Effect of radioisotope sentinel node mapping in patients with cT1 N0 M0 lung cancer

Background: Application of the sentinel node concept to lung cancer is still controversial. Patients with peripheral small lung cancers would gain the most benefit from this concept, if it were valid. We sought to determine whether it is possible to choose between limited lymph node sampling and systematic lymphadenectomy from the distribution of sentinel lymph nodes in patients with node-negative disease on the basis of imaging. Methods: Sixty-five consecutive patients with M NO MO non-small cell lung cancer were enrolled. A radioisotope tracer (4 mCi of technetium-99m tin colloid, 2.0 mL) was injected in the vicinity of the tumor before surgical intervention with computed tomographic guidance. The radioactivity of each resected lymph node was measured separately with a hand-held gamma probe after complete tumor resection. Sentinel nodes were identified, and the accuracy of sentinel node mapping was examined. Whether the location of the sentinel node depended on the site of the primary tumor was also examined. Results: Of the 65 patients, 3 were excluded because of the final pathologic results. Successful radionuclide migration occurred in 39 (62.9%) of the 62 patients. There was 1 (2.6%) false-negative result among 39 patients with a sentinel node, and therefore the sensitivity was 90%, and the specificity was 100%. The most common sentinel lymph nodes were at level 12 (46.7%), followed by level 11 (18.3%), the mediastinum (16.7%), and level 10 (11.7%). Conclusion: The sentinel node concept is valid in patients with cT1 NO MO lung cancer. The lobar lymph nodes were identified as sentinel nodes more frequently than other lymph nodes. We need to make further efforts to increase the sentinel node identification rate. However, we believe that if sentinel nodes are identified, sentinel node mapping can allow the accurate intraoperative diagnosis of pathologic NO status in patients with cT1 NO MO lung cancer.