This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Vascular, MR-Angiography, MR, Embolisation
Authors:
N. K. Majewska, A. Kociemba, M. Pyda, M. Wykretowicz, M. Stajgis, K. Katulska; Poznan/PL
DOI:
10.1594/ecr2016/C-2297
Results
Of the 25 patients for whom data were evaluated,
7 were found to have high-flow malformations and 18 had low-flow malformations.
Signal voids on SE T1-weighted images were observed in only four of seven high-flow malformations (Fig.
4).
Signal voids were also observed in two of the 18 low-flow malformations.
The sensitivity and specificity of this method to distinguish between high- and low-flow malformations was 57% and 89%,
respectively.
Analysis of signal intensity on T2-weighted images showed increased signal intensity in 17 of 18 low-flow malformations,
and in two of the high-flow lesions (Fig.
5,
6).
The sensitivity and specificity of this method was 71% and 94%,
respectively.
Calculation of the artery–lesion time,
maximum enhancement time,
and slope revealed significant differences between the high- and low-flow groups.
There was no overlap between the two groups for slope values only (Table 3).
ROC analysis revealed that an artery–lesion time of 4.2 s can be used to distinguish between high-flow and low-flow lesions with 100% sensitivity and 57% specificity.
A maximum enhancement time of >27 s can be used as indicator of low-flow malformations,
with 94% sensitivity and 100% specificity.
Slope values can also be used to classify malformations as high or low flow with 100% specificity and sensitivity.
The cut-off value in the case of slope values is 921 s–1 (see Table 4).