Objective: To investigate the major risk factors for coronary heart disease (CHD) in elderly patients (≥75 years old) using coronary angiography and develop relevant diagnostic model. Methods: Retrospective analysis was conduc-ted on 548 elderly patients with suspected diagnosis of CHD(≥75 years old) underwent coronary angiography (CAG) treatment in Ruijin Hospital from June 2018 to December 2019. According to CAG results, the patients were divided into two groups :CHD group and control group. The risk factors were analyzed by multivariate Logistic regression, and the predictive value of these factors for the occurrence of CHD in the elderly was evaluated by receiver operator characteristic curve (ROC curve). Results: There were 408 cases in the CHD group and 140 cases in the control group based on results of CAG. The multivariate Logistic regression analysis after baseline analysis showed that elevated level of troponin I(OR=6.828, 95%CI:3.834-12.160, P<0.001), carotid plaque formation(OR=3.440, 95%CI: 1.780-6.650, P<0.001), elevated level of HbA1c(OR=1.532, 95%CI: 1.182-1.987, P=0.001) and white blood cell (WBC) counts (OR=1.187, 95%CI: 1.027-1.371, P=0.021) were risk factors for elderly patients with CHD, While female(OR=0.329, 95%CI: 0.201-0.538, P<0.001), elevated level of direct bilirubin(OR=0.800, 95%CI: 0.679-0.942, P=0.008) and elevated level of hemoglobin(OR=0.976, 95%CI: 0.960-0.992, P=0.003) were protective factors for CHD in elderly patients. The area under the curve of the ROC curve model established with above 7 factors for recognizing CHD in elderly patients was 0.825(P<0.05), with a good diagnostic performance. Conclusions:Unlike the risk factors established among the whole population, such as diabetes, high blood pressure, smoking, hyperlipidemia, obesity and high homocysteine, the increased level of troponin I and carotid artery plaque formation are major risk factors for CHD in the elderly. However, the female gender are potential protective factors for CHD. The corresponding diagnostic model is of certain value for identifying high-risk patients from patients suspected of CHD, and may help reduce excessive and unreasonable use of CAG.