Background

Vascular region of infarct is part of the International Classification of Diseases-10 (ICD-10) coding scheme for ischemic stroke. These data could potentially be used for studies about vascular location, such as comparisons of anterior versus posterior circulation stroke. The objective of this study was to evaluate the validity of these subcodes.

Methods

We selected a random sample of 100 hospitalizations specifying 50 with anterior circulation ICD-10 ischemic stroke (carotid, anterior cerebral artery [CA], middle CA) and 50 with posterior circulation stroke (vertebral, basilar, cerebellar, posterior CA). The gold standard primary vascular distribution was scored using imaging studies and reports, blinded to the subcode. We compared gold-standard distribution to coded distribution and calculated the operating characteristics of ICD-10 posterior circulation versus anterior circulation codes with the gold standard. We also calculated the kappa statistic for agreement across all 7 vascular regions.

Results

In our population of 100 strokes, mean NIHSS was 8 (SD, 8). Head CT was performed in 95 % (95/100) and MRI in 77 % (77/100). The gold standard classified 55 primary posterior circulation strokes (26 PCA, 16 cerebellar, 8 basilar, 5 vertebral), 44 primary anterior circulation strokes (35 MCA, 6 carotid, 3 ACA), and 1 stroke with no infarct on imaging. The accuracy of the ICD-10 classification for primary posterior circulation stroke versus anterior circulation/no infarct was: sensitivity 89 % (49/55); specificity 98 % (44/45); positive predictive value 98 % (49/50); negative predictive value 88 % (44/50). The reliability of the 7-region classification was excellent (kappa 0.85).

Conclusions

We found that ICD-10 classification of vascular location in routine practice correlates strongly with gold-standard localization for hospitalized ischemic stroke and supports validity in differentiating posterior versus anterior circulation. At a more granular vascular level, the location reliability was excellent, although limited data were available for some subcodes.

中文翻译：

---

根据国际疾病分类 10 验证急性缺血性中风的血管位置子代码

背景

梗塞的血管区域是缺血性中风的国际疾病分类 10 (ICD-10) 编码方案的一部分。这些数据可能用于有关血管位置的研究，例如前循环卒中与后循环卒中的比较。本研究的目的是评估这些子代码的有效性。

方法

我们随机抽取了 100 名住院患者，其中 50 名患有前循环 ICD-10 缺血性卒中（颈动脉、大脑前动脉 [CA]、中 CA），50 名患有后循环卒中（椎骨、基底动脉、小脑、后 CA）。使用成像研究和报告对金标准主要血管分布进行评分，对子代码不知情。我们将金标准分布与编码分布进行比较，并计算了 ICD-10 后循环与金标准前循环代码的运行特征。我们还计算了所有 7 个血管区域的一致性的 kappa 统计量。

结果

在我们的 100 例中风人群中，平均 NIHSS 为 8 (SD, 8)。 95% (95/100) 进行了头部 CT，77% (77/100) 进行了 MRI。金标准分类为 55 例原发性后循环中风（26 例 PCA、16 例小脑、8 例基底动脉、5 例椎体）、44 例原发性前循环中风（35 例 MCA、6 例颈动脉、3 例 ACA）和 1 例影像学上无梗塞的中风。原发性后循环卒中与前循环/无梗塞的 ICD-10 分类的准确性为： 敏感性 89 % (49/55)；特异性 98% (44/45)；阳性预测值 98% (49/50)；阴性预测值 88% (44/50)。 7 区域分类的可靠性非常好（kappa 0.85）。

结论

我们发现常规实践中血管位置的 ICD-10 分类与住院缺血性卒中的金标准定位密切相关，并支持区分后循环与前循环的有效性。在更细粒度的血管水平上，尽管某些子代码的可用数据有限，但定位可靠性非常好。