Intraoperative examination of sentinel lymph nodes (SLN) in breast cancer can avoid a new surgical procedure in case of positive SLN, but its value, efficacy and the methods used are still controversial. The aim of our study was to evaluate the imprint cytology intraoperative method of SLN analysis performed at our institution. We did a retrospective study of the sentinel lymph node procedures performed during a period of 24 months on cT1N0 unifocal breast cancers. Intraoperative procedure was mainly by imprint cytology (touch prep). A SLN procedure was performed on 187 women with 360 SLN. Two hundred and seventy-seven SLN among 156 women were analyzed intraoperatively by touch prep. 19/48 positive SLN were detected by intraoperative touch prep (sensitivity 39.6%; specificity 100%; positive predictive value 100%; negative predictive value 88.7%, accuracy 89.5%). False negative rate of cytological intraoperative examination of SLN was 11,2% by SLN and 18,3% by patient. By univariate analysis, this rate significantly increased with lymphovascular invasion, tumor size cT1b and c and histological SBR grade 2 or 3. By multivariate analysis, only lymphovascular invasion was a predictive factor of intraoperative touch prep failure (OR = 3.3; IC 1.3-8.4). Intraoperative imprint cytology of SLN in breast cancer is associated with a high rate of false negativity that questions its use in this setting.