The Comparison In Ischemic Stroke Subtype Classification By TOAST Criteria，A-S-C-O Criteria And CISS Criteria

Objective: The precise analysis of ischemic stroke subtype is important, which is thebasis for epidemiological investigation, risk factor screening, developingindividualized treatment plan and prognostic measures. The TOAST classification isthe most widely used and approved form for etiologic subtyping. The improvement ofbrain and cardio-cerebrovascular imaging technologies and increasing knowledgeabout stroke mechanisms have supported the promotion of the new classificationsystems, which is A-S-C-O phenotypic classification and Chinese ischemic strokesubclassification(CISS). The aim of this study was to investigate the differentiation ofetiologic classification to each subtype by TOAST, A-S-C-O and Chinese ischemicstorke subclassification, as well as description the agreement between them, whichprovide a reference in the clinical application for A-S-C-O phenotypic classificationand Chinese ischemic stroke subclassification.Methods: We assigned167consecutive first-time acute ischemic stroke patients fromthe Second Affiliated Hospital of Nanchang University between March and Octemberin2011. For each, the cause of ischemic stroke was classified according to theTOAST, ASCO and CISS criteria and assigned TOAST group, ASCO group andCISS group. With the different etiologoies, every group was divided into fivesubtypes: atherothrombosis, small artery disease, cardioembolism, other determinedcause and cause undetermined. Compare the similarites and differences between thesubgroups with the three approaches.Results: In167first-ever ischemic stroke patients, ASCO grade1assigned fewerpatients as small-vessel disease(S)(32.9%versus38.3%，relative decrease14.1%,P=0.002), and did not reduce the proportion of cause undetermined cases(34.1%versus28.1%, P=0.04)compared with TOAST. Except the other determined cause,agreement between TOAST and ASCO grade1ranged from good(κ=0.788for largeartery disease) and excellent(κ=0.931for cardio source, κ=0.831for small-vesseldisease). When Chinese Ischemic Stroke Subclassification(CISS) was applied, morepatients were classified as large artery disease(34.1%versus19.8%, relative increase 72.7%, P<0.001), and fewer patients were classified as small-vessel disease(28.1%versus38.3%, relative decrease26.6%, P<0.001). Agreement between TOAST andCISS ranged from moderate(κ=0.560for large artery disease, κ=0.468forsmall-vessel disease) and excellent (κ=0.719for cardiac source, κ=1.000for otherdetermined cause). Application of ASCO grades1to3indicated the proportion ofevery subtypes was increasing compared with ASCO grade1:23.4%increase to77.3%for large artery disease,32.9%increase to87.0%for small-vessel disease,10.2%increase to56.5%for cardiac source,1.2%increase to16.2%.Conclusion: The A-S-C-O classification allows the recording of important additionalinformantion. This may be helpful for a specific treatment adaptation in eachindividual patient and creation of different etiological profiles in view of adaptedclinical trials. The CISS has introduced a classification system that considers bothetiological and pathophysiological causes, it represents an innovative classificationsystem that more faithfully reflects the pathophysiology of stroke，and may beprovided new reference data for future clinical management.