Nomogram model for prognostic prediction in elderly CHF patients based on echocardiography and serology
10.3969/j.issn.1009-0126.2024.09.004
- VernacularTitle:基于超声心动图参数和血清学指标构建预测老年慢性心力衰竭患者预后的风险列线图模型
- Author:
Zhaomei DENG
1
;
Zhifa DENG
;
Lian XU
Author Information
1. 550001 贵阳,贵州中医药大学第一附属医院超声医学科
- Keywords:
heart failure;
echocardiography;
nomograms;
prognosis
- From:
Chinese Journal of Geriatric Heart Brain and Vessel Diseases
2024;26(9):997-1001
- CountryChina
- Language:Chinese
-
Abstract:
Objective To construct and validate a nomogram prognostic model for elderly patients with chronic heart failure(CHF)based on echocardiographic parameters and serological indica-tors.Methods A total of 110 elderly CHF patients admitted to the First Affiliated Hospital of Guizhou University of Chinese Medicine from January 2021 to January 2023 were recruited and randomly divided into a modeling group(80 cases)and a validation group(30 cases).According to the prognosis,the patients in the modeling group were divided into good prognosis subgroup(57 cases)and poor prognosis subgroup(23 cases).Echocardiographic parameters and serological indi-cators were collected to analyze the factors affecting poor prognosis in elderly CHF patients,and then a risk-prediction nomogram model was drawn and verified.Results The patients of poor prognosis had significantly higher values of left ventricular end-diastolic diameter(LVEDD)and left ventricular end systolic diameter(LVESD)and serum levels of soluble growth stimulation ex-pressed gene 2(sST2),N-terminal pro-B-type natriuretic peptide(NT-proBNP)and cardiac tro-ponin T(cTnT),and lower left ventricular ejection fraction(LVEF)value than the good progno-sis group(P＜0.01).Multivariate logistic regression analysis showed that LVEDD,LVESD,LVEF,sST2,NT-proBNP and cTnT were all factors influencing poor prognosis in elderly CHF patients(P＜0.01).The AUC value,sensitivity and specificity of the constructed risk nomogram model based on the above influencing factors were 0.788,83.61％and 80.59％,respectively.Hosmer-Lemeshow test reveal x2=4.115,P=0.894.The results of decision curve analysis(DCA)dis-played that the benefit was higher when the threshold probability ranged from 17％-96％,and the model could select larger range of the threshold probability.Conclusion LVEDD,LVESD,LVEF,sST2,NT-proBNP and cTnT are all associated with poor prognosis in elderly CHF pa-tients.Our risk nomogram model for poor prognosis in elderly CHF patients based on the above factors has good accuracy and discriminability.
