Andrew D Chang1, Gian C Ignacio1, Ronald Akiki1, Brian Mac Grory1, Shawna S Cutting1, Tina Burton1, Mahesh Jayaraman2, Alexander Merkler3, Christopher Song4, Athena Poppas4, Hooman Kamel3, Mitchell S V Elkind5, Karen Furie1, Michael Atalay6, Shadi Yaghi7. 1. The Warren Alpert Medical School of Brown University, Department of Neurology, Providence, Rhode Island. 2. The Warren Alpert Medical School of Brown University, Department of Neurology, Providence, Rhode Island; The Warren Alpert Medical School of Brown University, Department of Neurosurgery, Providence, Rhode Island; The Warren Alpert Medical School of Brown University, Department of Radiology, Providence, Rhode Island. 3. Departments of Neurology and Feil Family Brain and Mind Research Institute, Weill Cornell Medical College, New York, New York. 4. The Warren Alpert Medical School of Brown University, Department of Internal Medicine, Division of Cardiovascular Medicine, Providence, Rhode Island. 5. Department of Neurology, College of Physicians and Surgeons, Columbia University, New York, New York; Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York. 6. The Warren Alpert Medical School of Brown University, Department of Radiology, Providence, Rhode Island. 7. New York Langone Hospital, Department of Neurology, Brooklyn, New York. Electronic address: shadiyaghi@yahoo.com.
Abstract
BACKGROUND AND PURPOSE: While studies have stratified cardioembolic (CE) stroke risk by qualitative left atrial appendage (LAA) morphology and biomarkers of atrial dysfunction, the quantitative properties that underlie these observations are not well established. Accordingly, we hypothesized that LAA volume and contrast density (attenuation) on computerized tomography (CT) may capture the structural and hemodynamic processes that underlie CE stroke risk. METHODS: Data were collected from a single center prospective ischemic stroke database over 18 months and included all patients with ischemic stroke who previously underwent routine, nongated, contrast enhanced thin-slice (≤2.5 mm) chest CT. Stroke subtype was determined based on the inpatient diagnostic evaluation. LAA volume and attenuation were determined from CT studies performed for various clinically appropriate indications. Univariate and multivariable analyses were performed to determine factors associated with ischemic stroke subtype, including known risk factors and biomarkers, as well as LAA density and morphologic measures. RESULTS: We identified 311 patients with a qualifying chest CT (119 CE subtype, 109 Embolic Stroke of Undetermined Source (ESUS), and 83 non-CE). In unadjusted models, there was an association between CE (versus non-CE) stroke subtype and LAA volume (OR per mL increase 1.15, 95% CI 1.07-1.24, P < .001) and LAA density (4th quartile versus 1st quartile; OR 2.95, 95% CI 1.28-6.80, P = .011), but not with ESUS (versus non-CE) subtype. In adjusted models, only the association between LAA density and CE stroke subtype persisted (adjusted OR 3.71, 95% CI 1.37-10.08, P = .010). CONCLUSION: The LAA volume and density values on chest CT are associated with CE stroke subtype but not ESUS subtype. Patients with ESUS and increased LAA volume or attenuation may be a subgroup where the mechanism is CE and anticoagulation can be tested for secondary stroke prevention.
BACKGROUND AND PURPOSE: While studies have stratified cardioembolic (CE) stroke risk by qualitative left atrial appendage (LAA) morphology and biomarkers of atrial dysfunction, the quantitative properties that underlie these observations are not well established. Accordingly, we hypothesized that LAA volume and contrast density (attenuation) on computerized tomography (CT) may capture the structural and hemodynamic processes that underlie CE stroke risk. METHODS: Data were collected from a single center prospective ischemic stroke database over 18 months and included all patients with ischemic stroke who previously underwent routine, nongated, contrast enhanced thin-slice (≤2.5 mm) chest CT. Stroke subtype was determined based on the inpatient diagnostic evaluation. LAA volume and attenuation were determined from CT studies performed for various clinically appropriate indications. Univariate and multivariable analyses were performed to determine factors associated with ischemic stroke subtype, including known risk factors and biomarkers, as well as LAA density and morphologic measures. RESULTS: We identified 311 patients with a qualifying chest CT (119 CE subtype, 109 Embolic Stroke of Undetermined Source (ESUS), and 83 non-CE). In unadjusted models, there was an association between CE (versus non-CE) stroke subtype and LAA volume (OR per mL increase 1.15, 95% CI 1.07-1.24, P < .001) and LAA density (4th quartile versus 1st quartile; OR 2.95, 95% CI 1.28-6.80, P = .011), but not with ESUS (versus non-CE) subtype. In adjusted models, only the association between LAA density and CE stroke subtype persisted (adjusted OR 3.71, 95% CI 1.37-10.08, P = .010). CONCLUSION: The LAA volume and density values on chest CT are associated with CE stroke subtype but not ESUS subtype. Patients with ESUS and increased LAA volume or attenuation may be a subgroup where the mechanism is CE and anticoagulation can be tested for secondary stroke prevention.