The Correlation Between Laboratory Medicine Indexes And Coronary Heart Disease And Coronary Artery Lesion Counts In Qujing

Objective:To study The correlation between laboratory medicine indexes and coronary heart disease and coronary artery lesion counts in Qujing,establish disease prediction model,and explore whether laboratory medicine indexes can be used as an effective means of CHD prediction,so as to guide the clinical practice and achieve the health goals of early prevention,early detection and disease monitoring of CHD in qujingMethods:A retrospective study was conducted to select a total of 371 patients who were hospitalized for chest pain and underwent coronary angiography for the first time in the Cardiology department of qujing first people's hospital.General clinical data and relevant laboratory medicine indexes were collected:Age,gender,BMI,smoking history,drinking history,hypertension history,diabetes history,TC?TG?LDL-C?HDL-C?Non-HDL-C?APOA1/APOB?LP(a)?AIP?TBIL?DBIL?GGT?UA?CYS-C?ALP?BUN?WBC?NLR?PLR?LMR?APTT?PT?FIB,All continuous measurement data were tested by normal distribution test,and variables not subject to normal distribution model were transformed by logarithm.1.according to the result of coronary angiography,371 cases of the object of study can be divided into CHD group 253 examples and control group 118 examples,univariate analysis and binary Logistic analysis were used to compare the differences of the above indexes between the two groups,analyze the correlation between indexes and CHD,establish Logistic regression model and Fisher linear discriminant model,and draw ROC curve,evaluate the predictive value of relevant test indexes and statistical model to CHD.2.253 patients with CHD can be divided into single,two,three lesions,using single factor analysis and orderly Logistic comparing differences between the above indexes in three groups,using the Spearman correlation analysis the correlation of indexes and coronary lesion counts,establish the Logistic regression model and Fisher linear discriminant model,and draw ROC curve to evaluate the predictive value of relevant test indexes and statistical models for the number of coronary artery lesions and MVD.Results:1.In the CHD group and the control group,The general clinical data:age,gender,BMI,smoking history,hypertension and diabetes were all statistically significant,but no statistically significant difference in drinking history.Among the test indexes,LnTG,HDL-C,Non-HDL-C,LnLP(a),AIP,TBIL,DBIL,LnGGT,UA,CYS-C,WBC,LnLMR and FIB all showed statistically significant differences.There was no significant difference in TC,LDL-C,LnAPOA1/APOB,ALP,LnNLR,LnPLR,BUN,APTT and PT.Ln TG,Non-HDL-C,LnLPa,AIP,LnGGT,UA,CYS-C,WBC and FIB in CHD group were higher than those in control group,while HDL,TBIL,DBIL and LnLMR were lower than those in control group.Diabetes,Non-HDL-C,LP(a),AIP,CYS-C and FIB are independent risk factors for CHD,while TBIL is an independent protective factor for CHD.2.In different subgroups of coronary artery lesion counts,there were statistically significant differences in age and hypertension,but no statistically significant differences in gender,BMI,smoking history,drinking history and diabetes.Among the test indexes,the differences of DBIL,LnGGT,CYS-C,ALP,Ln NLR,LnPLR,LnLMR and FIB were statistically significant,while the differences of TC,LnTG,LDL-C,HDL-C,Non-HDL-C,LnAPOA1/APOB,LnLP(a),AIP,TBIL,UA,WBC,BUN,APTT and PT were not statistically significant.Age,GGT,CYS-C,ALP and FIB were independent risk factors for the increased number of coronary artery lesions.3.Drawn the ROC curve for CHD prediction.The area under the curve of non-hdl-c,LP(a),AIP,CYS-C and FIB was 0.711,0.569,0.784,0.763 and 0.710,respectively.The area under the curve of Logistic regression model and Fisher linear discriminant model was 0.862 and 0.861,respectively.The validation accuracy of the two models was 71.9%and 77.6%,respectively.4.Fisher's linear discriminant model had an accuracy of 49.8%in predicting the coronary artery lesions counts.The ROC curve for predicting MVD was drawn,and the area under the curve of GGT,CYS-C,ALP and FIB was 0.601,0.732,0.559 and 0.733,respectively.The area under the curve of Logistic regression model and Fisher linear discriminator model was 0.796 and 0.795,respectively,and the validation accuracy of both two models was 71.4%.Conculations:1.Non-HDL-C,LP(a),AIP,CYS-C and FIB are independent risk factors of CHD in Qujing.Non-HDL-C,AIP,CYS-C and FIB had good predictive value for CHD in Qujing,while the predictive value of LP(a)is limited;TBIL is independent protective factors of CHD in Qujing.2.GGT,CYS-C,ALP and FIB are independent risk factors of the coronary artery lesions counts in Qujing.Cys-c and FIB have good predictive value for MVD in Qujing,while GGT and ALP have limited predictive value for MVD in Qujing.3.WBC,NLR and PLR were not independent risk factors of CHD and coronary artery lesions counts in Qujing;LMR is not an independent protective factor of CHD and coronary artery lesions counts in Qujing 4 Compared with single medicine indexes,the combined predictive model has good predictive and discriminant ability for CHD and MVD in Qujing,but poor predictive and discriminant ability for the coronary artery lesions counts.5.part of the medicine indexes and multi-medicine indexes joint prediction model can be used as effective prediction method for CHD and MVD in qujing,which is of great significance for early prevention,early detection and disease monitoring of CHD in qujing;in clinical practice,a multi-medicine indexes joint prediction model should be established so that clinicians can better evaluate and predict the risk of CHD and MVD.