Font Size: a A A

Utility Of Inflammatory Biomarkers To Predict The Length Of Stay For Acute Decompensated Heart Failure Patients

Posted on:2021-03-27Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y WangFull Text:PDF
GTID:1364330614967808Subject:Clinical medicine
Abstract/Summary:PDF Full Text Request
Backgrounds:Heart Failure(HF)is becoming the most critical battlefield in cardiovascular diseases nowadays.Acute Decompensated Heart Failure(ADHF)is a kind of urgent state for heart failure patients,which could be the main reason to result in inpatient hospitalizations.By studying the relationship between inflammatory biomarkers and ADHF patients' Length of Stay(LOS),it is expected that LOS of ADHF patients can be distinguished by the first complete blood cell test of admission.After receiving the target of prolonged LOS of ADHF patients,stricter and more comprehensive heart failure management measures should be incorporated into the care system,so as to accelerate the time-period of recovery,to reduce the LOS,and to improve the utility of social public medical resourcesAims:To explore the relationship of Neutrophil-to-Lymphocyte Ratio(NLR),Monocyte-to-Lymphocyte Ratio(MLR),Platelet-to-Lymphocyte Ratio(PLR)and the LOS of ADHF patients,and to determine the predictive ability of these three indicators;to explore the relationship of clinical baseline variables and the LOS of ADHF patients;to explore the relationship between NLR,MLR,PLR,baseline clinical variables and the in-hospital death of ADHF patientsMaterials and Methods:This study was a single-center retrospective study.Patients with a diagnosis of ADHF from 2009 to 2019 in our heart failure database were recruited in our study,the patients' data including general information,causes of heart failure,medical history data,pre-hospital medication history,blood routine examination,cardiac function indicators,etc.The main structure of this study was divided into five parts:1)survival patients were divided into the Short LOS group and the Long LOS group based on the median of LOS,analyzed their correlation and determined the significant indicators cut-off values of LOS;2)subgroup's analysis in different Ejection Fraction(EF)groups and different causes of heart failure groups;3)comparing heart failure risk models predictive power to NLR;4)follow-up analysis of patients in the survival group to determine the predictive effect of main indicators on LOS;5)analysis of patients in the hospital survival group and the hospital death group.Non-parametric Mann-Whitney U test was used to compare continuous variables,Chi-square test was used to compare categorical variables,Pearson correlation was used to test correlation,Logistic regression was used to test causality,and the Youden index of the Receiver Operating Characteristic Curve(ROC)was used to determine the cut-off value,and the Area Under Curve(AUC)of ROC was used to compare the prediction capabilitiesResults:A total of 8616 ADHF patients(60.7%male,73(61,80)years)were included in this study.There were 151 deaths and 8465 deaths in the hospital.The median LOS was 6 days,the short LOS group(<6 days,N=3860),and the long LOS group(?6 days,N=4605).A total of 2382 patients(57.7%male,72(63,81 years)were re-admitted within a 1-year follow-up period.1)Comparing the Short LOS group with the Long LOS group,it was found that age,valvular cause of heart failure,neutrophils,the hypersensitive C-reactive protein(CRP),the red blood cell distribution width(RDW),N-terminal prohormone of Brain Natriuretic Peptide(NT pro-BNP)were independently related to Long LOS;hypertension,use of anti-hypertensive drugs,use of anti-platelet drugs,use of anti-lipid drugs,White Blood Cell(WBC),EF were independently related to Short LOS;NLR(OR=1.142,95%CI:1.058-1.233,p<0.05),MLR(OR=3.321,95%CI:1.332-8.281,p<0.05),PLR(OR=1.002,95%CI:1.000-1.003,p<0.05)were independently correlated of LOS;the comparison of the three predictive capabilities found that AUC(NLR):0.609(95%Cl:0.597-0.622),AUC(MLR):0.594(95%CI:0.582-0.606),and AUC(PLR):0.522(95%CI:0.540-0.565),predictive ability AUC(NLR)>AUC(MLR)(p<0.05),AUC(NLR)>AUC(PLR)(p<0.05),AUC(MLR)>AUC(PLR)(p<0.05),and Youden's index cut-off value were 4.22 for NLR,0.47 for MLR,and 162.54 for PLR;compared with the traditional inflammation index CRP,WBC,and Neutrophil,there was no difference between NLR and CRP,but AUC(NLR)>AUC(WBC)(p<0.05),AUC(NLR)>AUC(Neutrophil)(p<0.05).2)the subgroup analysis between different EF groups and different causes of heart failure groups,it was found that in heart failure with reduced EF subgroup(N=1804)and heart failure with mid-range EF subgroup(N=1804),there was no independent effect between NLR,MLR,PLR,and LOS.However,in the heart failure with preserved EF subgroup(N=5154),the predictive ability was shown,and there was no significant difference between NLR and MLR,but AUC(NLR)>AUC(PLR)and AUC(MLR)>AUC(PLR);in the subgroup of patients with heart failure caused by primary reasons(N=4155),NLR,MLR,PLR were independent predictors of LOS,predictive ability AUC(NLR)>AUC(MLR)(p<0.05),AUC(NLR)>AUC(PLR)(p<0.05),AUC(MLR)>AUC(PLR)(p<0.05);in the subgroup of patients with heart failure caused by ischemic disease(N=3304),NLR was an independent predictor of LOS,and AUC of ROC was 0.649(95%Cl:0.630-0.668),Neither MLR nor PLR were independent predictors;in the subgroup of patients with heart failure caused by valvular disease(N=507),NLR,MLR,and PLR did not show independent predictive effects.3)OPTIMIZE-HF risk model(?=0.370,p<0.05)and A2B Score risk Model(?=0.085,p<0.05)were independent predictors of LOS in ADHF patients.However,predictive ability AUC(NLR)>AUC(OPTIMIZE)(p<0.05),AUC(NLR)>AUC(A2B Score)(p<0.05).4)the follow-up data(N=2382)found that the first cut-off value could be used as an index to distinguish short and long LOS.The LOS(low NLR group):4(2,7)days<LOS(high NLR group):6(3,10)days(p<0.05);LOS(low MLR group):4(2,7)days<LOS(high MLR group):6(3,10)days(p<0.05);LOS(Low PLR group):4(2,7)days<LOS(high PLR group):5(3,9)days(p<0.05);the correlation found that the first index NLR and the second index NLR,the correlation was 0.198(p<0.05),the first index MLR and the second index MLR,the correlation was 0.126(p<0.05),the first index PLR and the second index PLR,the correlation was 0.295(p<0.05);there was no correlation between the first index NLR,MLR,PLR and LOS of second inpatient period.5)comparing the in-hospital death group(N=151)with the in-hospital survival group(N=8465),it was found that high diastolic blood pressure is related to survival,and the AUC of ROC:0.631(95%CI:0.583-0.682),high NT pro-BNP is associated with death,and the AUC of ROC:0.706(95%CI:0.663-0.749);NLR(?=0.062,p=0.23),MLR(?=1.324,p=0.08),PLR(?=0.001,p=0.55),none indicators were independently associated with nosocomial deaths.Conclusions:The indicators of NLR,MLR,and PLR were independent predictors for the initial episode LOS for ADHF patient,and the predictive power of NLR was better than MLR and PLR;Age,Valvular cause of heart failure,Neutrophils,CRP,RDW,NT pro-BNP were independently related to Long LOS;hypertension,use of anti-hypertensive drugs,use of anti-platelet drugs,use of anti-lipid drugs,WBC,EF were independently related to Short LOS;diastolic blood pressure and NT pro-BNP were independent predictors of nosocomial death in ADHF patients.
Keywords/Search Tags:inflammatory biomarkers, cardiovascular risk factors, acute decompensated heart failure, length of hospital stay
PDF Full Text Request
Related items