Return To Abstract Listing

Scientific Session 14 — SS14: Breast Imaging - Tomosynthesis and Ultrasound

Wednesday, May 8, 2019

Abstracts 1224-3064

3064. Architectural Distortion in the Tomosynthesis Era: How to Manage Cases with Nonmalignant Pathology at Biopsy

Bahl M*,  Lehman C. Massachusetts General Hospital, Boston, MA

Address correspondence to M. Bahl (

Objective: Because architectural distortion is more readily detected on digital breast tomosynthesis (DBT) than digital 2D mammography (DM), more patients are undergoing image-guided biopsy of architectural distortion in the modern era of DBT. Historically, many patients with architectural distortion underwent surgical excision regardless of needle biopsy pathology results, but the appropriate management of architectural distortion on DBT with nonmalignant biopsy pathology results remains unknown. The purpose of this study is to determine upgrade rates of biopsied architectural distortion in the modern era of DBT.

Materials and Methods: In this IRB-approved and HIPAA-compliant study, consecutive cases of architectural distortion on mammography recommended for biopsy from January 2016 to June 2017 were reviewed. During this period, our institution performed combined DBT and DM in all patients. Medical records were reviewed for patient factors, imaging findings, needle biopsy pathology results, and surgical outcomes.

Results: Over an 18-month period, architectural distortion considered suspicious for or highly suggestive of malignancy was seen in 182 mammographic exams in 182 women (mean age, 56 years; range, 23-88 years). Of the 182 women, 153 (84.1%) underwent both image-guided biopsy and surgical excision (n=138) or underwent image-guided biopsy only and had at least one year of imaging and clinical follow-up (n=15). These 153 women constitute the study cohort. At image-guided biopsy, 65 cases (42.5%) were malignant, 53 (34.6%) were radial scars, and 35 (22.9%) were other nonmalignant pathology (high risk or benign). Of the 53 radial scars, only one case (1.9%), which was associated with atypical ductal hyperplasia, upgraded to grade I DCIS at surgical excision. None of the radial scars without atypia (n=42) upgraded to malignancy, and none of the radial scars with or without atypia (n=53) upgraded to invasive carcinoma. Of the 35 cases of other nonmalignant pathology (high risk or benign), two cases (5.7%) upgraded to carcinoma, one to grade I DCIS (which was stromal fibrosis at biopsy), and one to grade 2 invasive lobular carcinoma (which had atypia at biopsy). Therefore, the overall upgrade rate to carcinoma of architectural distortion with nonmalignant needle biopsy pathology results was 3.4% (3/88). The overall upgrade rate to carcinoma of architectural distortion with atypia was 12.5% (2/16) and without atypia was 1.4% (1/72).

Conclusion: The overall upgrade rate to carcinoma of architectural distortion on DBT with nonmalignant needle biopsy results is low at 3.4% (3/88). None of the architectural distortion cases with biopsy result of radial scar without atypia were upgraded to carcinoma. Given the overall low risk of upgrade, imaging surveillance rather than surgical excision could be considered for architectural distortion with nonmalignant core needle biopsy results in the modern era of DBT, particularly if there is no associated atypia.