Propensity score-matched comparison of robotic- and video-assisted thoracoscopic surgery, and open lobectomy for non-small cell lung cancer patients aged 75 years or older

摘要

Introduction Although robot-assisted thoracoscopic surgery (RATS) has been widely applied in treating non-small cell lung cancer (NSCLC), its advantages remain unclear for very old patients. The present study compared the perioperative outcomes and survival profiles among RATS, video-assisted thoracoscopic surgery (VATS), and open lobectomy (OL), aiming to access the superiority of RATS for NSCLC patients aged ≥75 years. Methods Pathological IA-IIIB NSCLC patients aged ≥75 years who underwent RATS, VATS, or OL between June 2015 and June 2021 in Shanghai Chest Hospital were included. Propensity score matching (PSM, 1:1:1 RATS versus VATS versus OL) was based on 10 key prognostic factors. The primary endpoints were perioperative outcomes, and the secondary endpoints were disease-free (DFS), overall (OS), and cancer-specific survival (CS). Results A total of 504 cases (126 RATS, 200 VATS, and 178 OL) were enrolled, and PSM led to 97 cases in each group. The results showed that RATS led to: 1) the best surgical-related outcomes including the shortest operation duration ( p < 0.001) and the least blood loss ( p < 0.001); 2) the fastest postoperative recoveries including the shortest ICU stay ( p = 0.004), chest tube drainage duration ( p < 0.001), and postoperative stay ( p < 0.001), and the most overall costs ( p < 0.001); 3) the lowest incidence of postoperative complications ( p = 0.002), especially pneumonia ( p < 0.001). There was no difference in the resection margins, reoperation rates, intraoperative blood transfusion, and volume of chest tube drainage among the three groups. Moreover, RATS assessed more N1 ( p = 0.009) and total ( p = 0.007) lymph nodes (LNs) than VATS, while the three surgical approaches dissected similar numbers of N1, N2, and total LN stations and led to a comparable incidence of postoperative nodal upstaging. Finally, the three groups possessed comparable DFS, OS, and CS rates. Further subgroup analysis found no difference in DFS or OS among the three groups, and multivariable analysis showed that the surgical approach was not independently correlated with survival profiles. Conclusion RATS possessed the superiority in achieving better perioperative outcomes over VATS and OL in very old NSCLC patients, though the three surgical approaches achieved comparable survival outcomes.