Video-assisted thoracic surgery (VATS) lobectomy for non-small cell lung cancer (NSCLC) is increasingly popular. However, the oncologic soundness of VATS for patients with NSCLC as measured by long-term survival has not been proven. The objective here is to determine the overall survival (OS) and disease-free survival (DFS) in two well-matched groups of patients with NSCLC resected by VATS or thoracotomy. We conducted a retrospective review of a prospective database to identify patients who had a lobectomy for NSCLC. A propensity score-matched analysis was done with variables of age, sex, smoking history, Charlson comorbidity index, forced expiratory volume in 1 second, lung diffusing capacity for carbon monoxide, histology, and clinical T and N status. Medical records were reviewed and survival was analyzed. After matching, there were 208 patients in each group. Patient and tumor characteristics were similar. The VATS group had a shorter length of stay. More nodes (14.3 versus 11.3; p=0.001) and more nodal stations (3.8 versus 3.1; p<0.001) were removed by thoracotomy. No differences were seen in OS and DFS. Median follow-up was 36 months. More than 90% of patients had clinical stage I disease, with 3- and 5-year OS of 87.4% and 76.5%, respectively, for VATS, and 81.6% and 77.5%, respectively, for thoracotomy (p=0.672). Both the incidence and distribution of recurrence were similar. Multivariate Cox regression analyses of OS and DFS confirmed the noninferiority of VATS. For patients with clinical stage I NSCLC, VATS lobectomy offered similar OS and DFS compared with thoracotomy. Thoracotomy offers a more thorough lymph node evaluation, and may be appropriate for patients with more advanced clinical disease.