Prognostic value of N terminal pro B type natriuretic peptide combined with inflammatory factors in patients with heart failure with preserved ejection fraction

CHEN Li-fei; LI Rong; LIANG Hong-juan

doi:10.3969/j.issn.1672-5301.2023.12.008

Abstract：Objective To explore the prognostic value of N terminal pro B type natriuretic peptide(NT-proBNP)combined with inflammatory factors in patients with congestive heart failure with preserved ejection fraction(HFpEF).Methods A totle of 148 cases of HFpEF patients admitted to the Third People's Hospital of Nanning from March 2020 to April 2022 were selected as research objects.According to the occurrence of end point events,the patients were divided into two groups:the good prognosis group(n=82)and the poor prognosis group(n=66).The general data,medication,serum NT-proBNP and inflammatory factor levels between the two groups were compared;the factors affecting the prognosis of HFpEF patients were analyzed by multivariate analysis;an artificial neural network model to predict the prognosis of HFpEF patients was constructed and its predictive value was evaluated.ROC curve was used to analyze the predictive effect of NT-proBNP and inflammatory factors on poor prognosis of HFpEF patients.Results Atrial fibrillation,cardiac function grading,serum creatinine(Scr),troponin I(cTnI),NT-proBNP,hypersensitivity C reactive protein(hs-CRP)and red blood cell volume distribution width(RDW)were the risk factors for the poor prognosis in patients with HFpEF,but hemoglobin(Hb)was the protective factor for the poor prognosis(P<0.05).Artificial neural network model showed that the weight of cardiac function classification,NT-proBNP and hs-CRP was higher.ROC curve and cumulative gain graph showed that this model had a good predictive ability.ROC curve results showed that the AUC predicted by combined Scr,cTnI,NT-proBNP and hs-CRP was 0.821(95%CI 0.798-0.894),far higher than the AUC predicted by each index alone.Conclusion NT-proBNP combined with inflammatory factors has a higher predictive value for the prognosis of patients with HFpEF.The artificial neural network model constructed to predict the prognosis of the patients may provide reference value for the identification of clinical risk factors.