N末端B型利钠肽原对老年非心力衰竭住院患者预后的预测价值

Predictive value of N-terminal B-type natriuretic peptide on outcome of elderly hospitalized non-heart failure patients

目的探讨N末端B型利钠肽原(NT-proBNP)对老年非心力衰竭(心衰)住院患者预后的预测价值。方法前瞻性入选2018年9月至2019年2月在北京医院住院的年龄≥65岁的老年患者,排除临床诊断心衰或左心室射血分数(LVEF)<50%的患者。收集入选患者的临床资料。测定入选患者血NT-proBNP水平,并据此将患者分为2组,即低NT-proBNP(<125 ng/L)组和高NT-proBNP(≥125 ng/L)组。分别在患者入组后3、6和12个月进行门诊或电话随访,记录主要不良事件发生情况。终点事件包括心血管事件、全因死亡、各种原因再次入院或急诊室治疗以及包括上述所有事件的复合终点事件。心血管事件定义为因心原性休克、心肌梗死、心绞痛、心律失常、心衰、卒中/短暂性脑缺血发作、血运重建导致的死亡、再入院或急诊室治疗。采用Kaplan-Meier法绘制生存曲线,用Log-rank法比较不同组间生存曲线差异。采用多因素Cox回归模型分析NT-proBNP是否为老年非心衰住院患者终点事件的独立预测因子。结果共入选老年非心衰住院患者600例,年龄(74.9±6.5)岁,其中男性304例(50.7%)。随访344(265,359)d,随访期间复合终点事件发生178例(29.7%),全因死亡19例(3.2%),失访12例(2.0%)。入选患者中,低NT-proBNP组286例(47.7%),高NT-proBNP组314例(52.3%)。与低NT-proBNP组比较,高NT-proBNP组患者年龄较大、合并心房颤动和心肌梗死的比例较高、认知水平和日常活动能力较低、白蛋白和内生肌酐清除率较低(P均<0.05)。1年随访结果显示,高NT-proBNP组患者复合终点事件发生率高于低NT-proBNP组[33.4%(105/314)比24.8%(71/286),P=0.02],心血管事件发生率亦高于低NT-proBNP组[17.5%(55/314)比8.4%(24/286),P=0.001],Kaplan-Meier生存分析结果亦显示出同样的趋势,高NT-proBNP组患者复合终点事件(Log-rank P=0.016)和心血管事件发生率(Log-rank P=0.001)均高于低NT-proBNP组。高NT-proBNP组患者的全因死亡率高于低NT-proBNP组[4.8%(15/314)比1.4%(4/286),P=0.020],但Kaplan-Meier生存分析结果则显示二者差异无统计学意义(Log-rank P=0.052)。校正了年龄、性别、体重指数、舒张压、肌酐清除率、血红蛋白、白蛋白、高敏C反应蛋白、心肌梗死、心房颤动、躯体生活活动能力评估评分、工具性日常生活活动能力评估评分后,多因素Cox回归分析结果显示NT-proBNP是复合终点事件(HR=1.376,95%CI 1.049~1.806,P=0.021)和心血管事件(HR=1.777,95%CI 1.185~2.664,P=0.005)的独立预测因子,但不是全因死亡的独立预测因子(P=0.206)。结论NT-proBNP对老年非心衰住院患者的预后特别是再入院和心血管事件具有较好的预测价值,测定此类患者的NT-proBNP水平有助于快速识别远期预后不良的高危人群。

Objective To investigate the predictive value of N-terminal type B natriuretic peptide(NT-proBNP)on the prognosis of elderly hospitalized patients without heart failure(non-heart failure).Method Elderly patients aged 65 years or older,who were admitted to Beijing Hospital from September 2018 to February 2019,were enrolled in this study.Patients with clinical diagnosis of heart failure or left ventricular ejection fraction(LVEF)<50%were excluded.The patients were divided into 2 groups based on the serum NT-proBNP level:low NT-proBNP group(<125 ng/L)and high NT-proBNP group(≥125 ng/L).Patients were followed up at 3,6,and 12 months after enrollment,and the major adverse events were recorded.The composite endpoint events included all-cause mortality,readmission or Emergency Department visits.Cardiovascular events include death,readmission or emergency room treatment due to cardiogenic shock,myocardial infarction,angina pectoris,arrhythmia,heart failure or stroke/transient ischemic attack.Results A total of 600 elderly patients with non-heart failure were included in the analysis.The average age was(74.9±6.5)years,including 304(50.7%)males.The median follow-up time was 344(265,359)days.One hundred and seventy-eight(29.7%)composite endpoint events were recorded during the follow-up,19(3.2%)patients died,and 12(2.0%)patients were lost to follow-up.There were 286(47.7%)cases in low NT-proBNP group and 314 cases(52.3%)in high NT-proBNP group.Patients were older,prevalence of atrial fibrillation and myocardial infarction was higher;MMSE scores and ADL scores,albumin and creatinine clearance rate were lower in high NT-proBNP group than in low NT-proBNP group(all P<0.05).At 1-year follow-up,the incidence of composite endpoint events was significantly higher in high NT-proBNP group than in low NT-proBNP group(33.4%(105/314)vs.24.8%(71/286),P=0.02).Cardiovascular events were more common in high NT-proBNP group than in low NT-proBNP group(17.5%(55/314)vs.8.4%(24/286),P=0.001).Kaplan-Meier survival analysis showed both composite endpoint events(Log-rank P=0.016)and cardiovascular events(Log-rank P=0.001)were higher in high NT-proBNP group than in low NT-proBNP group.All-cause mortality was also significantly higher in highNT-proBNP group than in lowNT-proBNP group(4.8%(15/314)vs.1.4%(4/286),P=0.020),and Kaplan-Meier survival analysis demonstrated borderline statistical significance(Log-rank P=0.052).Cox proportional hazard regression analysis showed that after adjusting for age,sex,creatinine clearance rate,myocardial infarction,and atrial fibrillation,NT-proBNP remained as an independent risk factor for composite endpoint events(HR=1.376,95%CI 1.049-1.806,P=0.021),and cardiovascular events(HR=1.777,95%CI 1.185-2.664,P=0.005),but not for all-cause mortality(P=0.206).Conclusions NT-proBNP level at admission has important predictive value on rehospitalization and cardiovascular events for hospitalized elderly non-heart failure patients.NT-proBNP examination is helpful for risk stratification in this patient cohort.