Introduction: Sentinel node (SN) biopsy to predict axillary involvement in breast cancer patients is a common practice. After a positive SN, additional metastases are present in an unpredictable percentage, as variable as 13-66%, of axillary clearances. Our aim is to define variables associated with non-sentinel node (NSN) metastases and to determine the predictive value of a derived clinical decision rule.
Methods: A consecutive series of patients with a positive SN submitted to SN biopsy plus axillary dissection was evaluated from June 1999 to December 2004 (n=143). Patient-, tumour- and SN-related variables were analysed in relation to the presence of additional metastasis. Univariate and multivariate analyses were done and predictive values of a clinical decision rule based on significant variables were estimated.
Results: A total of 66 patients had metastasis in non-sentinel axillary nodes. No significant differences were present between this group and those with only the SN metastasised. Significant and independent association was found between NSN positivity and increasing tumour size, the presence of multifocality and the presence of peritumoral lymph channel invasion.
Conclusions: A first derivation of a simple rule based on tumour-related variables concurs to define the presence of NSN metastasis. Care should be taken when including SNrelated variables in these algorithms.