Study Related To Short-term Prognostic Risk Assessment Of Acute Decompensated Heart Failure

Acute decompensated heart failure(ADHF)is one of the most prevalent critical illnesses worldwide with high morbidity,readmission,and mortality rate,and the poor prognosis usually occur inhospital or in the short term after discharge.Appropriate treatment strategies for ADHF varies depending on the patient’s specific ethnic,cultural,and economic background.In order to improve the treatment strategy and prognosis of patients with ADHF,it is necessary to predict the short-term prognosis.Although some risk scoring models for assessing patients with heart failure(HF)have emerged in recent years,it is unclear whether these scores can be directly applied to patients with ADHF,especially for the Chinese patients.In this study,we propose to externally validate previously published risk scores in patients with ADHF and to construct a new risk score system specifically to evaluate their predictive value for short-term prognosis of Chinese patients with ADHF according to the corresponding characteristics.At the same time,this study specifically investigated the early diagnostic value of some biomarkers for Cardiorenal syndrome type 1(CRS-1)in ADHF patients,and evaluated the predictive value of these biomarkers and CRS-1 for short-term prognosis of ADHF patients,with the aim of providing an early assessment of the short-term prognosis for ADHF patients and taking appropriate preventive and curative measures to help improve the short-term prognosis of ADHF patients.Part One External validation of the 3A3B,AHEAD,and OPTIME-CHF risk scores for the short-term prognosis of patients with acute decompensated heart failureObjective:At present,the poor prognosis of patients with ADHF is still one of the serious challenges faced by medical workers.Studies have shown that most adverse outcomes occur within a eraly period,and an effective prognostic scoring system can improve this situation.In this study,we intend to conduct external validation of three previously published risk scores(3A3B,AHEAD and OPTIME-CHF)in the Chinese ADHF population,and observe their predictive efficacy for short-term prognosis of Chinese ADHF patients.Methods:The subjects of this study were all obtained from the Heb-ADHF registry research database.From March 2016 to December 2018,inpatients with ADHF were recruited from the department of Cardiology of 13 tertiary hospitals hospitals in 9 sub-districts of Hebei Province.The primary end point of follow-up was defined as the composite outcome of in-hospital death,30-day all-cause readmission after discharge,or 30-day all-cause death after discharge.The risk scores to be validated were the 3A3B score based on the Japanese heart failure(HF)with preserved ejection fraction(HFp EF)patients with a median follow-up of 5.7 years,the AHEAD score based on the European acute HF population with a follow-up of 1-5 years,and the OPTIME-CHF score based on the US decompensated HF population with a follow-up of 6 months.The area under the receiver operating characteristic curves(AUROCs)with 95%confidence intervals(95%CIs)were used to evaluate the predictive value of the 3A3B,AHEAD,and OPTIME-CHF score for the endpoint.Z-test was used to compare AUROCs differences among them,and the sensitivity and specificity of each score were calculated.Results:A total of 4,550 patients with ADHF were eventually included in this study.The results showed that the predictive discrimination of 3A3B,AHEAD,and OPTIME-CHF score was all poor,with AUROCs of 0.55(95%CI 0.53-0.57),0.54(95%CI 0.53-0.56),and 0.56(95%CI 0.54-0.57),respectively.Only the specificity of 3A3B score was acceptable(83.5%),but the sensitivity was not satisfactory(22.0%).We also used 3A3B score,AHEAD score,and OPTIME-CHF score to predict in-hospital all-cause death,30-day all-cause readmission after discharge,or 30-day post-discharge all-cause death individually.The results showed that the independent predictive power of the three outcomes was also poor,with the measured AUROC failing to exceed 0.60.Summary:Using data from 4,550 inpatients with ADHF in the Heb-ADHF registry,this study externally validated 3 previously published HF risk prediction scores,3A3B,AHEAD,and OPTIME-CHF.All the 3 scores were found to be poor discrimination for the composite outcome of all-cause death in hospital,30-day all-cause readmission,or 30-day all-cause death after discharge.Therefore,a risk scoring system for short-term prognosis of ADHF in China population needs to be developed.Part two Construction of a new short-term prognostic risk score for patients with acute decompensated heart failure Objective:Accurate prediction of the prognosis of ADHF patients is very important for the treatment and the improvement of treatment decisions during and after hospitalization.However,there are few prognostic risk scoring systems for patients with ADHF.Therefore,this study intends to create a new risk scoring model to investigate its predictive effect on the composite end points of all-cause death in hospital,30-day all-cause readmission,or 30-day all-cause death after discharge in hospitalized ADHF patients in China.Methods:A total of 4,550 eligible patients with ADHF enrolled from the Heb-ADHF were randomly divided into the derivation cohort and the validation cohort at a ratio of 3:2.The primary endpoint was defined as a composite outcome of all-cause death in hospital,30-day all-cause readmission,or 30-day all-cause death after discharge.Firstly,univariate and multivariate logistic regression were used to screen risk predictors and to create the new risk scoring model.Secondly,AUROCs with 95%CIs were used to evaluate the predictive discrimination ability of the new risk score,and the Hosmer-Lemeshow goodness-of-fit test was used to evaluate the calibration of the model.Finally,the Mantel-Haenszel trend test and Pearson correlation test were used to analyze the correlation between the stratification of the risk score and outcomes.Results:The results showed that 2,745 patients were enrolled into the derivation cohort and 1,805 patients were enrolled into the validation cohort.Eight variables(P<0.05)maintained independent correlation with the composite endpoints of all-cause death in hospital,30-day all-cause readmission,or 30-day all-cause death after discharge by logistic regression analysis:diastolic blood pressure(DBP),lymphocyte count,serum creatinine(s Cr),blood urea nitrogen(BUN),brain natriuretic peptide(amino-terminal brain natriuretic peptide precursor)[BNP(NT-pro BNP)],QRS fraction of electrocardiogram,angiotensin converting enzyme inhibitor/angiotensin receptor antagonist enzyme inhibitor/angiotensin receptor blocker(ACEI/ARB)and recombinant human brain natriuretic peptide(rh BNP),which were finally incorporated into the new risk model.The formula for the new risk score(points)was 1×(DBP<80 mm Hg)+2×(lymphocyte>1.11×10~9/L)+1×(s Cr>80μmol/L)+2×(BUN>21 mg/d L)+2×[BNP 500 to<1,500 pg/m L(NT-pro BNP 2,500 to<7,500pg/m L)]or 3×[BNP≥1,500(NT-pro BNP≥7,500)pg/m L]+3×(QRS fraction<55%)+4×(ACEI/ARB not used)+1×(rh BNP used).The AUROC value for the new risk score was 0.67(95%CI 0.64-0.70),with good calibration(Hosmer-Lemeshow testχ~2=3.366,P=0.186).The results of the validation cohort validated the above findings.In the whole cohort,a linear relationship was showed between risk stratification and the incidence of prognostic events(Mantel-haenszel testχ~2=38.14,P<0.001).The Pearson correlation test showed that the incidence of outcome events increased with increasing risk stratification(r=0.480,P<0.001).Summary:A new risk score was created to predict the composite endpoint of all-cause death in hospital,30-day all-cause readmission,or 30-day all-cause death after discharge for ADHF patients in China from the Heb-ADHF registry.The formula was 1×(DBP<80 mm Hg)+2×(lymphocyte>1.11×10~9/L)+1×(s Cr>80μmol/L)+2×(BUN>21 mg/d L)+2×[BNP 500 to<1,500 pg/m L(NT-pro BNP)2,500 to<7,500 pg/m L]or 3×[BNP≥1,500(NT-pro BNP≥7,500)pg/m L]+3×(QRS Fraction<55%)+4×(ACEI/ARB not used)+1×(rh BNP).The new scoring system has moderate predictive value for short-term prognosis of ADHF patients,and the model fits well.Special attention should be paid when the risk score is of≥9 points,and clinical decisions should be made and adjusted more aggressively for high-risk stratified patients.However,the predictive accuracy of the new risk score needs to be improved,and more new variables and larger multicenter studies are need to make it as a practical and reliable tool for assessing short-term adverse events in ADHF patients.Part three The early prediction of urine NT-pro BNP and plasma proenkephalin for cardiorenal syndrome type 1 and vulnerable-phase prognosis in patients with acute decompensated heart failureObjective:The incidence of CRS-1 in patients with ADHF is high,which affects the prognosis.The current diagnostic criteria can cause a delay in the diagnosis of CRS-1,and the problem of its early diagnosis needs to be addressed urgently.Poor prognosis in patients with HF usually occurs within 90 days after discharge,which is known as the"vulnerability period"for HF.Using real-world data,this study sought to assess the predictive value of plasma proenkephalin(p PENK)and urine NT-pro BNP(u NT-pro BNP)for early diagnosis of CRS-1,and to explore the prognostic value of p PENK,u NT-pro BNP,and CRS-1 for vulnerable-phase in patients with ADHF.Methods:A total of 121 hospitalized patients with ADHF were included(31 with CRS-1 and 90 without CRS-1).The plasma NT-pro BNP(p NT-pro BNP),p PENK,and u NT-pro BNP were measured on admission.The plasma neutrophil gelatinase-associated lipocalin(p NGAL)was taken as the reference biomarker.To correct the possible variability in urine concentrations,the u NT-pro BNP/urine creatinine(u Cr)ratio(pg/nmol)was calculated in each sample.The follow-up period was 90 days after discharge.The logistic regression analysis was used to determine the predictors of CRS-1,the AUROC was calculated to assess the early diagnostic value of p NGAL,p PENK,and u NT-pro BNP/u Cr for CRS-1.The Cox regression analysis was carried out to evaluate the prognostic risk of factors for the 90-day clinical outcomes of all ADHF patients and the cumulative risk curve was plotted.Results:We found that p PENK[OR=8.503(95%CI,1.683-42.967),P=0.010;AUROC=0.899(95%CI 0.831～0.946)]and u NT-pro BNP/u Cr ratio[OR=5.153(95%CI 1.012-26.232),P=0.048;AUROC=0.934(95%CI 0.874～0.971)]but not p BNP,p NT-pro BNP,or p NGAL could independently predict the CRS-1occurrence in hospitalized patients with ADHF.The p PENK[HR=1.014(95%CI 1.000～1.028),P=0.044]and u NT-pro BNP/u Cr[HR=0.998(95%CI0.997～1.000),P=0.045]were also independent predictors of the risk of HF readmission or all-cause death 90 days after discharge in ADHF patients.Summary:The p PENK and non-invasive test of u NT-pro BNP/u Cr ratio(pg/nmol)on admission may be two promising novel predictive biomarkers for early diagnosis of CRS-1 and vulnerable-phase outcomes in ADHF patients,which may be useful tools for clinical treatment and disease monitoring in patients with ADHF.Conclusion:1.The previously published 3A3B,AHEAD,and OPTIME-CHF score were were found to be poor discrimination for the composite outcome of all-cause mortality in hospital,30-day all-cause readmission,or 30-day all-cause death after discharge.2.The new risk score has moderate predictive value on predicting the composite endpoint of all-cause mortality in hospital,30-day all-cause readmission,or 30-day all-cause mortality after discharge for ADHF patients.The formula was 1×(DBP<80 mm Hg)+2×(lymphocyte>1.11×10~9/L)+1×(s Cr>80μmol/L)+2×(BUN>21 mg/d L)+2×[BNP 500 to<1,500 pg/m L(NT-pro BNP)2,500 to<7,500 pg/m L]or 3×[BNP≥1,500(NT-pro BNP≥7,500)pg/m L]+3×(QRS Fraction<55%)+4×(ACEI/ARB not used)+1×(rh BNP).3.The p PENK and non-invasive test of u NT-pro BNP/u Cr ratio(pg/nmol)on admission may be two promising novel predictive biomarkers for early diagnosis of CRS-1 and vulnerable-phase outcomes in ADHF patients.