The Source for Neurovascular News and Education

May 22, 2019


Moving quickly trumps any additional information obtained with MRI, one expert stresses.


The use of MRI to select stroke patients for endovascular therapy is associated with greater workflow time delays compared with CT angiography (CTA), but it may help reduce rates of symptomatic intracranial hemorrhage (ICH), according to a retrospective registry analysis published recently in Stroke. Clinical outcomes did not differ based on the type of imaging used for selection, but patients who underwent MRI had a more favorable clinical profile at baseline.


Joon-Tae Kim, MD, PhD (Chonnam National University Hospital, Gwangju, South Korea), and colleagues analyzed data from a prospective, nationwide, multicenter stroke registry on 1,265 patients (mean age 69 years; 55% men); median NIHSS score was 15. Comparisons of  workflow timelines between patients selected for endovascular therapy based on MRI versus CTA included the time from symptom onset (last seen well) to arrival at the hospital, decision imaging, puncture, and reperfusion, as well as all points in between.


Overall, 78% of patients were selected for endovascular therapy based on MRI and 22% based on CTA. Perfusion imaging was performed in 67% of all patients and was more frequently performed in the MRI group (79%). Those who underwent MRI had a lower median NIHSS score at baseline and were more likely to be treated with a modern device and have an ASPECTS below 7 than those who underwent CTA, while the CTA group was more likely to receive IV tPA.


Workflow time metrics leading up to imaging were significantly delayed among patients who underwent MRI versus CTA.


Workflow Times by Imaging Modality



(n = 279)


(n = 986)

P Value

Arrival to Imaging, median min (IQR)

21 (15-36)

47 (33-64)

< 0.001

Arrival to Puncture, median min (IQR)

90 (41-94)

109 (44-83)

< 0.001

Arrival to Reperfusion, median min (IQR)

133 (100-185)

167 (133-220)

< 0.001

Imaging to Puncture, median min (IQR)

85 (41-94)

73 (44-83)


Imaging to Reperfusion, median min (IQR)

126 (77.5-153.5)

139 (86-172)



Among the 1,031 patients who arrived within 6 hours of symptom onset, there was no difference between imaging modalities in the rate of 3-month mRS 0-2 (38.1% with MRI vs 38.5% with CTA; P = 0.92). However, the MRI group had lower rates of symptomatic ICH (3.8% vs 7.7%; P = 0.01) and mortality (15.4% vs 20.9%; P = 0.04).


Multivariate analysis revealed that decision imaging was not significantly associated with 3-month functional outcomes ( P > 0.1 for all) or mortality (P = 0.051), but the MRI group remained less likely to develop symptomatic ICH (OR 0.34; 95% CI 0.15-0.77).


“Based on these results,” conclude Kim et al, “it might be suggested that MRI-based selection is not helpful in selecting patients who will most likely benefit from [endovascular therapy], but in selecting patients who are least likely to be harmed by [endovascular therapy]. Further efforts to reduce the acquisition time for decision imaging in [the] early window are warranted.”


Italo Linfante, MD (Herbert Wertheim College of Medicine and Baptist Hospital, Miami, FL), who commented on the findings for Neurovascular Exchange, said large database analyses like this one are important.


Linfante questioned, however, whether the delays caused by MRI negate the modality’s advantage. “There is a big debate now of CT versus MRI,” he said. “The problem is that nowadays, we need to be superfast. MRI is not as fast as CT.”


According to Linfante, the ability to move quickly trumps any additional information obtained with MRI. “The faster we can take patients with blood vessel occlusions to the angio room, the better,” he stressed.



Kim J-T, Cho B-H, Choi K-H, et al. Magnetic resonance imaging versus computed tomography angiography based selection for endovascular therapy in patients with acute ischemic stroke. Stroke. 2019;50:365-372.



Kim and Linfante report no relevant conflicts of interest.