Development of Machine Learning Tools for Predicting Coronary Artery Disease in the Chinese Population.

Purpose: Coronary artery disease (CAD) is one of the major cardiovascular diseases and the leading cause of death globally. Blood lipid profile is associated with CAD early risk. Therefore, we aim to establish machine learning models utilizing blood lipid profile to predict CAD risk. Methods: In this study, 193 non-CAD controls and 2001 newly-diagnosed CAD patients (1647 CAD patients who received lipid-lowering therapy and 354 who did not) were recruited. Clinical data and the result of routine blood lipids tests were collected. Moreover, low-density lipoprotein cholesterol (LDL-C) subfractions (LDLC-1 to LDLC-7) were classified and quantified using the Lipoprint system. Six predictive models (k-nearest neighbor classifier (KNN), logistic regression (LR), support vector machine (SVM), decision tree (DT), multilayer perceptron (MLP), and extreme gradient boosting (XGBoost)) were established and evaluated by the confusion matrix, area under the receiver operating characteristic (ROC) curve (AUC), recall (sensitivity), accuracy, precision, and F1 score. The selected features were analyzed and ranked. Results: While predicting the CAD development risk of the CAD patients without lipid-lowering therapy in the test set, all models obtained AUC values above 0.94, and the accuracy, precision, recall, and F1 score were above 0.84, 0.85, 0.92, and 0.88, respectively. While predicting the CAD development risk of all CAD patients in the test set, all models obtained AUC values above 0.91, and the accuracy, precision, recall, and F1 score were above 0.87, 0.94, 0.87, and 0.92, respectively. Importantly, small dense LDL-C (sdLDL-C) and LDLC-4 play pivotal roles in predicting CAD risk. Conclusions: In the present study, machine learning tools combining both clinical data and blood lipid profile showed excellent overall predictive power. It suggests that machine learning tools are suitable for predicting the risk of CAD development in the near future.

CYP2C19 metabolizer phenotypes may affect the efficacy of statins on lowering small dense low-density lipoprotein cholesterol of patients with coronary artery disease.

Background: Dyslipidemia is a major cause of arteriosclerotic cardiovascular disease (ASCVD), and low-density lipoprotein cholesterol (LDL-C) is the profile to be reduced to prevent disease progression. Small dense low-density lipoprotein cholesterol (sdLDL-C) has been proven to be a more effective biomarker than LDL-C for ASCVD primary and secondary prevention. CYP2C19 is an important drug metabolism gene. This study aimed to investigate the relationship between sdLDL-C and coronary artery disease (CAD) risk factors and explore the influence of CYP2C19 metabolizer phenotypes on the sdLDL-C lowering efficacy of statins. Methods: This study recruited 182 patients with CAD and 200 non-CAD controls. Baseline laboratory indices of fasting blood were detected, including blood lipids, glucose, and creatinine. In addition, LDL-C subfractions were separated and quantified. Gene polymorphisms of SLCO1B1 and CYP2C19 were detected in patients with CAD. The LDL-C subfractions levels of patients with CAD were followed up after statin drug treatment. Results: Total cholesterol, LDL-C, LDLC-2, LDLC-3, LDLC-4, LDLC-5, LDLC-6, LDLC-7, and sdLDL-C levels of patients with CAD were significantly higher than those in non-CAD controls. Meanwhile, sdLDL-C (AUC = 0.838) and LDLC-4 (AUC = 0.835) performed outstandingly in distinguishing patients with CAD from controls. Based on CYP2C19 metabolizer phenotypes, 113 patients with CAD were divided into the extensive metabolizer (EM, n = 49), intermediate metabolizer (IM, n = 52), and poor metabolizer (PM, n = 12) groups. The patients with IM and PM metabolizer phenotypes had better sdLDL-C lowering efficacy after taking statin drugs than patients with EM phenotype (P = 0.0268, FDR = 0.0536). The SLCO1B1 genotype had no significant impact on the efficacy of statins (P = 0.1611, FDR = 0.1611). Conclusion: sdLDL-C and LDLC-4 outperformed other blood lipids such as LDL-C for CAD risk screening. CYP2C19 metabolizer phenotypes had the potential to predict the efficacy of statins in lowering sdLDL-C.

Effects of Xuezhikang and pravastatin on circulating endothelial progenitor cells in patients with essential hypertension.

OBJECTIVE: To investigate the impacts of Xuezhikang (XZK) or pravastatin combined with antihypertensive drugs on circulating endothelial progenitor cells (CEPCs) in essential hypertensive (EH) patients. METHODS: Eighty-eight EH patients were enrolled into the study and randomly assigned to the antihypertensive drug treatment group (ATH group, 29 cases), the pravastatin treatment group (PRA group, 29 cases) and the Xuezhikang treatment group (XZK group, 30 cases). Patients in the 3 groups were treated with routine antihypertensive drugs. In addition, pravastatin and Xuezhikang were given to the patients in the PRA group and XZK group, respectively. After an eight-week treatment, CEPCs were counted using a laser scanning confocal microscope, and their proliferation function was evaluated by the MTT colorimetric assay and the adherent cell number was counted to estimate the adhesion function. RESULTS: After the treatment, CEPCs in the PRA group (116.60+/-5.70) and XZK group (114.40+/-6.55) was significantly higher than that in the ATH group (88.00+/-6.32, P<0.01). CEPCs proliferation capability and the adhesion function in the PRA group (0.406+/-0.016, 33.60+/-4.26) and XZK group (0.415+/-0.018, 34.30+/-3.77) were obviously superior to those in the ATH group (0.333+/-0.021, P<0.01; 23.30+/-3.19, P<0.01). No significant difference was found between the pravastatin group and the XZK group. CONCLUSIONS: Combined use of XZK or pravastatin with the anti-hypertensive therapy could increase the CEPCs number and improve their function in EH patients with the blood pressure controlled by antihypertensive drugs, leading to benefits independent of pressure-lowering effects.