Doris Dodig1, Zrinka Matana Kaštelan1,2, Nina Bartolović1, Slaven Jurković3,4, Damir Miletić1,2, Zoran Rumboldt2. 1. Radiology Department, Clinical Hospital Centre Rijeka, Croatia. 2. Department of Radiology, University of Rijeka, Croatia. 3. Department of Medical Physics and Biophysics, University of Rijeka, Croatia. 4. Department for Medical Physics and Radiation Protection, Clinical Hospital Centre Rijeka, Croatia.
Abstract
BACKGROUND: Virtual monoenergetic (VM) dual-energy computed tomography (DE-CT) enables grey-to-white matter contrast-to-noise ratio optimization, potentially increasing ischaemic brain oedema visibility. The aim of this study was to compare the diagnostic accuracy of VM and standard DE-CT reconstructions for early stroke detection. METHODS: Consecutive patients with non-contrast DE-CT of the brain scanned within 12 h of stroke symptom onset were prospectively included in the study. Patients with other significant brain pathology were excluded. Two radiologists jointly evaluated standard and VM reconstructions (from 40 to 190 keV at increments of 10 keV) for early stroke signs on a four-point Likert scale: (a) stroke definitely present, (b) stroke probably present, (c) probably no stroke, and (d) definitely no stroke. Follow-up imaging and clinical data served as the standard of reference. Diagnostic accuracy was evaluated by receiver operating characteristic analysis. RESULTS: Stroke incidence among 184 patients was 76%. In 64 patients follow-up imaging served as the standard of reference: ischemic brain oedema detection was significantly more accurate on VM reconstructions at 80 keV compared with standard DE-CT reconstructions (area under the curve (AUC) = 0.821 vs. AUC = 0.672, p = 0.002). The difference was most prominent within the first 3 h after symptom onset (at 11%, AUC = 0.819 vs. AUC = 0.709, p = 0.17) and in patients with National Institutes of Health Stroke Scale above 16 (at 37.5%, AUC = 1 vs. AUC = 0.625, p = 0.14). CONCLUSION: VM DE-CT reconstructions at 80 keV appear to be the optimal non-contrast CT technique for diagnosing early ischaemic stroke, particularly within the first 3 h after symptom onset and in severely ill patients.
BACKGROUND: Virtual monoenergetic (VM) dual-energy computed tomography (DE-CT) enables grey-to-white matter contrast-to-noise ratio optimization, potentially increasing ischaemic brain oedema visibility. The aim of this study was to compare the diagnostic accuracy of VM and standard DE-CT reconstructions for early stroke detection. METHODS: Consecutive patients with non-contrast DE-CT of the brain scanned within 12 h of stroke symptom onset were prospectively included in the study. Patients with other significant brain pathology were excluded. Two radiologists jointly evaluated standard and VM reconstructions (from 40 to 190 keV at increments of 10 keV) for early stroke signs on a four-point Likert scale: (a) stroke definitely present, (b) stroke probably present, (c) probably no stroke, and (d) definitely no stroke. Follow-up imaging and clinical data served as the standard of reference. Diagnostic accuracy was evaluated by receiver operating characteristic analysis. RESULTS: Stroke incidence among 184 patients was 76%. In 64 patients follow-up imaging served as the standard of reference: ischemic brain oedema detection was significantly more accurate on VM reconstructions at 80 keV compared with standard DE-CT reconstructions (area under the curve (AUC) = 0.821 vs. AUC = 0.672, p = 0.002). The difference was most prominent within the first 3 h after symptom onset (at 11%, AUC = 0.819 vs. AUC = 0.709, p = 0.17) and in patients with National Institutes of Health Stroke Scale above 16 (at 37.5%, AUC = 1 vs. AUC = 0.625, p = 0.14). CONCLUSION: VM DE-CT reconstructions at 80 keV appear to be the optimal non-contrast CT technique for diagnosing early ischaemic stroke, particularly within the first 3 h after symptom onset and in severely ill patients.
Entities:
Keywords:
Stroke; dual energy CT; virtual monoenergetic imaging
Authors: Victor Neuhaus; Nuran Abdullayev; Nils Große Hokamp; Gregor Pahn; Christoph Kabbasch; Anastasios Mpotsaris; David Maintz; Jan Borggrefe Journal: Invest Radiol Date: 2017-08 Impact factor: 6.016
Authors: Gregory W Albers; Michael P Marks; Stephanie Kemp; Soren Christensen; Jenny P Tsai; Santiago Ortega-Gutierrez; Ryan A McTaggart; Michel T Torbey; May Kim-Tenser; Thabele Leslie-Mazwi; Amrou Sarraj; Scott E Kasner; Sameer A Ansari; Sharon D Yeatts; Scott Hamilton; Michael Mlynash; Jeremy J Heit; Greg Zaharchuk; Sun Kim; Janice Carrozzella; Yuko Y Palesch; Andrew M Demchuk; Roland Bammer; Philip W Lavori; Joseph P Broderick; Maarten G Lansberg Journal: N Engl J Med Date: 2018-01-24 Impact factor: 91.245
Authors: S C Patel; S R Levine; B C Tilley; J C Grotta; M Lu; M Frankel; E C Haley; T G Brott; J P Broderick; S Horowitz; P D Lyden; C A Lewandowski; J R Marler; K M Welch Journal: JAMA Date: 2001-12-12 Impact factor: 56.272
Authors: Miriam Brazzelli; Peter Ag Sandercock; Francesca M Chappell; Maria Grazia Celani; Enrico Righetti; Nicholas Arestis; Joanna M Wardlaw; Jonathan J Deeks Journal: Cochrane Database Syst Rev Date: 2009-10-07
Authors: Christopher A Potter; Achala S Vagal; Mayank Goyal; Diego B Nunez; Thabele M Leslie-Mazwi; Michael H Lev Journal: Radiographics Date: 2019-10 Impact factor: 5.333
Authors: Fasco van Ommen; Jan Willem Dankbaar; Guangming Zhu; Dylan N Wolman; Jeremy J Heit; Frans Kauw; Edwin Bennink; Hugo W A M de Jong; Max Wintermark Journal: Neuroradiology Date: 2020-07-29 Impact factor: 2.804