The Study Of Clinical Prediction Model Of Chinese And Western Integrative Medicine For Coronary Heart Disease Complicated With Chronic Heart Failure

ObjectiveThe combination of traditional Chinese and western medicine plays a very important role in the treatment of coronary heart disease combined with heart failure(HF).In order to better improve the prognosis of patients with coronary heart disease(CHD),prevent,identify and treat its complications of HF as soon as possible,it is very necessary to establish the clinical prediction model of CHD complicated with HF with integrated traditional Chinese and western medicine,which is helpful for doctors to better carry out the clinical medical activities of integrated traditional Chinese and western medicine.At the same time of establishing the model,the traditional Chinses medicine(TCM)etiology and pathogenesis of CHD combined with HF were summarized,and then applied to the prescription drugs for the disease.MethodsThe medical record database of Shenzhen Hospital of Traditional Chinese Medicine from January 2010 to June 2020 was retrieved,and the medical record data of patients diagnosed with CHD were selected according to inclusion and exclusion criteria.The general clinical data,clinical examination and test results of the patients were extracted,and the source of model training set was established.Group case-control study design was applied.R software was used for statistical analysis.Lasso regression analysis was used to screen the variables used to construct the model.Logistic regression analysis method was used to establish the clinical prediction model,and the nomogram was used to display it.The C index of the model is calculated and the correction curve is plotted to evaluate the prediction efficiency of the model.The clinical practicability of the model was evaluated by calculating the net benefit ratio and drawing the decision curve.Internal validation,random sampling in the data set,and the effectiveness of the model verification.ResultsA total of 610 patients with CHD were included in this study,among which 254 patients with chronic HF were included in the case group.356 patients without chronic HF were included in the control group.Medical records were collected and 25 variables were recorded.Through Lasso regression analysis,the λ value of the mean value of the minimum objective parameter is 0.005616864,and the λ value of the simplest model is 0.0248862.Select 16 variables to build the model,respectively is:"CHD progression","cause","antiplatelet agents","heart rate control medication","tongue","pulse" and "ECG","pathological changes of coronary","coronary artery stenosis degree","history of myocardial infarction(AMI)","hyperlipidemia","diabetes","valvulopathy","arrhythmia","chronic lung disease","chronic kidney disease".After the establishment of the model,the prediction of "antiplatelet agents","hyperlipidemia" and "valvulopathy" were not in line with the clinical reality,so the three variables were eliminated and finally included into the remaining 13 variables to establish the model.The calculated C index of the model is 0.942(95%CI:0.9224-0.9616),which indicates that the prediction efficiency of the model is very good.The correction curve indicates that the prediction effect of this model is good.Clinical decision curve analysis shows that the probability of>is in the range of 1%and<85%,and using this model to predict the risk of complicated chronic HF has a good benefit.Through the Bootstrapping method sampling verification,the calculated C index is 0.916,still greater than 0.9,which proves that the prediction efficiency of this model is very good.ConclusionCHD patients with "CHD progression","cause","heart rate control medication","tongue","pulse" and "ECG","pathological changes of coronary","coronary artery stenosis degree","history of myocardial infarction","diabetes","arrhythmia","chronic lung disease","chronic kidney disease" as the risk factors were included in the construction of a clinical prediction model.The clinical prediction model of integrated traditional Chinese and western medicine for CHD complicated with chronic HF established in this study has been tested and proved to be effective.The model suggests that patients with deficiency of Qi and blood,deficiency of Yang,blood stasis,and dampness of phlegm are at significantly higher risk of chronic HF.TCM intervention can be carried out in advance for patients with high risk of CHD to prevent the occurrence of chronic HF.