Axillary lymph node dissection in patients with ductal carcinoma in situ (DCIS) of the breast is not warranted because DCIS has no metastatic potential. However, the risk of microinvasive carcinoma (MIC) exists in large DCIS treated by mastectomy. The aim of this series is to evaluate the incidence of lymph node metastases in DCIS and DCIS-MIC. We analyzed retrospectively patients treated in six French cancer centers for pure DCIS or DCIS-MIC. Surgical procedures were lumpectomy or mastectomy associated with an axillary sentinel node (SN) procedure. We included 161 patients suffering from pure DCIS (116/161, 72%) or DCIS-MIC (45/161, 28%). Mean age was 56 years (32-78). We observed underestimation between core biopsy and histological result in 43/142 cases (30%). These data show an association between lesion size, solid subtype, high-grade DCIS, and underestimation. Forty-eight breast conservative procedures were performed and 113 mastectomies (70%). SN procedure was performed using blue dye, technetium, or both. In our series, we selected patients with a high risk of occult invasive carcinoma: high grade (55%), mean size (27 mm), and mastectomy (112). Six SN were found positive (3.7%). In the five patients treated with complete axillary dissection, the SN was the only positive node. SN in DCIS is an interesting procedure but not necessary for all patients. We need to focus on the subgroup with or a high risk of occult MIC: extensive calcifications or palpable mass, DCIS diagnosed by core biopsy and underestimation, multifocality, high grade, large tumor size, MIC, and mastectomy.