TOAST Classification System for Differentiating Cardioembolic from Large-Artery Atherosclerotic Stroke
The TOAST (Trial of ORG 10172 in Acute Stroke Treatment) classification system is the standard scoring system used to differentiate cardioembolic from large-artery atherosclerotic infarcts in adults with acute ischemic stroke. 1
Overview of TOAST Classification
The TOAST system categorizes ischemic strokes into five distinct subtypes based on clinical features and diagnostic test results: (1) large-artery atherosclerosis, (2) cardioembolism, (3) small-vessel occlusion, (4) stroke of other determined etiology, and (5) stroke of undetermined etiology. 1 This classification system was specifically designed to allow investigators to report treatment responses among important subgroups of patients with ischemic stroke and has demonstrated high interobserver agreement. 1
Diagnostic Criteria for Large-Artery Atherosclerosis
Large-artery atherosclerosis requires imaging evidence of ≥50% stenosis or occlusion of a major cerebral artery or branch cortical artery due to atherosclerosis. 2, 3 Key diagnostic features include:
- Cortical infarcts in the distribution of a large cerebral artery on brain imaging 3
- Vascular imaging (CT angiography, MR angiography, carotid duplex ultrasound, or transcranial Doppler) demonstrating significant stenosis 2
- Often preceded by TIAs in the same arterial distribution 3
- Most common subtype in anterior cerebral, middle cerebral, vertebral, and anterior/posterior inferior cerebellar artery territories 4
Diagnostic Criteria for Cardioembolic Stroke
Cardioembolism requires identification of a high-risk cardiac source on ECG, rhythm monitoring, or echocardiography. 2 Distinguishing features include:
- Cortical or large subcortical infarctions on brain imaging 3
- Identifiable high-risk cardiac source (atrial fibrillation being most common) 3
- Associated with the highest mortality rate among ischemic stroke subtypes 3
- Most common cause in superior cerebellar artery territory 4
- Frequently causes multiple-vascular-territory infarction in carotid territory (44.2% of cases) 4
Required Diagnostic Workup
To accurately apply TOAST classification, a comprehensive evaluation must include:
Imaging Studies
- Non-contrast CT or MRI to confirm acute ischemic stroke and exclude hemorrhage 2
- Cervical and intracranial vessel imaging (CTA, MRA, carotid duplex, or transcranial Doppler) to identify large-artery stenosis 2
- Brain imaging should be completed within 25 minutes of ED arrival for thrombolytic candidates 2
Cardiac Evaluation
- 12-lead electrocardiogram to detect atrial fibrillation and arrhythmias 2
- Continuous cardiac rhythm monitoring (telemetry or Holter) for paroxysmal atrial fibrillation 2
- Transthoracic echocardiography at minimum to assess cardioembolic sources 5
- Cardiac troponin measurement to assess concurrent myocardial ischemia 2
Laboratory Studies
- Complete blood count with platelet count 2
- Prothrombin time/INR and activated partial thromboplastin time 2
- Serum electrolytes and renal function tests 2
- Blood glucose to rule out hypoglycemic mimics 2
- Fasting lipid profile and hemoglobin A1c 2, 5
Critical Pitfalls and Clinical Caveats
A major limitation is that clinical practice TOAST diagnoses are accurate in only 61% of patients when compared to expert reference standards, with particularly poor sensitivity for large-artery atherosclerosis (33%) and positive predictive value for multiple subtypes. 6 This suboptimal performance occurs despite the system's theoretical high interobserver agreement. 6
Misclassification has direct therapeutic implications: A patient with atrial fibrillation and a small subcortical infarct requires anticoagulation (cardioembolic mechanism), not just antiplatelet therapy (small vessel disease). 5 The clinical syndrome alone is insufficient—imaging confirmation of infarct size and location plus exclusion of alternative causes is mandatory. 5
Do not rely solely on clinical features without completing the full diagnostic workup, as stroke diagnostics have advanced substantially since TOAST was designed 30 years ago. 6 The American Heart Association emphasizes that accurate classification requires comprehensive diagnostic testing including both brain imaging and vascular studies to guide treatment approaches. 3
Emerging Adjunctive Tools
The BOCS2 scale, which combines TOAST classification with angiographic occlusion type patterns and histopathological findings from extracted thrombi, has shown 93.5% sensitivity and 100% specificity for identifying cardioembolic stroke in patients with undetermined etiology. 7 This represents a potential adjunctive diagnostic tool when mechanical thrombectomy is performed. 7