Abstract
OBJECTIVES: We aimed to compare the outcomes of segmentectomy with those of lobectomy in T1c (>2-3 cm) radiological pure-solid non-small-cell lung cancer detected on thin-section computed tomography. METHODS: This retrospective review compared the survival outcomes, causes of death and recurrence patterns between the segmentectomy and lobectomy in patients with c-T1cN0M0 radiological pure-solid non-small-cell lung cancer. Multivariable analysis was performed to control for confounders of survival. The overall survival (OS) and recurrence-free survival were analysed using the Kaplan-Meier method. Differences in cumulative incidence of recurrence between groups were assessed using the methods of Gray. RESULTS: Of the 426 patients, lobectomy was performed in 381 patients and segmentectomy in 45 patients. Nodal metastasis was noted in 104 (24.4%) patients. Multivariable analysis revealed that lobectomy was an independent prognosticator of better OS (hazard ratio 0.596, 95% confidence interval 0.366-0.969; P = 0.037). Lobectomy arm showed favourable 5-year OS and recurrence-free survival (OS: 72.9% vs 59.7%, log-rank test P = 0.007; recurrence-free survival: 64.4% vs 48.7%, P = 0.034) (median follow-up: 53 months). Approximately 14% of the patients in the lobectomy group and 27% in the segmentectomy group died of lung cancer. Furthermore, 5-year cumulative incidence of loco-regional recurrence rate was significantly higher in the segmentectomy group (35.5% vs 15.8%, P < 0.001). CONCLUSIONS: In T1c radiological pure-solid non-small-cell lung cancer, segmentectomy was significantly associated with worse survival and insufficient loco-regional cancer control. Lobectomy remains the standard surgical treatment; meanwhile, segmentectomy should be applied with great caution.