Nipple-areola sparing mastectomy (NSM) is increasingly used in patients with non-locally advanced breast carcinoma. Literature data on the preoperative assessment of the nipple-areola complex (NAC) are inconsistent.Out of 1359 patients submitted to total mastectomy between 2001 and 2010, we selected 61 patients whose pre-operative mammogram (MX) was available (MX group) and 39 patients who underwent preoperative breast magnetic resonance imaging (magnetic resonance imaging (MRI) group). The rate of NAC involvement, the value of MX and MRI to predict NAC involvement and the performance of the Schecter's and Loewn's algorithms for the prediction of NAC involvement were evaluated.In the combined MX and MRI groups, NAC involvement was found in 14% of the cases. At univariate analysis, tumour stage (p value: 0.03), central tumour location (p value: 0.004), presence of NAC retraction (p value: 0.001) and tumour-NAC distance (p value: 0.006) were associated with NAC involvement, but only the latter parameter retained statistical significance at multivariate analysis (p value: 0.05). Tumour-NAC distance was a key predictor of NAC involvement, with a negative predictive value of 94% for MX and of 100% for MRI when the cut-off was set at 10mm. Overall, the performance of Schecter's and Loewn's algorithms was respectively lower and similar as compared to the original series.Occult tumour involvement of the NAC is detected in a minority of breast cancer patients submitted to mastectomy. A tumour-NAC distance ≥ 10 mm by MRI may help select patients candidate to NSM.
