Contrast-enhanced chest CT (CECT) is more sensitive than non-contrast-enhanced chest CT (NCECT), but NCECT may have comparable efficacy in detecting new primary lung cancer among stage I NSCLC survivors after two years of surveillance. This study aimed to evaluate the efficacy of NCECT versus CECT for surveillance among stage I NSCLC patients surviving two years after curative resection without disease recurrence. We conducted a retrospective cohort study of patients with stage I NSCLC who underwent curative-intent lung resection between January 2009 and December 2017 using the Registry for Thoracic Cancer Surgery at the Samsung Medical Center, Seoul, Korea. Overall survival, recurrence-free survival, and cost effectiveness were compared between patients undergoing surveillance with CECT and NCECT. Among 3248 patients, 1002 (38.8%) patients underwent NCECT surveillance. During a median follow-up of 2.3 years (interquartile range, 1.5-3.9 years), a total of 208 deaths were observed. Although patients undergoing NCECT surveillance had 0.04 more deaths per 100 person-year compared with patients undergoing CECT surveillance (95% CI -0.36 to 0.44), this difference did not reach statistical significance (1.27 vs. 1.31 per 100 person-years
HR, 1.10
95% CI 0.81-1.50). Regarding cost effectiveness, CECT group had a gain of 0.024 quality-adjusted life-year but 85 higher total cumulative cost per patient compared to NCECT. There was no difference in recurrence and mortality between NCECT and CECT for surveillance among stage I NSCLC patients who survived two years after surgery without disease recurrence. Further randomized clinical trials are required to confirm the findings.