Objective: To evaluate surgical margins in cases of ductal carcinoma through a histopathological exam using frozen sections. Materials and Methods: Retrospective study encompassing 242 conservative surgeries, 179 of which included intraoperative frozensection histopathology and 63 intraoperative nonfreezing techniques (macroscopy/gross examination and cytology). The results of such analyses were compared with those of the histology processing following paraffin embedment and hematoxylin and eosin (H & E) staining. A margin was deemed free when the distance between the tumor and the surgical border was equal to or greater than two millimeters. The factors given consideration for possibly affecting the results were: age, surgical aspects (skin removal and widening of surgical margins), histopathological findings (size, affected lymph nodes, and angiolymphatic invasion), and extensive intraductal and immunohistochemical components (estrogen, progesterone, Ki-67, and HER-2 receptors). In the statistical analyses, the chi-square test was used and negative predictive values were calculated. Results: The negative predictive values were 87.1% and 79.3% for frozen and nonfrozen sections, respectively. There was no significant difference between the two groups (p = 0.14). The factors under consideration had no influence on the results of the intraoperative exam of the margins. Conclusion: The present study allowed to conclude that the intraoperative exam of the surgical margins by frozen section is not superior to a macroscopy and / or cytology exam.