The Trainee Day preceding the joint BSNR/ISNR annual meeting in Dublin was held at the Beaumont Hospital, and, drawing on the excellent interventional service based there, was focused on neurovascular pathology. Below is a summary of the day with some of the learning points I took away and useful references – any errors are mine rather than the speakers!
The day began with a comprehensive overview of current evidence in stroke imaging by Dr Grant Mair, specifically looking at the recent trials and the appropriate imaging modalities for patient selection in thrombectomy. Advanced imaging techniques were discussed, particularly 4D CTA (useful, for example, in differentiating occlusion and high grade stenosis) and CT perfusion; although there does not seem to be a clear consensus on quantification or which of the many perfusion parameters to consider, practically cerebral blood volume (CBV) is the most useful. He also highlighted the recent article from the HERMES collaborators looking at baseline imaging characteristics and response to thrombectomy, particularly the fact that infarcts with all ASPECTS values may have response to intervention. Another key point for me was the potential value of dual energy CT to better visualise early ischaemic change and to remove contrast to differentiate between contrast staining and haemorrhage, which can be difficult in practice.
Next was Dr Tilak Das on parenchymal haemorrhage, again taking a usefully practical approach to the investigation and causes of haemorrhage. He highlighted the prognostic utility of measuring haematoma volume (with >30cc a cutoff over which patients are unlikely to be functionally independent, albeit based on data from the 1990s), and imaging features to predict haematoma expansion (margin irregularity, fluid-fluid levels, heterogeneity and the spot sign – for more on this see the April journal round-up). Regarding causes, in young children the vast majority will have a vascular malformation, and in adults a secondary cause is more likely to be found with lobar haemorrhage (particularly with intraventricular involvement). Specific causes discussed included AVMs (for which the Spetzler-Martin classification provides a useful checklist of things to report), cerebral amyloid angiopathy, infarcts, and tumours (which should be considered if the haematoma has an irregular margin, too much oedema or too much enhancement).
Dr Stavros Stivaros covered paediatric vascular disease, highlighting the difference in mechanism and outcome in vascular injury between preterm and term infants, and within term infants whether the injury is prolonged, with preservation of metabolically more active areas, or acute and profound in which they are affected early. The CASCADE criteria are a useful framework for thinking about the potential causes of paediatric stroke. Infection is an important cause of stroke in older children, particularly HSV or VZV, and initial vascular imaging may be normal but CSF PCR should be performed in children with embolic infarcts, and perfusion imaging is often essential to identify occult hypoperfusion in the setting of vasculopathy. The importance of identifying features suggesting underlying genetic disorders was also stressed.
The workshops built on the themes of the lectures, covering thrombectomy case selection, patterns of subarachnoid haemorrhage and investigation pathways, vasculopathies and traumatic injuries with a great selection of interesting cases.
Dr Seamus Looby talked about subarachnoid haemorrhage, discussing the need to screen certain groups with MRA (for example those with two or more first degree relatives with SAH and adults with polycystic kidney disease) and the sensitivity of imaging compared to CSF analysis (xanthochromia is present from 6 hours to 2-3 weeks). Appropriate imaging strategies depend on the distribution of blood (a useful reference with flowcharts is available here). Vasospasm occurs typically between 3 and 14 days of SAH, peaking at 7-10 days, and only 40-70% of patients will have imaging evidence, while conversely those with imaging evidence may not be symptomatic. When considering monitoring of unruptured aneurysms, surveillance should be balanced with the available evidence on risk of rupture.
In the final lecture, Dr Sarah Power reviewed vasculopathies, dividing them into inflammatory and non-inflammatory causes. Useful classifications include the 2012 Chapel Hill consensus classification (available here as part of an excellent wider review of CNS vasculitis) which uses the size of the vessel but also considers associated pathologies (i.e. systemic disease or iatrogenic causes). As neuroradiologists we need to be confident with the entity of primary angiitis of the central nervous system (a helpful imaging study of PACNS is available here). In imaging terms it is useful to consider the brain, the vessel lumen and the vessel wall in turn to come to a diagnosis. Information regarding cause can be inferred from the size, timing and distribution of infarcts, the presence of haemorrhage, white matter lesions and presence of enhancement (particularly around infarcts, perivascular or leptomeningeal). Luminal abnormalities on angiographic imaging demonstrate the extent and distribution of the vasculopathy and vessel wall imaging can differentiate vasculitis and atherosclerotic disease (see the March journal round-up for more on VWI).
Overall it was a very educational day and our thanks go to Dr Power, the organiser, and all of the speakers and workshop leaders for their work. The next trainee days are currently being planned for spring 2019 – hope to see you there!
British Society of Neuroradiologists - Trainee Blog 