Recently, skin-sparing mastectomy (SSM) with nipple-areola complex (NAC) preservation has been promoted as an oncologically safe procedure in practice for selected patients. The criteria of selection have not been yet defined precisely. The focus of this study was to investigate predictive factors of NAC-base neoplastic involvement to define the indications for NAC preservation. A prospective clinical study was conducted of 108 randomly selected female patients with invasive breast cancer. Analyzed markers of NAC involvement were tumor-nipple distance (TND), tumor size, localization, histologic type, grade, lymphovascular invasion (LVI), site, and axillary lymph-node status. The definitive histologic findings of the NAC base were compared with analyzed markers and the frozen section results. NAC base was positive in 23.15% patients at definitive histology with false-negative results in 4.63% patients at intraoperative frozen section. Significant differences were found in TND, tumor size, axillary lymph-node status, and LVI. There were no significant differences in tumor grade and site and not enough cases for statistical evaluation in histologic type and localization. Clinical indications for NAC preservation, according to this study, include tumors < or =2.5 cm, TND >4 cm, negative axillary lymph node status, and no LVI. Considering the possibility of pre- or intraoperative measurement, tumor size, and TND evaluation will result in the lowest possible mistakes in NAC preservation. Frozen section analyses of the NAC base, because of the "false-negative" possibility, could be deemed as a relative prognostic factor until definitive histologic findings. The presence of an extensive intraductal component (EIC) in the "borderline" cases of these criteria could be an additional argument for NAC removal.