Article Text

Download PDFPDF
P-023 MR Diffusion for Stroke Triage: When is it most useful?
  1. F Hui,
  2. D Wisco,
  3. N Obuchowski,
  4. E Cheng-Ching,
  5. K Uchino,
  6. S Hussain
  1. Cerebrovascular Center, Cleveland Clinic, Cleveland, OH


Introduction MR Diffusion Weighted Imaging is known to be very sensitive to acute ischaemic change and probably represents one of the best clinically available methods to estimate the size of an ischaemic core. CT is more widely available, is faster, but may be relatively insensitive.

Given that sensitivities vary even as ischaemic core is changing, there may be periods during which MR and CT may agree more or less as the infarct progresses.

Methods The institutional database for MRI was queried to select patients that underwent MRI and CT during the triage process for acute ischaemic stroke. Patient data was collected and CT and MR ASPECTS scores were assessed for agreement and variance as a function of time from symptom onset.

Results 71 patients were found with complete data and having MRI and CT within 60 minutes. 32 females (45.1%), mean age 70.7 years (range26–95). The mean time of MRI after LSW was 362.5 minutes (range 72 minutes to 22.3hours). The mean time between CT and MRI 38.2 min (range 4–60 minutes, 91% > 20 minutes.

When using a theoretical discriminator of ASPECTS score 8 or greater for intervention, we found that MR and CT had a higher frequency of disagreement among patients with MRIs 3–6 hours after LKW compared with patients 6 hours (29.3% vs 10%, p=0.049). In patients with CT ASPECTS 7 or less, there were no disagreements with MR Diffusion. In patients with CT ASPECTS 8 or greater, there were 14/41 disagreements (34.2%) p<0.001.

Conclusions Based on this data, there is some cause to believe that MRI may be more useful during an intermediate period and less useful early in the time course of a stroke as well as in the late course of an acute stroke.

Abstract P-023 Table 1 Frequency of Disagreement in patient Management Using Cutpoint of 7
Abstract P-023 Figure 1

Difference between CT and MR ASPECTS

Disclosures F. Hui: 3; C; Penumbra, Microvention. D. Wisco: None. N. Obuchowski: None. E. Cheng-Ching: None. K. Uchino: None. S. Hussain: None.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.