Abstract Archives of the RSNA, 2013
Asif Iqbal MBBS, Presenter: Nothing to Disclose
Michael J. Michell MBBCh, Abstract Co-Author: Nothing to Disclose
Rema Wasan MBBCh, Abstract Co-Author: Nothing to Disclose
Abdel Douiri PhD, Abstract Co-Author: Nothing to Disclose
David Evans MBBS, Abstract Co-Author: Nothing to Disclose
Clare Peacock MBBS, Abstract Co-Author: Nothing to Disclose
Juliet Clare Morel MBChB, MRCP, Abstract Co-Author: Nothing to Disclose
In preoperative assessment of local extent of breast cancers, full field digital mammography (FFDM) and
ultrasonography (US) imaging are routinely used. Digital breast tomosynthesis (DBT), an application of FFDM which displays the breast structures as a series of thin slices, overcomes the problem of overlying tissues and displays more clearly the tumour outline. We compare DBT with FFDM and US in the size measurement of invasive cancers.
A retrospective analysis was carried out on 139 invasive breast cancers examined by DBT, FFDM and US. Only soft tissue lesions were included. Separate tumour size measurements on three imaging modalities and histological tumour size confirmed following surgical resection, were recorded. Statistical analysis was done by following Bland-Altman method and concordance correlation coefficient (CCC); the size differences in millimetres between imaging modality measurements and histology measurements were plotted.
Histologically, 108 lesions (77.7%) were diagnosed as invasive ductal carcinoma, 22 (15.8%) as invasive lobular carcinoma and 9 (6.5%) as other invasive types. For the whole dataset, on Bland-Altman analysis, the DBT showed the lowest bias 1.216±7.99; 95% CI [-14.44 to 16.87], whereas the FFDM was 1.88±9.387; [-16.51 to 20.28] and US was 5.647±9.424;[-12.82 to 24.11]. In the subanalysis by mammographic features, the CCC values demonstrated significantly high (p <0.0001) correlation values for parenchymal distortion [DBT rho=0.925, FFDM rho=0.704, US rho=0.422]; asymmetric density [DBT rho=0.892, FFDM rho=0.685, US rho=0.534]; spiculate masses [DBT rho = 0.652, FFDM rho=0.556, US rho=0.523] and circumscribed masses [DBT rho =0.519, FFDM rho=0.519, US rho=0.452].
This study has shown that DBT has a higher accuracy for determining invasive cancer size than FFDM and US. The most significant effect is seen with parenchymal distortion and asymmetric density, where measurement using FFDM and US is difficult and unreliable.
Use of digital breast tomosynthesis in the diagnostic workup of breast cancers for accurate pre-surgical assessment of tumour size could help in choosing appropriate treatment.
Iqbal, A,
Michell, M,
Wasan, R,
Douiri, A,
Evans, D,
Peacock, C,
Morel, J,
Measurement of Invasive Breast Cancer Using Digital Breast Tomosynthesis, Full Field Digital Mammography and Ultrasonography. Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL.
http://archive.rsna.org/2013/13015480.html