Format

Send to

Choose Destination
Radiol Med. 2011 Mar;116(2):264-75. doi: 10.1007/s11547-010-0602-4. Epub 2010 Oct 6.

Diffusion-weighted magnetic resonance imaging in focal breast lesions: analysis of 78 cases with pathological correlation.

[Article in English, Italian]

Author information

1
Department of Radiology, Ospedale Scaligero Veronese, Via Fontanelle, 1, 37047, S. Bonifacio, Verona, Italy. francescafornasa@libero.it

Abstract

PURPOSE:

This study assessed the usefulness of magnetic resonance diffusion-weighted imaging (DWI) in distinguishing between benign and malignant breast lesions.

MATERIALS AND METHODS:

Gadolinium-enhanced magnetic resonance imaging (MRI) and DWI with determination of the apparent diffusion coefficient (ADC) were performed on 78 women, each with a focal breast lesion at least 7 mm in diameter, which was studied by cytology or histology.

RESULTS:

Final diagnoses were obtained by cytology in 29 cases and histology in 49 (11 percutaneous biopsies and 38 surgical specimens). There were 43 benign lesions (13 fibrocystic disease, eight fibroadenoma, seven adenosis, five normal breast tissue, four inflammatory lesions, three intramammary lymph nodes, two scleroelastosis and one fat necrosis) and 35 malignant lesions (30 invasive ductal carcinoma, two invasive lobular carcinoma, one ductal carcinoma in situ, one carcinomatous mastitis and one metastasis from neuroendocrine carcinoma). The mean ADC values were 1.677±0.151 for benign lesions and 1.298±0.129 for malignant lesions (p<0.001). With an ADC cutoff value of 1.48, DWI had 88.6% sensitivity [confidence interval (CI) 78.1%-99.1%] and 95.3% specificity (CI 88.9%-100%), with 31 true positives, four false negatives (three invasive ductal carcinoma and one carcinomatous mastitis), 41 true negatives and two false positives (one fat necrosis and one fibroadenoma). With the cutoff value set at 1.52, DWI sensitivity (35 true positive, no false negative) was 100% and specificity was 86% (CI 75.7%-96.3%) due to 37 true negatives and six false positives (an additional two fibroadenoma and two fibrocystic disease compared with those recorded with the cutoff set at 1.48). The overall accuracy of DWI considering both cutoff values (72 correct evaluations out of 78 cases) was 92.3% (CI 86.4%-98.2%).

CONCLUSIONS:

DWI is a reliable tool for characterising focal breast lesions.

PMID:
21076884
DOI:
10.1007/s11547-010-0602-4
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center