The use of ultrasound in pre-operative assessment of the axilla in breast cancer.

Abstract

Abstract Abstract #1019 Introduction
 Sentinel lymph node biopsy is now standard practice in axillary staging in breast cancer. It is associated with less morbidity than an axillary node clearance. A drawback of sentinel node biopsy is the need for a second surgical procedure if the sentinel node shows metastases. Clinical examination of the axilla has been the standard pre-operative assessment but this has been shown to be unreliable. More recently ultrasound has been used to identify axillary metastases pre-operatively. The aim of our study is to assess the role of pre-operative ultrasound and fine needle aspirate cytology in refining the selection of patients for whom sentinel node biopsy is appropriate.
 Methods
 Three hundred patients with primary operable invasive breast cancer had axillary ultrasound preoperatively. If the ultrasound was normal the patient was offered a sentinel node procedure. If it identified equivocal nodes a fine needle cytology was performed. If the cytology was benign a sentinel node biopsy was performed. If the cytology was malignant then an axillary node clearance was performed. In cases where pathological nodes were identified on imaging, cytology was performed to confirm malignancy. An axillary node clearance was then performed.
 Results.
 Eighty three percent of our patients (n=249) had a normal axillary ultrasound pre-operatively. Of these 74.5% had a benign sentinel node biopsy. Sixty two percent of patients with equivocal nodes on pre-operative imaging had metastases on final histology (n=19). Of these 63% (n=12) were correctly identified pre-operatively with fine needle aspirate cytology of the equivocal node and had an axillary node clearance performed form the outset. Ultrasound identified pathological nodes in 7% of our patients (n=20). Malignancy was confirmed on cytology and an axillary node clearance was performed.
 Using ultrasound and fine needle aspirate cytology to assess the axilla pre-operatively sentinel node biopsy was performed in 265 patients (88%). Of these, 74.8% had a benign final histology. Thirty two patients had a pre-operative diagnosis of nodal metastases and had an axillary node clearance. All 32 had lymph node metastases on final histology. Factors that predicted for a malignant sentinel node were lymphovascular invasion (p<0.001), tumour grade (p<0.008) and size of tumour (p<0.001).
 Conclusion
 Eighty eight percent of our patients had a sentinel lymph node biopsy to stage the axilla. Pre-operative ultrasound combined with cytology correctly determined the status of the node in 78% of cases. Our rate of second axillary operations was reduced by 32%. Importantly all patients diagnosed with node metastases pre-operatively and advised to have an axillary node clearance did have metastases on final histology. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 1019.