Value of the US BI-RADS final assessment following mastectomy: BI-RADS 4 and 5 lesions

Abstract

BACKGROUND: Clinical examination is not entirely sufficient for evaluation of the postoperative site for follow-up of patients with mastectomy. A few studies have reported that postoperative follow-up US evaluation allows early detection and proper management of local tumor recurrence.

PURPOSE: To evaluate the diagnostic performance of the American College of Radiology (ACR) ultrasonographic (US) Breast Imaging Reporting and Data System (BI-RADS) categories 4 and 5 breast lesions at the mastectomy site. MATERIAL AND METHODS: Our institutional review board approved this study and waived the need for informed consent. We retrospectively reviewed the consecutive post-mastectomy US exams for palpable and non-palpable lesions in the post-mastectomy chest wall that were categorized as BI-RADS 4 (subcategorized 4a, 4b, and 4c) or 5 between January 2007 and April 2010. The positive predictive value (PPV) for final assessment was evaluated. RESULTS: From 2681 post-mastectomy US examinations, we obtained a study population of 50 patients with 50 lesions (20 palpable, 30 non-palpable). There were nine (45%) malignancies among the palpable lesions and six (20%) malignancies among the non-palpable lesions. The palpability showed no significant correlation with malignancy in overall subcategorization (P >0.05). The PPVs of categories 4 and 5 were 14.3% for category 4a, 62.5% for category 4b, 100% for category 4c, and 100% for category 5. CONCLUSION: The ACR US BI-RADS categorization of 4a, 4b, 4c, and 5 for breast lesions at the mastectomy site is a feasible method for predicting local recurrence. All lesions should be managed according to US characteristics during evaluation of local recurrence at the mastectomy site, regardless of palpability.