Prediction of pathologic node-negative clinical stage IA lung adenocarcinoma for optimal candidates undergoing sublobar resection

Objective: Patients with pathologic node-negative early lung cancer may be optimal candidates for sublobar resection. We aimed to identify predictors of pathologic lymph node involvement in clinical stage IA lung adenocarcinoma. Methods: The data from a multicenter database of 502 patients with completely resected clinical stage IA lung adenocarcinoma were retrospectively analyzed to determine the relationship between the lymph node metastasis status and tumor size on high-resolution computed tomography (HRCT) or maximum standardized uptake value (SUVmax) on [18F]-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography (FDGPET/CT). Revised SUVmax was used to correct interinstitutional discrepancies. Results: In multivariate analyses, either a solid tumor size on HRCT (P = .001) or an SUVmax on FDG-PET/CT (P = .049) was an independent predictor of lymph node metastasis. The predictive criteria of pathologic node-negative early lung cancer were a solid tumor size of less than 0.8 cm or an SUVmax of less than 1.5. Patients who met the predictive criteria of pathologic node-negative disease had less pathologic invasiveness, such as lymphatic, vascular, or pleural invasion (P < .001), and better disease-free survival (P < .0001) than those who did not, and 86 (40.4%) of the 213 patients with T1b (2-3 cm) tumors met the predictive criteria. Conclusions: Either a solid tumor size or an SUVmax was a significant independent predictor of nodal involvement in clinical stage IA lung adenocarcinoma. The pathologic node-negative status criteria of a solid tumor size of less than 0.8 cm on HRCT or an SUVmax of less than 1.5 on FDG-PET/CT may be helpful for avoiding systematic lymphadenectomy for clinical stage IA lung adenocarcinoma, even in cases of T1b (2-3 cm) tumor. (J Thorac Cardiovasc Surg 2012; 144:1365-71)