脑静脉窦血栓形成患者预后量表的有效性检验

Validation of a Predictive Outcome Score of Cerebral Vein and Dural Sinus Thrombosis

目的调查脑静脉窦血栓形成(cerebral venous thrombosis,CVT)患者的近期预后,验证CVT预后量表的有效性。方法在单中心观察性研究中,连续性入选2000年5月至2010年5月北京天坛医院CVT患者,进行回顾性分析。通过病例报告表收集患者的人口学信息,症状和体征,诊断方法,血栓的部位等。患者在住院后72 h内完成颅脑磁共振(magnetic resonance imaging,MPI)。评定者依据影像学结果结合基线情况(住院当时症状体征)并完成CVT预后量表评价。随访在患者出院前1 d完成。结局通过改良的Rankin量表(modified Rankin Scale,mRS)评定,mRS≤2分定义为神经功能结局良好。计算CVT的敏感性、特异性、阳性预测值、阴性预测值、阳性似然比和阴性似然比,来验证CVT预后量表的有效性。应用受试者工作曲线(Receiver operating characteristic,ROG)分析量表预测的准确性。结果 168名住院CVT患者纳入研究。165名患者在发病17 d(8～45 d)完成量表评价。随访时,130名患者(78.8%)预后良好(mRS≤2),CVT预后评分选择≥3分为分界点,量表预测的敏感性、特异性和ROC曲线下面积分别为0.75[95%可信区间(confidence interval,CI)0.67～0.82],0.38(95%CI0.22～0.55)和0.56(95%CI 0.45～0.68,P>0.05);当选择≥4分为分界点时,敏感性、特异性和ROc曲线下面积分别为0.85(95%CI 0.78～0.91),特异性0.35(95%CI 0.19～0.51)和0.60(95%CI0.49～0.71,P=0.074>0.05)。结论 CVT患者出院时预后相对良好,仍有部分患者预后不良(21.2%)。在本研究中,CVT预后量表准确性不理想,需要进一步前瞻性验证。

Objective The purpose of this study was to determine the short-term prognosis of patients with cerebral venous thrombosis （CVT） and to validate a predictive outcome score for CVT. Methods We retrospectively investigated consecutive patients who were diagnosed as having CVT admitted to one stroke center in this observational study. Demographic data, symptoms and signs from onset to diagnosis, imaging methods used, location of the thrombus and therapy were retrieved with case report form by a neurologist. Patients were followed up on the day before discharged. Primary outcome was independent as assessed by modified Rankin Scale （mRS） score ≤2 at the end of follow-up. Sensitivity, specificity, positive predictive values, negative predictive values, negative likely ratio and positive likely ratio were calculated to validate the previous predictive outcome score. Receiver operating characteristic （ROC） curves were used to analyze the accuracy. Results From May 2000 to May 2010, we recruited 168 adult patients with CVT. 165 patients were evaluated with CVT predict score on 17d （8-45d） from onset. At the end of follow-up, 130 patients （78.8%） were independent by mRS ≤2. Using the risk score with a cut-off of ≥3 points, sensitivity, specificity and C statistics were 0.75 （95% confidence interval （CI） 0.67- 0.82）, 0.38 （95%CI 0.22- 0.55） and, 0.56 （95%CI 0.45- 0.68） respectively. And 0.85 （95%CI 0.78- 0.91）; 0.35 （95%C1 0.19 - 0.51 ） and 0.60 （95% CI0.49 - 0.71, P=0.074 〉0.05） respectively with a cut-off of ≥ 4 points.Conclusion The prognosis of CVT is relative benign A subgroup （21.2%） of CVT patients is at increased risk of bad outcome. The CVT risk score showed poor predictive value in our study and need to be evaluated prospectively.