Felipe Torres Pacheco1, César Augusto Pinheiro Ferreira Alves2, Rubens Jose Gagliardi3, Antônio Jose da Rocha2. 1. Division of Neuroradiology, Santa Casa de São Paulo School of Medical Sciences, São Paulo, SP, Brazil. Electronic address: felipetorrespacheco@hotmail.com. 2. Division of Neuroradiology, Santa Casa de São Paulo School of Medical Sciences, São Paulo, SP, Brazil. 3. Division of Neurology, Santa Casa de São Paulo School of Medical Sciences, São Paulo, SP, Brazil.
Abstract
BACKGROUND: More insights in the etiopathogenesis of thrombi could be helpful in the treatment of patients with acute ischemic stroke (AIS). One of the most confident and early imaging findings of stroke includes arterial hyperdensity. The purpose of this study was to determine whether thrombi's density and length would be useful for predicting their origin. METHODS: We evaluated 68 consecutive patients with AIS to correlate the presence of thrombi and their imaging features with the stroke subtype. RESULTS: After excluding patients with small-artery occlusion mechanism and undetermined and other causes, the stroke etiologic subtypes were large-artery atherosclerosis (LAA) in 59.0% of the patients, cardioembolism in 31.0%, and cervical artery dissection (CAD) in 10.0%. CAD more often caused thrombi with the longest length and highest attenuation, while thrombi that originated from the LAA had the smallest length and lowest attenuation. The mean Hounsfield unit (HU) values of all thrombi (with and without hyperdensity) on noncontrast computed tomography were 62.4 (50.0-70.0) in CAD, 53.8 (42.0-65.0) in cardioembolism, and 48.6 (27.0-65.0) in LAA. The length measurements were 28.5 mm (12.0-52.0) in CAD, 13.7 mm (5.0-31.0) in cardioembolism, and 10.8 mm (3.0-25.0) in the LAA subtype. The minimum cutoff value of 60 HU and a length greater than 20 mm were able to discriminate the CAD thrombi with an accuracy of 86.8% and 92.6%, respectively. CONCLUSION: Our study findings show how important thrombus analysis is in patients with AIS. Thrombus analysis can allow early suspicion of CAD before dedicated imaging of the cervical arteries is performed.
BACKGROUND: More insights in the etiopathogenesis of thrombi could be helpful in the treatment of patients with acute ischemic stroke (AIS). One of the most confident and early imaging findings of stroke includes arterial hyperdensity. The purpose of this study was to determine whether thrombi's density and length would be useful for predicting their origin. METHODS: We evaluated 68 consecutive patients with AIS to correlate the presence of thrombi and their imaging features with the stroke subtype. RESULTS: After excluding patients with small-artery occlusion mechanism and undetermined and other causes, the stroke etiologic subtypes were large-artery atherosclerosis (LAA) in 59.0% of the patients, cardioembolism in 31.0%, and cervical artery dissection (CAD) in 10.0%. CAD more often caused thrombi with the longest length and highest attenuation, while thrombi that originated from the LAA had the smallest length and lowest attenuation. The mean Hounsfield unit (HU) values of all thrombi (with and without hyperdensity) on noncontrast computed tomography were 62.4 (50.0-70.0) in CAD, 53.8 (42.0-65.0) in cardioembolism, and 48.6 (27.0-65.0) in LAA. The length measurements were 28.5 mm (12.0-52.0) in CAD, 13.7 mm (5.0-31.0) in cardioembolism, and 10.8 mm (3.0-25.0) in the LAA subtype. The minimum cutoff value of 60 HU and a length greater than 20 mm were able to discriminate the CAD thrombi with an accuracy of 86.8% and 92.6%, respectively. CONCLUSION: Our study findings show how important thrombus analysis is in patients with AIS. Thrombus analysis can allow early suspicion of CAD before dedicated imaging of the cervical arteries is performed.