Search

Objectives Treatment and outcomes of acute stroke have been revolutionised by mechanical thrombectomy. Deep learning has shown great promise in diagnostics but applications in video and interventional radiology lag behind. We aimed to develop a model that takes as input digital subtraction angiography (DSA) videos and classifies the video according to (1) the presence of large vessel occlusion (LVO), (2) the location of the occlusion, and (3) the efficacy of reperfusion. Methods All patients who underwent DSA for anterior circulation acute ischaemic stroke between 2012 and 2019 were included. Consecutive normal studies were included to balance classes. An external validation (EV) dataset was collected from another institution. The trained model was also used on DSA videos post mechanical thrombectomy to assess thrombectomy efficacy. Results In total, 1024 videos comprising 287 patients were included (44 for EV). Occlusion identification was achieved with 100% sensitivity and 91.67% specificity (EV 91.30% and 81.82%). Accuracy of location classification was 71% for ICA, 84% for M1, and 78% for M2 occlusions (EV 73, 25, and 50%). For post-thrombectomy DSA (n = 194), the model identified successful reperfusion with 100%, 88%, and 35% for ICA, M1, and M2 occlusion (EV 89, 88, and 60%). The model could also perform classification of post-intervention videos as mTICI Conclusions Our model can successfully identify normal DSA studies from those with LVO and classify thrombectomy outcome and solve a clinical radiology problem with two temporal elements (dynamic video and pre and post intervention). Key Points • DEEP MOVEMENT represents a novel application of a model applied to acute stroke imaging to handle two types of temporal complexity, dynamic video and pre and post intervention. • The model takes as an input digital subtraction angiograms of the anterior cerebral circulation and classifies according to (1) the presence or absence of large vessel occlusion, (2) the location of the occlusion, and (3) the efficacy of thrombectomy. • Potential clinical utility lies in providing decision support via rapid interpretation (pre thrombectomy) and automated objective gradation of thrombectomy outcomes (post thrombectomy).

The last 10 years have seen a major shift in management of large vessel ischemic stroke with changes towards ever-expanding use of reperfusion therapies (intravenous thrombolysis and mechanical thrombectomy). These strategies ‘open the door’ to acute therapeutics for ischemic tissue, and we should investigate novel therapeutic approaches to enhance survival of recently reperfused brain. Key insights into new approaches have been provided through translational research models and preclinical paradigms, and through detailed research on ischemic mechanisms. Additional recent clinical trials offer exciting salvos into this new strategy of pairing reperfusion with neuroprotective therapy. This pairing strategy can be employed using drugs that have shown neuroprotective efficacy; neurointerventionalists can administer these during or immediately after reperfusion therapy. This represents a crucial moment when we emphasize reperfusion, and have the technological capability along with the clinical trial experience to lead the way in multiprong approaches to stroke treatment.

Background Endovascular treatment of large complex morphology aneurysms is challenging. High recanalization rates have been reported with techniques such as stent-assisted coiling and balloon-assisted coiling. Flow diverter devices have been introduced to improve efficacy outcomes and recanalization rates. Thromboembolic complications and in-device stenosis are certainly more worrisome when treatment of bilateral internal carotid arteries has been performed. This study aimed to report our experience with mid-term imaging follow-up of staged bilateral Pipeline embolization device placement for the treatment of bilateral internal carotid artery aneurysms. Methods We reviewed the clinical, angiographic, and follow-up imaging data in all consecutive patients treated with bilateral internal carotid artery aneurysms who underwent elective Pipeline embolization. Results Six female patients were treated, harboring a total of 13 aneurysms. Of these, 60% were asymptomatic. Diplopia and headache were the most common symptoms. The most common location was the paraclinoid segment (6/13), including by cavernous segment (4/13) and ophthalmic segment (2/13). Successful delivery of the device was achieved in 12 cases. Difficult distal access precluded the deployment of the device in one case. The treatment was always staged with at least eight weeks' difference between the two procedures. All aneurysm necks were covered completely. There were no periprocedural complications. Angiographic follow-up ranged between 3 and 12 months, and computed tomography angiogram follow-up ranged between 2 and 24 months. Complete aneurysm occlusion was achieved in all cases. Conclusion In our series, Pipeline deployment for the treatment of bilateral internal carotid artery aneurysms in a staged fashion is safe and feasible. Mid-term imaging follow-up showed permanent occlusion of all the treated aneurysms.

Purpose of reviewSpinal dural arteriovenous fistulas (SDAVFs) are abnormal connections between 1 or more radiculomeningeal arteries and a single radiculomedullary vein draining into the perimedullary venous system. SDAVFs present in older patients with a progressive myelopathy caused by diffuse spinal venous hypertension. The discrepancy between the focal nature of the arteriovenous shunt and the extent of the induced myelopathy is a classic feature of SDAVFs related to the coexistence of diffuse spinal venous drainage impairment.Recent findingsWe describe 3 cases of cervical SDAVFs (at C1, C4, and C7) presenting with a myelopathy that spared the cervical spinal cord and, in 2 instances, the upper thoracic cord. This is to our knowledge the first observations of cervical SDAVFs with MRI showing absent or subtle flow voids and presenting remote thoracolumbar myelopathy without cervical cord involvement.SummaryA considerable distance may separate low-flow spinal arteriovenous fistulas from the spinal cord damage they produce. These observations emphasize the importance of performing a complete spinal angiogram when investigating a vascular myelopathy of any location and extent.

Traumatic arteriovenous fistulas (AVFs) involving the middle meningeal artery (MMA) are uncommon lesions and rare in the pediatric population. An adolescent with headaches developed a bruit and a rhythmic “swooshing” noise in the left ear 9 weeks after a traumatic brain injury involving a left temporal bone fracture. An MRA of the brain demonstrated an enlarged left MMA and arterialized flow in the left middle meningeal vein (MMV), suggestive of an AVF. Cerebral angiography confirmed and characterized the left MMA-MMV fistula, which drained into the pterygoid plexus. The patient underwent transarterial coil embolization, with successful obliteration of the fistulous connection, and her tinnitus resolved. This case illustrates the importance of recognizing a post-traumatic AVF in a pediatric patient with pulsatile tinnitus, and that endovascular treatment by coil embolization is a safe and effective treatment option.

PURPOSE: To evaluate the impact of craniospinal irradiation (CSI) dose and molecular sub-type on medulloblastoma relapse rates. METHODS: 115 medulloblastoma patients were treated with 18- ≥ 36 Gy CSI. Molecular subtyping data was done by nanostring assay [WNT, SHH, and Group 3/4 non-WNT/non-SHH (NWNS)]. A CSI dose schema based on prognostic implications of molecular subtype was devised: ≥36 Gy for M+ or Group 3; ≤18 Gy for non-disseminated WNT; 23.4 Gy all others. Relapse rates were assessed and correlated with CSI dose. RESULTS: After excluding patients with incomplete data, the final cohort included 36 subjects with median age 8.7 years (range 3.3–16.9) and median follow-up 4.1 years (0.2-3.7). Molecular subtypes were as follows: 2-WNT, 6-SHH, 28-NWNS. CSI dose was 18 Gy(11 cases), 23.4 Gy(9), and ≥36 Gy(16). CSI dose was divergent from the proposed schema in 18 cases (14-lower, 4-higher). There were 14 relapses (by molecular subtype: 0-WNT, 1-SHH, 13-NWNS; by histology: 8-classic, 5-LC/A, 1-EN), with median time to relapse 18.5 months (0.43-62). Seven of 14 tumors treated to a lower dose relapsed (1 SHH; 6 NWNS); only 1 (NWNS) of 4 patients with higher divergent dose relapsed. Six of 18 patients treated to similar dose relapsed; all were NWNS, and 4/6 had M3 disease. Eleven of 14 patients who relapsed died. CONCLUSION: Molecular subtyping may help avoid under-treating some NWNS tumors but innovative strategies for metastatic NWNS are needed. We confirm that histological subtypes may hold some predictive value as 5/7 LC/A relapsed and no desmoplastic tumors relapsed. In this small sample, no WNT relapse occurred and only 1 SHH relapsed, suggesting that dose de-escalation can be cautiously explored for these subtypes. Future research directions include applying an MRI image-based algorithm to subtype the entire cohort, and combining molecular data with other institutions to examine relapse patterns by subtype.