ObjectiveTo explore the clinical efficacy of two procedures in thoracoscopic anterior mediastinal tumor resection. MethodsA retrospective study was conducted on patients who underwent thoracoscopic anterior mediastinal tumor resection at the Department of Thoracic Surgery, the 910th Hospital of Joint Logistics Support Force from October 2016 to January 2024. Patients were divided into two groups according to the surgical approach: a modified approach group (bilateral intercostal ports+two subcostal ports) and a classic subxiphoid approach group (one subxiphoid port+two subcostal ports). Perioperative data and postoperative improvement of myasthenia gravis (MG) subgroup were compared between the two groups. ResultsA total of 55 patients were included, including 27 males and 28 females with a mean age of (49.4±15.1) years. There were 23 patients in the modified approach group and 32 patients in the classic subxiphoid approach group. The modified approach group had shorter operation time [(129.0±20.5) min vs. (148.9±16.7) min, P<0.001], less intraoperative blood loss [(63.0±16.6) mL vs. (75.0±10.8) mL, P<0.001], shorter postoperative drainage tube removal time [(3.1±0.4) d vs. (3.9±0.6) d, P<0.001] and shorter postoperative hospital stay [(4.2±0.4) d vs. (5.0±0.6) d, P<0.001), and lower proportion of intraoperative cardiac dysfunction [4 (17.4%) vs. 14 (43.8%), P=0.040]. There was no statistical difference in maximum diameter of tumor resected [(4.5±1.7) cm vs. (4.0±0.9) cm, P=0.193] and postoperative drainage volume [(396.4±121.5) mL vs. (399.9±161.3) mL, P=0.932]. There was 1 patient of perioperative collateral injury in the modified approach group (pericardial injury), and 6 patients in the classic subxiphoid approach group (1 patient of diaphragm injury, 1 patient of liver contusion, 4 patients of pericardial injury). There was no statistical difference in pain scores at 24 h, 48 h and 72 h after surgery (P>0.05). The postoperative improvement of MG symptoms in the modified approach group was better than that in the classic subxiphoid approach group at 1 year after surgery (complete stable remission rate: 77.8% vs. 50.0%; effective rate: 100.0% vs. 91.6%). No conversion to open chest surgery occurred in either group, and there were no postoperative rehospitalizations or deaths related to surgery within 30 days after surgery in both groups. ConclusionThe modified approach is safe and controllable with more open surgical field and more reliable complete resection range than the classic subxiphoid approach group.