The Causative Classification of Stroke system: An international reliability and optimization study

A.A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02129, USA.
Neurology (Impact Factor: 8.29). 10/2010; 75(14):1277-84. DOI: 10.1212/WNL.0b013e3181f612ce
Source: PubMed


Valid and reliable ischemic stroke subtype determination is crucial for well-powered multicenter studies. The Causative Classification of Stroke System (CCS, available at is a computerized, evidence-based algorithm that provides both causative and phenotypic stroke subtypes in a rule-based manner. We determined whether CCS demonstrates high interrater reliability in order to be useful for international multicenter studies.
Twenty members of the International Stroke Genetics Consortium from 13 centers in 8 countries, who were not involved in the design and development of the CCS, independently assessed the same 50 consecutive patients with acute ischemic stroke through reviews of abstracted case summaries. Agreement among ratings was measured by kappa statistic.
The κ value for causative classification was 0.80 (95% confidence interval [CI] 0.78-0.81) for the 5-subtype, 0.79 (95% CI 0.77-0.80) for the 8-subtype, and 0.70 (95% CI 0.69-0.71) for the 16-subtype CCS. Correction of a software-related factor that generated ambiguity improved agreement: κ = 0.81 (95% CI 0.79-0.82) for the 5-subtype, 0.79 (95% CI 0.77-0.80) for the 8-subtype, and 0.79 (95% CI 0.78-0.80) for the 16-subtype CCS. The κ value for phenotypic classification was 0.79 (95% CI 0.77-0.82) for supra-aortic large artery atherosclerosis, 0.95 (95% CI 0.93-0.98) for cardioembolism, 0.88 (95% CI 0.85-0.91) for small artery occlusion, and 0.79 (0.76-0.82) for other uncommon causes.
CCS allows classification of stroke subtypes by multiple investigators with high reliability, supporting its potential for improving stroke classification in multicenter studies and ensuring accurate means of communication among different researchers, institutions, and eras.

Download full-text


Available from: William Meurer, Oct 06, 2015
60 Reads
  • Source
    • "In previous studies, the inter-observer agreement on ischemic stroke subtypes classified by the TOAST system was moderate (0.4<k<0.6).43 The computer-based algorithm increased the inter-observer agreement significantly (0.8<k<1).44 A retrospective medical and imaging review, however, showed that the inter-observer reliability differed depending on the stroke subtype: the reliability was high for strokes due to large-artery atherosclerosis (k=0.80) or cardioembolism (k=0.80), "
    [Show abstract] [Hide abstract]
    ABSTRACT: Proper classification of the causative mechanism of stroke is important for optimizing stroke treatment and assessing prognosis. The primary etiology of stroke differs according to race and ethnicity: emboli originating from the heart or extracranial large arteries are common in Western populations, whereas small-vessel occlusion or intracranial atherosclerosis is more prevalent in Asians. Intracranial atherosclerosis frequently leads to stroke by branch-artery occlusion, and the degree of stenosis in these cases is often <50%. Mild intracranial atherosclerotic stenosis may cause distal embolization, if the atherosclerotic plaque is sufficiently vulnerable. Moreover, high-resolution magnetic resonance imaging studies have identified small plaques causing infarction, even in patients with normal-appearing vascular findings. Such cases, which are prevalent in Asia, could not be classified as large-artery atherosclerosis by previous classification systems. Additionally, single subcortical infarctions, which are usually attributed to lipohyalinotic small-vessel disease, can have other causes, including microatheroma of perforators and atherothrombotic lesions at the parental artery. Single subcortical infarctions associated with parental artery disease or those bordering on the main vessel more often have atherosclerotic characteristics than do those associated with lipohyalinosis of the penetrating artery. In countries where intracranial atherosclerosis is common, such atherosclerotic single subcortical infarctions are predicted to be prevalent. These cases, however, could not be appropriately classified in previous systems. Further effort should be devoted to formulate ischemic stroke classification systems that adequately incorporate results of recent studies and reflect the underling pathologic mechanisms, especially in patients with single subcortical infarction and intracranial atherosclerosis.
    01/2014; 16(1):8-17. DOI:10.5853/jos.2014.16.1.8
  • [Show abstract] [Hide abstract]
    ABSTRACT: Previous data suggest that Dickkopf-1 (Dkk-1), an inhibitor of the canonical/β-catenin cascade of the Wnt pathway, is upregulated in carotid atherosclerosis and acute myocardial ischemia. It is currently unclear if such upregulation also occurs in cerebral ischemia. We measured plasma levels of Dkk-1 in patients with acute ischemic stroke (n=57) within 24h from symptom onset, in patients with clinically stable cerebrovascular disease (n=29) and in healthy controls (n=29). Stroke severity on admission was determined by the National Institutes of Stroke Scale (NIHSS). The modified Rankin Scale (mRS) served to define outcome at day 90. Ischemic stroke subtype and cause was determined by the Oxfordshire Community Stroke Project (OCSP) criteria and the Causative Classification of Stroke System (CCS). Dkk-1 plasma levels were significantly higher in acute stroke patients (median 727.1 pg/ml) as compared to patients with stable cerebrovascular disease (median 534.2 pg/ml; p=0.017) or healthy controls (median 371.3 pg/ml; p<0.001). The difference of Dkk-1 levels between patients with stable cerebrovascular disease and healthy controls was also significant (p=0.005). No significant differences in Dkk-1 plasma levels were found between different causes or subtypes of ischemic stroke. No correlation of Dkk-1 levels was found with stroke severity on admission and outcome at day 90. Our study provides for the first time evidence for a release of Dkk-1 into the circulation in patients with acute ischemic stroke and also in patients with clinically stable cerebrovascular disease.
    Atherosclerosis 05/2011; 218(1):233-7. DOI:10.1016/j.atherosclerosis.2011.05.015 · 3.99 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Ischemic stroke has a strong familial component to risk. The Siblings With Ischemic Stroke Study (SWISS) is a genome-wide, family-based analysis that included use of imputed genotypes. The Siblings With Ischemic Stroke Study was conducted to examine the associations between single-nucleotide polymorphisms (SNPs) and risk of stroke and stroke subtypes within pairs. The Siblings With Ischemic Stroke Study enrolled 312 probands with ischemic stroke from 70 US and Canadian centers. Affected siblings were ascertained by centers and confirmed by central record review; unaffected siblings were ascertained by telephone contact. Ischemic stroke was subtyped according to Trial of Org 10172 in Acute Stroke Treatment criteria. Genotyping was performed with an Illumina 610 quad array (probands) and an Illumina linkage V array (affected siblings). SNPs were imputed by using 1000 Genomes Project data and MACH software. Family-based association analyses were conducted by using the sibling transmission-disequilibrium test. For all pairs, the correlation of age at stroke within pairs of affected siblings was r=0.83 (95% CI, 0.78-0.86; P<2.2×10(-16)). The correlation did not differ substantially by subtype. The concordance of stroke subtypes among affected pairs was 33.8% (kappa=0.13; P=5.06×10(-4)) and did not differ by age at stroke in the proband. Although no SNP achieved genome-wide significance for risk of ischemic stroke, there was clustering of the most associated SNPs on chromosomes 3p (neuronal nitric oxide synthase) and 6p. Stroke subtype and age at stroke in affected sibling pairs exhibit significant clustering. No individual SNP reached genome-wide significance. However, 2 promising candidate loci were identified, including 1 that contains neuronal nitric oxide synthase, although these risk loci warrant further examination in larger sample collections.
    Stroke 09/2011; 42(10):2726-32. DOI:10.1161/STROKEAHA.111.620484 · 5.72 Impact Factor
Show more