This study analyzed the relation between the tumor size and the lymph node metastasis in adenocarcinoma of the lung with a size of 30 mm or smaller.

Four hundred thirteen patients who had undergone curative resection for lung adenocarcinoma were enrolled. If the tumor presented ground-glass opacities on the preoperative high-resolution computed tomography, both the total size including ground-glass opacities and the solid size alone were measured. To calculate the rates of lymph node metastasis by the tumor size, the tumors were divided into six groups by their sizes: 5 mm or less, 6 to 10 mm, 11 to 15 mm, 16 to 20 mm, 21 to 25 mm, and 26 to 30 mm.

The average numbers of dissected lymph nodes and dissected lymph node stations were 17 and 5, respectively. Seventy-five patients (18%) were postoperatively discovered to have positive nodes. The rates of node metastasis in each total size group were 0/1 (0%), 0/29 (0%), 5/77 (7%), 17/121 (14%), 27/101 (27%), and 26/84 (31%), respectively. The rates of node metastasis in each solid size group were 0/37 (0%), 1/53 (2%), 9/88 (10%), 17/104 (16%), 23/78 (30%), and 25/53 (47%), respectively. The area under the curve of receiver operating characteristic curves for the total size was measured as 0.701, and that for the solid size was measured as 0.777. By multivariate analysis, solid size, maximum standardized uptake value, and lymphovascular invasion were independent significant predictive factors.

Solid size, maximum standardized uptake value, and lymphovascular invasion were independent predictors for lymph node metastasis of lung adenocarcinoma. The size of the solid component explained the relation between the tumor size and the lymph node metastasis more accurately than that explained by the total tumor size on high-resolution computed tomography.