The accuracy of sentinel node mapping according to T stage in patients with gastric cancer

Background: Sentinel lymph node (SLN) mapping has been recently introduced to the field of gastric cancer. To the best of our knowledge, no study has dealt with the accuracy of SLN mapping according to the T stage of the primary tumor. The aim of the present study was to evaluate SLN status according to the T stage of the primary tumors. Methods: Eighty patients with gastric cancer underwent SLN mapping with patent blue dye during gastric resection. Results: Forty-seven patients underwent distal subtotal gastrectomy; 17 patients, proximal gastrectomy; 14, total gastrectomy; and 2, gastric stump resection. SLNs were stained in 61/80 patients (76.3%). The number of stained SLNs varied from 1 to 16 (mean, 3.3). Patients undergoing proximal gastrectomy had a mean of 3 stained SLNs, whereas patients undergoing distal subtotal gastrectomy had a mean of 2.8 stained SLNs. In 55/61 patients (90.2%) with stained SLNs a positive correlation was found between the presence of metastases and stained or non-stained SLNs. Ten out of 11 patients (90.9%) with T1 tumors (mean, 3.27 SLNs per patient) and 15/17 patients with T2 tumors (88.2%; mean, 3 SLNs per patient) had stained SLNs as compared to only 33/48 (68.8%) of patients with T3 tumors (mean, 3.3 SLNs per patient). The positive predictive value of the SLN mapping was 100%, the negative predictive value was 76.9%, and sensitivity was 85.4%. Conclusion: While in T1 and T2 tumors sentinel node mapping may be of assistance in the decision-making process regarding the extent of lymphadenectomy (sensitivity, 100%; negative predictive value, 90%-100%), SLN mapping in patients with T3 tumors will be misleading in a third of the patients and should not be attempted.