Establishment And Verification Of The Prognosis Evaluation Model Of Elderly Stable Coronary Heart Disease With Integrated Traditional Chinese And Western Medicine

With China's gradual entry into an aging society,the popularity of secondary prevention of coronary heart disease and the continuous improvement of treatment effect the number of elderly patients with stable coronary heart disease are increasing.Scholars from all over the world pay more and more attention to the factors that affect the prognosis of patients with cardiovascular diseases,such as coronary heart disease,and establish a suitable local prognosis evaluation model.However,at present,there are few researches on the prognosis of elderly patients with stable coronary heart disease(ESCAD),and fewer researches on the influence of TCM factors on the prognosis of ESCAD patients.Traditional Chinese medicine believes that tongue image plays an important role in judging the prognosis of diseases.Tongue image is more closely related to the prognosis of patients with heart disease.So whether it can improve the accuracy of prognosis evaluation of ESCAD patients for clinicians by adding tongue image into the model of prognosis evaluation of ESCAD to establish the model of integrated of Chinese and Western medicine?This study focuses on the above issues and explores the following three parts.Part ? Literature studyLiterature study I:A new trend in the study of prognosis of coronary heart disease:a scientific econometric analysis based on CiteSpace softwareResearch purpose:In order to explore the research frontier and future development trend,the related literatures in the field of prognosis of coronary heart disease were collected to conduct bibliometrics and visual analysis.Research methods:Literatures about the prognosis of coronary heart disease were searched in WOS database,and the time span was from 1989 to 2018.R software was used to analyze the citing literatures.CiteSpace software was used to conduct co-cited analysis.The top 100 cited literatures were selected with a two-year interval.Spectral clustering algorithm was used to cluster the cited literatures and draw the related knowledge map.The method of pathfinder was used to prune the map.The clustering view and the timeline view were used to show the relationship and time span of each cluster.Term frequency*inverse document frequency(TF*IDF)weighting algorithm,log likelihood rate(LLR)algorithm and mutual information(MI)algorithm were used to calculate the clustering labels of each cluster respectively.Modularity and average silhouette were used to evaluate the structure and clarity of views.Cited frequency,betweenness centrality,citation burst and clustering mean year were used to discover research frontier and future study trendResults:1 General situation of citing literaturesThere were 43264 citing literatures about the prognosis of coronary heart disease.There were 107240 authors in this field,and and there were no Chinese scholars among the top 10 authors.The literature in this field covered 95 countries and regions.Whether it was published in a single country or in cooperation with multiple countries,the United States had the highest total number of publications,which was 12175.The second was Japan(3453),the third was Italy(3068),the fourth was Germany(2259),and the fifth was Britain(2259).China's total number of publications ranked the ninth with 1241,accounting for 1/10 of the total number published by the United States,accounting for 4%of the total number of publications in the top 10 countries and 2.9%of the total number of publications in this field.2 Analysis of the research frontier in the field of prognosis of coronary heart diseaseThere were 391568 cited literatures in the field.There were 862 nodes and 1260 lines in the cluster view and Timeline view.Its Modularity was 0.87 and the average silhouette was 0.40.There were 76 clusters,including 18 clusters with more than 10 nodes.The largest cluster was cluster 0#with 72 nodes.The most recent cluster at the time point was cluster 17#with 11 nodes.Among the top 10 literatures cited,3 were from cluster 6#,1 from cluster 3#,and cluster 0#,cluster 2#and cluster 8#have two articles respectively.The literatures of high betweenness centrality were mainly clustered in cluster 13#,cluster 3#and cluster 7#,which were respectively 3,3 and 4.There were 744 nodes with high burst,accounting for 86.3%of the total nodes.Three of the 10 articles with the highest burst were from cluster 6#,two from cluster 0#,one from cluster 3#and cluster 2#and cluster 8#had two articles respectively.It was concluded that cluster 0#,cluster 6#and cluster 8#could show the research frontier in the field of prognosis of coronary heart disease,and cluster 17#might be a new research trend in the future.Conclusion:1 There was a certain gap between China and Western countries in the field of prognosis of coronary heart disease,which needs to be further strengthened;2 The influence of biomarkers,such as high sensitive C-reactive protein,on the prognosis of coronary heart disease(cluster 0#),the exploration of prognostic factors of thrombotic events in patients with coronary heart disease after drug-eluting stent implantation(cluster 8#),and the influence of bleeding events on the prognosis of patients with acute coronary syndrome(cluster 6#)were the frontier issues in the field of prognosis of coronary heart disease;3 Paying attention to the secondary prevention of coronary heart disease(cluster 17#)might be a new research trend in the future.Part ? clinical researchClinical research ?:Relationship between tongue image and prognosis of elderly patients with stable coronary heart diseaseResearch purpose:To explore the relationship between tongue image and the incidence of cardiovascular events in ESCAD patients in one yearResearch methods:The data source was the prospective cohort study data collected by 5 hospitals funded by the national major basic research program(973),"systematic study on the etiology of cardiovascular thrombotic diseases' blood stasis and toxin",and the real world data collected by 18 hospitals funded by subproject from the major project of Beijing Science and technology plan "Research on the clinical individualized diagnosis and treatment evaluation system of Chinese medicine for the prevention and treatment of major diseases" and the project of Beijing Science and Technology Commission "data mining and validation research on the laws of integrated diagnosis and treatment of Chinese and Western medicine for coronary heart disease".ESCAD patients with age?60 years were screened.There were 55 candidate prognostic factors,which were divided into two parts.One part was 38 conventional prognostic factors,the other part was 17 prognostic factors related to tongue image.The predicted outcomes were cardiovascular events,including unexpected revascularization,hospitalization for unstable angina(without revascularization),nonfatal myocardial infarction,stroke,and cardiac death.Conventional prognostic factors were included in lasso-logistic regression model.The minimum penalty value(? min)and one time of standard error penalty value(?lse)were calculated by the cross validation method,and the corresponding prognostic factors were screened out,and three models were established.In model 1,the conventional prognostic factors were selected when the adjusted ? value was 1 times the standard error;In model 2,the conventional prognostic factors were selected when ? was adjusted to the minimum value;In model 3,we included the parameters of model 2 and the variables with p<0.05 in univariate analysis or the conventional prognostic factors with potential confounding effect recorded to the literature.The prognostic factors related to tongue image were added to the above three models one by one for analysis,and the interaction between tongue image and conventional prognostic factors was considered.Logistic regression was used to reveal the relationship between each tongue image and the prognosis of patients with ESCAD.In order to further analyze the interaction between tongue image and conventional prognostic factors,subgroup analysis was carried out.Results:Finally,3256 ESCAD patients were included,with an average age of 70 years and a maximum age of 89 years.52%of them were male patients.349 patients had cardiovascular events within one year of follow-up1 Relationship between tongue image and prognosis of ESCAD patientsThere was a significant correlation between pale tongue color and the risk of cardiovascular events in patients with ESCAD(unadjusted OR value was 2.29,95%CI(1.36-3.69),p<0.01;after model 1 adjustment,OR value was 1.95,95%CI(1.14-3.20),p=0.01;after model 2 adjustment,OR was 1.82,95%CI(1.06-3.01),p=0.02;after model 3 adjustment,OR was 1.82,95%CI(1.06-3.02),p=0.02;adjusted after considering interaction based on model 3,the OR value was 12.69,95%CI(2.3-75.64),p<0.01).There was a significant correlation between red tongue color and the risk of cardiovascular events in patients with ESCAD(unadjusted OR value was 1.48,95%CI(1.00-2.14),p=0.04;after model 1 adjustment,OR value was 1.67.95%CI(1.11-2.44),p=0.01;after model 2 adjustment,OR value was 1.66,95%CI(1.10-2.44),p=0.01;after model 3 adjustment,OR value was 1.67,95%CI(1.10-2.46),p=0.01;adjusted after considering interaction based on model 3,the OR value was 4.55,95%CI(1.16-17.44),p=0.03).There was a significant correlation between white tongue-coating and the risk reduction of cardiovascular events in patients with ESCAD(unadjusted OR value was 0.62.95%CI(0.47-0.80),p<0.01;after model 1 adjustment.OR value was 0.73,95%CI(0.55-0.97),p=0.03;after model 2 adjustment,OR value was 0.72,95%CI(0.53-0.96).p=0.03;after model 3 adjustment,OR value was 0.71,95%CI(0.53-0.95).P=0.03;adjusted after considering interaction based on model 3,the OR value was 0.12,95%CI(0.04-0.31),P<0.01)2 Adjusted interaction and subgroup analysisAfter the adjustment of model 3,the interaction was more obvious:light dark tongue and grade 3 hypertension(p=0.03),dark red tongue and atrioventricular block(p=0.04),tongue ecchymosis and atrioventricular block(p=0.03),tooth mark tongue and clopidogrel(p=0.01),crack tongue and clopidogrel(p=0.04),yellow tongue-coating and aspirin(p=0.04),yellow tongue-coating and CCB(p=0.01).greasy tongue-coating and aspirin(p=0.02),thick tongue-coating and aspirin(p=0.02)In ESCAD patients with grade 3 hypertension,there was a significant correlation between light dark tongue color and the risk of cardiovascular events(OR:1.95,95%CI:1.21-3.06,p=0.005).In ESCAD patients with atrioventricular block,dark red tongue color was significantly associated with increased risk of cardiovascular events(OR:229,95%CI:12-14536,p=0.002).In ESCAD patients taking clopidogrel orally,there was a significant correlation between dentate tongue and the risk of cardiovascular events(OR:2.76,95%CI:1.58-4.75,p<0.001).In ESCAD patients taking CCB orally,there was a significant correlation between decreased risk of cardiovascular events and yellow tongue-coating(OR:0.44,95%CI:0.26-0.72,p=0.001)Conclusion:1 Compared with ESCAD patients without pale white and red tongue color,ESCAD patients with pale white and red tongue color had higher risk of cardiovascular events;compared with non white tongue-coating ESCAD patients,white tongue-coating ESCAD patients had a lower risk of cardiovascular events2 No significant relationship was found between purple tongue,ecchymosis,fat tongue,cracked tongue,thick coating,greasy coating,slippery coating,dry coating,exfoliated tongue-coating,no coating and the risk of cardiovascular events in ESCAD patients3 Dark tongue and grade 3 hypertension,dark red tongue and atrioventricular block,tongue ecchymosis and atrioventricular block,tooth mark tongue and clopidogrel,cracked tongue and clopidogrel,yellow tongue-coating and aspirin,yellow tongue-coating and CCB,greasy tongue-coating and aspirin,thick tongue-coating and aspirin had potential interactions in the process of cardiovascular events in patients with ESCAD4 Subgroup analysis:In the patients with ESCAD of grade 3 hypertension,those with light dark tongue color had higher risk of cardiovascular events;in ESCAD patients with atrioventricular block,those with dark red tongue had a higher risk of cardiovascular events;among the patients taking clopidogrel for ESCAD,those with tooth marked tongue had a higher risk of cardiovascular events;among the ESCAD patients taking CCB preparation,those with yellow tongue-coating had a lower risk of cardiovascular events.However,the sample size of cardiovascular events in the subgroup is small,and its clinical significance needs further studyClinical research ?:Establishment and Verification of the Prognosis Evaluation Model of Elderly Stable Coronary Heart Disease with Integrated Traditional Chinese and Western MedicineResearch purpose:Based on the data of previous studies,establish and validate the prognosis evaluation model of integrated traditional Chinese and western medicine in ESCAD;explore the influence of tongue image on the prognosis evaluation model of integrated traditional Chinese and western medicine in ESCAD;develop a ESCAD risk calculatorResearch methods:The data set in Study 2 was divided into training set and verification set according to 8:2 The routine prognostic factors,routine prognostic factors+tongue image and routine prognostic factors+tongue image(considering interaction factors)in the training set were respectively included in lasso-logistic regression model for screening,and the optimal penalty value(?)was calculated by cross validation.Three models were compared from the following five aspects:AIC,c-statistic,calibration,NRI,IDI and DCAResults:The c-statistic of model 1 was 0.68(0.65-0.72)in the training set and 0.67(0.60-0.73)in the verification set.The c-statistic of model 2 was 0.69(0.66-0.73)in the training set and 0.67(0.61-0.73)in the verification set.The c-statistic of model 3 was 0.70(0.67-0.74)in the training set and 0.67(95%CI:0.61-0.74)in the verification set.Compared with model 1,NRI and IDI of model 2 increased by 1.91%and 0.54%(p=0.04);compared with model 1,NRI and IDI of model 3 increased by 3.58%and 1.28%(p<0.01);compared with model 2,NRI and IDI of model 3 increased by 3.58%and 0.74%(p=0.02)Model 1(p=0.683/p=0.276),model 2(p=0.620/p=0.263)and model 3(p=0.635/p=0.238)had good calibration in training set and verification setWhen the cut-off value of the prediction probability is 10%,in the training set,the net benefit of model 1 is 2.7%,model 2 is 3.2%,and model 3 is 3.3%;in the validation set,the net benefit of model 1 is 3.6%,model 2 is 3.7%,and model 3 is 4.0%According to the model 3,ESC AD risk calculator(Android mobile app)was madeConclusion:1 For the first time,the prognosis evaluation model of elderly stable coronary heart disease with integrated traditional Chinese and western medicine for ESCAD patients was established,and the risk calculator of ESCAD was developed;2 Adding the tongue image to the model could further simplify the model,reduce the over fitting phenomenon of the model,and improve the ability of correct classification and comprehensive discrimination.Part ? Development of cloud data platformEstablishment of a practice-research-integration cloud data platform for cardiovascular disease treated with integrative medicineResearch purpose:In order to further study the prognosis of cardiovascular disease,initially construct a practice-research-integration cloud data platform for cardiovascular disease treated with integrative medicine.Function introduction:The initial target of the platform was clinicians with clinical research needs.The user layer includes patients,clinicians,physician teams and quality control personnel.The terminal system had Web and mobile terminals,and mobile terminals include IOS version,Android version and WeChat official account to meet the needs of various scenarios.In addition,considering the age of the patients,the patient's terminal was only WeChat official accoun,and it only kept the function of communication and submission of questionnaires.The platform was a cardiovascular specialized platform,with a medical record information collection system and a doctor-patient communication system.It had made a lot of exploration and improvement in structural,convenient,practical and quality monitoring.The main purpose of the early stage of the platform was to collect scientific research information.In the later stage,internet phone and internet video would be set up and online payment process will be improved,so that doctors and patients could communicate with each other at any time with compensation,forming a long-term mechanism for doctors to benefit and patients to benefit.We would improve the doctor communication system,and increase the modules of case discussion,remote consultation,scientific research and clinical training,and technology promotion,so as to improve the clinical practicability and scientific research application value.