Stroke mimic diagnoses presenting to a hyperacute stroke unit

Editor – The study that showed that access to magnetic resonance imaging (MRI) facilitated the diagnosis and management of stroke mimics¹ strikes a parallel with another study evaluating MRI after a negative computerised tomography (CT) scan.² In that study, as many as 11.5% of patients who had a non-diagnostic CT scan after presenting with atypical stroke symptoms were subsequently shown to have subacute infarcts when MRI was performed within 24 hours of the negative CT scan.² These observations are strong arguments for an MRI-first policy along the lines demonstrated in a recently reported ‘real world’ study.³ In that study, among 314 patients with suspected stroke who were screened solely by MRI, 73 proceeded to intravenous thrombolysis. Thanks to a concurrent quality improvement (QI) strategy, door-to-needle time (DNT) significantly improved during the three phases of implementation of the MRI-first policy such that median DNT amounted to 83 minutes, 68 minutes and 54 minutes in phases I, II and III, respectively (p<0.001). Exceptions to MRI-first included contraindications to MRI (one patient) and poor general condition (two patients).

The QI process included pre-notification by the emergency medical service, limiting the MRI sequence and introduction of a rapid examination tool.³ The disadvantage of MRI being more time consuming than CT scanning can be mitigated by applying novel strategies to reduce the time taken by high computation processes, such as diffusion-weighted imaging. High angular resolution diffusion imaging is one such strategy. In theory, this strategy is capable of reducing computation time by about 50% without any reduction in result quality.⁴

Conflicts of interest

The author has no conflicts of interest to declare.