁及以下的女性很少有冠状动脉疾病（97例中有7例）。尽管存在这些相关性，但使用判别函数分析将患者正确分为患病组和未患病组所获得的收益有限。这表明，虽然某些因素与冠状动脉造影结果有显著相关性，但其预测冠状动脉疾病存在的价值有限。
