Skoglund Per H, Arpegård Johannes, Ostergren Jan, Svensson Per
Karolinska Institutet, Department of Medicine, Solna, Internal Medicine Unit and Emergency Department, Karolinska University Hospital Solna, Stockholm, Sweden.
Am J Hypertens. 2014 Mar;27(3):363-71. doi: 10.1093/ajh/hpt278. Epub 2014 Jan 27.
Patients with peripheral arterial disease (PAD) are at high risk for cardiovascular (CV) events. We have previously shown that ambulatory pulse pressure (APP) predicts CV events in PAD patients. The biomarkers amino-terminal pro-B-type natriuretic peptide (NT-proBNP), high-sensitivity C-reactive protein (hs-CRP), and cystatin C are related to a worse outcome in patients with CV disease, but their predictive values have not been studied in relation to APP.
Blood samples and 24-hour measurements of ambulatory blood pressure were examined in 98 men referred for PAD evaluation during 1998-2001. Patients were followed for a median of 71 months. The outcome variable was CV events defined as either CV mortality or any hospitalization for myocardial infarction, stroke, or coronary revascularization. The predictive values of log(NT-proBNP), log(hs-CRP), and log(cystatin C) alone and together with APP were assessed by multivariable Cox regression. Area under the curve (AUC) and net reclassification improvement (NRI) were calculated compared with a model containing other significant risk factors.
During follow-up, 36 patients had at least 1 CV event. APP, log(NT-proBNP), and log(hs-CRP) all predicted CV events in univariable analysis, whereas log(cystatin C) did not. In multivariable analysis log(NT-proBNP) (hazard ratio (HR) = 1.62; 95% confidence interval (CI) = 1.05-2.51) and log(hs-CRP) (HR = 1.63; 95% CI = 1.19-2.24) predicted events independently of 24-hour PP. The combination of log(NT-proBNP), log(hs-CRP), and average day PP improved risk discrimination (AUC = 0.833 vs. 0.736; P < 0.05) and NRI (37%; P < 0.01) when added to other significant risk factors.
NT-proBNP and hs-CRP predict CV events independently of APP and the combination of hs-CRP, NT-proBNP, and day PP improves risk discrimination in PAD patients.
外周动脉疾病（PAD）患者发生心血管（CV）事件的风险很高。我们之前已经表明，动态脉压（APP）可预测PAD患者的CV事件。生物标志物氨基末端前B型利钠肽（NT-proBNP）、高敏C反应蛋白（hs-CRP）和胱抑素C与CV疾病患者的不良预后相关，但它们与APP相关的预测价值尚未得到研究。
对1998年至2001年间因PAD评估而转诊的98名男性进行了血样采集和24小时动态血压测量。对患者进行了中位数为71个月的随访。结局变量为CV事件，定义为CV死亡或因心肌梗死、中风或冠状动脉血运重建而住院。通过多变量Cox回归评估log(NT-proBNP)、log(hs-CRP)和log(胱抑素C)单独以及与APP一起的预测价值。与包含其他显著危险因素的模型相比，计算曲线下面积（AUC）和净重新分类改善（NRI）。
在随访期间，36名患者至少发生了1次CV事件。在单变量分析中，APP、log(NT-proBNP)和log(hs-CRP)均能预测CV事件，而log(胱抑素C)则不能。在多变量分析中，log(NT-proBNP)（风险比（HR）=1.62；95%置信区间（CI）=1.05-2.51）和log(hs-CRP)（HR=1.63；95%CI=1.19-2.24）独立于24小时脉压预测事件。当将log(NT-proBNP)、log(hs-CRP)和平均日间脉压相结合，并添加到其他显著危险因素中时，可改善风险辨别能力（AUC=0.833对0.736；P<0.05）和NRI（37%；P<0.01）。
NT-proBNP和hs-CRP独立于APP预测CV事件，并且hs-CRP、NT-proBNP和日间脉压的组合可改善PAD患者的风险辨别能力。
