Correlation between angioarchitectural characteristics of brain arteriovenous malformations and clinical presentation of 183 patients

• Autor(a) principal: Batista,Ulysses Caus
• Data de Publicação: 2022
• Outros Autores: Pereira,Benedito Jamilson Araujo, Joaquim,Andrei Fernandes, Tedeschi,Helder, Piske,Ronie Leo
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Arquivos de neuro-psiquiatria (Online)
• Texto Completo: http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0004-282X2022000100003
• topic: Central Nervous System Vascular Malformations; Intracranial Arteriovenous Malformations; Seizures; Stroke

ABSTRACT Background: The correlation between angioarchitecture and clinical presentation of brain arteriovenous malformation (bAVM) remains a subject of debate. Objective: The main purpose of the present study was to assess the correlation between angioarchitectural characteristics of bAVM and clinical presentation. Methods: A retrospective review of all consecutive patients presenting a bAVM who underwent a cerebral angiography at Beneficencia Portuguesa Hospital in São Paulo between January 2006 and October 2016 was carried out. Patients were divided in five groups: group 1 - hemorrhage; group 2 - seizure; group 3 - headache; group 4 - progressive neurological deficits (PND); group 5 - incidental). Results: A total of 183 patients were included, with group 1 comprising 56 cases, group 2 49 cases, group 3 41 cases, group 4 28 cases, and group 5 9 cases. Regarding hemorrhage presentation, a statistical correlation was observed with female gender (P < 0.02), Spetzler-Martin 3B (P < .0015), and lesions with low flow (P < 0.04). A positive association was found between group 2 and age less than 36 years (P < 0.001), male sex (P < 0.018), presence of superficial lesions not classified as SM 3B (P < 0.002), presence of venous ectasia (p <0.03), and arterial steal phenomenon (P < 0.03). Group 4 was associated with older age (P < 0.01). Conclusions: Angioarchitectural characteristics can be correlated with some clinical presentations as well as with some clinical data, making it possible to create predictive models to differentiate clinical presentations.