Construction of a Nomogram prediction model for prognosis of chronic heart failure patients based on peripheral blood miR-150 and IGF-3 levels

Cui Rong; Zhang Shu; An Xiaofang

doi:10.3969/j.issn.1674-4055.2025.04.10

Abstract：Objective To construct a Nomogram prediction model for prognosis of chronic heart failure(CHF)patients based on peripheral blood microRNA-150(miR-150)and insulin-like growth factor-3(IGF-3)levels.Methods Selected 182 cases of CHF patients admitted to the cardiovascular department of a hospital in Heilongjiang Province from January 2017 to January 2019.Follow-up after discharge,with cardiac death or the end of June 2023 as the endpoint of follow-up,those who occurred and did not occur major adverse cardiovascular events(MACE)were respectively included in the occurrence group(n=62 cases)and the non-occurrence group(n=98 cases).The general data and peripheral blood miR-150 and IGF-3 levels of the two groups were compared.The cox proportional hazard regression model was used to analyze the influencing factors of CHF prognosis.Nomogram prediction model was constructed based on the influencing factors of CHF prognosis,and its efficacy was verified.Results The MACE incidence rate of CHF patients was 38.75%.Compared with the non-occurrence group,the age,anemia,lesion vessel number≥3 percentage in the occurrence group were higher,serum Hb,LVEF level and regular outpatient follow-up,regular medication outside hospital percentage were lower(P＜0.05).Compared with the non-occurrence group,the peripheral blood miR-150 and IGF-3 levels in the occurrence group were lower(P＜0.05).Cox proportional hazard regression model showed that advanced age,coexisting anemia,and the involvement of≥3 vessels are risk factors for the prognosis of chronic heart failure(CHF),(P＜0.05),while a high left ventricular ejection fraction(LVEF),regular outpatient medication use,high levels of miR-150,and high levels of IGF-3 are protective factors for the prognosis of CHF(P＜0.05).Using the aforementioned factors as predictive variables to establish a Nomogram prognostic model,the total score ranges from 0 to 500 points,corresponding to the probability of poor prognosis in CHF ranging from 0.05 to 0.95.Hosmer-Lemeshow test showed that there was no significant difference between the prediction probability of the Nomogram model and the correction curve of the actual probability(P=0.475).ROC curve showed that Nomogram prediction model predicted CHF poor prognosis with a sensitivity of 91.53%(95%CI:87.7%～94.4%),specificity of 81.90%(95%CI:73.2%～88.7%),AUC of 0.916(95%CI:0.884～0.941),and the prediction efficacy of the model was good.Conclusion High peripheral blood miR-150 and IGF-3 levels are protective factors for prognosis of CHF patients,and the Nomogram prediction model constructed based on them,age,anemia and number of diseased vessels has good prediction efficacy.