基于血清生物标志物的对低卒中风险房颤病人预测模型的研究

Stroke is the main complication of atrial fibrillation (AF). With the advent of novel anticoagulant, guidelines have recommended that there should be a clinical practice shift so that the first step is to identify ‘low risk’ patients who do not need any antithrombotic therapy. However, the CHA2DS2-VASc score which is now widely used in the clinical practice has only modest predictive value for predicting “low risk” subjects. A novel stroke risk stratification model for refining true low risk patients with AF is urgently needed. We have conducted a comprehensive meta-analysis and found that elevated circulating plasminogen activator inhibitor-1 and thrombin-antithrombin were significantly associated with increased risk of stroke in patients with AF. In addition, we have carried out a case-control study on 2709 AF patients. We compared the levels of serum biomarkers between AF patients complicated with stroke and those without stroke, and found that the levels of C-reactive protein, fibrinogen and neutrophils/ lymphocytes ratio were significantly higher in AF patients with stroke than those without. These evidences indicate that adding inflammatory, hemostatic and cardiac markers to the stroke risk prediction model will improve the predictive accuracy for refining “true low risk” patients. We have already set up a prospective cohort among 965 eligible AF patients, and all of the baseline information has been collected. We plan to develop a novel stroke risk stratification model for refining true low risk patients with AF which is suitable for Chinese population, and then validate this model externally in another cohort. This study will have great significance on clinical intervention of stroke in AF patients.

卒中是房颤病人最主要的并发症，新型口服抗凝药的出现，使识别出低卒中风险人群成为卒中预测模型的发展方向，但现有的CHA2DS2-VASC风险评分对低风险人群的预测能力有限。本项目组前期开展了一项Meta分析，发现在房颤病人众多的生物标志物中，高水平的纤溶酶原激活物抑制剂和凝血酶-抗凝血酶复合物能增加卒中的风险。在我们开展的一项2709名房颤病人的病例对照研究中，合并脑卒中的房颤病人外周血C反应蛋白、纤维蛋白原、中性粒细胞/淋巴细胞比值都显著高于未合并脑卒中的房颤病人。提示将炎症、止血、心功能相关生物标志物纳入风险预测模型，将有助于识别低卒中风险的病人。本项目组前期已在房颤病人中建立起了前瞻性队列，收集到房颤病人965例，基线资料已全部收集完成。本研究拟基于此队列，建立适合中国人群的、能识别出低卒中风险的房颤病人的预测模型，并在外部人群中对模型进行验证。本研究将对卒中的干预有重要的临床意义。

房颤是临床上最常见的快速性心律失常，如何准确的评估房颤患者发生卒中的风险，并根据评估风险进行临床干预意义重大。本课题的研究内容包括三个方面：1）在房颤病人中建立前瞻性队列；2）探索血清生物标志物的对房颤患者卒中风险的预测作用；3）建立基于外周血生物标志物的对房颤病人卒中风险的新预测模型。主要研究进展包括：1）探索血清生物标志物的对房颤患者卒中风险的预测作用：a. 外周血炎症因子对房颤患者卒中终点的预测。外周因子中高水平的HsCRP、PLR、NLR、D-dimer及低水平的ALB、LMR是房颤患者卒中风险的危险因素。b. 外周血系统性炎症评分对房颤患者卒中终点的预测。高水平系统炎症评分卒中风险高于中等水平高于低等水平。c. 外周血Th17等相关因子对房颤患者卒中终点的预测。外周因子中高水平的IL-17A, IL-21, IL-9和INF-γ是房颤患者卒中风险的危险因素。2）建立基于外周血生物标志物的对房颤病人卒中风险的新预测模型。将CRP, ALB, D-dimer, Fib, PLR, LMR，NLR以上7个标志物和CHA2DS2-VAS评分一起放进Cox回归模型中，采用后退法筛选出与卒中最相关的标志物，得到最终的对房颤病人卒中风险的新预测模型。最终新模型的变量为CHA2DS2-VAS评分和NLR。在CHA2DS2-VAS评分的基础上增加NLR，能显著增加C统计量和NRI及IDI，说明新模型的预测能力更强。.本研究探索的外周血生物标志物，主要是外周血常见炎症因子，相对简单、便捷、成本较低，此外本研究首次探讨了Th17相关细胞因子对房颤预后的影响，作为一种新的辅助性Th细胞Th17正逐步引起关注。基于HsCRP、PLR、NLR、D-dimer、ALB、LMR建立的系统性炎症评分能够作为反映房颤患者的炎症强度指标，且对房颤患者的预后产生影响。 .本研究在传统的CHA2DS2-VAS评分基础上，探索建立了基于外周血生物标志物的新预测模型。发现在CHA2DS2-VAS评分基础上增加NLR形成的新模型，相比起单独的CHA2DS2-VAS评分，预测房颤患者发生卒中的能力显著增加。NLR属于临床工作中的常规测量因子，基于此的预测模型在临床应用中方便易行，以期为临床的预后治疗提供治疗或预防依据。