Meta-analysis of tumour burden in pre-operative axillary ultrasound positive and negative breast cancer patients.

BACKGROUND: Management of the axilla in breast cancer is becoming increasingly conservative. Patients identified with a low axillary nodal burden (two or fewer involved nodes) at sentinel node biopsy (SNB) can avoid completion axillary node clearance (cANC). 'Fast track' to ANC in patients with involved nodes on pre-operative ultrasound may be over-treating a subgroup of these patients with low nodal burden, which would have precluded their need for ANC. This systematic review assesses the proportion of patients with involved nodes on pre-operative axillary ultrasound, which would fit low axillary burden criteria. METHODS: Meta-analysis of studies comparing axillary burden of breast cancer patients identified as pre-operative ultrasound negative versus positive was performed. The primary outcome measure was the number of patients with two or fewer involved nodes (macrometastases only). Pooled odds ratio (OR), 95% confidence intervals (CIs), means and probabilities of identifying two or fewer involved nodes versus greater than two were calculated. RESULTS: Six studies reported the axillary burden in 4271 patients who were either directed straight to ANC or cANC after SNB. There was a significantly greater axillary burden in the ultrasound positive versus negative groups (OR 5.95, 95% CI 5.80-6.11) with mean nodal retrieval values of 2.9 [standard error (SE) 0.2] and 1.6 (SE 0.2) nodes, respectively. Cumulative probabilities identified 78.9% of ultrasound negative and 43.2% of ultrasound positive patients possessed low axillary burden. CONCLUSIONS: Pre-operative ultrasound positive patients have significantly higher axillary burden. However, nearly half do fit the criteria of low axillary burden and could be considered for omission of ANC.