Research on Constructing a Prognostic Prediction Model for Chronic Cardiac Insufficiency Based on Hematology and Echocardiography Examination Indicators

LIANG Guisheng; DENG Maobing; PAN Yurong; LU Mo; ZHANG Hui; LI Ming; WANG Jun

doi:10.14033/j.cnki.cfmr.2026.05.001

Abstract：Objective:A prognostic prediction model for chronic heart failure based on hematological and echocardiographic examination indicators was constructed,and the model was verified to analyze its predictive value.Method:A total of 100 patients with chronic cardiac insufficiency admitted to Shiyan People's Hospital of Bao'an District,Shenzhen City from August 2023 to August 2024 were selected as the research subjects.The patients were randomly divided into the modeling group(n=80)and the validation group(n=20).The modeling group was divided into the good prognosis group(NYHA improved by≥grade 1 after treatment and remained stable for more than 3 months)and the poor prognosis group(NYHA showed no improvement 6 months after treatment)based on the prognosis of the patients.The general information,echocardiographic examination parameters and hematological examination indicators of all patients in the modeling group were collected.The differences between the patients in the good prognosis group and the poor prognosis group were compared.logistic regression equation was used to conduct multivariate analysis of the influencing factors affecting the prognosis of patients with chronic heart failure,and a risk nomogram model was constructed.Result:There was no statistically significant difference between the two groups in terms of gender,age,disease duration,smoking history,drinking history,coronary heart disease history,hypertension history,diabetes history and hyperlipidemia history(P＞0.05).There was no statistically significant difference in cardiac output,stroke volume and cardiac index between the two groups(P＞0.05).The left ventricular end-diastolic and end-systolic diameters in the good prognosis group were both smaller than those in the poor prognosis group,and the left ventricular ejection fraction was higher than that in the poor prognosis group.The differences were statistically significant(P＜0.05).There was no statistically significant difference between the two groups in terms of fasting blood glucose,hemoglobin,total cholesterol,triglycerides,uric acid and serum creatinine(P＞0.05).The levels of troponin and N-terminal pro-B-type natriuretic peptide in the good prognosis group were both lower than those in the poor prognosis group,and the difference was statistically significant(P＜0.05).logistic regression equation calculation revealed that enlarged left ventricular end-diastolic diameter,enlarged left ventricular end-systolic diameter,decreased left ventricular ejection fraction,elevated troponin,and elevated N-terminal pro-B-type natriuretic peptide are all risk factors for poor prognosis in patients with chronic heart failure.After constructing a nomogram prediction model based on the above risk factors and conducting ROC curve analysis,it was found that the AUC,sensitivity and specificity of the model were 0.778,83.26％and 80.41％,respectively.Decision curve analysis(DCA)shows that the benefits are higher when the threshold probability is 16％to 95％.Conclusion:Enlarged left ventricular end-diastolic diameter,enlarged left ventricular end-systolic diameter,decreased left ventricular ejection fraction,elevated troponin,and elevated N-terminal pro-B-type natriuretic peptide are all risk factors affecting the poor prognosis of patients with chronic heart failure.Constructing a predictive model based on the above factors has good predictive efficacy.