Elsevier

Academic Radiology

Volume 16, Issue 3, March 2009, Pages 305-312
Academic Radiology

Original investigation
Diffusion-weighted Imaging in Ischemic Stroke: Effect of Display Method on Observers' Diagnostic Performance

https://doi.org/10.1016/j.acra.2008.09.012Get rights and content

Rationale and Objectives

When evaluating ischemic stroke on diffusion-weighted magnetic resonance imaging (DWI), the display method has not been investigated. The purpose of this study was to determine whether standardization of the display method for DWI affects observers' diagnostic performance in detecting ischemic stroke on DWI.

Materials and Methods

Twenty-six observers evaluated 40 DWI studies in 20 patients with acute (<6 hours) middle cerebral arterial strokes and 20 controls for the presence of hyperintense lesions in 10 areas using the Alberta Stroke Programme Early CT Score (ASPECTS) system and one area in the corona radiata using a modified version of the ASPECTS system (ASPECTS-DWI). The images were reviewed using a standardized display method (SDM) and a conventional display method (CDM). The reading time was recorded for each session. The observers' performance was evaluated with receiver-operating characteristic analysis.

Results

In all observers with ASPECTS-DWI scores of ≤8 points, the value of the mean average area under the receiver-operating characteristic curve was slightly higher for the SDM than the CDM, but the difference was not statistically significant. In the insular ribbon, diagnostic accuracy was significantly higher with the SDM than the CDM (P = .036). In the other locations, there were no significant differences. With the SDM, the mean reading time was reduced by 7.5 seconds (P = .024).

Conclusion

The SDM improved diagnostic accuracy for the insular ribbon and shortened the reading time, although it did not improve observers' performance with the ASPECTS-DWI system.

Section snippets

Selection of Patients and Controls

A total of 30 magnetic resonance imaging (MRI) studies on 30 patients who underwent MRI between January 2001 and July 2003 were collected from seven Japanese centers for this retrospective study of the effect of the DWI display method on the diagnosis of acute cerebral infarction. The inclusion criteria were acquisition of MRI scans within 6 hours of ischemic stroke onset; an age range of 40 to 85 years; the use of a 1.5-T MRI scanner featuring a head coil, echo-planar capability, and identical

Diagnostic Performance

With respect to the detection of a modified ASPECTS score of ≤8 points, we compared the mean Az values obtained with the CDM and the SDM (Table 1). In all observers, the mean Az value was slightly higher with the SDM than the CDM (0.979 ± 0.026 vs 0.971 ± 0.037); the difference was not statistically significant (P = .33) (Fig 4). There was no significant difference in the mean Az values obtained with the CDM and the SDM by readers grouped according to their specialties. On the other hand,

Discussion

The diagnostic accuracy of the insular ribbon was significantly increased when the images were displayed with the SDM. The highest number of false-positive assessments per case per session was recorded for the insular ribbon. On DWI, the brain cortices manifest physiologic regional signal variation; in healthy subjects, the insular and the cingulate cortices show higher SI than the other cerebral cortices (17). This phenomenon is probably due to T2 shine-through (18). In addition, DWI display

Acknowledgments

We are grateful to Shoji Morishita, Yoshiko Hayashida, Ichiro Ikushima, Yoshinori Shigematsu, Mika Kitajima, Tomoko Okuda, Masayuki Yamura, Koichi Kawanaka, Yoshiharu Nakayama, Masanori Imuta, Takanori Negishi, Shinichiro Kusunoki, Shinya Shiraishi, Masaki Watanabe, Tomoo Hirahara, Sadahisa Okamoto, Akihiko Ueda, Motohiro Morioka, Keishi Makino, Hideo Takeshima, and Shigetoshi Yano for participating in our observer performance study. We gratefully thank the investigators and collaborators of

References (21)

  • P.A. Barber et al.

    Validity and reliability of a quantitative computed tomography score in predicting outcome of hyperacute stroke before thrombolytic therapy

    Lancet

    (2000)
  • M.G. Lansberg et al.

    Comparison of diffusion-weighted MRI and CT in acute stroke

    Neurology

    (2000)
  • J.B. Fiebach et al.

    CT and diffusion-weighted MR imaging in randomized order: diffusion-weighted imaging results in higher accuracy and lower interrater variability in the diagnosis of hyperacute ischemic stroke

    Stroke

    (2002)
  • D. Saur et al.

    Sensitivity and interrater agreement of CT and diffusion-weighted MR imaging in hyperacute stroke

    AJNR Am J Neuroradiol

    (2003)
  • W. Hacke et al.

    Association of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials

    Lancet

    (2004)
  • W. Hacke et al.

    The Desmoteplase in Acute Ischemic Stroke Trial (DIAS): a phase II MRI-based 9-hour window acute stroke thrombolysis trial with intravenous desmoteplase

    Stroke

    (2005)
  • A.J. Furlan et al.

    Dose Escalation of Desmoteplase for Acute Ischemic Stroke (DEDAS): evidence of safety and efficacy 3 to 9 hours after stroke onset

    Stroke

    (2006)
  • A. Dávalos et al.

    The clinical-DWI mismatch. A new diagnostic approach to the brain tissue at risk of infarction

    Neurology

    (2004)
  • J. Prosser et al.

    Clinical-diffusion mismatch predicts the putative penumbra with high specificity

    Stroke

    (2005)
  • H. Tei et al.

    Clinical-diffusion mismatch defined by NIHSS and ASPECTS in non-lacunar anterior circulation infarction

    J Neurol

    (2007)
There are more references available in the full text version of this article.

Cited by (27)

  • Endovascular Therapy for Acute Ischemic Stroke in Patients with Large-Vessel Occlusion due to Atherosclerotic Stenosis

    2021, Journal of Stroke and Cerebrovascular Diseases
    Citation Excerpt :

    The data obtained from the patients’ records included age, sex, smoking history, mRS before stroke onset, NIHSS score, time from onset to door (of the hospital), and use of intravenous recombinant tissue plasminogen activator (IV-tPA). The extent of infarction was evaluated using the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) on non-contrast computed tomography (CT) images or using diffusion weighted image (DWI)-APECTS on diffusion-weighted magnetic resonance imaging (MRI) scans.13 ASPECTS ranges from 0 to 10, with higher scores representing smaller early ischemic change.

  • FLAIR vascular hyperintensity with DWI for regional collateral flow and tissue fate in recanalized acute middle cerebral artery occlusion

    2021, European Journal of Radiology
    Citation Excerpt :

    Patients with simultaneous symptomatic acute ischemic lesions in areas outside of occluded M1 were excluded. In the presence of neurological deficit, patients with a score ≥6 on the DWI-Alberta stroke program early CT score (ASPECTS) [16] were selected to receive MT < 8 h from known symptom onset or <24 h from last known well time if the time of symptom onset was unknown. IV-tPA was administered within 4.5 h of symptom onset according to the Japanese Guideline for the Management of Stroke [17].

  • C-reactive Protein is A Predictor of Deterioration of Acute Internal Carotid Artery M1 Qcclusion Following Recanalization

    2020, Journal of Stroke and Cerebrovascular Diseases
    Citation Excerpt :

    In order to identify early recanalization in patients with large artery occlusion and hemorrhagic cerebral infarction within 24 hours, follow-up MRI was performed 24 hours after t-PA infusion, including DWI, FLAIR images, T2-weighted imaging, and MRA. The DWI-ASPECTS score was evaluated from high-intensity areas on DWI images taken at the time of hospitalization.5 The Susceptibility Vessel Sign (SVS) was defined as low-intensity on MRI T2-weighted images at the time of hospitalization.6

View all citing articles on Scopus

Departments of Diagnostic Radiology (T. Hirai, H.F., A.S., Y.Y.), Neurology (T. Hirano), Neurosurgery (Y.K.), and Radiation Oncology (R.M.), Graduate School of Medical Sciences, and Kumamoto University Hospital (R.I.), Kumamoto University, 1-1-1 Honjo, Kumamoto 860-8556, Japan; the Advanced Medical Research Center, Iwate Medical University, Uchimaru, Morioka, Japan (M. Sasaki); the Department of Radiology, Mie University School of Medicine, Tsu, Mie, Japan (M.M.); the Department of Radiology, Ebara Hospital, Tokyo, Japan (M.I.); the Department of Radiological Technology, Kumamoto University School of Health Sciences, Kumamoto, Japan (S. Katsuragawa); the Department of Rehabilitation, Hatsudai Rehabilitation Hospital, Tokyo, Japan (M. Sakoh); the Department of Radiology, Fukuoka University, School of Medicine, Fukuoka, Japan (K.T., S.A.); and the Department of Radiology, University of Occupational and Environmental Health, School of Medicine, Kitakyushu, Japan (S. Kakeda). This work was partly supported by a Grant for Standardization Projects from the Japanese Society for Magnetic Resonance in Medicine and by a Research Grant for Cardiovascular Diseases (17C-3) from the Ministry of Heath, Labour, and Welfare of Japan

View full text