血管性血友病因子预测非瓣膜性房颤患者卒中风险的评估

Study on the von Willebrand factor for assessing the stroke risk in the patients with atrial fibrillation

目的研究血管性血友病因子抗原（vWF：Ag）水平预测非瓣膜性房颤患者发生卒中风险评估价值。方法选择2009至2011年于天津医科大学总医院就诊的非瓣膜性房颤患者180例进行回顾性队列研究，男112例，女68例，年龄61—87岁。采用ILACL-9000型血液凝固仪测定vWF：Ag水平；用ROC分析vWF：Ag的诊断性能，用Cox回归模型分析vWF：Ag对预后的影响，用χ^2检验分析vWF：Ag与临床病理因素间有无关联性。患者CHADS2评分组和CHA2DS：VASc评分组间的数据比较采用t检验。结果vWF：Ag水平分别为对照组（112±34）％、阵发性房颤组（119±31）％、持续性房颤组（179±47）％、永久性房颤组（217±56）％、房颤合并卒中组（235±104）％，其中阵发性房颤组vWF：Ag水平与健康对照组间差异无统计学意义（g=1．75，P〉0．05）；持续性房颤组高于阵发性房颤组（q=10．10，P〈0．01）；永久性房颤组高于持续性房颤组（q=5．21，P〈0．01）。vWF：Ag诊断房颤患者卒中风险的临界值为188．5％，曲线下面积为0．843（95％CI：0．785～0．901）。Cox比例风险模型多因素分析显示，血浆vWF：Ag（HR=0．405；95％CI=0．268～0．716；P=0．026）。vWF：Ag水平与临床病理因素的，检验分析显示，慢性心力衰竭／左室功能障碍（χ^2=8．227，P〈0．01）、高血压（χ^2=3．305，P〈0．05）、年龄（χ^2=7．581，P〈0．01）、糖尿病（χ^2=6．730，P〈0．01）、卒中／TIA／血栓栓塞史（χ^2=4．825，P〈0．05）、血管疾病（χ^2=4．126，P〈0．05）。CHADS2评分的1分组患者vWF：Ag水平高于CHA：DS：VASe评分的1分组，差异有统计学意义（t=4．283，P〈0．01）。结论vWF：Ag水平与房颤患者主要病理因素相关，并随病情而变化，高水平vWF：Ag是独立的卒中风险预测因素，对评估房颤患者卒中发生具有较高的参考价值。（中华检验医学杂志，2013，36：233-237）

Objective To investigate the value for the level mensuration of von Willebrand factor antigen （vWF： Ag） in stroke risk assessment in the patients with non-valvular atrial fibrillation （AF）.Methods 180 non-valvular AF patients were selected from the Tianjin medical university general hospital from the 2009 to 2011 for retrospective cohort study, 112 males and 68 females in the group, age 61 -87 years. Using the IL ACL-9000 blood coagulation instrument assay the level of vWF： Ag. Using R0C curve to analyze the diagnosis performance of vWF： Ag, using Cox regression analysis model to evaluate the of vWF ：Ag effect on prognosis, using χ^2 test to analyze the relevance between vWF： Ag and clinical pathological factors. Compared the patients group with CHADS2 score with the patients group with CHA2DS2VASc score date using t test. Results vWF： Ag levels were control group （ 112 ± 34） %, paroxysmal AF group （ 119 ±31 ） %, the persistent AF group （ 179 ± 47 ） %, permanent AF group （ 217 ± 56 ） %, atrial fibrillation associated with stroke group（ 235 ± 104 ）% respectively. There was no difference between the paroxysmal AF group and control group （ q = 1.75, P 〉 O. 05 ） ; vWF： Ag level was higher in persistent atrial fibrillation associated with stroke group （235 ± 104）% respectively. There was no difference between the paroxysmal AF group and control group （ q = 1.75, P 〉 O. 05 ） ; vWF： Ag level was higher in persistent atrial fibrillation group than in paroxysmal AF group （ q = 10. 10, P 〈 O. 01） ; permanent atrial fibrillation group was higher than that of the persistent AF group （ q = 5.21, P 〈0.01 ）. The optimum cut-off point with vWF： Ag for stroke diagnosis was 188.5% , the area under ROC curve = 0. 843 （ 95% confidence interval： 0. 785 -0. 901）. In Cox regression multianalysis, the vWF： Ag （ HR = 0. 405; 95% CI = 0. 268 - 0. 716; P = 0. 026 ）, the congestive heart failure（ HR = 2. 901 ; 95% CI = 1. 837 - 3. 951 ; P = 0. 001 ）, stroke/transient ischemic attack （ HR = 4. 665 ; 95 % CI = 2. 837 - 7.291 ; P = 0. 000）, age （ HR = 0. 474 ; 95 % CI =0. 211 - 0. 765; P =0. 039）, the Cox analysis showed that vWF： Ag was the independent prognosis factor for stroke in AF patients. In χ^2 analysis, there was the relationship between the level of the vWF： Ag and the congestive heart failure/LVdysfunction （χ^2= 8. 227, P 〈 0. 01 ）, hypertension （χ^2 = 3. 305, P 〈 0. 05 ）, age （χ^2=7. 581, P 〈 0.01） , diabetes mellitus （χ^2 = 6. 730, P 〈 0.01） , stroke/ transient ischemic attack/ thromboembolism （χ^2 = 4. 825, P 〈 0. 05 ） , vascular disease （χ^2 = 4. 126, P 〈 0. 05 ） . Compared the subjects with CHADS2 （ score = 1 ） with the CHA2DS2VASc（ score = 1 ）, the level of the vWF： Ag was higher in patients with CHADS2 score = 1 （ t = 4. 283, P 〈 0. 01 ）. Conclusion There was relationship between the A level of vWF： Ag and main pathologic factors in patients with AF, and changed with the condition, high vWF： Ag level was an independent predictor of stroke risk, and had superior reference value for in assessment of stroke in patients with atrial fibrillation. （ Chin J Lab Med,2013,36：233-237）