Wang Wei-Ying, Sang Wen-Wen, Jin Di, Yan Shuang-Mei, Hong Yuan, Zhang Huan, Yang Xu
The Graduate Student Training Base-AeroSpace Center Hospital of Liaoning Medical University, Beijing, China.
Department of Neurology, Aerospace Center Hospital, Beijing, China.
J Stroke Cerebrovasc Dis. 2017 Jun;26(6):1233-1238. doi: 10.1016/j.jstrokecerebrovasdis.2017.01.013. Epub 2017 Feb 22.
Disability and mortality represent the most relevant clinical outcomes after acute ischemic stroke. Recently, a number of prognostic models of acute ischemic stroke have been developed, but they have not been extensively validated. In this study, we evaluated the ability of 3 prognostic models including the iScore, the PLAN score, and the ASTRAL score in predicting clinical poor outcomes or mortality at 6 months in patients with acute ischemic stroke.
A total of 323 patients were divided into a good-prognosis group and a poor-prognosis group based on the modified Rankin Scale. Model discrimination was quantified by calculating the area under the receiver operating characteristic (ROC) curve, and calibration was assessed by Hosmer-Lemeshow goodness of fit test and Pearson correlation coefficient.
We identified 96 (29.7%) patients with poor prognosis, including 21 who were dead. All 3 models showed good ability in predicting poor prognosis and mortality in patients with acute ischemic stroke (all ROC > .70). There was no difference between these 3 models in terms of sensitivity and accuracy (all P > .05).
The results of this study suggest that the iScore, the PLAN score, and the ASTRAL score were equal in predicting 6-month poor prognosis and mortality in patients with acute ischemic stroke. Overall, there was a very high correlation between observed and expected outcomes at the risk score level.
残疾和死亡率是急性缺血性卒中后最相关的临床结局。最近，已经开发了一些急性缺血性卒中的预后模型，但尚未得到广泛验证。在本研究中，我们评估了包括iScore、PLAN评分和ASTRAL评分在内的3种预后模型预测急性缺血性卒中患者6个月时临床不良结局或死亡率的能力。
根据改良Rankin量表将323例患者分为预后良好组和预后不良组。通过计算受试者操作特征（ROC）曲线下面积对模型辨别力进行量化，并通过Hosmer-Lemeshow拟合优度检验和Pearson相关系数评估校准情况。
我们确定了96例（29.7%）预后不良的患者，其中21例死亡。所有3种模型在预测急性缺血性卒中患者的不良预后和死亡率方面均表现出良好的能力（所有ROC>0.70）。这3种模型在敏感性和准确性方面没有差异（所有P>0.05）。
本研究结果表明，iScore、PLAN评分和ASTRAL评分在预测急性缺血性卒中患者6个月时的不良预后和死亡率方面相当。总体而言，在风险评分水平上，观察到的结局与预期结局之间存在非常高的相关性。
