Establishment And Verification Of Coronary Heart Disease Prediction Model Based On Ultrasonic Myocardial Work Parameters

Background:Coronary atherosclerotic heart disease(coronary heart disease,CHD)is a major health problem that endangers human health and increases the social and economic burden.China Cardiovascular Disease report 2018 shows that the prevalence and mortality of cardiovascular disease are rising in China.In 2018,the number of cardiovascular patients is 290 million,and coronary heart disease has reached 11million^[1].Coronary heart disease has become a major social and public health problem.Early diagnosis and prevention of coronary heart disease Brooks no delay.How to diagnose CHD early is the focus and hotspot of current research by ultrasound doctors and cardiovascular doctors.At present,the preoperative examinations for coronary heart disease mainly include electrocardiogram,CTA,MRI and echocardiography.The degree of early coronary heart disease is mild,the clinical symptoms are not typical,the ECG is often lack of specificity,there are false positive and false negative,the clinical missed diagnosis rate of CHD patients with false negative ECG is high,the diagnosis is difficult,and it is easy to cause cardiovascular events and affect the prognosis,while unnecessary invasive examination is added to the false positive patients,which increases the economic and psychological burden of the patients.Coronary artery CTA is a non-invasive examination method with relatively simple operation,high time resolution,short examination time and low risk.this technique can reconstruct coronary artery images from multiple directions and angles to judge the location,degree of coronary artery stenosis and the nature of plaques,and has a high negative predictive value.However,it is easily affected by contrast medium allergy,obesity,coronary artery diffuse calcification,arrhythmia and the degree of cooperation of patients^[2].MRI has superior soft tissue contrast,and has certain clinical value in displaying three-dimensional structure of coronary artery,detecting high-risk plaque in coronary artery,observing myocardial perfusion,detecting myocardial scar and viable myocardium,etc.But at present,the spatial resolution of MRI is not enough to show the degree of coronary artery stenosis and blood flow changes,and the repeatability is poor,only when the degree of coronary artery stenosis over 85%,the abnormal myocardial perfusion can be detected^[3].Coronary angiography is the gold standard for the diagnosis of coronary heart disease,which can determine the location and degree of stenotic lesions,but it is invasive and expensive,and the acceptance degree of patients is relatively low.and can cause complications such as stroke,coronary artery dissecting aneurysm,puncture site bleeding,lower limb venous thrombosis and arrhythmia,so it is not suitable for routine screening of coronary heart disease.Conventional echocardiography mainly shows segmental motion abnormalities of ischemic and necrotic myocardium,which has high clinical value for patients with old myocardial infarction or severe myocardial ischemia,but the effect of early CHD diagnosis is often not ideal,and the rate of missed diagnosis is high.Three-dimensional speckle technique can be used to evaluate ventricular wall dysfunction and myocardial function caused by early myocardial ischemia in coronary heart disease,but the temporal and spatial resolution is low,it is easy to be affected by acoustic window conditions,and it is load-dependent.The increase of left ventricular afterload will lead to the decrease of left ventricular long axis strain(GLS),affecting the accurate evaluation of cardiac function^[4].Left ventricular myocardial work is a method to study the relationship between strain and afterload by combining strain index with non-invasive dynamic left ventricular pressure.This method studies the relationship between myocardial strain and afterload by global constructive work(GCW),myocardial work efficiency(GWE),global waste work(GWW)and global work index(GWI).This method shows great advantages in diagnosis of coronary heart disease(CHD),evaluation of cardiac resynchronization therapy and prediction of left ventricular myocardial fibrosis^[5-9].In precision medicine,accurate risk assessment is a prerequisite for risk screening and preventive treatment.Research and development of coronary heart disease risk prediction model is helpful to the early detection and treatment of coronary heart disease.At present,the clinical research on the prediction model of coronary heart disease based on the new ultrasound technology is very rare.Therefore,the establishment of coronary heart disease prediction model based on ultrasonic myocardial work parameters has more important clinical significance.Part ? The application value of ultrasonic myocardial work technique in early diagnosis of coronary heart diseaseObjective:To explore the application value of myocardial work technique in early diagnosis of coronary heart disease.Methods:Myocardial work and coronary angiography were performed in 245patients with suspected coronary heart disease who were hospitalized in the second affiliated Hospital of Nanchang University from January 2019 to April 2020.They were divided into coronary heart disease group and normal control group according to the results of coronary angiography.The differences of cardiac ultrasound parameters between the two groups were compared,the ROC curve was drawn and the optimal diagnostic boundary value,sensitivity and specificity of each parameter were calculated.Results:217 patients were included in the study,including coronary heart disease group(n=84)and normal control group(n=133).The values of GLS,GWI,GCW and GWE in the CHD group were lower than those in the normal control group,while the values of PSD and GWW in the CHD group were higher than those in the normal control group(P<0.05).Statistical analysis showed that PSD,GLS and GWW had diagnostic value for coronary heart disease,and the area under the curve was0.610,0.717 and 0.793,respectively.The sensitivity and specificity of GWW was higher than that of GLS and PSD,with 168mm Hg%as GWW cutoff value of 72.6%and 75.9%,respectively.Conclusions:Myocardial work parameter GWW has high sensitivity and specificity for the diagnosis of coronary heart disease,and is expected to become an important clinical tool for early detection of coronary heart disease.Part ? Establishment of Coronary Heart Disease Prediction Model based on Ultrasonic Myocardial work parametersObjective:To explore the risk factors of(CHD)in coronary heart disease,establish Logistic regression model and scoring system,and verify the predictive model and scoring system internally,so as to provide reference basis for the diagnosis and treatment of coronary heart disease.Methods:A total of 245 patients with suspected coronary heart disease who underwent coronary angiography in the second affiliated Hospital of Nanchang University in January 2019 and April 2020 were selected as the subjects.The general data,preoperative laboratory results and preoperative ultrasound data of all patients were collected.According to the results of coronary angiography,patients were divided into two groups:CHD group and control group.Univariate analysis was used to analyze the related risk factors of coronary heart disease and multivariate Logistic regression analysis was used to determine independent risk factors,and a binary Logistic regression model was established.According to the regression coefficientof each variable in the model,the corresponding score is calculated,and the scoring system is established.The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the calibration ability of the prediction model,and the receiver working characteristic curve(receiver operating characteristic,ROC)was used to evaluate the prediction ability of the prediction model and scoring system.Results:Results:1.Univariate analysis showed that sex,age,history of hypertension,history of diabetes,carotid plaque,HCY,TG,HDL,?-GT,GLS,PSD,GWI,GCW,GWW and GWE were risk factors for coronary heart disease.2.Multivariate analysis showed that grade 2 hypertension,HCY>11.575?mol/L,single and multiple carotid plaques,HDL<1.265mmol/L,GWW>168mm Hg%were independent risk factors for coronary heart disease.3.The established predictive model was Logit P=-3.333+1.123 X₁(hypertension grade 2)+1.245 X₂(HCY>11.575 umol/L)+1.805 X_3-1(single carotid plaque)+2.633 X_3-2(multiple carotid plaques)+1.278 X₄(HDL<1.265mmol/L)+1.711 X₅(GWW>168mm Hg%).The H-L test results show that the model has good calibration ability.The AUC of the model is 0.914,95%CI(0.878-0.950),P<0.001),and the prediction ability of the model is good.4.According to the regression coefficient?of each variable in the model,a scoring system was established,including hypertension grade 2 risk score 1,HCY>11.575?mol/L 1,single carotid plaque 1,multiple carotid plaque 2,HDL<1.265mmol/L1,GWW>168mm Hg%1,and the total score range was 0-6.The total score of coronary heart disease was divided into diagnostic variables,and the ROC curve of the scoring system was analyzed.The results showed that AUC=0.876,95%CI(0.831-0.921),indicating that the diagnostic value of the scoring system was good.The scoring system takes the total score of 3.5 as the best diagnostic threshold,?4 as the high-risk group of coronary heart disease,and<4 as the low-risk group of coronary heart disease.the diagnostic sensitivity,specificity and accuracy are 73.80%,84.96%and 80.64%,respectively.Conclusions:1.Sex,age,history of hypertension,history of diabetes,carotid plaque,HCY,TG,HDL,?-GT,GLS,PSD,GWI,GCW,GWW and GWE were risk factors for coronary heart disease.2.Grade 2 hypertension,HCY>11.575?mol/L,single and multiple carotid plaques,HDL<1.265mmol/L,GWW>168mm Hg%were independent risk factors for early coronary heart disease.3.The risk prediction model of coronary heart disease established in this study was Logit(P)=-3.333+1.123 X₁(hypertension grade 2)+1.245 X₂(HCY>11.575umol/L)+1.805X_3-1(single carotid plaque)+2.633X_3-2(multiple carotid plaques)+1.278 X₄(HDL<1.265mmol/L)+1.711 X₅(GWW>168mm Hg%).The scoring system takes the total score of 3.5 as the best diagnostic threshold,?4 as the high risk group of coronary heart disease,and<4 as the low risk group of coronary heart disease.the sensitivity,specificity and accuracy of the diagnosis of coronary heart disease are 73.80%,84.96%and 80.64%,respectively.4.The coronary heart disease risk prediction model and scoring system established in this study are relatively simple and applicable,have good calibration ability and prediction ability,and can provide a certain reference basis for the diagnosis and treatment of early CHD.Part ? Clinical Verification of Coronary Heart Disease Predictive Model and Scoring System based on Ultrasonic Myocardial work parametersObjective:The newly built coronary heart disease prediction model was applied to clinical practice to test the accuracy and practicability of the prediction model.Methods:A total of 108 patients with suspected coronary heart disease who underwent coronary angiography in the second affiliated Hospital of Nanchang University from May 2020 to July 2020 were selected as the object of study.The general data,preoperative laboratory results and preoperative ultrasound data of all patients were collected.The risk early warning model was used for external verification of 108 suspicious cases,and the risk of coronary heart disease in each case was calculated and compared with the final results of coronary angiography to test the sensitivity,specificity and accuracy of the risk early warning model.Results:Ninety six patients were included in the study,including 50 cases of coronary heart disease group and 46 cases of normal control group.The specific data of 96 clinical cases were substituted into the Logistic regression early warning model.The results of H-L test show that:?²=9.166,P=0.328>0.05,The calibration ability of the model is good.The predictive probability was used as a diagnostic variable to display the AUC=0.880,95%CI(0.809-0.950),P<0.001.The predictive ability of the model was good.The clinical case risk score was used as a diagnostic variable to show the predictive sensitivity of the coronary heart disease scoring system by 78%.The specificity is 93.48%,and the total correct rate is 85.41%.The AUC is 0.888.Conclusions:The prediction model of coronary heart disease based on ultrasonic myocardial work parameters is simple and practical,and has high prediction sensitivity,specificity and accuracy.It is believed that with the continuous improvement and improvement of the risk prediction model,it will provide a more useful reference for the early diagnosis of coronary heart disease.