Senologie - Zeitschrift für Mammadiagnostik und -therapie 2019; 16(02): e7
DOI: 10.1055/s-0039-1687954
Abstracts
Georg Thieme Verlag KG Stuttgart · New York

Application of DWI in abbreviated breast MRI (ABM): quantitative tissue analysis increases diagnostic-performance within a short examination time

M Dietzel
1   University Hospital Erlangen, Department of Radiology, Erlangen, Deutschland
,
E Wenkel
1   University Hospital Erlangen, Department of Radiology, Erlangen, Deutschland
,
S Ellmann
1   University Hospital Erlangen, Department of Radiology, Erlangen, Deutschland
,
M Uder
1   University Hospital Erlangen, Department of Radiology, Erlangen, Deutschland
,
R Schulz-Wendtland
1   University Hospital Erlangen, Department of Radiology, Erlangen, Deutschland
,
P Baltzer
2   Medical University of Vienna, Division of Molecular and Gender Imaging, Department of Biomedical Imaging and Image-Guided Therapy, Vienna, Österreich
› Author Affiliations
Further Information

Publication History

Publication Date:
28 May 2019 (online)

Also available at
 

Objectives:

Abbreviated breast-MRI (ABM) optimizes accessibility and costs of breast-MRI. However, this comes on the price of a reduced diagnostic-accuracy. We investigated whether fast diffusion-weighted-imaging (DWI) could increase diagnostic-accuracy without sacrificing rapidity of ABM.

Methods:

In 132 consecutive patients, breast-MRI according to international recommendations was performed (10:10 min). The MRI-protocol consisted of three section:

  1. ABM: one pre- and one postcontrast scan (2:40 minutes)

  2. Postinitial: Five additional postcontrast scans (+5 minutes).

  3. DWI (+2:30 minutes).

Histopathological verification served as referenced-standard.

ROIs (region-of-interest: 10 – 15 mm2) were drawn within the enhancing part of the lesion at the first postcontrast scan. ROIs were automatically transferred to the ADC-maps (apparent diffusion coefficient) of the DWI as well as to the precontrast, the first and the last postcontrast scan.

This gave one semi-/quantitative parameter for every section of the protocol (ABM/wash-in, Postinitial/wash-out; DWI/ADC).

We investigated whether the following extensions of ABM could increase diagnostic-accuracy.

I. ABM+: ABM+DWI

II. Curve: ABM+Postinitial

III. Curve+: Curve+DWI

Diagnostic accuracy of I. to III was quantified (area under the curve: AUC) and compared intra-individually (AUC-comparison; logistic-regression, alpha = 5%).

Results:

145 lesions were included (malignant/benign: 101/44). ABM+ (AUC = 0.877) outperformed both ABM (AUC = 0.743) and Curve (AUC = 0.788; P < 0.03). Further inclusion of Postinitial (Curve+) did not change diagnostic performance compared to ABM+ (P = 1).

Zoom Image
Fig. 1: Three sections of the MRI protocol. The centerpiece was the abbreviated breast MRI (ABM), consisting of the pre- and the first scan post Gadolinium (Gd) injection. Considering an average injection time of 10 seconds and a subsequent delay of 30 seconds, this gave an average acquisition time of 2:40 minutes. The ABM was expanded by five additional post-contrast scans (+5 minutes: Postinitial) and by a fast DWI sequence (+2:30 minutes).

Conclusions:

Inclusion of DWI into ABM added significant diagnostic-value, but did not sacrifice rapidity. Accordingly, this approach should be further investigated in order to optimize accessibility and costs of breast-MRI.