REVIEW Cerebral venous thrombosis: state of the art diagnosis and management Adam A. Dmytriw 1 & Jin Soo A. Song 1 & Eugene Yu 1 & Colin S. Poon 2 Received: 2 February 2018 /Accepted: 3 May 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract Purpose This review article aims to discuss the pathophysiology, clinical presentation, and neuroimaging of cerebral venous thrombosis (CVT). Different approaches for diagnosis of CVT, including CT/CTV, MRI/MRV, and US will be discussed and the reader will become acquainted with imaging findings as well as limitations of each modality. Lastly, this exhibit will review the standard of care for CVT treatment and emerging endovascular options. Methods A literature search using PubMed and the MEDLINE subengine was completed using the terms Bcerebral venous thrombosis,^ Bstroke,^ and Bimaging.^ Studies reporting on the workup, imaging characteristics, clinical history, and manage- ment of patients with CVT were included. Results The presentation of CVT is often non-specific and requires a high index of clinical suspicion. Signs of CVT on NECT can be divided into indirect signs (edema, parenchymal hemorrhage, subarachnoid hemorrhage, and rarely subdural hematomas) and less commonly direct signs (visualization of dense thrombus within a vein or within the cerebral venous sinuses). Confirmation is performed with CTV, directly demonstrating the thrombus as a filling defect, or MRI/MRV, which also provides superior characterization of parenchymal abnormalities. General pitfalls and anatomic variants will also be discussed. Lastly, endovascular management options including thrombolysis and mechanical thrombectomy are discussed. Conclusions CVT is a relatively uncommon phenomenon and frequently overlooked at initial presentation. Familiarity with imaging features and diagnostic work-up of CVT will help in providing timely diagnosis and therapy which can significantly improve outcome and diminish the risk of acute and long-term complications, optimizing patient care. Keywords Venous stroke . Venous thrombosis . Neurovascular . Algorithm . Review Introduction Background Cerebral venous thrombosis (CVT) is an uncommon condi- tion that accounts for 0.5 to 1% of all strokes in the adult population [1]. It is frequently underdiagnosed, as clinical manifestations are not specific and include a wide range of symptoms such as headache, focal neurological deficits, sei- zures, and altered mental status [2]. Therefore, high clinical suspicion is required to avoid misdiagnosis, and imaging is fundamental in accurately detecting CVT and possible asso- ciated complications that could modify the prognosis and the therapeutic approach [1, 3]. The diagnosis of CVT can be challenging because of the diversity of the underlying risk factors, variable clinical presentation, and non-specific imag- ing findings on initial routine head computed tomography (CT). Prompt diagnosis is essential as it can lead to a more effective treatment, whereas a delayed recognition of this condition is often associated with high morbidity and mor- tality [3, 4]. There are numerous imaging techniques that can be used to detect CVT, such as non-enhanced head CT (NECT), CT ve- nography (CTV), magnetic resonance imaging (MRI), MR venography (MRV), ultrasound (US), and angiography. Angiography was previously considered the gold standard; however, the contemporary utility is often limited to cases where thrombolysis is required, having been largely replaced by CTV, MRI, and MRV as the modern gold standard [2, 3, 5]. Ultimately each imaging modality has different advantages, and specific investigations should be tailored to the clinical * Adam A. Dmytriw adam.dmytriw@uhn.ca 1 Department of Medical Imaging, University of Toronto, 263 McCaul St, 4th Floor, Toronto, ON M5T 1W7, Canada 2 Department of Diagnostic Radiology, Yale University School of Medicine, New Haven, CT, USA Neuroradiology https://doi.org/10.1007/s00234-018-2032-2