The present study aims to evaluate the accuracy of sentinel lymph node (SLN) mapping performed by intratumoral injection of blue dye in a large series of patients with papillary thyroid cancer (PTC). 153 consecutive patients were enrolled in the study. All patients had a preoperative cytological diagnosis of PTC, and none had clinical or ultrasonographic (US) evidence of nodal involvement. At surgery, vital patent V blue dye was injected into the malignant thyroid nodule. Subsequently, total thyroidectomy, central compartment (CC) node dissection, and median inferior jugulocarotid node dissection of laterocervical compartment, ipsilateral to the primary tumour, were performed. The excised thyroid, the blue-positive SLN and blue-negative lymph nodes were sent for frozen section and definitive histophatologic analysis. At surgery, blue-positive SLN were found in 107/153 patients (69.9%), of whom 36 (33.6%) had micrometastasis in SLN; moreover, in 13 of these 36 patients (36.1%), other nodes were found to be metastatic. In the remaining 71/107 blue-positive SLN patients, both the SLN itself and the other removed nodes were found negative for the presence of metastatic disease. In 4 cases, a normal parathyroid gland and in 3 cases fibro-adipous tissue were blue-stained and mistakenly removed as SLN (7 false positive results). On the other hand, SLN was blue-negative in 46/153 patients (30.1%), of whom 7 patients (15.2%) had micrometastases in blue-negative lymph nodes. On the basis of these data, the blue dye procedure for SLN detection appears inappropriate as a standard of care in PTC due to a relatively high number of false negative and false positive results.