Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)The sub-nipple tissue (SNT) examination has been used by surgeons to preserve, or not, the nipple in nipple-sparing mastectomy. However, it is uncertain whether SNT evaluation can predict nipple involvement. The aim of this study was to evaluate the accuracy, sensitivity, specificity, PPV, and NPV of the intraoperative frozen section and imprint cytology, and permanent histology of SNT to predict the involvement of the nipple in breast carcinoma and to compare the three exams. A prospective study was performed with 68 consecutive breast carcinoma women who had undergone mastectomy or central segmentectomy (removing nipple-areolar complex). After surgery, the nipple-areolar complex was dissected simulating a nipple-sparing flap (ex vivo). The SNT was subsequently removed and submitted to frozen section, imprint cytology, and permanent histology. The nipple was examined separately by paraffin histopathology and was considered the gold standard. The occult nipple involvement rate was 11.7 %. The frozen section, cytology, and permanent histology of SNT presented accuracy 86.8, 76.5, and 86.8 %; sensitivity 50, 37.5, and 62.5 %; specificity 91.7, 81.7, and 90 %; PPV 44.4, 21.4, and 45.5 %; and NPV 93.2, 90.7, and 94.7 %, respectively. The accuracy of the frozen section was similar to that of permanent histology (p = 0.77) and both were better than cytology (p = 0.01). False negative rates were 6.8 % for frozen section, 9.3 % for cytology and 5.3 % for paraffin. SNT evaluation is a good method for predicting occult nipple involvement; the outcomes showed a good accuracy and low false negative rate for the frozen section, cytology, and permanent histology exams. When we compared the exams, the frozen section was similar to permanent histology and more accurate than imprint cytology.1533 557563Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)FAPESP [2013/03019-4]