Effect Of NT-proBNP And Galectin-3 Level On Prognosis Of Patients With Ischemic Stroke

Objectives To explore the association between N-terminal pro-brain natriuretic peptide(NT-proBNP)and galectin-3 in acute phase and the adverse clinical outcomes in 3 months and 12 months after stroke onset,and to explore the combined effect of NT-proBNP and galectin-3 level on prognosis of patients with ischemic stroke.Our study could provide evidence for early risk stratification and making clinical therapeutic plan for patients with ischemic stroke.Methods We employed a prospective cohort study design,and participants with ischemic stroke came from China Antihypertensive Trial in Acute Ischemic Stroke(CATIS)(clinicaltrials.gov Identifier: NCT01840072).Baseline data and blood samples of acute ischemic stroke patients were collected at the time of enrollment,and we measured NT-proBNP level and galectin-3 in serum.Study participants were categorized into four groups(Q1,Q2,Q3 and Q4)according to the four quartiles of serum NT-proBNP or galectin-3 levels and galectin-3 level.The hospitalization outcomes were evaluated at 14 days or hospital discharge,and patients were followed up in person at 3 months and 12 months after hospitalization.The clinical outcomes included death,composite outcome of death or major disability [(modified Rankin Scale,mRS)3-6],cardiovascular events,and composite outcome of death or cardiovascular events.We compared incidence rate with Chi-square trend analysis or use Kaplan–Meier curves to compare the cumulative risk of outcomes by log-rank test.We used Cox proportional hazard models to adjust traditional risk factors and calculate hazard ratios(HRs)and 95% confidence intervals(CIs)to evaluate the relationship between serum NT-proBNP or galectin-3 and prognosis,or used logistic regression models to calculate odds Ratios(ORs)and 95% CIs for clinical outcomes.We also carry out subgroup analysis to assess the robustness of these associations.The receiver operating characteristic curve(ROC)was utilized to analyze the prognostic values and we use the area under the curve(AUC)to make quantitative evaluation.Based on NT-pro BNP and galectin-3 level,participants were categorized into four groups: low galectin-3/low NT-proBNP,low galectin-3/high NT-proBNP,high galctin-3/low NT-proBNP,and high galectin-3/high NT-proBNP.A design of prospective cohort study was used in this study.We analyzed the association between these four status and prognosis of ischemic stroke.Results Association between serum NT-proBNP level and prognosis of patients with ischemic stroke A total of 3216 individuals with an average age of 62.5 years were included into analysis,and the male was accounted for 63.7%.After 3-months follow-up,93 patients died and the cumulative incidence rates of death was 2.89%,804 patients occurred composite outcome of death or major disability and the cumulative incidence rates of composite outcome of death or major disability was 25.00%.1.Association between NT-proBNP level and prognosis of patients in 3 month after stroke onset Within 3 months,the cumulative incidence rates of death among patients in four quartiles of serum NT-proBNP(Q1,Q2,Q3 and Q4)were 1.4%,2.2%,3.7% and 4.2%(P-trend<0.05),respectively.After adjustment for confounding factors,the OR(95%CI)for death compared with the lowest quartile(Q1),Q2,Q3 and Q4 were 1.41(0.62-3.21),2.10(0.98-4.49)and 1.84(0.86-3.93),it were not statistically significant.Within 3 months,the cumulative incidence rate of composite outcome of death or major disability among four quartiles of serum NT-proBNP were 20.6%,25.5%,27.3% and 26.6%,respectively.Logistic regression analysis indicated that the OR of death and major disability in Q2,Q3 and Q4 were 1.15(0.88-1.50),1.09(0.83-1.43)and 0.92(0.70-1.22)respectively to compare with Q1 group.2.Association between NT-proBNP level and prognosis of patients in 12 month after stroke onset Within 12 months,the cumulative incidence rates of death were 2.83%,4.55%,7.09% and 8.54% respectively among four quartiles of serum NT-proBNP(log-rank test,P<0.001).After adjustment for confounding factors,the HR associated with Q2,Q3 and Q4 were1.18(0.67-2.10),1.70(95% CI: 1.00-2.86,P<0.05)and 1.61(95% CI: 1.11-2.33,P<0.05)for death,respectively,compared with Q1.The AUC for NT-proBNP was 0.613(95% CI: 0.572-0.654,P<0.001)in prediction of death within 12 months.Cox proportional hazard models analysis showed that adjusted HRs with Q2,Q3 and Q4 groups were 1.42(0.96-2.09),1.53(1.05-2.26)and 1.61(1.11-2.33)for composite outcome of death or cardiovascular events to caompare with Q1 group,respectively;the AUC was 0.568(95% CI: 0.534-0.603,P<0.001)in prediction of composite outcome of death or cardiovascular events within 12 months.Logistic regression analysis showed that the adjusted ORs associated with Q2,Q3 and Q4 were 1.20(0.90-1.61),1.28(0.96-1.70)and 1.47(1.11-1.95)for the composite outcome of death or major disability,respectively.The AUC was 0.569(95%CI: 0.545-0.593,P<0.001)in prediction of composite outcome of death or major disability within 12 months.The mRS score in 12-month follow-up was different across quartiles of NT-proBNP,and the adjusted ORs associated with Q2,Q3 and Q4 were 1.34(1.11-1.63),1.26(1.03-1.51)and 1.47(1.22-1.78)for 1-U higher mRS score,respectively.Association between Galectin-3 level and prognosis of patients with ischemic stroke A total of 3082 individuals with an average age of 62.26 years were included into analysis,and the male was1956(63.5%)and female was 1126(36.5%).1.Association between Galectin-3 level and prognosis of patients in 3 months after stroke onset Within 3 months,the incidence rate of death were 1.3%,1.2%,1.4% and 4.4%(P-trend<0.05)among four quartiles of serum galectin-3,respectively.Logistic regression analysis showed that adjusted odds ratios(OR)of Q2,Q3 and Q4 were 0.70(0.27-1.84),0.80(0.32-1.99)and 2.44(95% CI:1.13-5.27)for death to compare with Q1,and the AUC for galectin-3 was 0.654(95%CI: 0.577-0.731,P<0.05)in prediction of death.2.Association between Galectin-3 level and prognosis of patients in 12 months after stroke onset Within 12 months,the cumulative incidence rates of death increased with an increase of galectin-3 levels in serum(P-trend<0.05).Cox proportional hazard models analysis showed that adjusted HR of Q2,Q3 and Q4 groups were 0.76(0.46-1.27),0.86(0.53-1.39)and 1.58(1.02-2.43)for death compared with Q1 group.Subgroup analysis showed that smoking and galectin-3 had an interaction effect on death.The AUC for galectin-3 was 0.614(95% CI: 0.569-0.659,P<0.001)in prediction of death.Cox proportional hazard models analysis showed that adjusted HR of Q2,Q3 and Q4 were 0.97(0.60-1.56),1.27(0.81-1.98)and 1.64(1.06-2.52)for the cardiovascular events to compare with Q1 group.Subgroups analysis showed that use of antihypertensive medications before stroke onset statistically interacted with galectin-3 on the cardiovascular events.The AUC for galectin-3 was 0.571(95% CI: 0.527-0.615,P<0.05)in predicting cardiovascular events.Combined effect of NT-proBNP and Galectin-3 level on prognosis of patients with ischemic stroke We include 2790 participants(1769 men and 1021 women;mean age 62.37±10.75 years)with values of both NT-proBNP and galctin-3 in this analysis.1.Combined effect of NT-proBNP and Galectin-3 level on prognosis in 3 months after stroke onset Compared with low galectin-3/low NT-pro BNP,Logistic regression analysis showed that multivariate adjusted ORs in high galectin-3/high NT-proBNP were 3.18(1.46-6.93)and 1.42(1.03-1.96)for death and composite outcome of death or major disability in 3 months,respectively.2.Combined effect of NT-proBNP and Galectin-3 level on prognosis in 12 months after stroke onset Compared with low galectin-3/low NT-proBNP,Cox proportional hazards models analysis showed that multivariate adjusted HR of high galectin-3/high NT-proBNP was 3.08(1.81-5.24),1.51(1.19-1.91),and 1.81(1.17-2.80)for death,composite outcome of death or major disability,and composite outcome of death or cardiovascular events in 12 months,respectively.In addition,NT-proBNP and galectin-3 interacted in association with composite outcome of death or major disability in 12 months after stroke onset.Conclusions 1.Serum NT-pro BNP levels in acute phase were independently associated with the risk of death,composite outcome of death or vascular events,composite outcome of death or major disability in 12 months after stroke onset,suggesting that NT-proBNP might be a potential prognostic biomarker for ischemic stroke patients.2.Baseline elevated galectin-3 level was an independent risk factor for death in 3 months,and it was also associated with death,cardiovascular events in 12 months after stroke onset,suggesting that galectin-3 could be as potential biomarker for prognosis of ischemic stroke.3.High galectin-3/high NT-pro BNP was associated with death,composite outcome of death or major disability in 3 months after stroke onset.High galectin-3/high NT-proBNP was also associated with death and cardiovascular events in 12 months after stroke onset.These finding suggested that the combination of galectin-3 and NT-proBNP could provide more precise information for prognosis of patients with ischemic stroke.4.NT-proBNP interacted with galection-3 on the composite outcome of death or major disability in 12 months after ischemic stroke onset,and the combination of these two biomarkers could provide more accurate information for predicting 12-month composite outcome of death or major disability.