P3-07-32: The Role of Axillary Ultrasound in the Detection of Metastases from Primary Breast Cancers.

Abstract

Abstract Background: The value of ultrasound (US) in the preoperative evaluation of axillary nodes has yet to be completely clarified. Preliminary experience with this technique in our institution was examined. Methods: Patients with a radiographic or palpable abnormality of the breast had simultaneous breast and axillary US. The exams were performed by dedicated breast radiologists using a 12 MHz linear array transducer (HDI 5000: Philips Ultrasound). Results were reviewed for all patients with invasive cancers who were cN0 and had definitive surgical procedures between June 2006 and May 2008. Criteria for abnormal lymph nodes were loss of reniform shape, focal or diffuse cortical thickening, or eccentric/replaced fatty hilum. US-guided biopsies were done using a 16g spring-loaded core biopsy device (16g MD TECH SuperCore). Patients with positive axillary node biopsies bypassed sentinel lymph node biopsy (SLNB) and had axillary dissection, whereas those with sonographically normal nodes or benign/non-diagnostic biopsy results had SLNB at the time of definitive surgery. Results: Of 128 patients diagnosed with invasive cancer, 23 (18%) had abnormal axillary US at the time of initial diagnosis. Biopsies were performed in 18 of the 23, of which 12 (67%) were malignant and 6 (33%) were benign. Ultrasounds were negative in 105 (82%) patients. SLNB was done in 110 patients: 103 with negative US; 4 patients with abnormal US but negative axillary biopsies; 2 patients with abnormal US but no core biopsies; and 1 patient with a positive US biopsy. SLNB was negative in 91 (83%) patients and positive in 19 (17%). The node-positive status was N1a in 14 patients and N1mic in 5. Axillary dissection was done in 32 (25%) of 128 patients, comprised of 11 patients with US-guided positive biopsies, 12 with positive sentinel nodes, 2 with US-guided negative biopsies, 2 with negative ultrasounds, 3 with unbiospied abnormal ultrasounds, and 2 with a false-negative SLNB. For determining axillary metastases with US, sensitivity was 16/31 (52%), specificity was 90/97 (93%), positive predictive value was 16/23 (69%), and negative predictive value was 90/105 (86%). Conclusions: US examination was a valuable method of evaluating the axilla in newly diagnosed breast cancers. Of 32 patients having an axillary dissection, abnormal US eliminated the need for SLNB in 17 (53%). Patients with US-guided positive nodes were submitted to axillary dissection without SLNB. Therefore, we were unable to determine how often US identified an abnormal non-sentinel node, thus upstaging the axilla relative to SLNB alone. This question should be the topic of further clinical study. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-07-32.